This article presents germanium telluride (GeTe)-based switches for radiofrequency (RF) applications, capable of reversible switching between their ON and OFF states through optical activation by irradiation. Unlike previous studies, the transition is induced by infrared laser pulses at a wavelength of λ = 915 nm, which is highly promising for future integration of laser sources and the proposal of fully integrated optical activation of phase change material (PCM) switches. This represents a novel approach compared to the existing literature, which primarily focuses on the ultra-violet spectrum, less suitable for on-chip optical integration. Our work also provides combined optical and thermal simulations to elucidate the challenges associated with actuating small PCM switches and demonstrates the effectiveness of PCMs at this wavelength. The study achieves bistable switching at high frequencies up to 40 GHz, with a figure of merit of 31.5 fs, despite the low GeTe conductivity of only 1.85·105 S/m. Additionally, significant advancements over the literature have been made by surpassing 30,000 cycles with optical actuation.

Future telecommunication systems are set to revolutionize connectivity, driven by advancements in technologies like 6 G, artificial intelligence, and the Internet of Things (IoT). However, this evolution brings significant challenges. Traditional silicon-based transistors struggle to meet demands for efficiency and power handling. Indium Phosphide (InP)-based Double Heterojunction Bipolar Transistors (DHBTs) deliver excellent performance at sub-mm-wave frequencies while minimizing power loss and heat generation. Additionally, achieving reliable large-signal performance in high-frequency applications requires accurate large-signal modelling and advanced testing techniques, such as load-pull measurements. In this paper, we report the comparison between two InP/GaAsSb Double Heterojunction Bipolar Transistors (DHBTs) with different collector epitaxial designs in terms of their small- and large-signal performance. The effect of the epitaxial design on the small- and large-signal performances is investigated and load-pull measurements in G-band are performed to assess the great power-handling and efficiency capabilities of the InP/GaAsSb DHBT technology. For both of the designs, THz cut-off frequencies with Power-Added Efficiency (PAE) > 30% are achieved. Moreover, the value of PAE = 39.2% reached in G-band represents the highest among any technology. Finally, the two different epitaxial designs are thermally characterized to investigate the effect of different layers on the thermal and RF-performances.

This paper presents the design, implementation, and characterization of a compact two-stage analog phase-shifter for the Ka-band, based on thin-film technology. The design utilizes a reflective-type configuration, employing four metal-insulator-graphene diodes as reflective loads. The fabricated prototype is realized on an 8-µ m-thick flexible polyimide substrate and occupies less than 0.7 mm2 of chip area including the contact pads. Performance evaluation of the fabricated circuit reveals an S11 of better than −13 dB and an S21 of −3.3 dB with a tolerance of ± 0.5 dB across the frequency band from 28 to 36 GHz, along with a tunable phase difference ($\Delta\phi$) exceeding 70∘. The introduced flexible thin-film technology promotes the realization of flexible cost-effective beam steering for smart surfaces implementations for communication and biomedical applications.

A wideband harmonic rejection (HR) voltage-domain mixer using resistive scaling is presented featuring excellent linearity and high intermediate frequency (IF) bandwidth. Thin-oxide devices with constant gate-to-source voltages (VGS) are utilized to maximize the switching linearity. A novel switching core topology providing low-impedance IF outputs is proposed to support wideband in-phase (I) and quadrature (Q) mixer outputs when capacitively loaded by an analog-to-digital converter (ADC). Eight LO clock phases, each with a 25% duty cycle, are on-chip generated for quadrature down-conversion and HR. By cleverly activating and organizing the mixer branches, the mixer's input impedance at radio frequency (RF) can be kept perfectly constant throughout all eight clock phases, enhancing the mixer’s linearity. The TSMC 40 nm-CMOS realized mixer reaches 20.9 dBm OIP3 at an IF of 50 MHz with a conversion loss of 22.5 dB. It offers an 800 MHz 3-dB IF bandwidth when connected to a differential capacitive loading of 0.15 pF, with a total power consumption of 40.7 mW drawn from a 1.1 V supply. The mixer targets linear wideband base station observation receiver applications.

This paper provides a profound perspective on the neuronal property of radiofrequency (RF) power amplifiers (PAs). The nonlinearity of the PA is studied for the first time by solving differential equations in a bio-inspired neuromorphic model, i.e. Hodgkin–Huxley model, with Bayesian estimation. This study demonstrates that the nonlinearity of RF components is biomimicking and thus can enhance neuromorphic computing capacities of a communication network under the framework of a joint communication and computation (J2C) scheme. Overall, our work contributes to incorporating artificial intelligence (AI) and automation into the communication network, a key trend in 6G.

