This paper repots on the performance of an InP chiplet heterointegration on SiGe-BiCMOS carriers for mm-wave applications. A flip-chip type of assembly with indium-based microbumps allows for seamless integration with minimal losses up to more than 300 GHz. A single-stage W-band heterointegrated power amplifier shows more than 5 dB small signal gain. A saturated output power of +13 dBm is achieved with 42 mW DC power consumption at 80 GHz, which corresponds to 17 % Power Added Efficiency (PAE). The results were used to assess performance metrics and demonstrate the effectiveness of the heterointegration approach.

In this paper, a novel design method for PIN diodes waveguide phase shifter (WPS) is proposed based on the loaded line theory. As examples, 22.5, 45, 90-degree state phase shifters of compact size are designed. Simulation results indicate that, compared to traditional electronic WPSs, the present WPS can provide digital phase shifting with low insertion loss and high phase shifting stability.

This paper presents the design of a 16-way power combiner featuring high efficiency and high power capacity. High combining efficiency is achieved through a radial power-combining topology, while the power capacity is enhanced by an optimized matching structure. Based on this design approach, a 16-way power combiner prototype operating in X band was fabricated and measured. Measurement results show that bandwidth of 40% with average combing efficiency more than 90% is realized. Furthermore, high-power tests indicate that the prototype can handle a peak power exceeding 5 kW.

This paper proposes a deceptive interference suppression method for FDA-MIMO radar based on range frequency compensation. Firstly, a compensation vector is constructed, which is based on the difference in the number of delayed pulses and the principle range to distinguish the true target from the false target in the spatial domain. Finally, an adaptive beamforming filter is used to suppress the interference for range dimension mismatch.

In this paper a multiprobe reflectometer having a tunable measurement uncertainty distribution is proposed. The presented topology is simple, being based on a transmission line with seven weakly-coupled probes. It incorporates seven coupled-line sections, a branch-line coupler with a reference one-port reflective device and a measured device under test. The circuitry has been designed and fabricated to operate in a narrowband at center frequency of 5.8 GHz. The proposed reflectometer has been measured and its performance has been verified through uncertainty distribution analysis.

The electric permittivity of dielectric resonators of any axisymmetric shape can be measured precisely applying radial mode matching solution. The measurements of the dielectric resonators with conical shapes of the inner hole are presented. The measurements have been done twice. First ring type dielectric resonators have been measured. Then the inner holes have been drilled to realize conical shape and the measurements have been repeated for changed resonators. The results show that mode matching method is precise and produce results with the same level of accuracy in both cases.

The use of OFDM modulated signals significantly minimizes the time required for the Q-factor extraction compared to conventional frequency-swept network analyzers. This, in turn, increases the frequency of measurement results' updates, in the order of hundreds or thousands per second, which is of high importance in some applications of microwave resonant sensors. This article discusses some of the capabilities and limitations of this technique.

This paper presents the design, fabrication, and experimental verification of a Fabry-Pérot resonance-based spaceplate operating in the Ku-band. Two prototypes with different dimensions and metallization thicknesses were manufactured using wet etching technology on an FR4-DE104 substrate. A significant part of the study is dedicated to the analysis of manufacturing tolerances and their impact on the device's performance, specifically the compression factor and numerical aperture. Experimental results obtained through free-space measurement using a vector network analyzer (VNA) revealed frequency shifts compared to initial nominal simulations. Through microscopic analysis and mechanical characterization, these discrepancies were identified as a result of substrate volumetric expansion due to chemical absorption and thermal expansion during etching and geometric deviations in the copper pattern. By incorporating these real-world parameters into a refined numerical model, an acceptable agreement between simulation and measurement was achieved. The findings provide critical guidelines for the precise fabrication of metasurface-based spaceplates using low-cost PCB processes.

This paper presents the results of a 3D simulation of the human head model, encompassing the skull and brain struc- tures, to assess electromagnetic field within the FR3 band. The model was created using polygonal surface modeling, employing Blender modifiers to achieve realistic morphology. Subsequently, mesh retopology was performed to reduce the polygon count while preserving essential geometric features, which is crucial for the stability of simulation computations. Furthermore, the spatial relationships between individual structures were verified, with particular attention given to fitting the skull model within the volume of the head model. The obtained results were compared with current literature to validate the model's performance.

The abstract will be disclosed

This paper presents a practical, resonator-based microwave workflow for characterization and validation of thin films and functional coatings, semiconductors, and highly conductive carbon-based electrodes. The approach is organized as a set of complementary measurement instruments selected according to targeted electromagnetic response of the material. Measurement methods are combined with simulation-referenced calibration to enable repeatable extraction of electromagnetic parameters of the material. For conductive layers and electrode structures, resonators extend the workflow to materials that are not reliably accessible with dielectric-only techniques, enabling conductivity evaluation. The methods are applied to ion-implanted materials and carbon-based thin-film technologies and battery-oriented coatings, demonstrating a consistent route for process verification. We formalize the procedure as a standardized operating workflow aligned with MODA/CHADA principles, improving traceability and reuse of resonator-based characterization data.

Low-loss dielectric laminates commonly used for printed circuit boards (PCBs) are known to be anisotropic. The in-plane and out-of-plane components of dielectric permittivity have typically been determined using one of several available methods. One method for measuring the out-of-plane permittivity component is based on a TM0n0 cavity. In this article, we address the main weakness of this approach: the relatively high measurement uncertainty of the dielectric loss (Df) caused by limited repeatability of the contact resistance between the resonator body and its lid. We propose inserting the sample into the resonator not through a removable lid, but by splitting the cavity along its vertical axis. This approach is superior because it does not interrupt the surface-current paths, thereby stabilizing the total Q-factor of the cavity even when it is opened and closed multiple times. As a result, the total measurement uncertainty of Df is improved at least by one order of magnitude, because the conductive losses of the cavity can be established more precisely.

Accurate electromagnetic characterization of low-loss dielectrics in the sub-terahertz (sub-THz) band is increasingly critical for next-generation high-frequency electronics, yet existing standardized methods are essentially limited to microwave frequencies up to about 10 GHz (i.e., IPC-TM-650 and IEC 61189-2). Above 300 GHz, industrial practice relies mainly on transmission terahertz time-domain spectroscopy (THz-TDS) and related approaches [3], which often exhibit poor sensitivity to thin, low-loss films, leading to non-physical permittivity spectra and large uncertainties, especially in dielectric loss [4]. In contrast, resonant techniques provide a much stronger interaction between the field and the material under test. Fabry--Pérot open resonators (FPORs) have been successfully used for low-loss dielectrics up to the mm-wave range, but systematic material characterization above approximately 330 GHz remains largely unexplored.

The abstract will be disclosed

A new approach for analyzing waveguide junctions containing conductive cylindrical objects is proposed. The algorithm is based on mode matching technique using local projection functions, which improves the numerical conditioning of the problem. Moreover, the approach enhances computational efficiency by reducing the boundary where the numerical integration is required. Both convergence and accuracy of the method were tested by using several examples, including multi-section microwave filters, and validated through comparison with results obtained using the finite element method implemented in a commercial full-wave simulator.

Due to the inappropriate estimation and inadequate awareness of scattering from complex substructures within ships, a reasonable, reliable, and complete interpretation tool to characterize ship scattering for polarimetric synthetic aperture radar (PolSAR) is still lacking. In this paper, a fine polarimetric decomposition with explicit physical meaning is proposed to reveal and characterize the local-structure-related scattering behaviors on ships. To this end, a nine-component decomposition scheme is first established through incorporating the rotated dihedral and planar resonator scattering models, which makes full use of polarimetric information and comprehensively considers the complex structure scattering of ships. In order to reasonably estimation the scattering components, three practical scattering dominance principles as well as an explicit objective function are raised, and a particle swarm optimization (PSO)-based model inversion strategy is subsequently presented. This not only overcomes the underdetermined problem, but also improves the scattering mechanism ambiguity by circumventing the constrained estimation order. Finally, a ship indicator by linearly combining the output scattering contribution is further derived, which constitutes a complete ship scattering interpretation approach along with the proposed decomposition. Experiments carried out with real PolSAR datasets demonstrate that the proposed method adequately and objectively describes the scatterers on ships, which provides an effective way to ship scattering characterization. Moreover, it also verifies the feasibility of fine polarimetric decomposition in a further application with the quantitative analysis of scattering components. Manmade targets are usually composed of dihedrals, which are the most prominent structures. For ships, the dihedral structures can be found between the hull and sea surface, as well as some upper structure. In traditional decomposition methods, dihedrals parallel to the radar flight path which exhibit strong co-pol power are referred to as double-bounce scattering. However, when the orientation angle shifts, the cross-pol power is unexpectedly stronger than co-pol power, which is mistakenly regarded as volume scattering in traditional decomposition. Therefore, a separate scattering model that can describe the cross-pol scattering from rotated dihedrals is necessary. In our previous work, a mathematical programming-based RD scattering model was proposed, which can not only partake the total cross-pol powers so as to overcome the scattering ambiguity, but also can reasonably assign the internal co-pol and cross-pol components in the scattering of rotated dihedrals. On the other hand, in classical radar polarization theory, it is widely accepted that three-dimensional helix structures in manmade targets can generate cross-pol energy that are insensitive to target orientation. Accordingly, the roll-invariant cross-pol power is only assigned to helix scattering. Nevertheless, the authors found that both three-dimensional helix structures and certain two-dimensional structures on ships can induce roll-invariant cross-pol response, and the two-dimensional roll-invariant cross-pol response is generally induced by a planar resonator (PR) scattering structure. The cognitive scattering mechanisms and the corresponding scattering contribution are the main outcomes of polarimetric decomposition, which can be applied to describe the global and local structure scattering of ships and are regarded as the primitive polarimetric features. Based on the proposed nine-component fine polarimetric decomposition and considering the scattering mechanism difference between the ship targets and other clutter interferences, a ship indicator with a linear combination of nine decomposed scattering contribution is proposed. Three datasets of real, high-resolution PolSAR data are involved in the evaluation of the proposed method, in which the decomposition effect, performance comparison, and intermediate process are discussed. Qualitative and quantitative experimental results demonstrate that the global and local scattering mechanisms of ships are adequately and objectively revealed by the proposed decomposition. While the scattering significance of ships are effectively announced by the proposed indicator, which is conducive to the following detection. These two products constitute a favorable ship scattering characterization tool, which verifies the feasibility of polarimetric decomposition in actual applications.

The Global Navigation Satellite System (GNSS) can be used not only for positioning, but also for environmental sensing. In this paper, we present a phased antenna array system driven by software-defined radio (SDR) GNSS receivers, capable of distinguishing between line-of-sight (LOS) and non-line-of-sight (NLOS) signals in urban environments. By measuring the direction-of-arrival (DOA) angles of these signals and combining this information with available building footprint data, the system enables building height estimation at a lower cost than traditional aerial or drone-based methods.

Radio frequency (RF) fingerprinting is a technique that exploits unique hardware-induced imperfections present in transmitted signals. Recent research has leveraged Convolutional Neural Networks (CNNs) to replace traditional statistical meth- ods. In this paper, we introduce a measurement methodology for automated RF fingerprinting, including signal extraction and packet collection based on information from transmitted signals. The approach relies on a highly lightweight CNN model designed to support optimization. Experimental results demonstrate device identification accuracies of 92.87% for samples acquired in an anechoic chamber and 89.42% in a real-world propagation environment.

In this contribution, we analyze the computational complexity of the finite-difference time-domain (FDTD) methods employing time-fractional (TF) constitutive relations. The relations employ fractional-order (FO) derivatives based on the Grünwald-Letnikov definition, capturing hereditary properties and memory effects of the media. Thus, the considered relations facilitate incorporating hybrid diffusion-wave processes -- modeled by the diffusion-wave equation -- into the dielectric response. Two FDTD methods based on Yee's grid implementing the TF constitutive relations are analyzed and compared in terms of their computational complexity. These include the standard leapfrog time-stepping FDTD method and the Crank-Nicolson FDTD method. We demonstrate that due to the infinite memory of the past in the mathematical model of TF media, the computational overhead of the considered FDTD methods increases with the square of the number of iterations, which limits the applicability of these methods.

High performance microwave filters are critical, performance defining components of modern space and terrestrial communications systems. The history of advances in this technology field is well documented and spans more than seventy years. In my talk, this will be discussed based on my personal involvement in this technology for a period of almost sixty years. Starting with Ragan's book, articles by Seymour Cohn and Ralph Levy and Matthaei's book we will move to what I consider the "Golden Era" of microwave filter technology driven by needs of communications satellite industry. Significant advances in this technology such as novel implementations, filter network synthesis, EM simulations, fabrication techniques and high power filters, to name a few, will be discussed. The impact of development of temperature stable, high dielectric constant materials, discovery of high temperature superconductive ceramics and MEMS technology will also be discussed. By year 2000, communications satellite filter and multiplexer technology matured and filter development shifted toward wireless terrestrial communications systems, which are constantly moving to higher frequencies for wider bandwidth availability. This presents additional challenges for filter industry such as low cost fabrication techniques, new materials etc. In conclusion, the future of microwave and now, millimeter wave filter technologies will be discussed.

This paper presents two realizations of extremely narrowband high-power dielectric resonator filters. The filter center frequencies are 1300 MHz and 2997.92 MHz. High quality filters are realized as three-pole filters with relative bandwidths as low as 0.035%. The center frequency insertion loss of filters is better than 2.1 dB. The filters are realized in copper housings using ring dielectric resonators of permittivity 37 at lower frequency and 30.3 at higher frequency. The resonator's temperature coefficient is +3 ppm/°C. The filters for 1300 MHz can handle up to 100 W of input power and the filters for 2997.92 MHz can handle up to 50 W of input power. The filters demonstrate good thermal stability better than 2 ppm/°C. They have a very wide spurious free band.

In our recent work [1], we presented the design process and experimental validation of a class of RF-input-quasi-reflectionless coupled-resonator bandpass filtering dispersive-delay structures (DDSs) based on complementary-diplexer architectures. The proposed circuits combine bandpass filter transmission response, arbitrary group-delay shape, and reflectionless behavior at the RF input, thus mitigating undesired out-of-band signal reflections that are inherent to conventional reflective-type filtering DDSs. The architecture consists of a main reflective-type bandpass filtering DDS channel and a resistively terminated auxiliary bandstop filtering channel with a quasi-complementary frequency response. As a result, the non-transmitted RF input power is largely dissipated in the auxiliary branch resistive load rather than being reflected back toward preceding RF stages, improving RF robustness and system-level performance. The main channel is implemented as an in-series cascade of coupled-resonator triplets with one frequency-dependent coupling (FDC) per triplet, which enables arbitrary shaping of group-delay profiles through the controlled placement of complex transmission zeros. The auxiliary channel is synthesized to exhibit a complementary bandstop response, ensuring broadband input-power absorption. A systematic coupling-matrix-based synthesis framework is developed, consisting of the initial characteristic polynomials/coupling matrix synthesis of the main and auxiliary channels, followed by an optimization of the overall quasi-reflectionless DDS network to obtain the target group-delay, minimum in-band transmission requirements, and prescribed input-reflection limits. The proposed design methodology is validated through fifth- and seventh-order synthesis examples with differently shaped group-delay responses, including linear- and stepped-type delay profiles, and demonstrating input-power-matching over wide frequency ranges. Furthermore, the extension of the approach to multi-band filtering DDSs via two types of single-to-multiband frequency transformation is considered, with particular emphasis on the resulting group-delay distortion mechanisms and their implications for practical synthesis. As experimental verification, a fifth-order RF-input-quasi-reflectionless bandpass filtering DDS is implemented in microstrip technology. The measured results show good agreement with target response, confirming effective RF input quasi-reflectionless level over prescribed bandwidth while maintaining the target group-delay characteristic and amplitude transmission.

Superconducting qubits rely critically on microwave engineering, and superconducting cat qubits in particular require carefully designed microwave filtering. In these architectures, the memory mode (a harmonic oscillator) is populated via buffer mode in a process of mixing. Consequently, the memory can be strongly isolated from environmental noise by employing a filter that provides a stopband at the memory frequency. This paper outlines the key performance requirements for such filters and presents their practical implementations and experimental characterization in superconducting quantum processors, with an emphasis on the microwave engineering aspects of the technology.

Dielectric materials are essential to the performance of microwave, millimetre-wave, and terahertz systems, where accurate knowledge of electromagnetic material properties directly impacts device modelling, circuit design, and system reliability. These properties are commonly described by the real part of the complex relative permittivity, Re(εᵣ), and the dielectric loss tangent, tan δ. Precise determination of these parameters is critical for accurate electromagnetic simulations and can significantly reduce design uncertainty, prototyping iterations, and production costs in radio-frequency (RF) and high-frequency integrated circuits. The increasing deployment of high-frequency technologies has intensified the demand for reliable material characterisation at millimetre-wave and terahertz frequencies. Applications such as 5G and emerging 6G wireless communication systems, automotive radar sensing, and satellite payloads require dielectric parameters to be known with high accuracy across broad frequency ranges. However, material characterisation in these frequency regimes remains challenging due to increased sensitivity to measurement uncertainty, sample preparation, and calibration accuracy. A wide variety of techniques have therefore been developed, including open resonators, free-space and quasi-optical methods, guided free-space approaches using commercially available Material Characterisation Kits (MCKs), as well as frequency- and time-domain spectroscopic techniques. Accurate material measurements are also a fundamental requirement for on-wafer RF and millimetre-wave metrology. Calibration substrates used based on-wafer measurements must be characterised with high confidence to ensure traceable calibration and reliable uncertainty estimation. This work is carried out within the framework of the ongoing EU project 23IND10 OnMicro, which aims to develop, validate, and compare accurate measurement techniques for determining permittivity and loss tangent of a broad range of materials, including bulk substrates, semiconductors, thin films, and emerging two-dimensional materials. Within this study, three complementary VNA-based measurement approaches are implemented and evaluated over overlapping frequency ranges. These methods include (i) a guided free-space technique employing commercially available MCKs, (ii) a resonator-based measurement method providing high sensitivity to dielectric losses, and (iii) an on-wafer-based approach using substrate materials with deposited metal structures representative of calibration and device environments. In this presentation results will be presented to provide practical insight into the accuracy, limitations, and applicability of each measurement approach. This work supports the development of improved material characterisation strategies for high-frequency applications and contributes to establishing reliable measurement methodologies for next-generation RF, millimetre-wave, and terahertz systems.

The design of the broadband magnetization dynamics measurement system for the determination of the magnetic properties of thin-films is presented. The setup is based on a dedicated coplanar-waveguide placed inside the electromagnet, with additional modulation coils. A microwave signal is supplied from a broadband signal generator, while detection is based on a zero-bias Schottky diode connected to the lock-in amplifier. The system is capable of investigating in-plane and perpendicularly magnetized thin films down to a single nanometer in thickness. We demonstrate its feasibility in determining the effective magnetic anisotropy of the thin Co-Ni super-lattice at the wafer level, together with magnetization damping.

This study presents a quantitative analysis of the complex permittivity of potato tubers as a function of structural state. Dielectric measurements were performed using an open-ended coaxial probe connected to a vector network analyzer (VNA) over the frequency range 500 MHz to 20 GHz. The real (ε^') and imaginary (ε^'') components of the complex permittivity were extracted for three structural conditions of the same tubers: intact, sliced, and mashed. The results show systematic variations in both ε^'and ε^'' depending on tissue integrity. Mechanical disruption of the cellular structure leads to measurable changes in dielectric dispersion and loss mechanisms. In particular, the puree state exhibits increased dielectric losses and modified frequency-dependent behavior compared to intact samples, reflecting alterations in water distribution, membrane integrity, and solute mobility. Quantitative comparisons highlight distinct permittivity trends between the three structural states over the entire microwave spectrum.

This part of the bootcamp introduces participants to the principles and practice of broadband dielectric measurements using the open-ended coaxial probe technique, centered around the industry-standard N1501A Dielectric Probe Kit. This hands-on session is designed to provide a comprehensive understanding of the entire measurement chain, from initial setup to final data analysis. Participants will explore the unique advantages of this method, such as its non-destructive nature, ease of sample preparation, and the ability to characterize materials across a continuous, wide frequency spectrum-capabilities that are vital for the development of 5G/6G technologies and broadband electronic components.

The workshop will candidly address the inherent trade-offs of the coaxial probe method, comparing its accuracy levels with resonant techniques and identifying the specific material types and frequency ranges where it excels. A significant portion of the session will be dedicated to the calibration process, specifically the Air-Short-Water routine, and the critical importance of handling the equipment with precision. Participants will learn to identify and mitigate common measurement errors, such as cable instabilities, air gaps between the probe and sample, and improper "short" termination. By interpreting real-world results, attendees will gain the skills to distinguish between actual dielectric relaxation phenomena and measurement artifacts, ensuring a robust and repeatable methodology for characterizing liquids, semi-solids, and flat solid materials.

A design methodology aimed at enhancing the beam scanning performance of a cascade of two rotatable transmitarrays (TAs) is presented. The resulting Risley prism antenna (RPA) enables simultaneous multibeam (SMB) radiation with mechanical steering, offering a substantial improvement over traditional single beam steering solutions. The proposed tailored approach Risley prism (TARP) synthesis technique mitigates undesired lateral lobes by applying a non-uniform compensation of the incident spherical wave. A fully-dielectric RPA prototype operating at 35 GHz is manufactured using an additive manufacturing process. The experimental results validate the theoretical framework and confirm the proposed synthesis strategy.

preliminary theoretically study concerning the possibility to exploit three-dimensional (3D) printing technology for the design of multilayer dielectric transmitarray (TA) antennas enabling efficient wavefront manipulation is addressed. Each TA unit cell comprises a stack of dielectric layers whose permittivity values are rigorously synthesized to implement the prescribed transformation of the incident electromagnetic wavefront. More in detail, independent control of both amplitude and phase of the transmitted electromagnetic field is achieved by tailoring the dielectric permittivity profile of each unit cell. To validate the methodology, a 10-layer dielectric TA that generates three concurrent main beams has been designed and compared with respect to a single-layer dielectric TA.

In this work, automatic generation of antennas with nature-inspired shapes is considered. The method exploits a generic point-based simulation model of the radiator and a proxy equation that enables generation of complex shapes using a limited number of design parameters. The method is demonstrated through the development of a proof-of-concept planar antenna. The resulting geometry features a quad-band operation with dual-polarization behavior and a maximum realized gain of almost 13 dBi.

RF energy harvesting using rectenna systems has been emerging as a promising solution for powering low-power IoT and wireless sensor nodes. This work presents the design of a WLAN-band (2.4 GHz) rectenna configuration comprising a sequentially rotated circularly polarized 4-port Square Patch Antenna Configuration (SPAC) integrated with a 4×1 Wilkinson power combiner and a high-efficiency Schottky-diode-based rectifier. The proposed left hand circularly polarized (LHCP) antenna design achieves around 10 dBi peak gain with excellent cross -polarized isolation greater than 22 dB and an optimum impedance bandwidth of 50 MHz around the frequency of operation 2.4 GHz. The best optimized rectifier design demonstrates a maximum RF-DC conversion efficiency of 70% and DC output voltage of nearly 1.4 V. With suitable power management circuitry, the proposed rectenna configuration is an excellent candidate for low-power IoT nodes in smart home platforms promoting sustainable indoor green energy harvesting from Wi-Fi routers.

