This paper presents an analog RF-domain implementation of a Vanilla Recurrent Neural Network (RNN) for real-time anomaly detection in 5G and beyond wireless networks. Real-time analysis is crucial to distinguish benign irregularities from true anomalies that may indicate malicious activity or network issues. The proposed design leverages passive components, such as switches and capacitors, enabling low-latency processing directly at RF frequencies. Circuit simulations validate the approach, achieving up to 89% accuracy across three anomaly classes, demonstrating the potential of RF-domain neural networks for fast and efficient wireless signal analysis.

This paper aims to achieve ultra-low phase noise for a voltage-controlled oscillator (VCO) through the adoption of a dual-tank configuration. The tank, based on a 1:1 transformer, incorporates an impedance scaling mechanism to reduce the equivalent parallel resistance of the tank without compromising its quality factor. Additionally, the tank provides a passive voltage gain from the drain to the gate of the core transistors, resulting in a reduction of the phase noise of the core transistors and aiding the start-up of the VCO. The tank also provides impedance peaks at the fundamental and second harmonic frequencies, leading to a flattened voltage waveform for half an oscillation cycle at the core transistors’ drains, further reducing their phase noise contribution. The VCO is fabricated in 180-nm CMOS technology. The VCO, with a die size of 0.14 mm2, demonstrates a state-of-the-art phase noise of -157.2 dBc/Hz at an offset frequency of 10 MHz, with the center frequency of 5.08 GHz. The oscillator’s peak Figure-of-Merit is 193.2 dBc/Hz.

This paper presents a powerline energy harvesting circuit to power wireless sensor nodes for powerline safety monitoring. The magnetic harvester is a ring-shaped nano-crystalline magnetic core, which transforms the ac powerline current to ac voltage. The major building blocks of the circuit are a buck-boost converter operating in discontinuous conduction mode (DCM) and a microcontroller unit (MCU) for maximum power point tracking (MPPT). The MPPT algorithm based on the perturb and observe senses the current flowing into the load and adjusts the duty cycle of the buck-boost converter to match the source impedance. The magnetic core delivers 6.98 W to an optimal$200\ \Omega$resistor directly attached to the core under the powerline current of 30 A. The output power of the proposed circuit is 4.86 W with the optimal load resistance of$R_{L}=250\ \Omega$, resulting in the conversion efficiency of 70%.

This paper presents an investigation into practical considerations in the minimization of Age of Information (AoI) for the system architectures with and without updating feedback. To study the impact of the critical characteristics of the energy harvesters on the computing accuracy of the average AoI, an RF energy harvester with a half-wavelength patch antenna array along with a 3-stage voltage rectifier is designed and prototyped with the center frequency of 2.655 GHz. A cryptography algorithm is also implemented on a$\mu$-controller unit to imitate the behavior of a practical processing unit. The paper shows measurement results and discusses the impact of the non-idealities in the energy harvester on the average AoI of the system. It also shows that the nonlinear power conversion profile of the energy harvester can increase the average AoI to over 300% compared to an ideal and linear energy harvester.

This paper investigates direct and indirect learning methods to develop deep learning digital predistortion (DL-DPD) models and apply the models to improve the linearity of a power amplifier (PA). The two methods are applied to class-AB and class-$\mathrm{F}^{-1}$ PAs designed with gallium nitride (GaN) on silicon carbide (SiC) high electron mobility transistors (HEMTs). The simulation results show that both direct and indirect DL-DPD methods improve the linearity of the class-AB PA by about 12dB and the class-$\mathrm{F}^{-1}$ PA by 11 dB, while the indirect method offers marginally better performance. The paper shows the direct learning method leads to significant improvement of the DL-DPD method based over the memory polynomial. It also presents the advantages of a BiLSTM based on the neural network architecture to design direct/indirect DPDs. Finally, it demonstrates that the DL-DPD can improve the linearity of class-AB and class-$F^{-1}$ PAs without architectural changes.

This paper proposes an M-PSK (M-ary Phase Shift Keying) polar transmitter, which features low power dissipation, low design complexity, and compact core size. The key blocks of the proposed transmitter are a ring oscillator and a charge control unit, which dynamically set the amount of charge to draw from the output node of the oscillator. The charge drawn from the output node effectively shifts the phase of the output voltage waveform of the oscillator. It enables the transmitter to perform the phase shift keying by controlling of the amount of the charge to draw. The transmitter is laid out in TSMC 180 nm CMOS process technology. Post layout simulations show that the transmitter can achieve the data rate of 40 Mbps with the error vector magnitude (EVM) of 3.7% for 16-PSK signals.

This paper presents a temperature compensated variable gain phase shifter (VGPS), composed of an active phase shifter followed by a variable gain amplifier (VGA). A temperature sensor adjusts the biasing of the phase shifter block and suppresses temperature effects on its insertion phase. Further, a self-compensation mechanism compensates the gain reduction at the VGA stage. The circuit is designed and laid out in 0.15 μm GaN HEMT technology. Post layout simulation results show that the proposed VGPS achieves over 180 °C of the relative phase shift for the frequency range of 4.5 to 5 GHz over the temperature ranging from 25 °C to 250 °C. The gain of the VGPS can be controlled from 5 dB to 17 dB.

This paper presents a compact backscatter modulator tag operating at 1.76 GHz. The modulator is composed of a series combination of a variable resistor and a varactor. These components modulate the backscatter signal by adjusting the input impedance or input reflection coefficient. A gallium nitride (GaN) on silicon carbide (SiC) high electron mobility transistor (HEMT) is adopted to implement the variable resistor. The modulator can support a variety of modulation schemes without increasing the size and design complexity. The tag can be continuously powered through ambient RF energy in the 2.45 GHz frequency band. A differential rectifier harvests RF energy to power a microcontroller eliminating the need for a battery. Experimental results show that the modulator can support 16-QAM with an error vector magnitude (EVM) of 1.73%. The measured RF to DC power conversion efficiency (PCE) of the energy harvester is 57.1% at a received power of 1.7 dBm.

This article presents a high temperature model for gallium nitride (GaN) high electron mobility transistor (HEMT) on silicon carbide (SiC). The proposed model for the channel resistance Rds is based on an empirical nonlinear model. The model is applied to design a resistive mixer of a high temperature transceiver for downhole communications through a systematic approach and estimate the performance of the mixer. The proposed model matches well with measurement results of the mixer and accurately estimates its performance at temperatures up to 250 °C. The model is also applied to obtain the optimal gate bias voltage of the mixer for a given temperature. The optimal bias voltage scheme reduces the conversion loss of the mixer by a factor of 8.4 dB at 250 °C under the local oscillator power of -10 dBm.