This paper presents a compact, dual-polarized, 32-element, 47 GHz phased array transceiver, fabricated in 55 nm CMOS technology with antenna in package (AiP) technology for 5G communications. The proposed transceiver employs an intermediate frequency (IF) phase-shifting architecture and a facing-up (FU) configuration AiP. The IF phase shifting is realized using a bidirectional IF vector sum phase shifter and IQ mixer with drain bias, achieves less than 1° phase resolution, with rms phase error of 0.047° and rms amplitude error of 0.063 dB. The output third-order intercept point is above +1.0 dBm with Txconversion gain of more than 16.5 dB over the radio frequency ranges from 46 to 49 GHz. To reduce heat concentration from the high integrated phased-array transceiver, a FU AiP-fan-out wafer-level packaging utilizes solder balls mounted on the package as a heat spreader, resulting in a thermal resistance of less than 0.3 K/W. The finalized AiP size is only 12.3 mm × 14.9 mm. Regarding the over-the-air measurement, the proposed transmitter can deliver an equivalent isotropic radiated power of 30.7 dBm with single polarization and demonstrates transmitter error vector magnitude less than 3.9% under 5G NR modulation scheme (256QAM, 100 MHz bandwidth).

Respiratory and cardiac rates can be estimated by analyzing a spectrum of linearly mixed phase fluctuations in a radar echo of an individual. However, there are high-order harmonics caused by time-varying respiratory rate, and the interference effect of the respiratory rate and its harmonics makes it difficult to estimate the cardiac rate with relatively low energy in a spectrum. To solve this problem, we exploit the independent component analysis method with dual-band distributed continuous wave radar for effective decomposition of phase fluctuations corresponding to respiratory and cardiac rates. In simulations and experiments, the respiratory and cardiac rates were successfully estimated by the proposed decomposition method, compared with conventional methods.

An original analysis of Fabry–Perot cavity antennas based on thick partially reflecting sheet (PRS) is presented in this work. The bandwidth enhancement of such radiating devices with respect to Fabry–Perot cavity antennas based on thin PRS has been investigated through a leaky-wave, transverse-equivalent-network approach, and a field matching technique. This analysis led to an optimal condition for considerably improving the gain-bandwidth figure of merit for this class of radiating devices on a sound physical basis. A Fabry–Perot cavity antenna based on a thick PRS working at 60 GHz is discussed as a case study. An excellent impedance matching is finally achieved by means of an efficient feeding network designed through a fast ad hoc, hybrid, analytical-numerical method. Theoretical results are in an excellent agreement with full-wave simulations corroborating the proposed methods.

In this paper, we report an on-wafer High-Electron-Mobility transistor characterization method over a large frequency band [250 MHz–1.1 THz]. The transistor’s coplanar accesses were optimized to enable high-frequency measurement of the devices up to 1.1 THz. The characterization method implements an on-wafer multiline Thru-Reflect-Line calibration kit fabricated on indium phosphide (InP) substrate validated using comparisons between simulations and measurements of coplanar waveguide (CPW) devices. An 80-nm gate length InAlAs/InGaAs/InAs InP-HEMT was fabricated using the optimized transistor accesses then it was measured over the [250 MHz–1.1 THz] frequency band using the developed on-wafer characterization method. It is important to mention that the same transistor was measured on six different on-wafer test benches [250 MHz–110 GHz], [140–220 GHz], [220–325 GHz], [325–480 GHz], [500–750 GHz], and [750–1100 GHz]. The S-parameter measurement results show good continuity between the six measured frequency bands. Observations on the extracted gain measurements and a comparison between the measured and [250 MHz–110 GHz] extrapolated cut-off frequencies are also provided.

This paper presents the design and characterization of a unit cell for dual-polarized liquid crystal (LC)-based reconfigurable intelligent surfaces (RIS), as well as an efficient, full-wave simulation methodology for the far-field beam-steering capabilities of large-scale LC-RIS. Within this framework, the unit cell relies on defected delay lines with a 4.6 μm thin LC layer aperture coupled to a patch antenna. This delay line architecture aims towards simultaneous optimization of loss, bandwidth and response time. Full-wave simulations of the unit cell in a periodic environment show an operating frequency between 26.5 and 29.5 GHz with wide angle radiation. Measurements of the unit cell in a 3 $\times$ 3 rectangular grid exhibit wideband impedance matching and overall good agreement with simulations. Furthermore, a simulation methodology is introduced that evaluates large-scale LC-RIS far-field beam-steering capabilities without requiring full-wave simulations of the entire structure, but just few unit cells. Within this scope, the LC-RIS achieves a maximum efficiency of 20.8% with a beam-steering range from −48° to +48°, despite the use of a lossy glass substrate and gold as a conductor. It exhibits a minimum bandwidth of 8.2% for an efficiency of at least 10% across all analyzed steering angles in E-Plane and H-Plane.