This work investigates the impact of potential biocompatible encapsulation strategies on near-field transdermal wireless transmission for biomedical implant applications at microwave frequencies. A pair of near-field coupled antennas is configured to represent the wireless transdermal link between implanted and external units. A low-profile microstrip-fed slot antenna topology is investigated for near-surface subcutaneous implantation. The transmitter-receiver antenna pair is designed with a nominal operating frequency of 6 GHz and evaluated under near-field transdermal coupling conditions. A dataset of measured frequency-dependent dielectric properties of porcine skin is used to model the tissue losses. To improve transdermal coupling, encapsulation strategies are evaluated for biocompatible materials polyimide, PTFE, SiO2 and Al2O3. A -10 dB transmission bandwidth of 3880 MHz and a minimum insertion loss of 3.62 dB have been achieved with an optimised PTFE encapsulation over the 2.19-6.07 GHz band. The encapsulation of the antennas is evaluated with a focus on transmission performance rather than their conventional role as hermetic barriers. The simulation results demonstrate that encapsulation layer design and material selection can significantly enhance the wideband transdermal coupling in near-field implanted antenna systems.

Sixth generation (6G) physical layer (PHY) is evolving beyond the legacy orthogonal frequency division multiplexing (OFDM)-based waveforms. In this paper, we compare the bit error rate (BER) performance of three beyond-OFDM waveforms, namely, orthogonal time-frequency-space (OTFS) modulation, affine frequency division multiplexing (AFDM), and orthogonal chirp division multiplexing (OCDM), which are particularly suitable for the highly mobile non-terrestrial network (NTN) vertical of 6G. In order to characterize the effect of mobility and Doppler shift in low Earth orbit (LEO) satellites, we performed BER comparisons over four different NTN tapped-delay-line (TDL) models, TDL-A, TDL-B, TDL-C, and TDL-D, as specified in the 3rd generation partnership project (3GPP) technical report TR 38.811. After channel equalization, a minimum mean squared error with successive detection (MMSE-SD) algorithm was used to enhance the BER performance. It was found that AFDM and OTFS consistently outperformed OCDM across all TDL models, while AFDM performed better than OTFS in TDL-B and TDL-C, in the high signal-to-noise ratio (SNR) regime. The complete simulation framework is made available as an open-source code for quick validation and further development.

This talk focuses on the recent advances related to the design of high-performance microwave sensors implemented in planar technology, with special emphasis on sensitivity and selectivity optimization. Strategies for sensitivity enhancement in phase- variation and magnitude-variation sensors will be analysed. The talk also deals with the issue of sensor selectivity, important to determine analyte composition in complex samples, and approached by means of machine learning. Examples of applications will be also given, including sensors for structural health monitoring and biosensors.

There are a number of medical applications that require knowledge of internal body temperature, such as brain temperature monitoring during aortic repair open-heart surgery, monitoring athletes and soldiers under heavy physical training or during hypothermia, circadian cycle related sleeping disorders, determining type of strokes, and measuring temperature of tissues during cancer hyperthermia and ablation. Currently, there is no commercially available non-invasive passive and potentially wearable internal-body thermometer. Microwave radiometry can potentially solve this problem - this work focuses on the fundamentals of operation, design, implementation and testing of radiometers for internal temperature measurements of the human body. In this approach, the total blackbody power from a tissue stack is received by a near-field antenna placed on the skin connected to a low-noise receiver. Since human tissues are lossy at microwave frequencies, a relatively low frequency (between 1 and 3 GHz) is used for sensing temperature several centimeters under the skin. The near-field antenna receives total noise power from all tissue layers and temperature retrieval for sub-surface tissue layers is performed using weighting functions, obtained by full-wave simulations with known tissue complex electrical parameters, combined with retrieval algorithms. Measurements are presented using several types of calibrated radiometers at 1.4GHz and 3GHz for various phantom tissues. In-vivo tracking in the human cheek is also shown to follow ground-truth thermocouple measurements [3]. With an integration time on the order of a second, temperature can be tracked within a fraction of a degree. Technical challenges include interference mitigation, variability of human tissues between people and parts of the body, limited spatial resolution and miniaturization. Some methods that address these challenge are: RFI cancellation by statistical processing; using reflectometry combined with signal processing for tissue calibration; multiple radiometers that use statistical methods to improve spatial resolution; and integrated radiometers in GaAs that reduce sensor size.

In this part of the bootcamp, participants will get to know the principles and practice of compact free-space dielectric measurements at sub-terahertz frequencies. The presented device is a custom material characterization kit based on a single-chip D-band FMCW transceiver integrated with a quasi-optical beam-guide fixture. This hands-on session is designed to give a comprehensive understanding of the whole measurement chain: from signal generation on the transceiver IC operating in the 114 -134 GHz band, through wave shaping and propagation along the beam guide, up to time-domain signal processing and extraction of dielectric constant and loss tangent from the round-trip transmission through the sample.

A multilayer chipless Radio Frequency IDentification (RFID) tag operating in the millimeter-wave frequency range is presented. The proposed solution exploits a one-dimensional photonic crystal with a central anisotropic defect layer. Encoding is achieved through polarization-dependent resonant frequencies. Simulations results validate the proposed approach.

This paper presents a compact, polarization-preserving Ku-band bandpass filter designed for 3-D-printing in metal. The filter is intended for integration with active antennas in satellite communication systems, which requires dual-polarization operation. The filter is composed of five cascaded sections made of conical ridge resonators with four-fold rotational symmetry (C4) and provides a passband for Ku-band receivers (13-14.8 GHz). A prototype was fabricated in one piece in aluminum alloy using selective laser melting and measured, exhibiting good agreement with simulation and very low insertion loss of 0.15 dB.

This paper presents results of tests and simulations of a novel S-band diplexer for space applications. The device is developed and manufactured in one of WiRan projects for European Space Agency. Design driving factors were immunity to corona discharge, Far Out of Band Rejection (FOOBR, defined as ratio of frequency of 1st image of filtering bandwidth to higher edge frequency of the filter, f_image/f_edge)), low loss, high isolation and compactness. The device was designed, manufactured and tested in WiRan. Results are promising, exceeding current state of the art in this field of applications.

Accurate component values are critical for RF system design. Typical component models rarely include substrate coupling and inter-component coupling effects. Therefore, air coils within phase shifters were evaluated. They were used in phase gradient supported electrical balance duplexers, which enable multi-band operation beyond the 3 dB insertion-loss limit of conventional architectures. This work analyzes the inductive behavior of the 5.45 nH commercial off-the-shelf air coils using manufacturer data, non-vendor-purchased models, measurements, and electromagnetic simulation. Deviations in inductance values across sources are observed and evaluated. Measurements of two coils connected by short transmission lines further reveal steep frequency-dependent inductance increases and resonance frequencies substantially below those of individual coils. EM simulations of coils were derived for comparison with measurements. These effects led to frequency-dependent changes in inductance, directly impacting the tunability and functional reliability of the pi-network based phase shifter.

Multistatic passive radar systems reduce ambiguities in target localisation due to the spatial diversity of transmit and/or receive elements. When dedicated synchronisation channels are unavailable, discrepancies between clocks manifest as errors in both time and frequency, and the receivers' sampling start times are inherently misaligned. In this paper, we present a simple software-based method for synchronising spatially diverse, unsynchronised receivers. Frequency errors are reduced by estimating a combined fractional and integer component of the carrier frequency offset. Time and frequency related errors are reduced through resampling to minimise differences between measured and expected OFDM symbol durations. Spatial diversity is restored by applying time-of-flight corrections derived from known transmitter and receiver positions. Results from a field trial using commercially available, low-cost hardware show that the proposed processing method preserves target information, enabling data fusion without dedicated synchronisation hardware.

This paper presents a study on the visualization of multi-static passive radar signal data. In multi-receiver networks, coherent detection provides both the ability do detect weaker and distant targets, as well as the ability to localize targets in Cartesian coordinate space. Furthermore, moving objects are also detected as velocity vectors, providing both speed as well as direction. Proper visualization of the two-dimensional space requires a search within each spatial dimension (X, Y, …) as well as the velocity dimension (dX, dY, …) per each pixel. To generate these plots, an appropriate algorithm, as well as suitable data is required. The algorithm selected is the Generalized Canonical Correlation Analysis (GCCA), and the data requires coherently synchronized and recorded data. Coherent data provides quite significant challenges including the need for time and frequency stability, as well as requirement of synchronization in both of these quantities.

This contribution deals with the determination of the radar channel transfer matrix in short-range radar applications. In these, the radar antennas are often located in the near field of an electrically large target, in which the scattering properties, in contrast to far-field relationships, depend on distance as well as the antenna. Based on physical optics and the antenna radiation pattern, a bistatic complex-valued transfer function between the feed of a transmit antenna and the output port of a receive antenna is derived. For multiple antennas, it extends to a channel matrix, which allows to predict the radar response of multiple-input multiple-output (MIMO) systems as well as antenna arrays. Measurements of the channel matrix in the case of a disk target conducted with horn and open-ended waveguide antennas at 24 GHz prove the validity of the proposed approach. It accurately predicts near-field effects such as power level fading and nonlinear phase distortion for MIMO radar configurations.

With growing automation in the automotive sector, the demand for integrated communication and sensing is increasing. This work proposes an integrated sensing and communication (ISAC) system based on index-modulated frequencymodulated continuous-wave (FMCW) that requires no modifications to the radar front-end. The system's performance was validated through ray-tracing simulations and radar measurements. Reliable communication was achieved with symbol error rates (SER) below 0.003 at signal-to-noise ratios (SNRs) above -15 dB and data rates up to 18 kbit/s, while radar sensing remains mostly unaffected. The observed effects result in targets appearing in adjacent range bins with varying power and higher side lobes. Overall, the results demonstrate that frequencymodulated continuous-wave (FMCW) radar hardware can be extended via index modulation to support robust communication with an additional receiver, making it a promising technology for future automotive ISAC systems with minimal additional effort.

Synthetic Aperture Radar (SAR) is highly vulnerable to coherent interference, especially Artificial Modulation Targets (AMT) generated by Digital Radio Frequency Memory (DRFM), which severely degrades imaging quality. Existing studies on AMT suppression are limited, and conventional methods lack effectiveness. This paper proposes a novel three-stage deep learning framework to suppress AMTs at the raw signal level. First, a Multi-Input Multi-Output Fusion Wavelet Neural Network (MIMOFWNN) detects AMT components and identifies their modulation types. Second, a Complex Signal Codec Neural Network (CSCNN) reconstructs and cancels AMT signals via coherent subtraction. Third, a Loss Complex Signal Recovery Neural Network (LCSRNN) restores useful signals lost during cancellation. Experiments validate the framework's performance: AMT recognition accuracy reaches 99.0%, and the structural similarity index between suppressed results and clean benchmarks achieves 0.85508.

This paper addresses the significant challenge of Artificial Modulation Targets (AMTs) in Synthetic Aperture Radar (SAR) systems, which are coherent interference signals generated by Digital Radio Frequency Memory (DRFM)-based devices. These AMTs exploit the SAR's matched filtering process to create deceptive false targets in the final image, severely compromising image interpretation and remote sensing applications. While extensive research exists on suppressing Radio Frequency Interference (RFI), AMT suppression remains largely unexplored. This study proposes a novel, neural network-based reconstruction method for AMT signals directly from raw SAR echoes, enabling their subsequent cancellation. The core of the proposed method is a Complex Signal Codec Neural Network specifically designed for the unique requirements of SAR signal processing. The network takes the raw, complex-valued total echo matrix and AMT modulation type as inputs and outputs the reconstructed complex-valued AMT signal. Its architecture incorporates several key innovations to overcome limitations of conventional optical image processing networks. Firstly, it processes amplitude and phase channels separately yet interactively, ensuring high-fidelity reconstruction crucial for subsequent SAR imaging. Secondly, a custom Large-interlayer Signal Reconstruction Link (LSRL) module is introduced. This module allows the network to process echo matrices of arbitrary, non-standard dimensions (e.g., 1928×220) without resizing, which is critical as resizing disrupts phase information and renders the signal unusable for imaging. The LSRL facilitates dynamic, small-scale size mappings between network layers, preserving signal accuracy. Furthermore, a novel Composite Signal Reconstruction Precision (CSRP) loss function is designed to optimize the training process. Recognizing that AMT signals are sparse (most matrix elements are zero), the CSRP loss combines a global mean absolute error component with a micro-cumulative-error component. The latter amplifies the impact of small errors in the predominantly zero-valued regions, significantly enhancing the reconstruction precision of the sparse AMT components. To support this research, a comprehensive 65.3 GB C/X-band SAR AMT Suppression Dataset was constructed. It includes SAR echoes from diverse real-world scenarios (urban, ocean, mountain, plain) based on data from satellites and UAV platforms. The dataset incorporates nine distinct CMF modulation types for AMT generation, four Jam-to-Signal Ratios (JSR), and five different AMT positions, ensuring robust network generalization. The method's effectiveness was rigorously validated through Hardware-In-the-Loop (HIL) simulations and extensive comparative and ablation tests. Quantitative evaluation using Structural Similarity Index Measure (SSIM), Peak Signal-to-Noise Ratio (PSNR), and Root Mean Square Error (RMSE) demonstrated superior performance. The reconstructed AMTs achieved metrics of 0.81054 SSIM, -21.6941 dB PSNR, and 11.3241 RMSE against ground truth. After cancellation, the repaired scene images showed 0.85508 SSIM, -13.3882 dB PSNR, and 4.4334 RMSE compared to AMT-free references. Ablation studies confirmed the critical contributions of both the LSRL module and the CSRP loss function to the network's performance, significantly outperforming configurations without them and existing conventional RFI suppression methods like notch filtering and subspace decomposition. The promising results, validated under HIL conditions, pave the way for future work, which will focus on computational efficiency optimization and extending the approach to other SAR frequency bands.

In most sensing settings, accurate knowledge of sensing geometry is critical: receiver coordinates are typically well known, whereas transmitter positions are harder to establish. Robust transmitter calibration is therefore required to enable reliable behaviour detection and tracking across diverse indoor environments. We address 2-D transmitter localisation for indoor passive bistatic radar that reuses Wi-Fi access points as opportunistic illuminators. Rather than typically used received-signal-strength or time-difference-of-arrival method that apply grid search or maximum likelihood, we develop a geometric pipeline: each range-difference induces a hyperbola with foci at the receiver and an anchor; we compute closed-form pairwise intersections, aggregate all intersections, and select the physically consistent cluster via target-pair coverage (with optional residual screening). The cluster centroid provides the estimate, optionally refined by a lightweight weighted least-squares step. The method is search-free with complexity O(m 2 ). Simulations with Wi-Fi quantisation (Delta q=0.94m) recover a unique solution in the noiseless case and compact intersection clusters under noise, enabling accurate, low-complexity localisation.

ASTERIX enjoys widespread adoption by radar networks, air traffic control, and cooperative identification systems. The protocol is evolving over time, with the addition of new message classes, called categories. Recently, it has been expanded to include two new categories, CAT. 015 and 016, in an attempt to better accommodate passive radars and multistatic systems in general. In this paper, we evaluate the suitability of the ASTERIX protocol for the transfer of target information originating from passive radars. The evaluation is based on the identification of the most suitable information level to be shared, and the corresponding selection of the information that needs to be exchanged at that level. A one-to-one correspondence between the selected information and the protocol's data items is established and reported. Finally, there is a flip side to this paper. It can serve as a guide on how to use ASTERIX for passive radar data transmission.

Asymptotic methods rooted in Geometrical Optics (GO) have long been employed for the analysis of lens antennas. When the GO framework, particularly its ray-tracing (RT) formulation, is combined with the inherent approximations of Physical Optics (PO), the result is a computationally efficient methodology that has proven to be highly adaptable in a wide range of engineering problems. Nevertheless, the combined RT-PO technique is frequently considered too approximate to accurately predict the radiation characteristics of lens antennas and radomes, and is thus presumed to be suitable only for a restricted range of scenarios. Contrary to this common assumption, our experience with RT-PO has demonstrated that its practical applicability is far broader than that generally acknowledged. Some practical examples will be summarized during the presentation. The central issue is not whether such approaches are approximate, but rather identifying the problem classes for which their accuracy is sufficient. Based on our experience, these classes are significantly more numerous than is often assumed.

The paper presents research on a low-profile antenna based on a cylindrical resonator, in which a novel temperature-stable high-permittivity (Mg1.02-xNix)TiO3-Ca0.8Sm0.4/3TiO3¬ composite ceramic was applied. Low-profile, temperature-stable antennas are promising in 5G, IoT and RF energy-harvesting applications, where small dimensions, weight and parameter stability are critical features

A full-wave method is presented for planar waveguide junctions enclosed by a non-perfect conductor, with rectangular and cylindrical ports. It is based on a 2-D scalar Finite Element Method (FEM) formulation, which seamlessly integrates the Leontovich boundary conditions for conductors with finite conductivity with minimal overhead. The technique exploits the specific geometry of the planar junction to divide the problem into TMz and TEz modal excitations. In addition, for each TMz/TEz excitation, the problem field is further divided by the order of the variation of the field along the axis of the main junction. This formulation is now extended to cylindrical regions where the classical circular wave expansion is used to represent the field. This allows to reduce the computational load for some problems of interest, while preserving the complete modal characterization of the problem.

The article addresses a computational benchmark for numerical methods used to assess the impact of thin lossy dielectric layers on the monostatic RCS (Radar Cross Section) over a microwave frequency range. It proposes a material model (lossy dielectric) and assumptions that can serve as a useful reference point in analyzing the impact of applying non-magnetic absorptive materials. Two numerical methods were used to develop the benchmark: FDTD (Finite Difference Time Domain) and MoM (Method of Moments). Simulation results were compared with an semi-analytical solution as a reference point. The obtained values of RCS for perfect electric conducting (PEC) plate and PEC plate with a layer of absorbent material applied were also presented. There are also presented measures for simulation result consistency investigation and thresholds proposed for their assessment.

This work presents advanced numerical and semi-analytical methodologies for the rigorous modal analysis of periodic electromagnetic structures supporting both bound and radiative phenomena. A full-wave numerical framework is first introduced to investigate guided and leaky modes in one-dimensional periodic chains composed of two-dimensional cylindrical inclusions. Building on this formulation, an efficient semi-analytical approach is developed for the study of modes with real and complex wavenumbers in three-dimensional woodpile electromagnetic bandgap structures formed by stacked periodic layers of circular dielectric rods.

This paper presents a novel method for the electromagnetic characterization of dielectric liquids using a Fabry-Perot open resonator (FPOR). In contrast to conventional FPOR approaches based on transverse electromagnetic (TEM) modes, the proposed method employs Laguerre-Gaussian modes, enabling the use of a standard test tube placed along the resonator axis without degrading the quality factor. The method is aimed in particular at characterizing dielectric coolants designed for immersion cooling applications. The measurement setup is described, and initial dielectric-constant results are reported and validated against a cavity-resonator reference method, showing good agreement.

Accurate knowledge of the intrinsic dielectric properties of powder grains is essential for the design of composite substrates used, for instance, in low temperature cofired ceramics. However, direct measurement of powders is difficult because the material is strongly porous and the electromagnetic response is dominated by the powder air mixture. This paper presents a two-step method for determining the complex relative permittivity of nonmagnetic dielectric powder grains from resonant measurements. In the first step, the effective complex permittivity of a powder air mixture is determined at two transverse electric modes present in a cylindrical dielectric resonator. The sample is placed in a fused silica tube inserted into the resonator, and the effective parameters are extracted from the measured changes of the resonant frequency and quality factor using a mode-matching model. The volume fraction of the solid phase is estimated from the mass of powder and the tube geometry. In the second step, a differential mixing equation is applied to recover the intrinsic dielectric properties of the grains from the measured mixture parameters. The method is experimentally validated on a polystyrene powder with an air volume fraction of about 45 percent. The extracted intrinsic dielectric constant and loss tangent show very good agreement with independent reference data. The proposed approach provides a practical and sensitive tool for microwave characterization of dielectric powders for composite material engineering.

Fabry-Perot open resonator allows for a broadband material testing in microwaves and millimeter waves. It can be used to characterize various types of materials including dielectric films, low-loss liquids, and electric conductors. However, the standard measurement procedure provides measurement results only for one position of the sample, typically in its center, which may be insufficient for samples exhibiting spatial non-uniformities. Therefore, in this paper a new automated two-dimensional positioning system dedicated to spatial dielectric testing in a Fabry-Perot open resonator is presented. The method is experimentally validated using fused silica and sapphire samples at 112-115 GHz.

Transmission-mode measurements of spin resonances in an yttrium iron garnet (YIG) film placed on the wall of a rectangular cavity were performed. Two modes excited in the YIG film were strongly coupled to the rectangular cavity modes. Between them, up to 7 higher-order modes were observed. The Q-factors of some of the modes were measured, allowing for the dtermination of the ferromagnetic linewidth of the film.

This paper presents a comparative study of two complementary techniques for dielectric liquid characterization: broadband transmission/reflection (T/R) measurements and a dielectric resonator cavity (DRC) method. Measurements were performed on 3M L-19906 Developmental Fluorochemical Fluid under identical thermal conditions using both techniques. The T/R method with a meniscus-removal algorithm utilizes a 7 mm coaxial line in a semi-open test cell configuration and the complex permittivity is measured in 0.2-18 GHz bandwidth. Dielectric resonator measurements were conducted using a dielectric resonator cavity (DRC) at 2.4 GHz and 5.1 GHz operating in TE01delta and TE02delta modes, respectively. The results of the dielectric constant and the loss tangent demonstrate a good agreement between the two methods. The study confirms the suitability of the combined approach for accurate permittivity characterization of medium-loss dielectric liquids.

The rapid advances in ultrabroadband digital and RF technologies are driving novel capabilities by integrating Radar, Electronic Support Measures (ESM), Electronic Countermeasures (ECM), and COMMs functions across an ultrawide frequency spectrum, all through a common Active Electronically Scanned Array (AESA). Such systems known as Multifunctional RF Systems (MFRFS) are not in operation today, however, they offer unprecedented operational versatility for airborne, shipborne, and ground-based assets in the future.

Ultrabroadband AESA architectures, which dedicate independent sub-apertures for reception, can pick up short radar emissions, making low-probability-of-intercept (LPI) signals far more detectable. Combined with the arrays' rapid beam steering and high directivity, plus independent transmit and receive sub-apertures, MFRFS deliver extended-range, multi-signal, multi-target jamming as well as better burn-through protection at closer range.

Deploying such systems in multisite, multistatic configurations-both on the ground and in the air-unlocks new ECM and ESM tactics essential for tomorrow's demanding scenarios: wider target coverage, instantaneous geolocation, coordinated ECM across platforms, and graceful degradation for greater resilience in a system-of-systems environment.

This keynote will briefly explore these cutting-edge concepts, highlighting how shared ultrabroadband AESA apertures along with their resp. RF architectures redefine detection, jamming, and electronic warfare across all domains.

Synthetic aperture radar (SAR) represents a fundamental means for Earth monitoring from space. An inherent limitation of SAR, however, is that the azimuth resolution constrains the swath width and thus hinder frequent observations at a global scale. This limitation can be overcome by staggered SAR, an acquisition mode that uses digital beamforming together with continuous variation of the pulse repetition interval and has been implemented in the recently-launched NASA-ISRO SAR mission. Besides complex systems, novel ambiguous SAR modes can circumvent the swath/resolution constrain without the need for digital beamforming and are effective for dedicated applications, such as maritime surveillance and ground deformation monitoring, also thanks to the exploitation of waveform diversity and innovative postprocessing techniques. The coherent combination of SAR images, taken from different angles, unlocks further opportunities, such as the generation of accurate digital elevation models and high-resolution tomograms that unveil the three-dimensional structure of vegetation, ice, and dry soil. Distributed systems based on clusters of small satellites flying in formation enable simultaneous collection of such multi-angular images with a significant impact on numerous applications. An affordable and versatile approach for their demonstration is based on swarms of drones equipped with lightweight radar sensors. These advances are forerunners for the development of future groundbreaking Earth observation missions that will offer remarkable societal benefits.