Proper management of mutual interference plays an important role in the successful simultaneous operation of automotive frequency-modulated continuous-wave (FMCW) radar sensors at different vehicles. Compared to traditional interference handling concepts such as detect-and-mitigate or detect-and-avoid, the detect-and-exploit paradigm turns the originally interfering signals into signals of interest and uses them to obtain information about the environment. Following this idea, a method that implements such an interference exploitation strategy in terms of joint passive spectral sensing and localization of surrounding objects is elaborated and presented in this work. In summary, the method consists of a dedicated radar operational mode and a corresponding signal processing chain including pre-processing, beam steering-based signal component separation, maximum likelihood (ML)–inspired signal parameter estimation, and joint direction of arrival (DoA)-time difference of arrival (TDoA) based object localization. The unique advantage of the presented concept compared to over-the-air synchronization (OTAS)-based solutions is that it can also deal with interferers that change their ramp parameters over time. The applicability of the concept is both theoretically analyzed as well as practically demonstrated by means of measurements in an anechoic chamber, where the position of the interferer and an additional object in the surrounding can be determined with an accuracy of a few centimeters.

The impact of misalignment errors, specifically yaw and pitch deviations, on dihedral reflectors’ scattering responses is studied for millimeter-wave polarimetric multiple-input multiple-output automotive radars. Through simulations and experiments at 77 GHz, it is demonstrated that significant radar cross-section (RCS) variations of up to 30 dB can occur within small misalignment ranges (0$^{\circ}$–2$^{\circ}$). The findings emphasize that larger dihedral dimensions can amplify sensitivity to misalignment in some specific misalignment scenarios, offering trade-offs between reflection strength and robustness to misalignment errors. The study also explores near-field effects, revealing notable discrepancies between the dihedral near- and far-field scattering response in misalignment scenarios. A polarimetric calibration method is applied to show how polarimetric channel phase response is affected under such conditions, achieving stable results in specific configurations (e.g., dihedral at 0$^{\circ}$ under yaw misalignment angle). This study addresses key challenges in calibration accuracy, including the high sensitivity of RCS to small angular misalignments, the trade-offs between reflector dimensions and robustness, and the influence of near-field effects in practical setups.

The design of a hexagonal six-ridged waveguide (H6RWG) phased array antenna (PAA) element featuring a wide scan angle matched slotted horn aperture is presented for Ka-band satellite downlink in low Earth orbit non-terrestrial network applications. The proposed PAA element is evaluated against an open-ended waveguide (OEWG) PAA element and achieves a very low active reflection coefficient (ARC) of less than -18 dB and a total antenna efficiency greater than 84% over a wide bandwidth from 17.3 to 20.2 GHz with a $\pm$ 50$^\circ$ scan range. Specifically, the aperture of the H6RWG was designed to limit the variations in ARC during scanning, thereby minimizing load pulling of integrated active devices, as demonstrated with a power amplifier (PA) in a co-simulation. As a result, the power-added efficiency, output power, and linearity of the PA remained stable over the bandwidth and scan range. Compared to the OEWG PAA, the co-polarized system efficiency and equivalent isotropic radiated power are improved for most scan angles within the bandwidth, especially at high scan angles.

This paper presents two distinct configurations of a GaN-based digital transceiver (TRx) to evaluate their performance and integration efficiency. The first configuration, features a novel low-noise amplifier with integrated switching capability (LNAiS) and a digital class-E power amplifier (PA) on a single compact chip. The LNAiS eliminates the need for an external antenna switch, reducing module complexity and chip size while maintaining high performance. It achieves a gain of 12.7 dB and a noise figure of 3 dB at 4.7 GHz in Rx mode and provides over 20 dB isolation in Tx mode across 4.7–7.4 GHz. The digital PA demonstrates flexibility and efficiency, achieving 46% and 23% efficiencies for 20 MHz LTE and OFDM signals, respectively, and 22% for a 240 MHz OFDM signal with 10 dB PAPR. The second configuration integrates the same concept of digital PA with a standard LNA and an SPDT switch (LNAsS), achieving a gain of 24.8 dB and a noise figure of 2.65 dB at 4.2 GHz. This work highlights the trade-offs between these two architectures and demonstrates that the LNAiS-based approach drives the development of greener, more flexible, compact,lower-complexity, and cost-effective transceivers for 5G networks.

Dielectric lens antennas provide significant advantages for various radar-based applications. While an ellipsoidal geometry is often utilized due to its beneficial antenna characteristics and its point-like feed into the lens, its length and antenna reflections are drawbacks for some applications. In this work, we investigate the design of a Fresnel-based lens antenna to overcome these limitations. A quasioptical design is presented, highlighting its ability to reduce internal reflections of lens antennas. Additionally, the effects of the number of Fresnel steps and manufacturing tolerances are analyzed. Measurements validate the design and demonstrate the reduction of internal reflections by more than 13 dB, leading to an increased measurement range in a Tank Level Probing Radar scenario and a size and weight reduction of 41% and 34%, respectively.