This paper presents the design and experimental evaluation of a compact wideband antenna intended for underground-to-aboveground (UG2AG) LoRa communication. The proposed antenna is integrated with a LoRa-based sensor module and evaluated under realistic conditions by burying the module in soil, while a LoRa gateway located above ground receives the transmitted signal. Wireless communication quality parameters are employed to assess the performance of the radio link.

The design, fabrication and testing of a metallic printed dual-ridged horn antenna for ground-penetrating radar applications is presented. In the frequency range between 250 MHz to 2.5 GHz the horn antenna is optimized for broad bandwidth and lightweight based on perforated metallic and 3-D printed parts with reinforced acrylic materials. With the same structural dimensions as a standard dual-ridged horn antenna, the proposed design achieves lesser weight while extending the lower frequency gain bandwidth by 250 MHz. Further analysis of the far-field radiation characteristics demonstrates good directional performance with minimized side lobe levels - making the proposed dual-ridge horn antenna potentially efficient for low frequency, high penetration depth GPR applications.

The concept for implementing an SIW-to-WR transition, originally designed for multilayer substrates, in a standard two-layer laminate is presented. The proposed transition eliminates the separating metallic wall beneath the air-filled waveguide aperture, which in original design required an additional substrate layers and thus enables its realization on a two-layered structure. In addition, a 3D printed pyramidal horn antenna with an angled feeding waveguide has been manufactured and supplied by the proposed transition in order to test its performance in real-life scenario and to demonstrate a feasible, low-cost approach to the realization of PCB-mounted pyramidal antennas with symmetrical radiation patterns in both principal planes.

In this work, a modulated metasurface (MMTS)-loaded antenna array architecture is proposed to achieve focused near-field energy localization for noninvasive microwave ablation applications. To evaluate the electromagnetic-thermal performance, a voxelized tissue model is positioned at an elevation of 0.08 λ 0 above the MMTS-loaded antenna surface, and fully coupled EM-thermal simulations are conducted to analyze the temporal evolution of temperature rise. The scalar amplitude-tapered metasurface occupies an overall aperture of 4.2 λ 0 × 4.2 λ 0 and is designed to operate at a resonant frequency of f r0 = 2.4 GHz.The metasurface is integrated with a 4 × 1 binomially distributed, series-fed microstrip patch antenna array. The antenna module is excited using a uniformly distributed 1 × 4 Wilkinson power divider, producing a broadside radiation pattern with a peak gain of 13.85 dBi and a sidelobe level (SLL) of 14.97 dB. With the incorporation of the metasurface, the realized gain is further enhanced, reaching a peak value of 16.02 dBi. Simulated and measured impedance-matching characteristics are presented, along with comprehensive thermal and specific absorption rate (SAR) analyses,

This paper presents the design, implementation, and experimental validation of a low-cost near-field measurement chamber for the characterization of antennas operating in the X-band. The proposed system enables measurements of both fixed-beam antennas and active electronically scanned array (AESA) antennas with electronically steerable beams, without the need for conventional far-field measurement ranges. The chamber is constructed using commercially available mechanical components and cost-effective measurement equipment, resulting in a significant reduction in system cost compared to commercial near-field facilities. The paper describes the key design assumptions, measurement setup configuration, and data processing methodology used to reconstruct far-field radiation characteristics from near-field measurements. The proposed approach is experimentally validated through measurements of a fixed-beam antenna array and an AESA antenna for multiple beam steering angles. The obtained results confirm that the developed low-cost near-field chamber provides reliable accuracy for practical antenna characterization in the X-band.

The growing availability of low-cost integrated radar chipsets allows for increasing penetration of radar in industrial and consumer applications. Especially in the field of contact-free medical monitoring, radar may in the future become a powerful and generic monitoring tool for human vital parameters and biomarkers such as heart and respiration activity. Beyond micromotions, however, additional parameters are accessible, for example the indirect determination of mental stress or indications of an upcoming epileptic seizure. This talk introduces a different perspective on this topic from the radar point of view, aside from commonly used Doppler or Frequency Modulated Continuous Wave (FMCW) radar architectures: millimeter-wave interferometry is a special interpretation of Continuous Wave (CW) radar, featuring superior range resolution below the micrometer regime with minimal hardware and algorithmic complexity. Clinical trials have demonstrated highly reliable heart action monitoring with accuracy and precision comparable to the medical gold standard. However, there are also challenges that must be addressed and will be discussed in this talk, such as largescale random body movement artifacts, as referred to by the medical radar community.

In today's short-cycle threat environments, radar engineering faces a familiar Cold War dilemma: systems must be fielded rapidly without sacrificing robustness, controllability, and operational relevance. Historically informed engineering helps by identifying recurring trade-offs, preventing the reappearance of old failure modes under new labels, and preserving experimentally grounded design habits that turn concepts into fieldable capability. Against this backdrop, we revisit West Germany as a NATO frontline state, where phased-array radar evolved from an antenna concept into a programmable, experimentally validated information source. We focus on the closed-loop triad of (i) phased-array sensing, (ii) adaptive target tracking as a knowledge-gathering process, and (iii) radar resource management as the control layer that makes multifunction radar operational. The narrative centres on Wulf-Dieter Wirth and his team, whose work established experimentally validated phased-array operation, tracking-driven radar control, and early sensor-management concepts-and offers lessons for current radar R&D under contested, software-intensive conditions.

Atoms have established their utility in numerous sensing and measurement applications: from high-precision clocks and frequency references, to inertial sensors for positioning and navigation, to gravity and magnetic field sensors, to RF detectors, the spectrum of atom-based solutions is increasing over time. Often these quantum sensors achieve their exquisite performance using atoms that are cooled using laser light to ultracold temperatures (without liquid helium or cryostats). "Ultracold" refers to temperatures near absolute zero, typically 1E-6 K and lower, where the behaviour of atoms is dominated by quantum mechanics rather than by thermodynamics. Correspondingly, quantum noise rather than thermal noise dominates sensor performance. This talk introduces a remarkably faithful and very fruitful analogue between microwave electronics and atom-based circuits. Such "atomtronic" circuits, as they are called, may serve as a foundation for future sensor and signal processing systems based on atoms. For this audience in particular, design principles based on tailoring impedance and the use of transistor action will be familiar, except they are applied here to atoms rather than to electrons. The foundations begin with what can be called the "laws of atomtricity", which are derived starting from seemingly abstract physics called "gauge field theory" applied to interacting identical particles. Applying that theory to electrons one can derive Maxwell's equations, taking as empirical fact that electromagnetic waves travel with speed c in vacuum, and are subject to a vacuum impedance of 377 . Though atoms are neutral, they nevertheless interact through collisions. When one applies gauge field theory to identical neutral atoms and take as empirical fact the matter-wave speed and impedance derived from the laws of quantum mechanics, a set of matter-wave duals to Maxwell's equations arise. From them one can derive the matter-wave duals to Maxwell's electromagnetic wave equations. Continuing the analogy leads to the duals of electric current and voltage, i.e., atom current and "tronic potential" and a characteristic impedance that relates the two - interestingly but unsurprisingly, that impedance involves Planck's constant. While the laws of atomtricity are in some ways more complicated than those of electricity, many electronic circuit design principles nevertheless carry over nicely to atomtronics. In particular there exists the concept of an atomtronic transistor as a device that enables a weak signal to control the impedance of a large signal. Thus one can consider the design of functional circuits such as amplifiers and oscillators, and other possibly useful circuits that utilize atoms rather than electrons. It is fascinating to consider, though, that atomtronic devices like transistors are not made of matter - they are made of light.

Contactless sensing is becoming an important direction in modern healthcare, mainly because it can improve comfort and hygiene while still providing clinically useful information. This paper briefly summarizes recent trends in radar-based monitoring of vital signs, from mmWave FMCW sensors for heart-rate and respiration tracking to high-resolution micro-Doppler solutions and emerging terahertz systems. We also describe cascaded radar architectures, where a wide-field, low-frequency stage is combined with a high-frequency fine scan to achieve high resolution without losing overall coverage. Finally, we present a custom software tool for loading, processing, and visualizing FMCW radar recordings, providing standard views such as range-Doppler maps, micro-Doppler spectrograms, RTI, and range FFT. The goal is to give a compact overview and a usable workflow that can support research and validation in contactless monitoring applications.

This paper presents a novel sensing setup designed for the broadband dielectric spectroscopy (BDS) of liquids, utilizing a combination of additive manufacturing and traditional PCB technology. The proposed solution addresses the common challenge of sensor contamination and the need for rigorous cleaning in liquid characterization by employing custom-designed, 3D-printed exchangeable test tubes. Each tube features a central metallic pin and, when inserted into a complementary spring-loaded jig, functions as an open-ended coaxial transmission line. This configuration ensures that the electromagnetic field is confined primarily within the liquid-under-test (LUT), providing high sensitivity to volume-averaged permittivity across a wide frequency range.

In this paper, a comprehensive investigation on six-port reflectometers' measurement uncertainty distribution is presented. For this purpose, five six-port reflectometers based on different power distribution networks are studied. Their uncertainties are determined using four different approaches. They comprise a general model and different solution models, including linear and non-linear ones, also with weighted approach. The obtained results clearly show how the final uncertainty distribution of a reflection coefficient measurement depends on chosen solution method.

In this paper, a dual-band four-channel multiport amplifier is presented. The circuit consists of two planar 2-D Butler matrices implemented with dual-band branch-line directional couplers and four gain blocks. To obtain high isolation in the frequency range between the operating bands, only two 180° phase shifters are introduced into the amplifier, which reduces the number of required shifters compared to a conventional four-shifter implementation. The proposed multiport amplifier operates in two independent frequency bands at 2.4 GHz and 5 GHz. The measured results demonstrate good electrical performance and show good agreement with the simulations.

The construction, simulation and measurement results of the microstrip antenna designed for X -band, feed by proximity-coupled line have been presented in the paper. The antenna was designed on the Rogers 3003 substrate, to obtain high efficiency and parameters stability in the operating frequency band.
Design process of the antenna in Simulia CST Microwave Studio simulation software was made. Based on that design, the prototype antenna was manufactured in local laboratory of the department. Basic parameters of the antenna were measured in laboratory conditions. Prototype antenna has better than -30 dB impedance matching for center frequency and 6,3 dBi antenna gain, 2 dB higher than simulation results. The proximity-coupled feeding method allows to build complex antenna arrays in simple form, without complex impedance matching, signal dividing and distribution circuit.

A frequency-continuous measurement of passive harmonic responses of 30 common electronic devices from 600 to 1600 MHz is presented. The results indicate strong broadband responses for UHF RFID tags, while smartphones and wearables exhibit only sporadic harmonic signals.

This paper proposes a novel reflecting surface architecture based on diffraction gratings for the 6G upper mid-band (7-24 GHz). The proposed solution leverages a global periodic framework to redirect incident waves through specific diffraction orders, bypassing the need for complex phase-shifting circuits. By optimizing the blaze angle and the spatial period of an echelette grating, the surface concentrates reflected power into targeted anomalous directions while suppressing specular components. This design is validated via a hybrid approach combining full-wave simulations and analytical pattern multiplication. The results demonstrate a high-efficiency, low-complexity, and passive strategy for establishing secondary line-of-sight links thus eliminating blind zones in smart radio environments.

The aim of this study is to analyze the propagation of radio signals inside a commuter train car at the Frequency Range 3, a potential band for future sixth-generation (6G) communication technologies. The study focused on the characteristics of radio signal propagation in the enclosed metallic space of the car, taking into account the effects of obstacles, reflections, and interior design on signal loss. Measurements were carried out in the interior of a commuter train. The results indicate that the interior exhibits relatively high attenuation under both LOS and OLOS conditions. The simulation results indicate the significant role of the rail car's interior environment in electromagnetic wave propagation.

This paper experimentally demonstrates backscatter modulation using a diode detector interfaced with an analog-to-digital converter (ADC). Backscatter is realized by switching the ADC input impedance through digital reconfiguration of its internal buffer and programmable gain amplifier, thereby inducing controlled variations in the reflection coefficient at the input. Measurements are performed using a commercially available PIN diode-based detector operating at 2.4~GHz in conjunction with an ADS1256 ADC. Binary impedance modulation is employed to maximize impedance contrast. The system is first characterized by measuring the ADC input directly to determine the intrinsic operational limits, followed by full over-the-air backscatter measurements using the same modulation scheme. Reliable operation is observed at low baud rates, e.g., 20~bps, with a measured BER of (\sim 10^{-4}), while higher baud rates exhibit increasing error probability, with operation near 500~bps resulting in a BER on the order of (\sim 10^{-1}).

Recent advancements in the field of communications and cryptology have attracted significant research efforts in studying the randomness of bit sequences. These studies have led to the development of different test methodologies, such as those developed by NIST (National Institute of Standards and Technology), whose main objectives are to verify the independence of the individual elements in the sequence and to test their distribution within the bitstream. This study presents a new gap-based approach for analysing bit sequences by introducing a burstiness factor. The proposed burstiness factor is used to measure the level of burstiness to indicate whether the commonly considered IID characteristics (independent and identical distribution) of random variables are violated. To validate the proposed approach, this study employs different polynomial and non-polynomial sequence generation methods. The results confirm that the proposed definition of burstiness effectively indicates the non-IID characteristics of randomised bit sequences.

This paper presents a unified, simulation-driven optimization framework for fast synthesis of multi-layer pixelated passive RF circuits, including power dividers, baluns, and duplexers. Candidate layouts are represented as stacked binary metallization grids across multiple layers and evaluated through full-wave electromagnetic (EM) simulation feedback, without relying on predefined topologies, templates, or geometric constraints. A discrete Gaussian-process surrogate defined in normalized Hamming space is combined with a Hamming-ball trust-region mechanism to achieve stable exploration under tight compute budgets. The formulation generalizes across circuit class, grid resolution, and number of metal layers and ports. Optimization uses coarse-mesh EM simulations for rapid iteration and periodically re-simulates promising candidates with a fine mesh for validation. We evaluate our approach on D- and V- band transformers, power dividers, and duplexers. Our results show that wideband and low-error designs can be obtained with around 2000 EM simulations compared to more than 5000 simulations on an evolutionary optimizer.

Various phase-shifter architectures have been presented in the literature. Different techniques prove useful for various applications. In some cases, the ability to apply a continuous 360° phase tuning is critical. One example of a phase-shifter architecture with infinite phase shift is a transversal filter-based solution.

Such devices are prone to errors due to component imperfections and manufacturing variation. This contribution presents the background and principles of operation of a novel calibration method for a transversal phase shifter based solely on scalar measurements. In the final paper, the theory of operation and the primary sources of error will be discussed. The proposed calibration method will be presented. In addition, an example of a prototype transversal phase shifter will be shown, along with its performance results before and after calibration.

This paper presents a unified analysis of microwave protective coatings based on absorbing composites and metamaterial/metasurface concepts. Fundamental electromagnetic and material-related limitations governing the relationship between reflectivity, operational bandwidth, and coating thickness are discussed. Special attention is paid to the roles of dielectric, conductive, and magnetic constituents in composite absorbers, as well as to impedance matching and wavefront manipulation enabled by metasurfaces. By combining material-science and electrodynamic perspectives, general constraints on broadband low-profile microwave protection are formulated, and design routes for advanced radar-absorbing and reflection-controlling coatings suitable for adaptive camouflage are outlined.

A novel compact high power waveguide four-port junction circulator is proposed. The circulator composed of two junctions with opposite circular directions. By choosing ferrite materials with appropriate line width, the power capability is improved. Meanwhile, the ferrite cavity is analyzed to expand the bandwidth. Based on the analysis above, a waveguide four-port junction circulator working in X band with section size of about 0.6 λ by 0.6 λ is designed, fabricated and measured. Measurements show that bandwidth of 11.7% with an insertion loss lower than 0.35 dB is realized. High power tests show that its peak power capacity can reach 40 kW.

To address the demand for high-performance power signal synthesis, this paper introduces a four-way combiner utilizing a ridge waveguide transmission structure. The device comprises a cascaded arrangement of a front-stage E-plane and a rear-stage H-plane power divider, effectively reducing its cross-sectional footprint. A load incorporating carbonyl iron material is designed to form a resonant circuit that absorbs differential-mode energy, thereby achieving high isolation characteristics. Based on this design, a high-performance combiner was fabricated with an approximate cross-sectional dimension of 0.6λ*0.6λ. Within a relative bandwidth of 11.7%, it exhibits an insertion loss better than 0.4 dB (corresponding to a combining efficiency exceeding 91%) and a port isolation greater than 15.5 dB. Simulations confirm that the combiner can withstand power levels above 30 kW.

This paper presents the design process of the low- noise oscillator operating at 1.3 GHz and stabilized with a dielectric resonator (DRO). The general aspects of the construction of negative-resistance-based DROs, together with the technical details of passive and active circuits, are included in Section II. The results of phase noise spectrum measurement are shown and discussed in Section III. The DRO presented in this paper showed a floor noise level of -166 dBc/Hz, an output signal power of 5.8 dBm and a loaded quality factor of 6 900.

Diversity of supported movements is an important factor for evaluation of prototype antenna far-field performance figures. A wide range of motion of a radiator under test w.r.t. the reference antenna can be maintained using robotic arms. However, loading of such systems with absorbing material without limiting the range of motion is difficult. In this work, the effects of antenna distance from the robot, as well as post-processing of measurements on the fidelity of radiation patterns are investigated. The study involves evaluation of the radiator at a few offsets from the effector of the robot and distances to the reference antenna. A comparison of a formal multi-setup correction against an approach based on aggregation of multiple spatially-separated measurements is also considered. The obtained results demonstrate that multi-setup post-processing has more pronounced effects on the fidelity of responses compared to the change of antenna offset.

This paper presents a novel microwave biosensor designed for the rapid and label-free detection of Escherichia coli (E. coli). The proposed device leverages a space-filling Hilbert curve geometry to enhance sensitivity while maintaining a compact footprint. By utilizing a differential transmission-mode measurement and surface biofunctionalization, the sensor achieves a limit of detection (LoD) of 10^2 CFU/mL, significantly outperforming conventional planar microwave transducers.

The paper presents a hetero-integration approach combining InP-HBT chiplets with BiCMOS chips through a microbump flip-chip scheme. The interconnects offer more than 200 GHz bandwidth, the InP process allows to realize MMICs with a transistor fmax of more than 500 GHz, which qualifies it for both D-band transceivers and high-data-rate drivers and transimpedance amplifiers. The platform is being developed in the framework of the APECS pilot line funded by the EU and the German government. It will be made accessible to external partners.

Wireless Power Transfer (WPT) has gained immense interest in the last decade due to its vast potential applications in existing and emerging domains such as in powering appliances, gadgets, electric vehicles, and a plethora of sensors for IOT applications. Resonant inductive (or capacitive) coupling provides a promising means of WPT due to its high efficiency over short and midrange distances. However, high range WPT technology require overall system design, which includes design of suitable antennas, rectifiers, power management units, channel estimations, and most importantly design of the feedback loop between the receiver and transmitter. Recently, Time Modulated Antenna Array (TMA) or timed array, is emerged as a potential candidate for enabling versatile, multi-user heterogeneous wireless systems, with each user uniquely powered by a switching harmonic frequency. In timed array, each switch at the antenna element feed is controlled with a Pulse Width Modulated (PWM) signal, carrying the phase information of a triangular wave and a reference signal. PWM timed array architecture offers precise control of the number of radiated harmonics, independent steering of individual beams and radiated beam power levels. This keynote talk would deal with arts and science of various WPT concepts, techniques and some recent developments by the speaker in his research group at IIST, India and RMC, Canada

Classical electromagnetics (EM) distinguishes two basic types of boundary conditions, Perfect Electric Conductor (PEC) and Perfect Magnetic Conductor (PMC), which are both "polarisation-dependent". For example, at a PEC boundary the tangential component of E-field is zero, while the normal component has zero normal derivative. In the context of antenna technology, excellent contributions were made to develop "polarisation-independent" boundaries, corresponding to the soft and hard surfaces in acoustics. For example, corrugated "soft" antenna horns produce beams of the same width of the radiation patterns in both E- and H-plane. In this presentation, we explore a new type of artificial boundary conditions, which we shall call "mode-selective", as they impose nearly-zero or nearly-unity reflection coefficient, depending on the incident mode. We focus on their use in high-Q resonators, as elaborated in the recent patent application. It has been shown there, that with such boundary conditions (which is practically realized with a combination of corrugations and lossy inserts) we can filter-out a class of unwanted modes while keeping the wanted ones unaffected. Thereby, the spectrum of modes inside the resonator is "cleaned" to contain the desired modes only. When applied in a high-Q resonator, the "mode-selective" surface is named Q-Choke [2]. Its design is performed with fullwave EM simulations, which will be used to illustrate this talk. Two examples of application to microwave material measurements will be discussed, by reference to two resonators manufactured and measured at QWED: Q-SCR (Q-choked Split Cylinder Resonator) and Q-SSR (Q-Choked Sandwiched Sapphire Resonator [4]). We hope that the presented idea will inspire members of the Microwave Community to extend the concept of "mode-selective boundaries" to other types of resonators and their applications.

Estimation of target angle in a radar system using a frequency-scanning array is considered. To characterize the problem's properties, an absolute performance bound in the form of Crámer-Rao lower bound is established. Then, behavior of the pseudo-monopulse estimator - a method based on the amplitude comparison principle - is characterized using analytic approach and computer simulations. Several expressions with varying levels of complexity and accuracy are obtained.

This paper investigates direction-of-arrival (DoA) estimation for amplitude-only multi-channel forward scatter radar (MC-FSR) systems using a compressed sensing approach based on orthogonal matching pursuit (OMP). The proposed approach employs a modified sensing matrix tailored to realvalued signals and incorporates multiple DC-removal strategies to suppress dominant stationary components. By exploiting spatial sparsity, the proposed OMP-based approach enables superresolution DoA estimation with reduced snapshot requirements, including operation in the single-snapshot case. Through comprehensive computer simulations, both qualitative and quantitative analyses are provided to characterize the performance of the proposed approach especially in comparison with an alternative MUSIC-based method. The effects of snapshot number, assumed source count, and limited array size were also examined. Overall, the results suggest that the proposed method represents a viable alternative to existing super-detection techniques with limited snapshot availability or processing resources.

Pulse-to-pulse waveform diversity is attracting ongoing interests in recent years. However, doing so normally introduces range sidelobes modulation (RSM) effect which greatly degrades clutter cancelation. To address this issue, we consider the optimization of unimodular sequences to minimize the RSM effect. This goal is achieved by generating diverse sequences of similar autocorrelations. The design problem is formulated and solved using conjugate gradient (CG) method. Particularly, we derive a way to employ fast Fourier transform (FFT) in calculating both the objective function and its gradient with respect to the sequences' phase, so that the CG method can be implemented efficiently. Simulation results show that the generated sequences yield a small RSM effect with the degree of diversity between pulses holding almost unchanged.

In this paper, we present a practical yet novel predictive analytics method to dynamically form and select adaptive electronic protection (EP) countermeasure radar waveforms against knowledge-based (KB) noise jammer signals using electronic support (ES) collection and interference estimation. While previous works assume a priori knowledge of the jammer to mitigate the threat, we introduce an improved radar-ES system that first senses the radio frequency (RF) environment and measures the prevailing interference through temporal and spectral estimation for cognitive EP applications. The estimates are then used to generate an aggregation of possible counter transmit-adaptive radar waveform options that are spectrally formed to null or avoid KB interference. Finally, counter-waveform predictive analytics-based decisions are applied to select the predicted best-performing EP strategy based on user-defined prioritization metrics that rank, score, and weigh the resulting probability of detection, bandwidth utilization, and processing time.

Dual-Function Radar and Communication (DFRC) systems or Integrated Sensing and Communication (ISAC) can be categorized in two groups. The first, primarily intended for high end Multi-Function Radar Systems (MFRS), uses common RF hardware and implements time, frequency or spatial sharing strategy between "radar" and "communication": the radar waveforms are optimized only for radar purpose and the communication waveforms are optimized only for communication performances. The second group is aimed more at simpler systems: a single waveform compatible with both uses is employed. This paper proposes a pulsed chirped waveform in which data is embedded through zero-mean orthogonal polynomial perturbations of the instantaneous frequency. The global frequency sweep and spectrum are preserved, ensuring pulse response comparable to a strictly Linear Frequency-Modulated chirp (LFM). The communication symbols are recovered via maximum-likelihood estimation of the residual phase after de-chirping.

Synthetic Aperture Radar (SAR) imagery of a target on or above a rough surface is strongly affected by the composite electromagnetic field between the target and the soil. To model this target-soil coupling for Radar Cross Section (RCS) simulations, deterministic or statistical methods have been proposed in the literature. The former requires a discretised, deterministic scene of interest to compute the scattering coefficient of a given realisation, whereas the latter leads to closed-form expressions of an average scattering coefficient for a given random process of the soil heights. For SAR image simulations, deterministic approaches have already been proposed and applied. In this paper, we adapt a statistical rough-soil model to SAR image simulation in the context of target-soil coupling. We combine it with a deterministic model for the target. Finally, using the presented semi-statistical model, the simulation results are analysed with respect to the rough soil parameters.

The accurate interpretation of synthetic aperture radar (SAR) images is crucial for mission planning and sensor design in both military and civilian applications. In common practice, image interpretation is standardized using the National Image Interpretability Rating Scale (NIIRS), while NIIRS levels are estimated through General Image Quality Equations (GIQE) using appropriate imaging parameters. Most of radar GIQE formulations in the literature use resolution and noise parameters, ignore image sharpness effects, and are derived from limited data sets. This study conducts a comprehensive performance comparison between the recently introduced Radar Image Quality Equation (RIQE) and existing radar GIQE models. The introduced RIQE differs from previous GIQEs by including the Relative Edge Response (RER) parameter to represent image sharpness. RIQE framework integrates three fundamental parameters-ground sample distance (GSD), signal-to-noise ratio (SNR), and relative edge response -to enhance the prediction of SAR image interpretability. The coefficients of all models are re-estimated using a common large-scale synthetic SAR dataset to ensure statistical consistency in coefficient estimation. Model accuracy and reliability are evaluated through multiple complementary metrics, including quadratic weighted Cohen's κ, Spearman's ρ, global and in-class RMSE and ROC-AUC. Experimental results demonstrate that the RIQE formulations achieve the highest goodness of fit (κ = 0.984), correlation (ρ = 0.979), and discrimination capability (ROC-AUC = 0.998), while maintaining stable interpretability mapping across all NIIRS levels. In addition to synthetic benchmarking, the models were further evaluated on a real SAR dataset consisting of 30 X-band Capella open-data images. Using the same complementary metrics, the proposed RIQE yields markedly lower prediction error (RMSE = 0.277 NIIRS) and near-zero bias (−0.047 NIIRS) while preserving strong rank consistency (ρ = 0.817). These findings confirm the robustness of the RIQE framework and provide a quantitative benchmark for assessing radar GIQE performance under consistent data and evaluation conditions.

Polarimetric backscattering characteristics of a low-vegetated land area containing a series of trees and bushes and reference targets including trihedral corner reflectors and metallic rods are investigated using ground-based synthetic aperture radar (GB-SAR) measurements and H/α ̅ scattering decomposition technique. Two distinct frequency bands, a low-frequency band (0.75-1.75 GHz) and a high-frequency band (2.25-3.25 GHz) are considered to assess the influence of frequency/wavelength on the observed scattering mechanisms. The polarimetric entropy H, mean scattering angle α ̅, and H/α ̅ classification images demonstrate that metallic reflectors and manmade structural elements exhibit pronounced changes in dominant scattering responses between frequency bands, while natural vegetated areas show frequency-dependent transitions between moderately random backscattering with double-bounce and dipole scattering contributions. These findings highlight the capability of polarimetric GB-SAR to capture wavelength-driven variations in scattering physics.

A consistent trend in Synthetic Aperture Radar (SAR) is towards a larger swath width and higher resolution, both parameters usually contradicting each other. While there are many imaging techniques that can increase swath width, they usually trade it for the resolution. This paper presents a new method, utilizing the Frequency Scan for Time-of-Echo Compression (f-STEC) imaging mode, that increases the total swath width by imaging the scene with multiple beams on transmit and receive, where all sub-swaths take advantage of the full system bandwidth, i.e., resolution. True to the principle of f-STEC, the beamforming is completely analog and can be realized with a simple antenna front end design.

Aerial Synthetic Aperture Radar (SAR) systems are increasingly used in applications such as autonomous navigation of Unmanned Aerial Vehicles (UAVs), where accurate Above-Ground-Level (AGL) altitude information is critical for both image formation and safe flight operations. Small UAVs often lack payload capacity or power budget for dedicated altimetry sensors, motivating alternative solutions leveraging existing SAR data. This paper proposes a novel method to estimate AGL altitude directly from SAR range-profile data, without any additional hardware. The approach exploits the characteristic energy step in the range profiles caused by the nadir ground echo to infer the slant range to ground using a robust edge detector. The method is conceptually simple, computationally efficient, and readily integrable into existing SAR processing chains. Experimental validation on real airborne X-band SAR data from a fixed-wing aircraft shows that the proposed approach achieves meter-level accuracy, with a root-mean-square AGL error below 1% of the nominal flight altitude and median absolute errors on the order of a single range-resolution cell, demonstrating its suitability for terrain-relative navigation and support of autonomous flight.

The increasing availability of very high-resolution (VHR) spaceborne Synthetic Aperture Radar (SAR) imagery is opening new opportunities for maritime surveillance applications, by enabling more detailed characterization of vessel scattering, including both primary returns and secondary multipath contributions (i.e., double- and triple-bounces returns). This paper presents a methodology for estimating ship vertical profiles by exploiting ships' multipath effects. A two-dimensional geometric model is derived to relate the slant-range separation of double- and triple-bounce signatures to the height of ship structures. To retrieve these features, a dedicated processing chain is presented, incorporating inverse SAR (ISAR) autofocus to compensate for motion-induced defocusing and adaptive sidelobe suppression to reveal fine multipath signatures that could otherwise be masked by the main target response. The framework is validated using real spotlight data from Capella Space. The analysis of two crude oil tankers of the same class demonstrates the method's capability to estimate vertical features, such as freeboard and mast height, and to potentially discriminate between different loading conditions (e.g., ballast vs. full load), thereby providing valuable inputs for enhanced maritime situational awareness.

As well-known, the unaccountable radial velocity of uncooperative targets in the scene causes them to appear displaced in conventionally focused SAR imagery. This work introduces a new technique which exploits properly mismatched filters for azimuth compression to retrieve a moving target's radial velocity from the generated relative displacement and thus to relocate the target in the imaged scene. The effectiveness of the proposed technique is demonstrated both on simulated and real spaceborne SAR data.

The full velocity vector of a surface moving target can be estimated from Synthetic Aperture Radar (SAR) observables by either using: (i) a single platform able to measure Doppler rate and along-track interferometric (ATI) phase or (ii) a constellation of two satellites simultaneously measuring only Doppler rates or ATI phases or both. This work derives and compares the estimation accuracies of the available possibilities and compares them. Results obtained for a spaceborne case study show that the single-platform configuration is sufficient to achieve a good estimation accuracy provided that it has the two receiving channels connected with antenna element with a sufficient ATI baseline and operates at low squint angles (below 10°). Unless this baseline is very large, a comparable accuracy can be obtained by a symmetric constellation of two single-channel satellites only measuring the Doppler rate, operating with higher squint angles (above 10°). This provides the basis for a trade-off between a single large-satellite SAR and a small coherent constellation of small-satellite SAR. In addition, the presented analysis shows that the best estimation performance can be obtained by exploiting a more complex constellation of (large) dual-channel SAR; however, most of the improvement with respect to the simpler cases is obtained using a constellation of a large dual-channel SAR and a small single-channel SAR.

In recent years, the miniaturization of space radar systems has been a key enabler for the deployment of Synthetic Aperture Radar (SAR) sensor constellations. This trend offers the potentiality to estimate kinematic parameters of moving targets exploiting the spatial diversity among platforms. In this context, a technique based on the exploitation of the relationship between velocity and residual Doppler rate has been developed in [9], with specific focus on maritime targets. However, since movers may present more complex motion, the technique's performance may be influenced by the presence of longitudinal accelerations. This study aims to analyse these effects, with particular emphasis on how increasing the coherent processing interval influences the sensitivity of velocity estimates to acceleration. To achieve this, the impact of acceleration is examined and the limits of applicability of the Doppler-rate-based velocity estimation technique are determined. Both simulated results and theoretical performance analyses are provided.

Methods of measurement of natural materials that exhibit negative permeability or negative permittivity at microwave frequencies are presented. They are based on rigorous electrodynamic analysis of resonance structures containing dispersive and anisotropic media. Special attention is given to measurements of gyromagnetic materials, both insulating and conducting (thin metallic ferromagnetic films). It is shown that samples made of such materials exhibit properties of magnetic plasmon resonators, which operate at negative permeability values. Several distinct phenomena of plasmonic resonators are analyzed, such as unequal electric and magnetic energies stored at the resonance, the small influence of dielectric and conductor losses on the Q-factors of magnetic plasmon resonances, and the role of radiation and dissipation losses on properties of plasmonic resonators. Examples of materials that exhibit negative permittivity values at microwave frequencies are also presented, namely gaseous plasma and superconductors. In theoretical analysis, free oscillation and Mie scattering theories are compared. Results of measurements of yttrium iron garnet samples, thin CoFeB films, and high-temperature semiconductors are presented. Most of the theoretical and experimental work shown in this presentation has already been published, but some new ideas are underlined.

Measurements of the dielectric properties of three grades of paraffin were made by a Materials Characterisation Kit (MCK) in conjunction with a Vector Network Analyser, and THz TDS system. The purpose of the work is to develop techniques for characterizing liquids that are used in the cooling circuits of data centres and solid-state transformers.

In this paper, we analyze the details and preparation of liquid samples for human-like phantoms. The key element in the preparation of such phantoms is the precise validation of dielectric parameters for the designed sample materials. In this paper, we discuss the process of creating a measurement based on a dedicated dielectric resonator fixture for liquids. The fixture operates close to 2.5 GHz, which matches the vial geometry and provides a practical compromise between sensitivity and sample handling. The outer diameter of the tube and its dispersion were propagated to bounds on the extracted properties.

Leaky-lens antennas (LLAs) count among the most effective (sub-)millimetre wave radiators for ultra-high-rate digital communications and deep-space astronomical instrumentation. In line with, practically, the entire literature on leaky-wave (LW) propagation, the operational principles of LLAs are described in the frequency-domain (FD). Remarkably, the time-domain (TD) perspective on the LLA functioning was missing until recently. While the FD approach yielded a thorough understanding of the radiator's steady-state, time-harmonic (TH) operation and, primarily, allowed developing extremely effective design tools, concerns surfaced about the adequacy of the FD models in ultra-high-rate systems, especially when extremely narrow beams and agile beam scanning are requisite for the application at hand. Specifically, it was far from obvious if a steady-state TH regime can be still assumed when the propagation consists of extremely short packages of sine-waves. To address this limitation, efforts were invested in the past years in developing a strictly causal description of the LW radiation. This contribution will then summarise the main findings of this research line, by insisting on specific characteristic aspects of the LW radiation. The conclusions will be drawn based on a careful numerical analysis of the typical, weaklydispersive, LW configuration underpinning LLAs. That study proves beyond any doubt that the investigated weakly-dispersive, LW radiation is the result of constructive or destructive interference of causal waves launched at regular intervals from a feeding point. This crucial observation falls completely outside the scope of any FD analysis that, inherently, cannot account for the origin of the field values at given locations. While our investigation did in no way question the validity of results inferred via steadystate FD instruments, we gave a first demonstration of the transient mechanism leading to the observed steady-state field behaviour. At the same time, our experiments cogently illustrate that the beam emerging from the feeding point can differ a lot during the transition to the steady-state from the expected, settled, steady-state profile, a matter that must considered in the design of ultra-high-rate communication channels making use of basic, on-off keying modulation. The reported research was carried out by a group comprising, apart from the author, Dr. Martin Štumpf of the Brno University of Technology, the Czech Republic, and Professor Andrea Neto of the Delft University of Technology. Their support and agreement to reporting our common results is hereby acknowledged.

This paper proposes a method for the generation of a Bessel beam. The beam is generated with the use of a graded refractive index (GRIN) lens and multiple-feed-per-beam (MFPB) technique. The lens is illuminated by a 2×2 antenna array where all radiating elements are placed off-axis, so each element produces beams tilted from the broadside direction. The combined radiation patterns, generated by individual radiating elements, result in creation of the Bessel beam. The concept has been experimentally verified with a biodegradable polyactic acid (PLA)-based GRIN lens antenna operating at 12 GHz. The lens antenna achieves a maximum realized gain of 17.2 dBi at 12 GHz, which corresponds to the improvement in gain of 8.3 dB over the 2×2 array. The maximum gain improvement in the non-diffractive region is 10.3 dB.

This paper presents a cost-effective flat lens waveguide antenna operating in 90 - 140 GHz frequency band. The antenna design utilizes a structure of parallel high resistivity silicon flat plates with air gaps positioned at the open end of a rectangular waveguide to achieve beam focusing. Simulation results show that this straightforward and compact construction can achieve gain up to 18.8 dBi. Contrary to other, higher gain flat silicon lens designs, presented solution does not require advanced processing of the silicon such as selective etching to create patterns which greatly reduces manufacturing costs.

Future lunar exploration requires advanced multi-frequency radar sounders capable of mapping geological structures and detecting volatiles. This paper presents the design of a compact, fully-digital radar sounder and the implementation of its demonstrator on the Xilinx ZCU111 Radio Frequency System-on-Chip (RFSoC) platform. Exploiting the device's Direct RF capabilities, we propose a parallelized, multiplier-free digital waveform generation architecture capable of synthesizing coherent Linear Frequency Modulated (LFM) chirps at HF (15, 60 MHz) and VHF (150 MHz) bands without requiring analog up-conversion. The design leverages a parallel Numerically Controlled Oscillator (NCO) approach to bridge the gap between the FPGA clock frequency and the mega sample-per-second RF sampling rates required for wideband operation. Experimental validation via hardware loopback confirms the high quality of the digitally generated LFM waveforms. Furthermore, the pulse compressed output shows the vertical resolution predicted by theory. Finally, the resource utilization table highlights that no DSP slices are used for signal generation, thus preserving significant logic resources for future on-board radar signal processing.

Experiments conducted by the authors on passive radar observation of Low-Earth Orbit (LEO) objects using digital radio and television transmitters as illuminators of opportunity, with sensitive LOFAR radio telescope receiver arrays capturing extremely weak reflections from orbital targets, have demonstrated the feasibility of detecting and estimating target motion parameters. During analysis of the measurement data, a characteristic blurring of strong echoes with high signal-to-noise ratio (SNR) on range-velocity maps was consistently observed, forming triangular patterns. In this paper, we show that this effect is caused by multipath propagation, where part of the illuminating signal reflects off the Earth's surface near the illuminator before reaching the satellite. Simulations replicating realistic measurement conditions, including identical satellite trajectories, confirm that interference between direct and reflected signal components produces the observed echo distortion. The results explain multipath-induced artifacts, highlight their impact on detection and tracking accuracy, and provide guidance for future mitigation strategies in passive bi- and multi-static observation systems.

Recent technological developments and evolving trends in radar systems have renewed interest in passive radar techniques. As an example, the LOFAR (Low-Frequency-ARray) radio telescope, with its large-scale distributed antenna array, represents a promising surveillance receiver for passive space radar applications, including the detection of Low-Earth Orbit (LEO) objects. This paper presents a large-scale electromagnetic simulation study of the entire LOFAR antenna array, with a particular focus on modeling array gain characteristics as a function of azimuth and elevation. The simulated gain patterns were subsequently used to assess passive radar performance through Signal-to-Noise Ratio (SNR) estimation. The predicted SNR values derived from the modeled array performance are compared with results obtained from real-world passive radar measurements, including the effects of an actual illuminator of opportunity and its radiation pattern. The electromagnetic simulations are performed using Ansys software, employing the Far-Field Array Data Discrete Method (FADDM), which enables computationally efficient full-array modeling of the LOFAR antenna system. The results demonstrate the feasibility and relevance of large-scale electromagnetic modeling for passive space radar performance assessment.

Over-the-horizon radars (OTHRs) are a novel category of long-range radar that can achieve ranges often out to ~3,500 km. This is achieved by using the reflective properties of the ionosphere in the High Frequency (HF) band to reflect radio waves back towards the ground and enable beyond line-of-sight (BLOS) propagation and target detection. Coordinate registration (CR) is an essential step required to convert OTHR measurements into meaningful geographical positions and has a significant impact on target parameter errors. As errors inevitably exist in the propagation model used in CR, it is imperative to understand how the errors degrade the performance of the CR system. We here investigate the error behavior of an inversion-based CR in response to bulk ionospheric errors of up to ±1.5 MHz in fof2 and ±30 km in hmf2 in a simulation environment. Radial velocity errors were shown to not exceed one solution bin except for cases with a +30 km hmF2 offset and fof2 of between 0- and -1.5 MHz. All range errors were shown to be less than 50 km including the most significant offset of -1.5 MHz and -30 km.

This paper presents a comprehensive comparison of all multistatic sensing configurations within Coordinated Multipoint (CoMP) Integrated Sensing and Communication (ISAC) systems for an automotive intersection scenario. An OFDM-based simulation framework is developed to evaluate monostatic, bistatic, and multistatic sensing schemes under binary integration using various m-of-n fusion rules. Simulation results demonstrate that full multistatic sensing significantly enhances sensing performance, improving coverage from 20% in the standalone monostatic configuration to 80%, corresponding to a fourfold increase, while the average probability of detection increases from 0.25 to 0.83 over the entire sensing area. These results confirm that cooperative sensing substantially improves detection reliability and coverage without incurring high communication overhead. Furthermore, the results highlight the trade-off inherent in the fusion logic, where smaller values of m favor coverage extension, while larger values of m improve detection robustness.

Occupancy monitoring in public transportation has attracted increasing attention from multiple stakeholders. In order to effectively comply with the European Union regulations on privacy protection, radio-frequency-based passenger localization and occupancy monitoring provide a promising privacypreserving solution. This study proposes a passive, device-free passenger positioning and occupancy recognition system based on a dual-antenna Ultra-Wideband (UWB) radar. The study presents a detailed processing pipeline that starts from the baseband data of the UWB radar, applies a Moving Target Indicator (MTI) based on Infinite Impulse Response (IIR) to suppress static clutter, and subsequently estimates Angle of Arrival (AoA) of the signal through Phase Difference of Arrival (PDoA) method. Furthermore, based on the estimated AoA, combined with K-means clustering and the Mahalanobis distance, the system achieves position and occupancy detection for single and double passengers in a single-row bus scenario. Expreimental results showed that the proposed system is able to recognize all positioning configurations considered within this study.

High resolution sensors can be beneficial for automated driving in complex urban environments. The increased requirements in terms of all-weather capability indicate that microwave measurements can be advantageous for ensuring reliable sensing of vehicle's surroundings. This paper presents a combined high resolution radar and lidar approach.

High-resolution imaging radars are central to advanced driver-assistance systems (ADAS) and autonomous driving. Fine angular resolution requires large apertures. Sparse MIMO array configurations achieve this in practice by synthesizing large virtual apertures from fewer physical elements. The design of such arrays involves balancing the number of transmit and receive channels against antenna placement. This is done to suppress grating lobes and ambiguities across the field of view (FOV) of interest. Standard array performance metrics, such as the peak side lobe level (SLL) of the array factor, assume far-field conditions and usually evaluate the response at boresight. These metrics do not capture imaging performance in the near-field, where the plane-wave approximation breaks down. In this regime, the second-highest imaging peak can appear at a range different from the main target location. In this paper, we introduce a range-angle imaging ambiguity function that extends into the near-field. We formulate a radially localized side lobe level (rLSLL) estimator and, by extracting the second peak across the complete FOV, the localized side lobe level (LSLL). It quantifies ambiguity for a given imaging position. Numerical results on a sparse planar MIMO array demonstrate the LSLL range migration across the near-field and converges to the classical far-field SLL at sufficient range. These estimators provide a practical tool for the design of high-resolution imaging radars and their performance characterization.

Interference in frequency modulated continuous waveform (FMCW) radars often results in bursts of corrupted samples within the received signal. While detecting and suppressing or reconstructing corrupted samples is a common mitigation strategy, the conventional threshold based detection can miss several corrupted samples, thereby degrading radar performance. This paper proposes a method to enhance interference detection algorithms by reducing false negatives.

In this study, a rectangular waveguide based wideband method is investigated to characterize the effective surface impedance of a superconducting thin film at X-band. Waveguide based broadband microwave data which is collected via simulations for S-parameters is used to determine the film's effective surface impedance, which is crucial for quantum computing applications where low losses are required with precise resonance frequency settlement in the design of microwave passive circuits operating at cryogenic conditions. This method is demonstrated to characterize a 10 nm thin film on a Silicon-Germanium substrate with thickness of 0.5 mm as well as permittivity of 12.9 and loss tangent of 0.0001. It gives more information along a wider spectrum all at once compared to other resonant methods that require fine frequency tuning. The study also discusses the importance of saving significant time and cost in cryogenic test setups and measurements.

This paper presents a microwave ring resonator based detection technique for the toxic additives in water. The proposed system operates in the 1--5 GHz frequency range and exploits resonance frequency shifts and amplitude variations induced by changes in the effective dielectric properties of the liquid under test. Unlike conventional sensing techniques, the proposed approach does not require direct contact with the liquid, enabling rapid, non-invasive, and real-time measurements. The planar ring resonator is designed using CST Microwave Studio and manufactured by the use of RT/duroid 5870 substrate with εr = 2.33 and tanδ = 0.0012. Multiple higher order resonance modes are utilized better to enhance the discrimination capability. Experimental results obtained using a compact portable vector network analyzer (VNA) platform demonstrate clear spectral separation between pure water and water contaminated with selected toxic substances. The proposed technique is suitable for field-deployable water safety monitoring and security-critical applications.