This article presents a highly integrated 300-GHz frequency-modulated continuous wave radar sensor using a custom-developed dual-function transceiver MMIC. The system can either be configured as a stand-alone ultra-wide-band radar sensor or as a flexible RF front-end, enabling up-conversion and down-conversion of modulated signals to and from the terahertz range. The transceiver MMIC is manufactured using a 90 nm SiGe BiCMOS process, featuring high-speed hetero-junction bipolar transistors with an $f_{\rm{T}}$ of 300 GHz and $f_{\rm{max}}$ of 520 GHz. Using on-chip antennas and a focusing lens, the EIRP of the system for radar operation is greater than 3.2 dBm in a bandwidth of 54 GHz. The full potential of the system’s 90 GHz tuning range is demonstrated in radar measurements. A calibration method is applied to expand the usable tuning range, achieving an extraordinary spatial resolution of 1.97 mm with a frequency sweep from 330 to 240 GHz in 5 ms for a target at a distance of 0.35 m. The potential industrial use of this spatial resolution is demonstrated in a plastic thickness measurement scenario. Additionally a 100 Mbd OKK communication link with a BER of 0.55% is presented using two systems at 0.3 m distance.

We developed a compact, ultra-wideband radar demonstrator for measurements of snow thickness. We designed the radar to be capable of reconfigurable operation over Ku- and S/C bands, and with a size, weight, and power compatible with a C-3 class unmanned aircraft system (UAS). We implemented the radar’s radio frequency frontend using low-cost laminate materials and employed 3D printed antennas for an inexpensive implementation. To demonstrate its performance and capabilities, we first conducted a series of laboratory tests, followed by tests of opportunity in Antarctica using a sled-based setup. Next, we integrated the radar demonstrator into an Aurelia X6 Pro system and completed a series of local flight tests over areas including grass-covered land and a wooded section with different seasonal foliage conditions. Lastly, we used our UAS-borne radar test bed to map seasonal snow accumulation to a depth close to ∼30 m in Greenland from 100-m altitude. In this paper, we offer a succinct description of the radar test bed electronics, a discussion of laboratory tests and integration considerations, and present sample results from various field scenarios.

Hand gesture recognition (HGR) has gained significant attention in human-computer interaction, enabling touchless control in various domains, such as virtual reality, automotive systems, and healthcare. While deep learning approaches achieve high accuracy in gesture classification, their lack of interpretability hinders transparency and user trust in critical applications. To address this, we extend MIRA, an interpretable rule-based HGR system, with a novel gesture onset detection method that autonomously identifies the start of a gesture before classification. Our onset detection approach achieves 90.13% accuracy on average, demonstrating its robustness across users. By integrating signal processing techniques, MIRA enhances interpretability while maintaining real-time adaptability to dynamic environments. Additionally, we introduce a background class, enabling the system to differentiate between gesture and non-gesture frames and expand the dataset with new users and recordings to improve generalization. We further analyze how feature diversity affects performance, showing that low diversity can suppress personalization due to early misclassifications. Using a foundational and personalized rule framework, our approach correctly classifies up to 94.9% of gestures, reinforcing the impact of personalization in rule-based systems. These findings demonstrate that MIRA is a robust and interpretable alternative to deep learning models, ensuring transparent decision-making for real-world radar-based gesture recognition.

The characterization of biological objects with microwave spectroscopy is getting increasing interests, as it is label-free and noninvasive. To perform their analysis, 2-port sensors are present in the literature, enabling only partial investigations of 3D biological samples, without taking their structural heterogeneity into account. Within this context, a 4-port microwave-based biosensor dedicated to microtissue characterization is proposed, in order to extend the sensing capabilities of microwave dielectric spectroscopy and provide electrical responses of 3D biological models subdivisions. An electrical model suitable for such a multiport device is established to extract the characteristics of the different sections of the 3D entity. The modeling methodology exploits the symmetry of the microwave component, while applying a common and differential modes approach derived from the measured 4 ports scattering parameters. After the mathematical validation of this approach, different elementary models are evaluated. Ethanol-based aqueous solutions are first used for their homogeneity within the fluidic channel. Polystyrene beads exhibiting two different diameter sizes are then numerically and experimentally investigated due to their 3D configuration and their uniform and known permittivity. This study demonstrates that the 4-port sensor and associated electrical model enable to consider electrical subdivisions of the 3D entity under test, based on the localization of the object on the different microwave electrodes. This constitutes the first step toward the analyses of complex and heterogeneous 3D biological models such as microtissues.