This paper introduces a new approach to multifunctional sensor interfaces, such as robotic e-skins and haptic sensors, using planar deep-subwavelength microwave resonator technology. This method is based on detecting changes in the resonance frequency and the magnitudes of S11 and S21 parameters caused by external factors, such as object proximity, pressure, strain, temperature, or magnetic fields. The proposed technology offers several advantages over piezoelectric, resistive, or capacitive sensors. First, planar microwave resonators display a wide dynamic range in perturbed resonance, enabling high-dynamic-range sensing. Second, their deep-subwavelength nature enables millimetre-scale spatial resolution and operation in the Super High Frequency band below 20-30 GHz, with affordable signal-generation and readout hardware. Third, multiple sensors at different resonant frequencies can be accessed via a single-line interface using orthogonal frequency-division multiplexing (OFDM), enabling large-area e-skins with simplified wiring, which constitutes a significant improvement over current designs that require two or more wires per pixel. Further enhancement of the sensing capabilities is proposed by merging microwave resonator technology with functionalised nanomaterials and metamaterials. In particular, this study demonstrates that loading a microwave resonator with functionalised carbon nanotubes yields a temperature sensor with a wide sensing range. The results of this study are important for advanced robotic and industrial sensing interfaces.

In this work, we investigate the impact of the sample container's geometry on the parameters of a low-cost, single-use microwave sensor designed for honey adulteration detection. The sensor features a nine-finger circular Interdigital Capacitor (IDC) geometry, which provides a multi-resonant response allowing for sensitive dielectric spectroscopy. The analysis focuses on the 1 to 2 GHz frequency band, due to the optimal penetration of the electromagnetic field into the sample in that range. To isolate geometric effects from substrate variability, a parametric study of sixteen 3D-printed container configurations was conducted using a single sensor unit loaded with off-the-shelf lime honey. Experimental results demonstrate that the sensor's response saturates at a sample height of approximately 5 mm, indicating full field containment. Furthermore, increasing the radius enhances performance, though this comes at the expense of a larger sample volume. Different behavior was observed between resonant modes at 1.19 GHz and 1.49 GHz, attributed to the trade-off between volumetric field penetration and dielectric losses. The results suggest that a thin dielectric floor can be introduced without performance degradation, paving the way for the cost-effectiveness of a design improvement, with a reusable sensor and disposable sample containers.

Our recent IEEE MWTL paper introduced a novel design of Q-Choked Sapphire Sandwiched Resonator (Q SSR), utilising Q-Chokes. Q SSR combines the advantages of the previously proposed Q SCR, where the choke ensures a spectrum of pure TE0np modes, with the dielectric focussing properties of classical SPDRs, which help stabilise resonant frequencies and field. Thereby, Q-SSR facilitates accurate and non-destructive measurements of dielectric samples at several nearly-equally-spaced and easily-identifiable modes. Its prototype in [1] was validated with samples from the industrial benchmarking. Q-SSR is well suited for materials' assessment in the FR3 band, a key spectrum layer for the deployment of 5G-A and 6G communications.

The development of new radar systems operating at sub-THz and THz frequencies with high-resolution imaging has led to new applications and enabled remote sensing for short-range applications, such as non-destructive testing or surface-roughness analysis. Furthermore, these remote-sensing systems open an opportunity to study downscaled real-life scenarios that involve long-range remote sensing in a closed laboratory environment.

In this paper, we introduce a method for studying the radar signatures of objects sensed in the far-field region using X-band radars, employing a scaled sub-THz 2D inverse synthetic aperture radar (ISAR) system that performs simultaneous monostatic and bistatic radar imaging. The method is experimentally studied via a VNA-based radar system operating at D-band with application to a downscaled model of a 3.3 m-long unmanned aerial vehicle (UAV). The results demonstrate that the proposed method provides insight into the radar signatures of physically large targets in the laboratory environment. Furthermore, the results demonstrate that monostatic and bistatic imaging provide complementary information on the radar signature of the object under test.

Drones play an important role in modern society due to their wide range of civilian and industrial applications. However, they also pose potential risks, as they can be misused to carry threatening materials and compromise security systems. Small drones, often less than one meter in size, are particularly difficult to detect and classify. Existing approaches struggle to reliably distinguish between drones and birds. Deep learning approaches can be effective in analyzing the structure and flight patterns of flying objects. In this research, an open-source dataset containing information on drones (DJI Mavic, DJI Phantom) and a bionic bird is utilized for classification. The dataset is used to train deep learning model with different classifiers, including Residual-ConvNet, VGG, and Inception. The model performance is evaluated using confusion matrices, which demonstrate the classification capabilities. Among the evaluated classifiers, Residual-ConvNet achieved the highest balanced accuracy of 0.98 and an F1 score of 0.97. Therefore, Residual-ConvNet is suggested for radar-based flying object classification

Wi-Fi orthogonal frequency-division multiplexing (OFDM) transmissions and receiver-estimated channel state information (CSI) enable multistatic Doppler sensing in complex indoor environments. Accurate device-free 3D localization inside tightly confined and multipath-rich platforms such as passenger buses remains challenging due to finite coherent observation windows, strong reflections, and the coupling between target position and velocity in Doppler-centric measurements. This paper proposes a bistatic Doppler localization framework that explicitly decouples velocity and position estimation to improve robustness under limited observation windows. A high-fidelity RF digital twin of a full-scale city bus is constructed in Blender and integrated with the SensingSP engine to synthesize OFDM-consistent Wi-Fi CSI with per-path delays, complex gains, and Doppler shifts. Six spatially distributed receivers observe a common transmitter and provide Doppler measurements that are fused via a two-stage weighted least-squares and Gauss-Newton estimator with residual- and conditioning-based quality control. Simulation results demonstrate sub-meter 3D localization under modeled Doppler estimation uncertainty, achieving median position errors below 0.30 m for an assumed Doppler standard deviation of 0.2 Hz and remaining below 0.45 m at 0.4 Hz. The proposed framework provides a physics-based, geometry-consistent testbed and processing pipeline for multistatic Doppler localization using commodity OFDM waveforms, supporting emerging integrated sensing and communication (ISAC) scenarios.

The emergence of new radar systems operating in the terahertz frequency range facilitates new applications such as material defect detection, material characterization, hyper-accuracy localization, and scattering analysis of rough surfaces. To locate defects inside the material, the synthetic aperture radar (SAR) principle can be realized using a radar system, and localization of defects can be implemented in SAR images. Since a material typically exhibits a non-unity refractive index, neglecting this factor during SAR image formation can cause smearing and displacement of internal defects in the reconstructed image. These effects become particularly critical for materials with high refractive indices or when defects are located deep within the medium. In this paper, we present a wave propagation model inside a non-unity refractive index material, and then the model is utilized to develop a backprojection algorithm for SAR imaging inside a non-unity refractive index material. The simulation results are provided to show the effects of non-unity refractive indices on SAR images, whereas the experimental results help us to verify the proposed wave propagation model in practice. The experiments are based on an SAR testbed based on a vector network analyzer operating in the frequency range 220$-$330 GHz and an electrical insulator with internal damages.

This work focuses on the comparison of the performance of a low-cost Time-of-Flight (ToF) sensor with a Frequency Modulated Continuous Wave (FMCW) radar in the context of applications such as human movement observation. To demonstrate the capabilities of the ToF sensor, a radar-like application is proposed, which produces a range-velocity map, a typical processing output of actual radar data. The experiment with joint radar and ToF measurements of human target movement shows that, in the simplest functionalities, both sensors produce similar results. In particular, simultaneous measurement of target distance and velocity using ToF is consistent with radar-based reference, using significantly less computational effort compared to radar data processing. The conducted experiment suggests that under certain conditions and for simple applications, a ToF sensor can be a considerable substitute for an order of magnitude more expensive radar system.

This paper presents a semi-additive manufacturing process for broadband conical inductors with high design freedom. Several conical inductor designs are investigated, fabricated, and measured in a shunt configuration. The conical inductors operate over a frequency range up to 125 GHz. The deviation between measured and simulated inductance does not exceed 18.5%, while inductance values of up to 9nH are demonstrated. The maximum DC resistance per unit wire length remains below 0.021 Ω/mm. This work demonstrates the potential of semi-additive manufacturing for inductors, especially for biasing applications. It stands out due to its excellent high-frequency performance, easy assembly process, and low DC resistance, resulting in a higher quality factor and improved current carrying capability compared to previous reported additively manufactured inductors. Furthermore, the developed approach provides high design flexibility and tailoring of the inductor properties to specific application requirements.

Antenna miniaturization is a significant challenge, particularly when aiming to reduce physical size without compromising radiation efficiency. A promising approach to overcome these limitations involves the use of magneto-dielectric materials, such as hexaferrite, which exhibit both high permittivity and permeability to effectively reduce the guided wavelength. In this study, we employ Stereolithography (SLA) as an advanced additive manufacturing method to fabricate functional composites based on UV- curable resins with addition of ferrite particles. Unlike conventional subtractive techniques, SLA allows to produce high-resolution objects with minimal material waste and high dimensional accuracy. Test samples, containing fractions of barium hexaferrite (BaFe12O19) up to 10 wt.% were fabricated and characterized over a broad frequency range using coaxial transmission line fixture. The results indicate that for a low filler concentration, the interaction between the electromagnetic field and the magnetic particles is insufficient, resulting only in slight changes in the permittivity while permeability remains unity. This research analyzes the relationship between the filler loading and the resulting electromagnetic properties, concluding that while SLA is viable for low-load composites, alternative additive manufacturing methods are required to achieve high-loading fractions necessary for the next- generation antenna substrates.

We developed a WRD750 double ridge waveguide to handle power of 100 W and higher, using additive manufacturing with plastic plating technology, in the frequency band of 7.5 to 18 GHz. The additive manufacturing process is based on a laser sintering technology.

In this paper, we demonstrate operation of a 5-cm-long WRD750 double ridge waveguide. The waveguide exhibits an insertion loss of 0.1 dB and a return loss of 33.2 dB at a center frequency of 12 GHz, and high-power transmission availability using a 200 W continuous wave. The internal temperature was 41.4◦ C during successive 90 minutes periods.

This paper presents the differences between a conventional 3D printed plano-convex and a dual-lens design optimized for the 60 GHz antenna-on-chip pulsed coherent radar system A121 from Acconeer. The lens geometry is generated using an algorithm based on the principles of geometrical optics and is specifically optimized for the chosen orientation, where the radar system faces the planar side of the lens. Both configurations are simulated and experimentally evaluated. In addition, a distance variation measurement is conducted to investigate the effect of the spacing between the radar system and the lens on the beamforming performance.

The topic of sustainability is of great importance in current discussions, with PLA (Polylactic Acid) frequently used as a raw material in 3D printing due to its biodegradability. However, during 3D printing process, there could be a great deal of failed prints. It is possible to recycle and reprocess them to produce the new usable PLA filament for 3D printing. This study presents a transmission/reflection (TR) method by using the shims which construct a small rectangular waveguide to measure the permeability and permittivity of the samples printed with normal PLA filament and recycled reproduced PLA filament. Their outcomes will be compared to evaluate whether the permeability and permittivity using the processed PLA filament are changed.

This paper presents four GaAs-based voltage-controlled oscillator (VCO) designs operating in the C-band frequency range, although each topology has different frequency bands, they all commonly cover the 5.73-5.9 GHz range, optimized for low phase noise, high figure-of-merit (FoM), and competitive output power. The designs are employed the same GaAs pHEMT active device under identical conditions. Each topology differs in inductor design or VCO core number and their performances are evaluated in terms of phase noise, tuning range, power efficiency, and area. Harmonic balance simulations show that the proposed VCOs achieve phase noise levels below up to -136 dBc/Hz and FoM values that are superior to state-of-the-art results reported in the literature. The output power levels are between 3.1-18.3 mW and also sufficient for C-band applications. The results demonstrate that GaAs technology promises a strong candidate for high-frequency, low-noise VCO design, while the comparative analysis provides insights into the advantages and trade-offs among different oscillator topologies.

This contribution presents a 5-V 1.8-W GaAs HBT Doherty power amplifier MMIC conceived for Wi-Fi 6E applications in the 6-7 GHz band. The paper discusses the solutions adopted, at transistor and amplifier level, to deal with the challenges posed by the demanding design frequency, which is at the limit of the commercially available technologies. The proposed amplifier achieves more than 23 % efficiency in a 5-dB back-off range, as well as good linearity performance when tested with wideband Wi-Fi 6E modulated signals.

We report an S-band internally matched GaN HEMT power amplifier which is designed to deliver more than 155 watt of RF output power. In order to reduce the parasitics of device, single layer capacitor having high dielectric constant were used in both input and output. Here tuned input and output matching circuits are implemented using alumina substrate to reduce overall size. The entire IMFET is placed into a compact package and coupling capacitor along with bias circuit is implemented on Taconic pcb. The source and load impedances are optimized at fundamental frequency to operate in S band. Under pulse condition, the pulsed power GaN PA delivers high drain efficiency along with an average gain more than 11 dB. The fractional bandwidth of this PA is estimated to be more than 13%. The developed GaN power amplifier is planned to be used in TR modules of phased array Radar's.

This paper presents a theoretical analysis of the load modulation conditions in a Load Modulated Balanced Amplifier architecture as a function of the type of hybrid coupler employed. In particular, two cases are examined: the 90-degree hybrid coupler and the 180 degree hybrid coupler. Furthermore, the feasibility of controlling the load impedance at higher harmonic frequencies is also investigated.

A 0 to 1 watt (30 dBm) software adjustable class-AB radio frequency (RF) power amplifier (PA) operating in the very high frequency (VHF) band is presented with closed-loop power control. The system uses a directional coupler and an RMS (Root-Mean Squared) logarithmic power detector, in a feedback configuration, to adjust the RF output power by step sizes of approximately 0.5 dB.

This paper presents a radar centric multi sensor perception framework for real time collision risk assessment that integrates motion cues from Frequency Modulated Continuous Wave (FMCW) automotive radar with camera based deep learning detections and Light Detection and Ranging (LiDAR) based spatial ranging. The proposed approach explicitly uses Doppler derived radial velocity for detection verification and association stabilization through lightweight motion coherent tracking, instead of relying only on camera and LiDAR alignment and static range consistency. In order to isolate the contribution of motion information, a camera and LiDAR baseline is comparatively evaluated against a radar assisted configuration under the same experimental setup. Range Doppler representations and tracking outputs show clear separation of moving targets from static components near the zero Doppler region, confirming reliable extraction of range and radial velocity through two stage Fast Fourier Transform (FFT) processing. The results indicate improved temporal consistency and association robustness without a meaningful increase in computational load on an embedded platform.

Frequency modulated continuous wave (FMCW) radar represents the current state-of-the-art in automotive radar systems, while orthogonal frequency-division multiplexing (OFDM) radar is regarded as a promising candidate for next generation systems due to its potential for joint sensing and communication. This work compares the performance of FMCW and OFDM radars within realistic traffic scenarios characterized by mutual interference from other radar-equipped vehicles. The analysis investigates a frequency-hopping interference mitigation technique within the allocated spectrum bands of 77-80 GHz and 140-143 GHz. Monte Carlo simulations are used to evaluate the performance of a front-mounted automotive radar in a six- lane highway scenario under both high- and low-density traffic conditions. The simulation framework integrates a radar receiver software tool with an open-source traffic simulator. The results indicate that, neglecting adverse weather effects, the 77 GHz and 140 GHz bands provide comparable performance, and that both systems benefit significantly from random frame-by-frame carrier frequency hopping.

Interference among frequency modulated continuous wave radars severely degrades target detection in automotive scenarios. This paper presents a lightweight U-Net based deep learning framework for radar interference mitigation that jointly performs interference detection and signal reconstruction. The network processes a dual-channel input comprising the corrupted radar signal and an initial interference mask, and outputs both a cleaned signal and a refined mask. Evaluation across 300 simulated frames demonstrates that the proposed method achieves the highest suppression of interference-induced noise floor among all tested algorithms, while maintaining an average peak strength loss below 1 dB and under 1° mean phase distortion.

A wide deployment of radars, such as a surveillance radar network and many vehicles equipped with automotive radars in traffic, will make the signal environment very challenging. If the radars are active simultaneously, one radar will interfere with other radars. In a critical case, two radars having similar parameters interfere with each other and the separation between them is double to the range to a target. In this case, the target can totally be hidden by the relatively strong interference. In this paper, an interference suppression method is introduced for this type of interference. The method is based on plane rotation in the range-time domain and interference detection and removal in the range-Doppler domain. The experiment shows the feasibility of the method.

This paper presents an approach to environmental perception for Advanced Driver Assistance Systems (ADAS) and autonomous vehicles by classifying road traffic participants using a combination of radar micro-Doppler signatures and RGB camera imagery. The primary contribution involves the development and validation of a deep learning-based classification framework integrated with a specialized digital signal processing (DSP) pipeline. The methodology evaluates deep convolutional neural network (CNN) architectures, including MobileNetV2 and EfficientNetB0, within a unimodal and multimodal context. Two distinct fusion strategies are investigated: Early Fusion, which concatenates RGB data with radar spectrograms into a four-channel input tensor, and Late Fusion, which leverages independent deep network backbones in a parallel dual-stream architecture where deep features are extracted independently from each modality before integration. Furthermore, the study explores autoencoder-based feature extraction, demonstrating that a Supervised Encoder Classifier (AEC) significantly outperforms unsupervised approaches. Experimental results indicate that for lightweight architectures, the proposed AEC-based models achieve superior classification accuracy compared to baseline CNNs. These findings highlight the efficacy of the proposed methodology for real-time deployment in resource-constrained automotive edge-computing environments.

Wireless systems have become present in every area of our lives. Proper management of radio resource access can be achieved using cognitive radio (CR) technology. However, despite its many advantages, CR may face challenges in effective sensing methods. In spectrum monitoring, an accurate channel-occupancy algorithm is essential for decision-making in the dynamically changing landscape of wireless communication. Three definite methods for spectrum sensing that do not require any a priori information about the signals are: the classical energy detection algorithm, the eigenvalues method based on the ratio of the maximal to the minimal eigenvalues, and the covariance absolute value (CAV) method, which compares elements of the covariance matrix. This paper reviews the simulation implementation of spectrum-monitoring methods to analyze the behavior of CR under various conditions and scenarios, with a UAV swarm.

Counter Unmanned Aerial Systems (C-UAS) is an emerging area of active research and development especially since the Ukraine war. A parallel stream of research and development is also taking place addressing the defensive survivability of friendly unmanned platforms against hostile C-UAS threats. While extensive literature addresses offensive C-UAS capabilities, the complementary challenge of defending autonomous platforms remains profoundly underdeveloped. We call this research and innovation direction Counter-Counter Unmanned Aerial Systems (C²-UAS). In this paper we aspects and directions in radio C²-UAS domains. For RF systems, we address radar cross-section reduction through geometric shaping and advanced materials, phase centre manipulation, communications link resilience against jamming through anti-jamming techniques and low probability of intercept/detection (LPI/LPD) waveform design, and navigation resilience through GPS anti-spoofing mechanisms and multi-constellation GNSS redundancy. We identify critical research gaps, emerging technologies, and future research directions. This work establishes the theoretical and practical foundations for C²-UAS as a distinct research discipline with profound strategic importance for military, civil, and commercial autonomous systems operators.

Clutter poses a significant challenge to radar system performance, impacting measurement accuracy and operational efficiency. Rural clutter channels, characterized by complex and dynamic amplitude, Doppler, and phase variations, are particularly difficult to model and simulate. This study addresses these challenges by analyzing transfer functions derived from real X-band radar measurements collected in diverse rural environments, including forested and vegetated areas, with the aim of developing realistic synthetic data for enhanced radar simulation. Two generative models, Generative Adversarial Networks (GANs) and Denoising Diffusion Probabilistic Models (DDPMs), are developed to generate realistic rural clutter channels. To evaluate the models, a 1D Convolutional Neural Network classifier is used to assess the accuracy of the generated samples, alongside a statistical analysis comparing synthetic data to the original measurements. The results demonstrate that both models effectively replicate the key statistical properties of the measured data. These findings highlight the potential of generative models to enhance radar system simulations, providing a foundation for further research into clutter modeling and mitigation.

This paper presents a method for determining the radar cross section (RCS) of simple shapes coated with a thin layer of an electromagnetic radar absorbing material (RAM) using electromagnetic simulations based on material parameters derived from measurements. A frequency-dependent description of the complex permittivity of a nonmagnetic RAM material was prepared based on measurements performed with a coaxial probe using the DAK-TL-P system and a vector network analyzer (VNA) in the frequency range from 10 MHz to 67 GHz. The obtained characteristics were incorporated into electromagnetic models using the method of moments (MoM) and applied to RCS calculations for selected conducting structures. Two simple geometries were analyzed, considered as perfectly conducting (PEC) objects and as surfaces coated with a thin RAM layer. The paper discusses the influence of the assumed description of complex permittivity dispersion, interpreted in terms of Debye-type relaxation mechanisms, on RCS calculations in the limited microwave range. The results indicate the important role of consistency between the measured data and the applied material model in electromagnetic scattering simulations.

The article describes the potential development of a next generation of electronic protection resources in the military domain for modern and future electronic warfare. The need for active and passive electronic protection systems for combat platforms' self or escorted protection against threats from hostile electronic RF sensing systems. Modern electronic warfare requires the deployment of various means and measures to effectively protect combat assets, employing various technologies, techniques, and methods.

All-metal leaky-wave antennas represent an appealing class of radiators for millimetre-wave applications due to their low profile, high power handling, robustness, and reduced losses compared with dielectric-based solutions. Their radiation characteristics are strongly governed by the dispersion of the supported leaky modes, making accurate modal analysis a key design requirement, particularly for controlling broadside radiation and mitigating the open stopband phenomenon. In this context, rigorous full-wave formulations based on the Method of Moments (MoM) provide an effective and physically insightful tool for the analysis of periodic open waveguides with complex propagation constants. Such approaches enable systematic modification of the unit-cell geometry to tailor modal dispersion and suppress undesired open stopband, thereby improving radiation efficiency and beam continuity across broadside.

In this study, electromagnetic design and optimization of a multilayered radar absorbing structure (RAS) are done with the aid of analytically derived equations for total reflection coefficient from N-layered planar lossy media, which is illuminated by horizontally polarized plane waves at different frequencies and incident angles. Layer thicknesses, layer numbers and layer orders of the multilayered RAS are determined for a design case with 2-8 GHz frequency range and 60°-90° incident angle, by using optimization tools such as genetic algorithm with appropriate object functions for both wide frequency bandwidth and angle range of incidence. In order to observe effectiveness of the designed RAS, it is curved and bended onto an electrically large aircraft simulation model, and it is shown to reduce radar cross section of the aircraft by an expected maximum level of around 11 dB, at the central frequency of 5 GHz.

The finite element method (FEM) is a well-established numerical technique for solving electromagnetic problems involving complex geometries and material distributions. In frequency-domain analysis, FEM requires solving large systems of linear equations, usually at many frequency points, resulting in high computational cost, particularly for broadband simulations. Model order reduction (MOR) techniques offer an effective approach to speeding up such analyzes by constructing reduced-order models that preserve the essential input--output characteristics of the original system. This paper concerns the selection of optimal parameters of the hybrid MOR technique, which combines reduced basis method and moment matching approaches.

When complex permittivity is treated as a design variable in CAD of microwave systems, synthesizing materials with prescribed dielectric properties becomes necessary. This contribution proves, via a modification of Carathéodory's Theorem, that at most three constituent materials are ever needed to achieve a specified complex permittivity at minimum cost when using power-law mixture formulas. An open-source convex optimization routine is described, and its outputs agree with published experimental data.

The patch coupler with the metamaterial AMC cover has been designed and measured. The presented coupler has been designed for use in K-band microwave multi-port networks. The use of metamaterials in the coupler cover allowed to build a low-profile structure resistant to external radiation, and to improve the transmission characteristics of the system. The results of the measurements confirmed an improvement of about 0.5dB in the considered frequency range from 18 to 24 GHz. In addition, the proposed solution has been compared with variants of the coupler without and with the PEC cover.

Radio frequency (RF) power amplifiers (PAs) are key components in almost all wireless transmitters, ranging from tiny sensors and mobile devices to high-power cellular base stations. Their essential function is to convert DC power into amplified RF signals, ensuring reliable wireless transmission. With 6G networks expected to support ultra-high-speed communications, massive IoT connectivity, and immersive user experiences, PAs must overcome critical challenges in energy efficiency, broadband operation, and linearity, particularly under high peak-to-average power ratios and dynamic traffic conditions. This talk discusses the major obstacles and emerging solutions in next-generation PA design. It will cover advanced PA architectures, novel device technologies such as GaN MMICs, operation in new frequency ranges (e.g., FR3), and the application of AI-driven optimization techniques. Approaches for linearity enhancement, tighter system integration, and cross-layer codesign will also be highlighted.

In this paper, a broadband integrated GaN Doherty power amplifier (PA) module for 5G massive MIMO base stations is reported. Benefiting from the dual-driver Doherty structure and the optimal biasing scheme, the lineup efficiency has been boosted without compromising the linearity. The adopted passive networks, including the combiner, interstage and input matching network, have been developed for broadband operation and eventually have been miniaturized into 12×8 mm² laminate. The PA module features compact dimensions and lightweight characteristics for the 5G base station scenario. It achieves an efficiency higher than 45.8% and a gain of more than 31 dB across the 3.6 - 4.0 GHz band under a 100 - MHz LTE signal with 8 - W average output power. Linearized adjacent channel power ratio of -54.5 and -50.5 dBc, is achieved through digital pre-distortion for 20 - and 100 - MHz signals, respectively.

This paper presents an application of non-uniform transmission lines (NUTLs) in the matching networks of a broadband high-power GaN-based amplifier. The S-band 10W driver-amplifier with GaN HEMT (CG2H40010F MACOM) was designed for use in transmit/receive (T/R) modules for active electronically scanned arrays (AESA) radar systems. Therefore, non-uniform microstrip line sections are shaped for maximum output power and efficiency (PAE) while minimising changes in transmittance during the RF pulse. Measured results demonstrate stable broadband performances from 3 GHz to 4 GHz, including an output power of 41 dBm ± 1 dB at 1 dB saturation, a small-signal gain of 16.3 dB ± 0.8 dB and a power-added efficiency exceeding 55% in CW operation.

This study presents the introduction of a Class-F amplifier operating at X-band, employing a GaN HEMT. Alumina thin film substrate was utilized in the design of matching networks. The final configuration of the amplifier is housed in a metal enclosure with connectors, forming a hybrid structure that incorporates a bare die GaN HEMT, bonding wires, and thin-film-based input and output networks. The design process considered the impact of the transistor's intrinsic parasitics and the effects of bonding wires to ensure Class-F waveforms at the output. The measurement results demonstrate an output power of 39.8 dBm, 57.4 % power-added efficiency (PAE), and a gain of 13.9 dB at a frequency of 7.76 GHz.

In this paper, we show that the fractional curl operator is essentially a rotation operation of components of the complex electromagnetic field formulated based on the Riemann-Silberstein vector. However, modifying the fractional curl definition allows one to introduce the fractional power of the helicity operator, which is a single-qubit phase-shift gate in quantum computing.

We extend the Schrodingerized framework for Maxwell equations to include both electric and magnetic sources within a quantum-circuit implementation. Starting from the dual-symmetric form of Maxwell equations, the fields are discretized on a Yee lattice and the resulting non-homogeneous system is converted into a homogeneous operator through operator homogenization. Warped phase lifting is then applied to embed the dynamics into a Hermitian Schrodinger evolution suitable for Hamiltonian simulation. The proposed formulation introduces an extended source-injection operator allowing simultaneous encoding of electric and magnetic currents and charges within a unified quantum circuit without modifying the underlying unitary evolution scheme. This enables the quantum simulation of electromagnetic boundary-value problems involving dual-symmetric responses, such as metasurfaces and reconfigurable intelligent surfaces, which cannot be represented within single-source Schrodingerized formulations.

In this work, a bi-stage procedure for post-processing of antenna far-field measurements performed in uncontrolled conditions is presented. The approach involves sequential time-domain refinement of a few spatially separated experiments using a Morlet-based algorithm followed by synchronization and aggregation of the experiments using a multi-setup routine. The results indicate that correction of responses using the proposed approach improves the fidelity of antenna responses by up to 12 dB w.r.t. tests in anechoic chamber. Ablation studies of the algorithm are also performed.

Tunable rectifier and detector front-ends must select tuning states to maximize rectified output or achieve a target output voltage, but exhaustive tuning atlases scale poorly with the number of adjustable parameters. This paper presents a measurement-efficient, per-band approach that treats the matching-plus-rectifier chain as a black box and trains a low-order response-surface surrogate model (RSM) from a compact face-centered central composite design (CCD) data set. The surrogate enables the prediction of band-specific output voltage extrema and inverse ``hit-the-target'' tuning to a specified output voltage. Hardware measurements on a varactor-tuned rectifier at 700/800/900~MHz demonstrate atlas-comparable target-hitting accuracy using substantially fewer measurements, reducing reliance on dense tuning atlases for low-power rectifier front-ends.

In this work, the uncertainty and bias of antenna measurements performed in uncontrolled conditions are evaluated for two configurations of the portable test system (featuring different distance between the reference antenna and the structure under test). At each position, the data points are acquired at five independent frequencies from ten consecutive experiments and post-processed using a Morlet-based method. The obtained results indicate that the utilized correction mechanism noticeably reduces both bias and uncertainty of responses w.r.t. measurements within an anechoic chamber.

Unmanned aerial platforms (UAPs) have emerged as a flexible alternative to fixed infrastructure for supporting wire- less sensing and vehicular applications, offering improved line- of-sight conditions and dynamic deployment capabilities. This work investigates an aerial platform-enabled bistatic multiple-in multiple-out (MIMO) radar architecture applied to Vehicular Ad Hoc Networks (VANETs), where an airborne node operates as an aerial Roadside Unit (RSU) to support sensing and localization of ground vehicles. By exploiting the inherent spatial separation between transmitter and receiver in bistatic configurations, the proposed framework enhances geometric diversity compared to conventional monostatic and terrestrial RSU-based sensing approaches. A signal model for the aerial-assisted bistatic MIMO radar scenario is formulated, and tensor-based signal processing techniques are employed to jointly estimate the direction of departure (DoD) and direction of arrival (DoA) of vehicle targets. The multidimensional structure of the received signals is naturally captured using tensor representations, enabling efficient parameter estimation with automatic angle pairing. Numerical simulations are conducted to assess the impact of the aerial RSU on angular estimation performance under different signal-to-noise ratio conditions. The results indicate that the proposed aerial platform-enabled bistatic MIMO radar frame- work achieves angular estimation performance comparable to that of fixed infrastructure while providing increased flexibility and adaptability, highlighting its potential as a complementary sensing solution for future intelligent transportation systems.

Spaceborne Synthetic Aperture Radar (SAR) is susceptible to interference from Artificial Modulation Targets (AMT) in complex electromagnetic environments, which severely affects imaging quality and interpretation performance. To address this issue, this paper proposes an AMT suppression method based on the Pearl Beaded Curtains Neural Network (PBCNN). Building upon the earlier Complex Signal Codec Neural Network (CSCNN), this method introduces a multi-layer residual structure, enhancing the capability to extract AMT signals with low-rank and sparse features, thereby effectively improving the reconstruction accuracy and cancellation performance. This paper constructs a large-scale spaceborne SAR echo dataset that includes multiple satellite platforms, diverse scenarios, and various modulation type, and validates the robustness and generalization capability of the proposed method under different intensities, locations, and modulation types through a Hardware-In-Loop (HIL) experimental system. Experimental results show that when combined with the Composite Signal Reconstruction Precision (CSRP) loss function, PBCNN improves the comprehensive evaluation metric by approximately 5% compared to the earlier CSCNN method in AMT suppression and scene restoration tasks, outperforming existing methods. This study provides an effective neural network solution for reliable imaging of spaceborne SAR in AMT interference environments.

This paper presents SignalTrack, a modular exper- imental platform for distributed and mobile wireless systems based on software-defined radio (SDR). The platform combines multiple SDR-equipped mobile units with a centralized controller responsible for coordination, synchronization, and experiment orchestration. A message-based publish-subscribe communica- tion model enables dynamic configuration, scalable integration of nodes, and algorithm-agnostic operation. The proposed ar- chitecture is designed to support experiments in distributed beamforming, adaptive transmission, and integrated sensing and communication scenarios under realistic conditions.

This paper presents a low-cost wireless testbed enhanced with a software defined radio (SDR) for experimental evaluation of interference and connectivity in the 2.4 GHz band. The proposed platform extends a previously introduced Bluetooth-based testbed by integrating software-defined radio capabilities using HackRF One and the Opera Cake RF switching module, enabling controlled, repeatable, and direction-dependent interference generation. The testbed supports received signal strength, noise-floor-estimation and derived signal-to-noise (SNR) and signal-to-interference-plus-noise ratio (SINR) metrics, and is operated through an automated command-line interface to ensure reproducibility. Validation experiments conducted with an Electronically Steerable Parasitic Array Radiator (ESPAR) antenna demonstrate the correct operation of the extended testbed and its ability to capture direction-dependent interference effects. The results confirm that appropriate antenna pattern selection can significantly mitigate interference and improve communication reliability. The proposed platform provides a scalable and cost-effective tool for benchmarking adaptive antennas and interference mitigation methods under realistic conditions.

A 5-bit switched-line phase shifter intended for ground-based IFF ESA antenna systems is presented. The proposed design operates in a shared transmit/receive path. The phase shifter employs a non-standard three-stage architecture providing 32 discrete phase states with a least significant bit of 11.25°. Each delay line incorporates a Non-Uniform Transmission Line (NUTL), resulting in a compact layout and enhanced phase stability. Switching function is realized using high-power GaN-based RF switches. The phase shifter handles power levels up to 250 W under pulsed operation, with a phase error lower than ±2° and total insertion loss below 3 dB at the IFF system operating frequencies. Fine phase adjustment is enabled by trimmer capacitors integrated into delay lines, allowing compensation of fabrication tolerances. The design is implemented in microstrip technology on a Rogers RT/duroid 6010LM substrate.

In 5G transmitters, Doherty power amplifiers (DPAs) are key to achieving high energy efficiency. In emerging pulsed operation schemes, the DPA is enabled only during active transmission slots within the frame structure, thereby significantly reduces average power consumption. This paper presents the effect of such frame-gated operation on the performance of digital predistortion (DPD) based on the Memory Polynomial (MP) model. Measurement results show that reducing the on-time can improve drain efficiency by approximately 60 %, while simultaneously degrading the in-band error vector magnitude (EVM) and adjacent channel power ratio (ACPR) for 64-QAM signals. These results highlight a fundamental efficiency-linearity trade off in pulsed Doherty operation and indicate that conventional MP DPD, designed for continuous operation, only partially compensates memory effects induced by frame-synchronized gating. Such frame level gating supports 5G Network Energy Saving (NES) goals by reducing base station power consumption during low traffic periods.

This paper presents developed power supply unit for compact radar system components. It may be advantageous for use in applications where low noise, high current capability, multi-channel output, modular construction with stacking capabilities, low weight and profile are of concern. It may be especially profitable in Frequency Modulated Continuous Wave small size mobile radar system setups, eg. hand carried or installed on aircraft or drone. Described circuitry has been designed, realized and tested with radar systems developed over recent years at our laboratory. Noise and interference reduction in Range-Doppler map obtained from example radar system is presented.

Vegetation-induced attenuation and temporal dispersion remain important impairments in millimeter-wave radio propagation, particularly in the frequency range 2 (FR2) band, where highly directional links are employed. Existing vegetation attenuation models, including ITU-R P.833, predominantly address scenarios involving trees or vegetation clusters and are typically derived for the main radiation direction of the transmitting antenna. This paper presents an experimental study of angular-dependent attenuation and temporal dispersion introduced by a single deciduous bush under winter conditions. Measurements were conducted on a university campus at 28 GHz and 38 GHz using a wideband channel-sounding testbed operating from 1 MHz to 44 GHz. The bush, approximately 3 m in height, was located at the center of a circular measurement trajectory with a radius of 10 m, along which the receiver was positioned at angular intervals of approximately 8 deg. Directional antennas with 7-9 deg half-power beamwidth were employed, and reception angles from 0 to 200 deg relative to the transmitter-bush axis were investigated. The results demonstrate significant angular variability in both excess attenuation and temporal dispersion, indicating substantial contributions from scattering outside the main propagation direction. Differences between the two carrier frequencies are analyzed, highlighting frequency-dependent attenuation mechanisms even in leaf-off conditions. The presented findings extend existing single-vegetation models by incorporating angular characteristics and temporal effects, providing valuable insights for the design and modeling of FR2 wireless links in sparse-vegetation environments.

This paper presents the work on sensing vital signs using electromagnetic imaging. The spatial response of the pulsing vital sign modelled with a dynamic torso phantom has been measured over 1 - 2.5 GHz frequency range with ten Vivaldi antenna using a Vector Network Analyser. The dynamic phantom composes of two pulsing targets emulating the heart and the lungs in the human body. The frequency and the location of the pulsing targets can be sensed and enables more medical applications using Electromagnetic Imaging,

Achieving low loss in high-GHz and mmWave interconnects increasingly depends on the real microwave performance of copper foils [1,2], not just their DC conductivity or nominal roughness class. In practice, adhesion driven surface treatments that improve lamination reliability can also raise conductor loss as the skin depth shrinks and current concentrates near the topography [3]. To support fast material screening and process optimization, this work presents a dedicated measurement workflow [4] that quantifies conductor loss on copper foils without building microstrip or stripline and without substrate dependent uncertainty. The method based on dielectric resonator method uses a configured to minimize dielectric and parasitic losses so that the measured dissipation is dominated by the tested conductive layers [5]. A cylindrical cavity with a dielectric resonator inside operates in a dual-mode regime enabling two measurement points from the same fixture (approximately 28 GHz and 44 GHz). The technique targets electrically thick foils (thickness greater than several skin depths), ensuring that the extracted surface resistance and effective conductivity represent high-frequency current flow conditions relevant to real circuits. The measurement fixture is connected to a vector network analyzer (VNA) through low-loss coaxial cables. The VNA records S-parameters around each resonance. Resonance parameters are derived from the measured response to obtain resonant frequency and the Q-factor. A dedicated PC application guides the operator, automates data capture, and converts resonance parameters into surface resistance and effective conductivity for each mode. With stable instrument settings, a single measurement cycle, including sample placement and software capture, is typically completed in under one minute. Copper foils are provided as sheets [6] and prepared into square samples. For each data point sample is mounted to form the top conductive boundaries of the resonator cavity. The samples are pressed flat against the dielectric head using controlled weight to ensure uniform contact and minimize variability from air gaps, folds, or uneven pressure. Measurements are performed on a defined foil side (the electrolyte or drum side) and recorded together with thickness and surface-finish identifiers so that loss metrics can be directly compared across treatments. Measurement procedure delivers substrate-independent, conductor metrics that can be used to rapidly compare foil treatments, monitor batch-to-batch variability, establish acceptance windows, and connect conductor loss trends with independently measured surface descriptors. By concentrating on fast, repeatable resonance measurements on foils, the workflow provides an efficient path from copper-foil manufacturing choices to quantified high-frequency loss performance, aligned with the needs of next-generation RF and mmWave hardware [7,8].

Bistatic configuration is one of the useful arrangements in radar applications. There are many benefits to bistatic geometry, one of which is the capability to intercept stealthy targets and reduce the vulnerability of the receiver side, stemming from the separation of the transmitter and receiver(s). In a network of radars, monostatic, bistatic, and even multistatic configurations can coexist, providing spatial diversity for better, more accurate detection and ranging. A key requirement for bistatic or multistatic configurations is the integrity of the transmitter and receiver signals. Specifically, for a network of FMCW radars, it is challenging to share the Local Oscillator (LO) over long distances due to the millimeter wavelength order. Additionally, other parameters can cause incoherence between independent FMCW radars, including ramp bandwidth, chirp duration, oscillator phase noise, carrier frequency, and phase. In this paper, a study of all these impacting parameters is conducted. An effective and low-cost technique to improve the coherence in the timing profiles of two radars is also proposed.

This paper presents a reconfigurable ultrawideband (UWB) code-modulated microwave radar platform is presented for practical comparison of pseudo-noise (PN) and complementary-sequence (CS) waveforms under identical hardware conditions. After outlining the fundamentals of pseudorandom binary sequences (PRBS) and complementary codes and their role in code-based spread-spectrum modulation, the system architecture and implementation are described. The platform leverages a high-speed digital-to-analog converter (DAC) and UWB Vivaldi-antennas to realize wideband transmission, and its software-defined waveform generation supports rapid switching between different sequence families. Initial range measurement experiments comparing PN and CS operation are presented, and the possible benefit of CS in certain fields are discussed.

Microwave resonator method used for characterisation of battery electrodes provides non-destructive access to surface resistance and effective conductivity, parameters that reflect losses. In this work we focus on single side coated LiCoO₂ deposited on aluminum film and investigate how ion implantation with different gases influences their conductivity at 13.75 GHz and 20.31 GHz. We apply a dual-mode ruby dielectric resonator, combined with automated acquisition and a dedicated post-processing application. During study we find that argon implantation yields the highest conductivity results, outperforming nitrogen and helium, and is therefore the best candidate for further developments of high-quality energy storage device.

Passive localization of airborne objects using radio emissions constitutes an attractive alternative to GNSS-based systems. One method enabling position estimation from radio signals is the Time Difference of Arrival (TDoA) technique, based on multilateration. This paper investigates a low-cost TDoA localization system utilizing 70-centimeter band LoRa (Long Range) transmissions based on Semtech SX1262 modules and a distributed network of synchronized receiver nodes. To obtain precise measurements of signal arrival time, an ScioSense AS6501 time-to-digital converter was employed, together with optional GPS-based receiver synchronization. The system was validated in field experiments using four receivers and a single transmitter. The results indicate that LoRa-based TDoA localization, when combined with precise timing measurements, can significantly outperform standard ground based multilateration methods. In the tested configuration, the achieved two-dimensional localization accuracy was approximately 9 m, demonstrating that meter-level passive localization is feasible using low-cost hardware and standard LoRa transmissions.

The sensor fusion of multi-static radar detections to tracks has been investigated for the case with moving receivers and transmitters. A use case is passive radar with broadcast transmitters and the receiver mounted onto a vehicle or in miniaturized form on drones. To compensate the additional bistatic Doppler shifts, the velocity vectors of the platforms have to be taken into account besides the position with their sensor measurement errors. The incorporation in Kalman filtering is derived in detail. The Doppler shift between transmitter and receiver can be measured, but may be also zeroed out related to the signal processing and frequency synchronization, which must be handled in the tracking.

We present the Turing Synthetic Radar Dataset, a comprehensive dataset to serve both as a benchmark for radar pulse deinterleaving research and as an enabler of new research methods. The dataset addresses the critical problem of separating interleaved radar pulses from multiple unknown emitters for electronic warfare applications and signal intelligence. Our dataset contains a total of 6000 pulse trains over two receiver configurations, totalling over 4 billion pulses, featuring realistic scenarios with up to 110 emitters and significant parameter space overlap. To encourage dataset adoption and establish standardised evaluation procedures, we have launched an accompanying Turing Deinterleaving Challenge, for which models need to associate pulses in interleaved pulse trains to the correct emitter by clustering and maximising metrics such as the V-measure. The Turing Synthetic Radar Dataset is one of the first publicly available, comprehensively simulated pulse train datasets aimed to facilitate sophisticated model development in the electronic warfare community.

Noncontact vital signs monitoring using radar technology offers a promising alternative to wearable sensors and camera-based systems, ensuring both comfort and privacy. This paper explores the capabilities of the 10 GHz ADALM-Phaser, a low-cost phased array system, for the remote detection of respiration and heartbeat. A novel signal processing pipeline is proposed to address hardware-specific constraints, particularly the discontinuity of data acquisition caused by frame-based transmission. The methodology introduces a hybrid heartbeat extraction technique that adaptively fuses the fundamental frequency obtained via standard band-pass filtering with high-frequency cardiac oscillations detected using Short-Time Fourier Transform (STFT). Additionally, an autoregressive prediction model based on the Yule-Walker method is applied to mitigate the impact of signal gaps. The extracted signals are analyzed in the time-frequency domain to estimate respiratory and heart rates. Experimental results demonstrate that the proposed pipeline effectively handles data discontinuity and successfully detects sub-millimeter chest displacements, validating the potential of the ADALM-Phaser for physiological sensing applications.

Signal decomposition is fundamental to digital signal processing, enabling applications from spectral analysis to spatial localization. Classical methods such as FFT and MUSIC, while robust and widely adopted, exhibit inherent limitations: FFT resolution is constrained by the number of samples, while MUSIC requires prior knowledge of source count and eigendecomposition. This paper presents a benchmark evaluation of the Steering-Model, a compact physics-informed neural network (PINN) designed for general signal decomposition. Central to this approach is a learnable complex-valued weight matrix that replaces fixed analytical bases with data-driven representations. Although applicable to signal decomposition broadly, we validate the method through Direction-of-Arrival (DoA) estimation using synthetic MIMO mmWave radar data. Benchmark comparisons against classical DSP and state-of-the-art machine learning methods demonstrate that the Steering-Model achieves high accuracy with 13k parameters, offering an effective trade-off between accuracy and computational efficiency. Keywords-signal decomposition, direction of arrival, physics-informed neural networks, array signal processing, MUSIC algorithm, benchmark.

Automatic characterization of hypersonic-like radar signatures is challenging due to scarce labeled data and the complex spectro-temporal morphology of high-velocity atmo- spheric echoes. Meteor radar observations provide a natural proxy for studying such phenomena, as meteoroid atmospheric entry produces ionized plasma trails that generate distinctive Doppler radar signatures. This work investigates the charac- teristics of radar echoes in both image and signal domains to better understand the information contained in different feature representations. Radar observations from the Belgian RAdio Meteor Stations (BRAMS) network were processed to detect and isolate individual meteoroid and aircraft events in spectrograms using morphology- based detection methods. From each detected object, two groups of descriptors were extracted: morphological features derived from spectrogram contours and statistical features computed from the corresponding signal waveforms. The resulting fea- ture spaces were analyzed using dimensionality reduction and clustering techniques to explore their structure and relation to expert-identified event types. The analysis reveals distinct patterns associated with meteoroid and aircraft echoes and highlights the relevance of spectrogram- based morphological descriptors for capturing key characteristics of these radar signatures. The presented approach provides a framework for exploring radar echo properties and supports the development of automated methods for analyzing high-velocity atmospheric observations in data-constrained environments.

We developed a method to process sequences of range-Doppler maps for the detection of dangerous or violent activities at public places. The detection is performed by applying a type of 3-dimensional convolutional neural network (CNN). This technique was applied within a field study at a train station. Based on the experiences we present further approaches for improving the detection rate and reducing the computational resources with the goal of implementing these approaches in an embedded system. The methods will be applied and tested as components of a mobile system that will be developed within a recently started research project.

Mutual interference among densely deployed automotive 76-81 GHz band FMCW (Frequency-Modulated Continuous Wave) radars elevates the noise floor in range-Doppler (RD) map, thereby degrading target detection performance. Inter-radar interference suppression is a critical yet challenging problem, particularly under non-stationary interference conditions caused by dynamic traffic environments. To address this issue, this paper proposes a dual sliding-window UCB (Upper Confidence Bound) based waveform adaptation method that jointly adjusts the center frequency and a hardware-friendly phase code of the radar waveform to mitigate time-varying interference without prior knowledge of the interference statistics. Specifically, a frequency sliding-window UCB is first employed to adapt the center frequency, and a phase-code sliding-window UCB is subsequently activated when residual interference remains after frequency adaptation. To balance waveform diversity and implementation complexity, a compact phase codebook consisting of eight BPSK code classes is adopted. Extensive simulation results under dynamic interference scenarios demonstrate that the proposed method achieves higher SNR, faster convergence, and lower cumulative regret compared with conventional UCB based baselines.

Deep neural networks are powerful models for radar detection tasks, but their high computational cost limits suitability for real-time inference on CPUs. In this paper, we address this limitation by proposing an approach to automatically search for optimized neural network architectures tailored to airborne target detection. Our method relies on neural architecture search with multi-objective Monte Carlo search that optimizes the trade-off between detection accuracy and latency constrained by arithmetic intensity, guiding the search towards CPU-efficient networks. Experimental results show that the proposed architecture matches the baseline detector's performance while being up to four times faster on CPU.

Global navigation satellite systems (GNSSs) underpin navigation across aviation, transportation, and unmanned platforms. However, in electronic-warfare environments, they may be deliberately degraded or deceived by jamming and spoofing. This paper reports an experimental study of their impact on a GNSS receiver, focusing on time-to-first-fix (TTFF) and the time required to regain a valid fix after an attack. We built a controlled laboratory setup based on Safran Skydel and a Universal Software Radio Peripheral (USRP) X310 software-defined radio (SDR), feeding a wired radio frequency (RF) path with a coupler, 60 dB attenuation, and a direct current (DC) block, disciplined by a 10 MHz rubidium reference. Real-time in-phase/quadrature (IQ) streams for GPS L1 and Galileo E1 were generated and injected under repeatable straight-line motion scenarios, including a dedicated spoofing segment. Using the prepared test-bed, we assess the sensitivity of the selected GNSS receiver to various spoofing variants.

The increasing demand for efficient spectrum utilization in modern wireless and defense systems has accelerated the development of Integrated Sensing and Communication (ISAC) technologies. While extensive research has been conducted on the classification of radar and communication signals individually, and their coexistence in shared spectral environments, no prior study has specifically addressed the classification of ISAC waveforms designed to perform both functions simultaneously. In this study, we investigate the effectiveness of multi-channel input-level time-frequency representations for ISAC waveform recognition. Specifically, short-time Fourier transform (STFT), continuous wavelet transform (CWT) and Wigner-Ville distribution (WVD) images were generated and used as inputs to convolutional neural networks, ShuffleNet and ResNet-18. A multi-channel input-level fusion strategy was employed, where complementary time-frequency representations were mapped to separate RGB channels and jointly processed by a single CNN. Experiments on complex ISAC environments containing LFM, Frank, Legendre, ISAC-LFM, and ISAC-ZC waveforms reveal that ISAC-LFM signals can be reliably classified even under low SNR conditions, while the proposed multi-channel structure significantly improves ISAC-ZC detection performance.

This work presents an enhanced RSSI-based direction-of-arrival (DoA) estimation technique for indoor wire- less systems using a frequency-beam scanning leaky-wave an- tenna (LWA). The inherent high directivity of the LWA enables the application of amplitude-monopulse techniques; however, the resulting angular response exhibits sidelobes and spurious radiation that lead to ambiguity peaks in the monopulse an- gular pseudo-spectrum. To address this limitation, we propose a lightweight ambiguity-suppression strategy that exploits the frequency-dependent beam steering of the LWA and applies a consistency filter across channels to isolate the true DoA. The method significantly improves DoA robustness without requiring multi-port processing, array calibration, or high computational load, in contrast to other signal processing approaches. Simulated radiation patterns are used to generate synthetic RSSI datasets for evaluation. The results show that the proposed technique reduces angular ambiguity errors by more than an order of magnitude and acquires one-degree accuracy under low-SNR conditions, demonstrating its suitability for low-cost ISAC and indoor IoT positioning applications.

Phase-conjugating (PC) cross-eye jamming has been proposed as an electronic countermeasure (ECM) capable of inducing systematic bias in monopulse angle estimates. While prior work established an analytical framework and tolerance analysis, the experimental evidence under realistic hardware constraints remains limited. This paper presents a hardware-in-the-loop (HIL) experimental assessment of PC cross-eye jamming implemented with software-defined radio (SDR) hardware in a controlled guided-wave environment. The jammer is realized with a dual-channel USRP B210 performing digital phase conjugation and enforcing a 175° phase offset on one channel, while a monopulse receiver is implemented using a second B210 SDR. A calibration procedure is used to compensate static inter-channel offsets, and a GPS-disciplined reference clock is used on the jammer to improve LO stability and reduce phase errors. Antenna pattern effects are emulated numerically prior to applying the monopulse processing. Experimental results confirm that the measured monopulse responses follow the trends predicted by the analytical framework under the tested phase and amplitude conditions, which confirms the robustness of the proposed framework.

This paper investigates the detection and classification of radar pulses in the electromagnetic spectrum. For radar pulse recognition, we employ a machine-learning approach based on semantic segmentation of spectrograms. We propose a neural network based on the U-Net architecture, that utilizes Atrous Spatial Pyramid Pooling and Residual Connections and compare the model against a standard U-Net. The results show that a semantic segmentation approach can effectively detect pulses and distinguish between 22 intrapulse modulation classes from densely occupied spectrograms.

The use of fiber-optic links for transporting radio signals in wireless networks is a well-established technology and the convergence of optical and wireless networks continues to evolve. Fiber-optic remoting of radio signals is used in a diversity of wireless networks, including indoor/in-building distributed antenna systems and outdoor cellular networks. Today the capabilities of wireless networks are progressing more rapidly than ever. The proliferation of connected high-capacity smart devices as well as the increase in the number of broadband multi-media services available to the consumer, has led to an escalating demand for wireless access to high-speed data communications. The next generation 5G+/6G network promises to deliver unprecedented data rates to the mobile user and the millimeter-wave and Terahertz frequency regions are being actively pursued for the provision of these services. This talk will describe some of the related challenges and opportunities for radio-over-fiber technologies in emerging Future G wireless systems.

Satellite communication links, including Low Earth Orbit (LEO) are a critical aspect in the 5G/6G landscape for providing mobile communication everywhere and minimizing the geolocational digital divide. Several companies are currently rolling out LEO-based systems to provide high data connectivity in rural and isolated locations throughout the world. This in conjunction with the use of GEO (Geostationary Earth Orbit) and mid-orbit satellite gateways promise to significantly improve global connectivity. It is due to this push to incorporate Non-Terrestrial Networks (NTNs) in our mobile communications world as well as the utilization of multiple bands (to enhance capacity) that we are seeing innovative antenna concepts to potentially provide efficient reconfigurable terminals. Imperative to all the proposed antenna solutions is cost as the application is a mass market. Another important aspect is versatility. These terminals may have to be able to not only communicate with LEO systems, but also GEO platforms and so they need to be agile and will utilize multiple frequency bands. A shared aperture, phased array approach would allow this kind of agility and potentially a low-cost solution. This talk will describe some approaches to develop printed antennas that can help realize high performance shared aperture antennas required for emerging Future G NTNs.

The problem of detecting and resolving UAV swarms using radar systems is considered in this paper. Conventional FMCW radars operating with fixed waveform configurations are limited by the trade-off between range resolution and spatial coverage within the maximum unambiguous range. To address this, an adaptive bandwidth selection approach based on Proximal Policy Optimization (PPO) is proposed within the cognitive radar framework. The radar adjusts its transmitted bandwidth on a per-CPI basis using closed-loop feedback to improve swarm detectability. The approach is validated using a dedicated FMCW radar simulator with realistic target motion and detection modeling. Results show that the learned policy via PPO consistently outperforms fixed-bandwidth baselines and approaches optimal-level performance (i.e., that achievable by access to ground-truth information) across multiple swarm scenarios.

In order to manage the increasing density of flight movements in airspace, an improved target classification can be crucial for Air Traffic Control (ATC). Our earlier research has already shown the benefits of Convolutional Neural Networks (CNNs) for distinguishing targets and clutter, particularly for single and dual beam radars with low spectral resolution. In this contribution, we investigate the benefit of spectral information of a triple beam radar with low spectral resolution for target classification in addition to scalar-valued features. After a brief introduction, we elaborate on the dataset generation and the data labeling strategy as well as on the used CNN architecture. Finally, numerical experiments underline the performance of the proposed network architectures and provide insights about feature importance.

To enable Integrated Communications and Sensing (ICAS) in a peer-to-peer vehicular network, precise synchronization in frequency and phase among the communicating entities is required. In addition, self-driving cars need accurate position estimates of the surrounding vehicles. In this work, we propose a joint, distributed synchronization and localization scheme for a network of communicating entities. Our proposed scheme is mostly signal-agnostic and therefore can be applied to a wide range of possible ICAS signals. We also mitigate the effect of finite sampling frequencies, which otherwise would degrade the synchronization and localization performance severely.

While deep learning thrives in automotive and SAR radar applications, its utility for surveillance pulse doppler radar classification remains underexplored. This paper addresses this gap by introducing an efficient CNN-based framework for classifying airborne targets from single Doppler spectra. We propose a novel split-architecture in which a lightweight, FPGA-ready CNN encoder extracts compact embeddings, minimizing data bandwidth for subsequent CPU-based fusion with traditional radar parameters. To identify suitable design trade-offs, we benchmark encoder backbones based on ResNet, MobileNetV2, and ShuffleNetV2. Furthermore, we investigate the utility of preserving inherent phase information by translating the best-performing real-valued architecture into the complex-valued domain. Experiments on a real-world dataset show that efficient real-valued architectures can match or exceed the performance of larger baselines, with MobileNetV2 achieving the best accuracy-efficiency balance. In contrast, complex-valued networks yield only a marginal performance improvement at substantially increased computational cost, limiting their practicality for resource-constrained deployment.

Radar performance is fundamentally constrained by the characteristics of the local oscillator, from which all timing and frequency stability are derived. While oscillator quality is known to affect clutter behaviour and target detectability, its influence on automatic target classification performance-particularly in bistatic radar networks-has received limited experimental attention. This paper presents an experimental study quantifying the impact of oscillator phase noise on bird-drone classification accuracy using a bistatic L-band staring radar network deployed in an urban environment. Two commercially available GPS-disciplined oscillators of differing performance are evaluated. Bird and multi-rotor consumer UAV spectrogram datasets collected with each oscillator are used to train and test a two-class ImageNet pre-trained AlexNet convolutional neural network via transfer learning. Results demonstrate that lower phase-noise oscillator performance leads to consistently improved classification accuracy, both when used for testing and when used to generate training data, highlighting the importance of oscillator selection in multistatic radar systems performing automatic target recognition.

Optimizing hardware costs and memory footprint represents a critical challenge in frequency-modulated continuous wave (FMCW) chirp radar systems for advanced driver-assistance systems (ADAS) and autonomous driving (AD) applications. Conventional approaches to reduce sampling rates of analog-to-digital converters (ADC) lead to a deterioration of key performance indicators, such as maximal range and range resolution. To overcome these limitations, a novel FMCW modulation scheme is proposed, which involves dividing the chirp sequence into blocks with dynamically increasing bandwidths, so that the lowest bandwidth ensures the required maximal range, while the highest provides the required range resolution. To avoid data loss, the anti-aliasing filter (AAF) cutoff frequency is dynamically increased, allowing aliasing, a phenomenon traditionally considered detrimental. This approach generates ambiguities, which are then effectively resolved through a decision tree methodology. Simulation results demonstrate the benefits of the proposed scheme: it achieves range resolution comparable to high-performance modulations while maintaining the low memory footprint of memory-efficient systems. A hardware proof-of-concept test, utilizing a radar prototype and a radar target generator (RTG), successfully validates these core principles.

Aperture level simultaneous transmit and receive (ALSTAR) technology enables concurrent transmission and reception within a shared digital array at the same frequency, removing or limiting the need for bulky analog isolation components. While typical ALSTAR solutions partition the aperture into adjacent transmit and receive sub-arrays to mitigate mutual coupling, this configuration limits the potential for system miniaturization. This paper presents a compact interleaved array configuration, where transmit and receive elements are arranged in a dense triangular lattice, effectively reducing the overall array size by approximately 50%. A comparative performance analysis between the partitioned and interleaved layouts is presented, which included the use of optimized digital transmit and receive beamforming weights, designed to maximize the Effective Isotropic Isolation (EII). Simulation results demonstrate that, despite the severe mutual coupling present in the interleaved layout, the optimized beamforming successfully suppresses the self-interference noise down to the thermal floor in both configurations, at the cost of a partial reduction in the array gain. Overall, the comparison highlights a favorable trade-off between isolation performance and size, demonstrating the potentials of the interleaved layout in size-constrained applications requiring limited transmitted power and antenna miniaturization.

The increasing congestion of radio frequency spectrum and the evolution of multifunctional radar systems have motivated the integration of radar sensing and communication within a single platform. In Air Traffic Control (ATC) applications, Electronically Scanned Array (ESA) radars provide the flexibility required for such integration. This paper presents a radar-centric analysis of joint radar and communication (JRC) operation in ATC ESA systems, focusing on waveform design, digital beamforming, and performance trade-offs under safety-critical constraints. The presented concepts are discussed in the context of operational ATC radar requirements and European regulatory frameworks. To simplify the problem only one aspect - radar-plane communication is discussed.

Conformal, lightweight radio-frequency (RF) components are critical for flexible antennas and radar-absorbing materials (RAM) in aerospace and wearable applications. We report a novel solvent-free approach combining physical sputtering of single-walled carbon nanohorns (SWCNHs) on poly(vinylidene fluoride) (PVDF) to achieve impedance-tunable RF skins. By controlling sputtering load (0.025-1.0 mg·cm⁻²), we tune sheet resistance across six orders of magnitude (10⁶ → 1 Ω/sq), enabling independent optimization for antenna conductors or broadband absorbers. Electromagnetic characterization reveals εr ≈ 5-15 (Ku-band), tan δe ≈ 0.3-1.0 (GHz), and σac ≈ 0.1-3 S·m⁻¹, outperforming pristine PVDF (εr ≈ 10-11, tan δ ≈ 0.02-0.04) and competing favorably with graphene (σac ≈ 1-20 S·m⁻¹) and MWCNT composites (σac ≈ 0.3-3.5 S·m⁻¹). Proof-of-concept antennas at 2.45 and 5.8 GHz achieved 76-79% radiation efficiency, while broadband absorbers demonstrated −14.7 dB reflection loss @ 12 GHz and −10 dB absorption across 8-15.8 GHz at <3 mm thickness. Mechanical durability (5000 bend cycles) and thermal stability (−10 to +60 °C) confirm practical viability. This work positions PVDF-SWCNH as a scalable, solvent-free alternative to solution-cast composites for next-generation conformal RF skins.

The paper presents an analysis of a cylindrical antenna array with calculations of circular array of tube antennas and simulations of a multi-sector array of patch elements. The parameters calculations of the array operating at 3.5 GHz were performed, including gain, level of radiation patterns intersection of antenna elements and operating bandwidth.

The presence of the ionosphere--an ionized layer of the atmosphere located between 50-1000 km altitude--enables electromagnetic signals to propagate far beyond the horizon, supporting both communication and radar applications. The polarization of an electromagnetic wave reflected from the ionosphere changes because the ionized, magnetized plasma acts as a birefringent medium, differentially delaying and rotating its orthogonal polarization components. Consequently, the reflected signal may undergo substantial polarization distortion, resulting in energy loss when received by a system with a single polarization. This letter introduces a time-frequency approach for monitoring the ionosphere in terms of its polarization characteristics and investigates its influence on the polarization of the received signal. Analysis of real-world radar data reveals a significant and continuous drift in time and frequency within the ionosphere, which negatively impacts signal quality.

This paper presents the design and simulation of a new curved monopole antenna optimized for skywave HF radar applications, with a systematic investigation of the effects of curvature and fixed-section length on antenna performance. The proposed design achieves improved impedance matching, broader bandwidth, and enhanced realized gain compared to a conventional quarter-wavelength monopole at 15 MHz. Parametric analysis shows that fully bending the monopole degrades performance, whereas introducing a straight section and carefully optimizing the curvature enables a 18.5% gain increase and a 400 kHz bandwidth expansion. The single- element design is further extended to a 12-element linear array with 0.45λ spacing (where λ is the wavelength), demonstrating stable embedded-element behavior and improved low-to- moderate elevation gain for skywave over-the-horizon radar operation. At θ = 30°, the proposed array achieves 14.04 dBi compared to 13.11 dBi for the reference array, corresponding to 24% gain enhancement, which is significant in high-power HF radar systems. These results confirm that the proposed curved monopole antenna provides a compact, broadband, and scalable solution for next-generation HF radar arrays.

The position and structure of the aurora oval play a central role in determining the performance of Over the Horizon Radar (OTHR) systems, as aurora activity can introduce severe propagation disturbances that degrade detection capability. This study analyzes the evolution of the aurora oval across two contrasting phases of the solar cycle: the low activity period of 2019-2020 and the high activity period of 2024-2025. A particular focus is placed on the extreme geomagnetic storm of May 2024, the strongest in more than two decades, which produced dramatic expansions of the oval. During the storm, the aurora boundaries shifted sharply equatorward, with the outer boundary reaching latitudes near 40°, and the oval broadened substantially. Concurrent measurements showed intense particle precipitation, with enhanced electron densities occurring at altitudes near 100 km. Together, these observations highlight the sensitivity of the aurora oval to geomagnetic forcing and the operational importance of monitoring its variability for reliable high latitude OTHR performance.

Skywave over-the-horizon radar (OTHR) enables long-range sensing via ionospheric reflection, yet propagation variability and sensor miscalibration often induce geometry-dependent measurement mismatch, making the effective measurement mapping only partially known. We propose a reference-aided measurement residual learning approach that captures the unknown component as a state-dependent residual added to a nominal physics-based mapping, and learns this residual without assuming an explicit low-dimensional bias parameterization. The learned correction is incorporated into a residual-corrected measurement mapping and applied in the unscented Kalman filter (UKF) update, providing mismatch compensation while keeping the state dimension unchanged. Monte Carlo simulations with biased skywave OTHR measurements and Automatic Dependent Surveillance-Broadcast (ADS-B) reference uncertainty show reduced measurement residuals and improved tracking accuracy relative to the Nominal-UKF and Joint-UKF baselines.

Traditional significant wave height (SWH) estima- tion from HF radar relies on the calculation of Doppler and ocean wave spectra. This was simplified by establishing a linear relationship between SWH and the standard deviation of received HF radar voltages under first-order scattering by the second author. Building on this concept, this paper presents a neural- network based model which incorporates second-order scattering effects. The method was evaluated using HF radar data collected in July 2018 at Argentia, Newfoundland, with buoy measurements as ground truth, achieving a minimum RMSE of approximately 19 cm.

There is a long list of unphysical results presented in papers and books on passive microwave circuits. Among them are negative impedances, amplification instead of attenuation, superluminal velocity of propagation etc. Such unphysical results usually have two sources: general circuit of two ports and model od coupled transmission lines. The paper reveals this sources and long-lasting presence of unphysical results in microwave circuit theory, especially in journals published by IEEE.

Gap waveguide technology has gained increasing attention as a low-loss and fabrication-tolerant solution for microwave and millimeter-wave systems. In this paper, a compact double-ridge transition based on gap waveguide technology is presented for wideband operation in the 18-30 GHz frequency range. The proposed transition employs a double-ridge configuration to enhance field confinement and impedance matching over a broad bandwidth. To accommodate space constraints typically encountered in practical integrated layouts, a bent ridge geometry is introduced at the transition interface. Full-wave electromagnetic simulations demonstrate stable wideband performance with reflection levels better than −15 dB across the entire operating band. The proposed transition preserves the inherent advantages of gap waveguide technology, including contactless operation and relaxed fabrication requirements.

We present a specialized directional coupler design intended for use in phase-reference distribution lines of linear particle accelerators. Such applications often have requirements that do not match well with commercially available models, motivating custom approaches. Apart from the usual requirements of coupling, directivity, return loss (or VSWR), insertion loss, and high power handling, the less-standard requirements include low phase drift and resistance to ionizing radiation. On the other hand, the intended applications are inherently narrowband systems; thus, there is additional flexibility in some design decisions.

The design is prototyped for an operating frequency of 162.5 MHz in two coupling coefficient variants: 25 dB and 19 dB. It can also be easily adopted to other frequencies and coupling coefficients.

Conventional microwave engineering education relies heavily on analytical methods, canonical circuit topologies, and intuition-driven design, which have proven effective at microwave frequencies. However, as systems increasingly operate in the millimeter-wave and terahertz regimes, parasitic effects, process-dependent electromagnetic interactions, and ultra-wideband performance requirements challenge both topology/layout-constrained traditional design methodologies and existing teaching paradigms. This paper proposes a pedagogical shift in microwave engineering education by introducing machine-learning (ML) and data-driven electromagnetic synthesis as a complementary design framework for microwave circuits such as power dividers and combiners, couplers, and baluns. Rather than emphasizing predefined topologies, the proposed approach enables topology-agnostic, performance-oriented exploration of the design space, allowing students to directly engage with electromagnetic behavior through specification-driven synthesis. By integrating machine-learning-based inverse design and multi-objective optimization into the curriculum, the framework enhances physical intuition, encourages design creativity, and better aligns microwave education with emerging industrial practices in high-frequency and ultra-wideband system design.

This paper presents the design and optimization validation of a compact 4x4 multiple-input multiple-output (MIMO) antenna system characterized by dual band and high isolation. To meet the increasing demands for high data rates in modern wireless communication, a four-port configuration is developed using microstrip monopole patch with slot elements. To mitigate the mutual coupling inherent in dense MIMO arrays, defected ground structures (DGS), and parasitic elements are strategically integrated, ensuring high port-to-port isolation without increasing the overall size. The proposed 4x4 MIMO array is designed on a FR-4 substrate with compact physical dimensions of 68 mm × 68 mm × 3.5 mm. Simulation results demonstrate that the antenna operates within the frequency range of 4.8-6.2 GHz, achieving a fractional bandwidth of 25.5%. Key MIMO performance metrics indicate a port isolation better than 28 dB and an Envelope Correlation Coefficient (ECC) lower than 0.08 across the entire operating band, which confirms excellent diversity performance. Furthermore, the antenna exhibits a peak gain of 4.51 dBi and a radiation efficiency exceeding 72%. The diversity gain (DG) is maintained near the theoretical limit of 6.45 dB. With its combination of compact size, robust isolation, and high-gain characteristics, the proposed 4x4 MIMO antenna is a suitable candidate for 5G NR, Wi-Fi 6E.

Precise localization of radar targets is a fundamental requirement in many applications, including geodetic calibration, infrastructure monitoring and military target detection. This paper presents a comprehensive hierarchical framework for the automatic detection and sub-pixel localization of diverse target geometries within spatially correlated clutter. We propose a two-stage methodology: first, a Constant False Alarm Rate (CFAR) adaptive thresholding algorithm identifies candidate Regions of Interest (RoI). Second, we evaluate two localization cases: 1) the high-precision positioning of isolated corner reflectors, and 2) the center-of-gravity estimation for larger extended targets such as ships or buildings. The sub-pixel target centers are refined using competing estimators, including Center of Gravity, Parabolic Interpolation, and Non-Linear Least Squares fitting to Gaussian Model. Numerical experiments demonstrate that while algebraic estimators suffer from significant bias in correlated clutter, the Gaussian Model Fitting method achieves precision approaching the Cramer-Rao Lower Bound. Furthermore, we show that for the cases of overlapping responses for extended structures, Super-Gaussian modeling are required to prevent localization collapse. The proposed workflow provides a robust, unified solution for automated target analysis in complex radiometric environments.

We present a method for measurement-to-target association in multistatic passive radar in the presence of missed detections and false alarms. Building on a generalized likelihood ratio cost, the association problem is formulated as a staged decision process solved by dynamic programming (DP). The DP explores only feasible associations, reuses partial evaluations across previously visited states, and stops when no further cost reduction is possible. Analytical results show that the brute-force method scales super-exponentially with the number of targets. In contrast, the computational cost of the proposed DP method grows near-exponentially, with a substantially smaller growth rate. Simulations based on real transmitter geometry and a digital elevation model for terrain height confirm this gap, showing that the DP method remains computationally effective in scenarios with target numbers well beyond the range feasible for brute-force association, achieving speedups of several orders of magnitude in the tested configurations.

During the free-fall phase of ballistic missiles, fragments separated from the missile can travel along trajectories and at velocities similar to those of the re-entry vehicle, causing interference in radar detection and tracking. Instantaneously strong fragment reflections may trigger false alarms, while multiple fragments can raise the detection threshold, leading to missed detections. These issues arise from CFAR-based detection processing. This paper proposes a method to mitigate false alarms and missed detections caused by fragments. The approach segments target regions using a flattened CFAR threshold that considers all samples within the receive gate, and then detects local maxima within each segmented region. The effectiveness of the proposed method is demonstrated through simulations and real data.

Forward scatter radar systems detect objects by measuring signals scattered in the forward direction, offering enhanced sensitivity to small targets. Various detection schemes have been proposed that exploit the amplitude modulation induced by the target on the direct signal from the transmitter, enabling low-cost implementations compared to conventional coherent radar systems. In this paper, we investigate several approaches within a unified framework. Specifically, we compare a theoretically optimal matched filter, which requires precise phase knowledge, with a sub-optimal marginalized likelihood ratio test approach. Theoretical and simulation results show that, while the optimal filter provides a 3 dB Signal-to-Noise Ratio (SNR) advantage under ideal conditions, it is extremely sensitive to unknown phase terms. In contrast, the sub-optimal scheme proves robust and effective, simplifying practical implementation. Moreover, our analysis shows that the sub-optimal approach naturally converges to the standard short-time Fourier transform (STFT) solution when short integration times are used. Therefore we analyze how the choice of the processing interval affects detection losses caused by amplitude and phase mismatches in realistic multi-frequency scenarios.

This study presents an edge-deployable Convolutional Neural Network (CNN) model designed for fast and efficient human presence detection using raw ultra-wideband (UWB) radar data. The objective is to minimize decision time and enable deployment on edge devices with limited computational resources through model quantization techniques. Data were collected from two human subjects positioned at various distances and angles relative to the radar. To enhance feature representation, frequency-domain and amplitude-based analyses were applied using the Fast Fourier Transform (FFT), Root Mean Square (RMS), and Hilbert envelope methods. In the final step, Occlusion Sensitivity and Integrated Gradients have been applied to reveal the temporal and spatial patterns that drive the model's human detection decisions.

This paper presents a dual-polarized Van Atta array operating in the millimeter-wave band, based on patch antennas and a substrate integrated waveguide (SIW) feeding network. The array, with a resonance frequency around 60 GHz, was fabricated using low temperature cofired ceramics (LTCC) technology and characterized in a millimeter-wave anechoic chamber. Good agreement between simulated and measured results was observed, with a maximum radar cross section (RCS) of −30 dBsm and an angular range of nearly 66 degrees in which the RCS does not drop more than 10 dB relative to the peak. To the best of the authors' knowledge, this is the first reported realization of a dual-polarized Van Atta array at millimeter-wave frequencies with patch antennas fed via an SIW network.

This paper introduces a new concept to significantly improve end-of-life (EoL) design for disassembly applied to 3D-printed dielectric overlays for electronically steerable parasitic array radiator (ESPAR) antennas miniaturization. A core-shell overlay is proposed, where a mechanically stable polylactic acid (PLA) shell encapsulates a polyvinyl alcohol (PVA) infill core with controlled porosity (70-90% infill). This allows to increase effective permittivity while reducing overlay mass and enabling a water-assisted EoL pathway through PVA dissolution, leaving only a limited solid PLA residue for conventional recycling/disposal routes (~13.7% of a solid PLA reference overlay mass). Bulk dielectric properties of printed PLA and PVA are characterized at 2.45 GHz using a split-post dielectric resonator (SPDR). Overlays are designed in Altair FEKO and fabricated by dual-material fused deposition modeling (FDM). Antenna performance is evaluated via reflection coefficient (S11) and elevation-plane radiation patterns in an anechoic chamber. Environmental robustness is assessed using room-temperature water immersion followed by air drying. Moisture exposure causes a temporary detuning, however, the S11 response largely returns after drying, indicating reversible humidity effects for the tested structures. The fabricated core-shell overlays reduced the as-printed mass relative to the solid PLA overlay by 12.1% for PVA 90%, 19.4% for PVA 80%, and 26.9% for PVA 70%. The results have high potential impact on sustainability by supporting lightweight, miniaturizing overlays aligned with innovative circular-by-design principles.

This paper presents the design of parabolic dielectric lenses integrated with a two-dimensional Van Atta array fabricated using low temperature cofired ceramics (LTCC) technology and operating at 24 GHz. The lens integration aims to enhance the read range of the array when employed as a chipless radio frequency identification (RFID) tag or as a component of precision landing support systems for unmanned aerial vehicles (UAVs), while also providing mechanical protection of the structure. The lenses were manufactured using 3D printing technology using Acrylonitrile-Butadiene-Styrene (ABS) mixed with ceramic powder. The proposed lenses with different geometric profiles increase the radar cross section (RCS) of the investigated array by up to 9.3 dB, corresponding to increase in read range by 71%. To the best of the authors' knowledge, this is the first work reporting the use of 3D-printed parabolic lenses to enhance the read range of a two-dimensional Van Atta arrays.

This work presents a novel approach for the implementation of a wake-up receiver for LoRa communication. It is designed to exploit the LoRa modulation based on chirp spread spectrum and is inherently less susceptible to false wake-ups caused by interfering signals. The theory of the concept is presented indicating key parameters. For validation, a proof-of-concept implementation is developed with exemplary measurement results.

As healthcare shifts toward prevention and wearable monitoring scales to millions of devices, the cumulative environmental footprint of disposable electrode materials and inefficient data processing becomes a critical design concern. We address this through two complementary approaches: sustainable bioimpedance electrode materials and resource-efficient on-device signal processing. Our evaluation of 20 screen-printed electrode configurations on PLA, recycled PET, and cellulose-based substrates shows that screen-printed copper paste electrodes match or exceed conventional Ag/AgCl in bioimpedance signal quality (regression 0.97 vs 0.88). An on-device signal processing block extracts heart rate and respiratory rate with Polar H10 reference-device disagreement of 2.66 BPM and 0.805 BrPM respectively, using only 3.5 KB Flash and 8.9 KB DRAM per extraction block on an ESP32-C3-Mini. These results demonstrate that practical vital sign extraction and sustainable electrode materials can together enable population-scale preventive monitoring with substantially reduced environmental impact.

Rain drop shapes and their fall speeds are examined for several events associated with significant turbulence. Drop shapes have also been used to compute their S- and C-band radar cross sections (RCS) for horizontal and vertical polarizations and hence the single particle differential reflectivity (ZDR). Both fall speeds and the computed ZDR show a gradual decrease with turbulence intensity beyond a certain threshold. For one event the adjusted rain fall rate algorithm for S-band radars was used to determine rain accumulation with time and found to have reasonable agreement.

Wind-related atmospheric phenomena remain one of the key sources of uncertainty and risk for aviation operations, especially under conditions of rapidly evolving weather. Advanced weather radar systems provide a variety of polarimetric observables whose sensitivity to atmospheric dynamics can be exploited for improved wind-related phenomena detection. This paper investigates a multivariate atmospheric sensing approach that combines polarimetric weather radar measurements with wind information derived from unmanned aerial vehicle (UAV) flight data within a common meteorological information space. The proposed concept supports data fusion from heterogeneous observation sources and enables complementary interpretation of atmospheric processes. A simulation of radar backscattering from non-spherical and vibrating raindrops is performed to analyze the behavior of co-polarized and cross-polarized signal components under wind-induced microphysical effects. The results demonstrate increased sensitivity of cross-polar returns to drop deformation and vibration, indicating their potential as indicators of dynamic atmospheric processes. In addition, real-time extraction of wind characteristics from UAV telemetry data is demonstrated, providing an independent source of wind information, particularly in the lower airspace. The combined use of polarimetric radar observables and UAV-derived wind data is shown to enhance the capability of advanced meteorological monitoring frameworks for aviation-oriented atmospheric sensing.

Accurate Doppler moment estimation in weather radars is challenging when multiple scatterer populations create multi-modal velocity spectra. In this paper, an adaptive framework that combines an Expectation Maximization (EM) algorithm with folded-Gaussian mixture components to handle velocity aliasing is presented. An adaptive transmission policy adjusts pulse repetition time and coherent pulse count to improve Doppler resolution while avoiding ambiguities. Monte Carlo tests across drift regimes show adaptive selection clearly lowers estimation error when compared to fixed settings, and effectively reduces aliasing artifacts, enabling more robust multi-modal Doppler estimation.

This paper presents experimental multi-frequency radar measurements of rain obtained simultaneously at K-, Ka-, and W-band frequencies using colocated vertically pointing radars operating at 24, 35, and 94 GHz. The analysis focuses on frequency-dependent behavior of radar reflectivity and mean Doppler velocity during an hour precipitation event. Systematic differences in reflectivity are observed across the three frequency bands, with increasing impact of non-Rayleigh scattering and rain-induced attenuation toward higher frequencies. Doppler velocity measurements reveal consistent reduction of mean fall velocity at Ka- and W-band due to enhanced sensitivity to small-drop populations. Surface rain-gauge observations indicate intermittent precipitation reaching the ground, highlighting the importance of vertical context and sub-cloud evaporation for interpretation of radar measurements. The results demonstrate the complementary nature of multi-frequency radar observations and their potential for improved characterization of precipitation processes

This paper describes in the form of a whitepaper recommendations for future development of sustainable and circular electronics. It describes policy recommendations that were created through workshops focusing on standards, regulations and future roadmaps withing the framework of collaborative ecosystem research project Sustronics. Two main themes emerged for enabling sustainability and circularity: eco-design and lifetime extension supported with conscious computing.

This study presents initiation studies on rubble penetrating radar (RPR) technology piloted at the Rubble Characterization Research and Application Laboratory at the Çiftlikköy Campus of Mersin University. The laboratory facility is introduced, and the experimental setup is presented. Real experiments for people under the debris are presented with the obtained detection two-dimensional (2D) range slow-time radar images. The constructed RPR images prove the success and validity of the proposed RPR approach in detection and locating the movement under the rubble.

This work proposes bi-static ultra-wide band (UWB) radar system approach to detect life-sign motions of trapped humans under heterogeneous and lossy wreckage medium. On this scope, an embedded SF/CW radar data acquisition and processing system targeting near-field and through-debris sensing is designed for potential use in collapsed structure inspection.

Moreover, electromagnetic wave reflections conducted in a realistic search and rescue training area are interpreted using advanced signal processing and deep learning techniques. The UWB amplitude and phase measurements acquired over programmable frequency sweeps are examined in a complex spectrum and transformed into range-domain signatures via inverse fast Fourier transform (IFFT) to obtain the range profile with good resolution. Here the debris-oriented CW frequency yields the most efficient Doppler measurements. Background subtraction and adaptive filtering techniques are employed to improve robustness in cluttered environments.

It is aimed to improve the human motion detection range and false alarm performance of the search and rescue radar by using both bi-static setup approach and benefiting advanced signal processing techniques that may overcome complex environmental challenges.

This study presents a comparative investigation into the electromagnetic attenuation characteristics of four construction materials extensively used in the Turkish building industry: C-30/37 reinforced concrete, aerated concrete, pumice block (Bims), and clay brick. Transmission-only S_21 measurements were conducted in the 0.8-2.8 GHz frequency band using a free-space setup with a Vector Network Analyzer (VNA). To overcome the limitations of finite measurement bandwidth and multipath reflections typical of outdoor experiments, a hybrid signal processing scheme is proposed. This method integrates raw time-domain analysis for qualitative assessment with a Gaussian-windowed time-gating technique for precise attenuation extraction. The results indicate significant variation in signal penetration, ranging from high-loss concrete (~80 dB/m) to low-loss clay brick (<5 dB/m). These findings, validated against recent localized studies, provide critical data for outdoor propagation modeling and through-wall radar applications.

This paper presents the design, production and usage of a practical, low-cost, low-profile pin-fed stacked rectangular patch antenna in through-the-wall imaging radar (TWIR). The proposed design consists of stacked V-shaped patches layered on top of each other to achieve ultra-wideband (UWB) operation and attain the required range resolution for TWIR. The antenna is designed for the 1.62-2.12 GHz band, enabling effective signal propagation through common building materials in TWIR applications. The antenna design and optimization were performed using CST Microwave Studio. The prototype was fabricated on a plexiglass substrate (εr = 2.6). Experimental evaluations of S11 demonstrate good agreement with simulated results, validating the proposed antenna design. Finally, the fabricated antennas were tested in a through-wall imaging scenario using a transmitter-receiver configuration. n this scenario, a person walking behind a wall was successfully detected, and their movement and range were monitored in real time, demonstrating the antenna's effectiveness for TWIR applications.

Millimetre-wave (mmWave) Frequency-Modulated Continuous-Wave (FMCW) radar systems are increasingly used in imaging, automotive, and industrial applications due to their compact size, high resolution, and robustness to environmental conditions. Linear MIMO radar arrays, while providing excellent angular resolution in one plane (typically azimuth), suffer from poor resolution in the orthogonal (elevation) direction due to limited aperture extent. In this work, we propose a general and hardware-efficient method to overcome this limitation by acquiring radar measurements at two orthogonal orientations using a simple mechanical rotation. The horizontal and vertical acquisitions offer complementary spatial resolution, and their coherent fusion yields a more isotropic point spread function (PSF), enabling improved 2D imaging quality and target separability. The proposed technique does not require any hardware modification and is applicable to a wide range of mmWave MIMO radar systems. Experimental validation is performed using a cascaded 4×AWR2243 FMCW radar platform, demonstrating the effectiveness of the approach in achieving high-resolution 2D imaging with minimal complexity.

This paper presents the real time passive SAR system demonstrator based on DVB-T/T2 illumination of opportunity. The system demonstrator was developed under the project and was designed to be used in uav applications, including imaging of the earth surface around the DVB-T2 terrestrial illuminators of opportunity in Passive Synthetic Aperture Radar (SAR) mode. The system was designed to work in real-time, so passive SAR images are obtained during the flight/mission. Experiments carried out in the laboratory and in open area environments, SAR trails using an airborne platform are presented and discussed from the perspective of future work and further improvements.

We demonstrate passive radar detection of small, low and slow flying targets using Digital Audio Broadcast (DAB) signals of opportunity with only minimal sensor complexity and cost. The sensors are based on cheap off-the-shelf components and do not require a separate reference antenna. We show that such sensors could be valuable for stand-in use in a contested area or as gap-fillers for larger passive radar networks at a very attractive price point.

This paper investigates the application of artificial neural networks to target localization in netted passive bistatic radar systems. It briefly describes passive bistatic radar configurations and the challenges inherent in determining target positions using non-cooperative emitters. A classical multilayer perceptron (MLP) is proposed as the core learning architecture. Custom datasets were synthesized to reflect a range of geometric and signal propagation conditions. The neural network models were trained and evaluated with respect to their accuracy in predicting the positions of objects. Simulation results demonstrate the feasibility of this approach and reveal dependencies between localization performance and variables such as the observer-target distance and target altitude. Designed for fixed sensor infrastructures, this ML-based approach offers a viable complement to traditional signal processing techniques in netted passive bistatic radar systems.

This paper demonstrates a DVB-T2 based passive radar system designed for real-time vehicle localization and traffic flow monitoring. Utilizing an existing DVB-T2 transmitter as an emitter of opportunity offers a cost-effective and non-intrusive solution for traffic monitoring applications. Incorporating road layout constraints enables precise vehicle position and velocity estimation based on measured bistatic target parameters. An experimental setup and a subsequent measurement campaign validate the system's performance in detecting and localizing multiple vehicles. Furthermore, the transition to an embedded computing platform demonstrates that real-time processing requirements are met for field deployment.

Results from field tests validating OFDM-based passive radar concept making use of the Digital Audio Broadcasting (DAB) signal are presented in this paper. Two low-cost software-defined radio (SDR) receivers connected to standard personal computers were checked in different localizations: the RTL-SDR stick with a 1-meter V-shape antenna working indoors and the RSPduo box with a 2-meter vertical antenna installed outdoors. In the performed experiments, broadcast DAB signals were recorded, and the states of DAB carriers were measured. Their reconstructed and observed values were then used to estimate the time-variant channel frequency response, time-variant channel impulse response, and the bistatic range-velocity maps (RVMs). Positions of DAB signal reflections from jet airplanes, visible in RVMs, were compared with references calculated from the airplanes' ADS-B data. As a result, good agreement was confirmed between the measured and reference coordinates.

Breathing detection can help to distinguish noisy reflections from actual persons when evaluating radar signals, because of their harmonic nature. In this paper, breathing detection is investigated systematically to evaluate a new tracking-based algorithm for person detection using a millimeter wave frequency modulated continuous wave radar. This algorithm uses the range-Doppler data of the radar to determine whether stationary or moving persons are present. Potential detections are found by using constant false alarm rate. Kalman tracking is used to track them over time. To decide if a detection is a person, the frequency components of its phase over time are reviewed.

In a first step, the detection algorithm is validated using moving radar reflectors, which are used to simulate normal adult breathing motion. The investigated parameters are breathing rate and chest displacement, distance, and position on the azimuthal plane. Two breathing movement modes were compared. The results show that the distance and breathing rate estimation works well for one and two reflectors. The direction of arrival estimation, however, still leaves room for improvement.

The accuracy of polynomial approximation for launch and impact point estimation of artillery projectiles using radar observations is evaluated from multiple points of view. The influence of measurement accuracy, update rate, and time span is assessed using computer simulations. The method is found to offer reasonable accuracy provided that a suitable part of the trajectory is observed with practically attainable update rates and accuracies.

Tracking targets in bistatic passive radar within Cartesian space relies on achieving fast convergence and reliable performance, which are critical in safety-sensitive applications such as airport surveillance. Initialization uncertainty can substantially degrade tracking accuracy that slows down the convergence time. Prior studies emphasize the importance of proper initialization in tracking, as filtering algorithms rely on prior states and motion models to predict current target positions. However, few works systematically quantify initialization effects under varying geometric and signal conditions. This study compares a standard Extended Kalman Filter (EKF) with a hybrid Particle Filter-EKF (PF-EKF) approach, in which the PF assists initialization by processing the same measurements to isolate the effects of uncertain initial conditions across different geometric dilution of precision (GDOP) and signal-to-interference-plus-noise ratio (SINR) scenarios. Simulation results show that the PF-EKF accelerates convergence and enhances track stability compared to the standard EKF. These findings validate the PF as an effective initialization assistant and provide a foundation for future experimental validation using a software-defined-radio-based testbed.

Range-only localization under limited observability arises in numerous sensing and surveillance applications, where nonlinear measurement models and poor geometric diversity challenge reliable state estimation. In such scenarios, achieving low estimation error alone is insufficient; the credibility of the reported uncertainty plays a critical role in downstream tracking and decision-making tasks. This paper presents a comparative statistical consistency analysis of four Bayesian estimation frameworks, the Extended Kalman Filter (EKF), Iterated Extended Kalman Filter (IEKF), Unscented Kalman Filter (UKF), and Particle Filter (PF) applied to the localization of an unknown static emitter using pseudorange-only measurements. Each filter is evaluated within a common simulation framework that isolates the effects of nonlinearity and limited observability. Consistency is assessed through ensemble-based error statistics, time-evolving covariance bounds, and the fraction of realizations contained within predicted confidence intervals. The results reveal that while Gaussian-based filters may achieve apparent convergence, they frequently exhibit overconfident or inconsistent uncertainty estimates under severe nonlinearity. In contrast, the Particle Filter demonstrates robust uncertainty representation and superior consistency, albeit at increased computational cost. These findings highlight fundamental trade-offs between estimator complexity, convergence behavior, and statistical reliability in range-only localization problems. The presented analysis provides practical insight into filter selection for tracking systems operating under weak observability and nonlinear measurement conditions.

Reducing false alarm rate and improving the accuracy of target coordinate estimation remain a critical challenge for air-defense surveillance radars operating under realistic clutter and interference conditions. This paper proposes a three-stage radar data processing framework that jointly addresses false alarm suppression and target coordinate accuracy improvement using real operational radar data. In the first stage, multiple representative elevation angle estimation strategies for clustered detections are systematically analyzed and evaluated, overcoming the limitations of conventional centroid-based approaches. In the second stage, an enhanced multi-target tracking scheme is developed by jointly exploiting position, ambiguous Doppler, and received power measurements, where Doppler ambiguity is effectively mitigated through consistency checks between measured and estimated Doppler values. In the final stage, target kinematic features, including velocity, acceleration, and heading rate, are analyzed to suppress noise- and interference-induced tracks after track formation. Experimental results obtained from a three-dimensional air-defense surveillance radar demonstrate that the proposed framework improves target coordinate estimation accuracy by up to 23% in terms of RMSE and eliminates approximately 40% of false tracks caused by noise and clutter, compared with a reference processing chain. These results confirm the effectiveness and practical applicability of the proposed framework for operational radar systems.

This presentation will focus on the impact of applied technology on military operations. In particular, it will highlight two key perspectives. First, it examines how emerging technologies influence military operations, including the balance between man-in-the-loop and man-on-the-loop approaches. Second, it explores how new technologies affect education, training, and exercises.

The presentation will demonstrate how radar, along with other technologies, is reshaping the conceptual understanding of military operations. Examples from current research at the Norwegian Defence University College (NDUC) and the Norwegian Defence Research Establishment (FFI) will be presented, illustrated through the sensor-to-shooter chain.

For the radar community, the presentation aims to place groundbreaking research within a broader ecosystem, spanning from fundamental science to operational military effects. Finally, the presentation will provide examples of current and emerging requirements driven by the evolving and increasingly challenging security and defence environment.