Yue Lin

This article presents a compact distributed power amplifier (DPA) for high-data-rate communication in 40-nm standard CMOS technology. By leveraging high-power-density cells, this DPA breaks the conventional compromise among output power, area, and bandwidth. With the adoption of the Ruthroff-type distributed balun, balanced signals are efficiently combined across a broad bandwidth in a compact area. Furthermore, the implementation of a twisted and embedded output matching network (OMN) ensures resonance with parasitic capacitance while avoiding extra area overhead. The DPA achieves a 16.6-dB gain and delivers an output power of 19 dBm with 21.4% power-added efficiency (PAE) at peak ${P} _{\text {sat}}$ , covering the bandwidth of 1.5–23.5 GHz from a 1.8-V supply. It reaches up to 150.5-Gb/s data rate with 14.4-dBm average output power in 128-quadrature-amplitude-modulation (QAM). In addition, this study supports 256-QAM at a data rate of over 110 Gb/s.

Millimeter-wave (mmWave) radar is increasingly used in smart environments for human detection due to its rich sensing capabilities and sensitivity to subtle movements. However, indoor multipath propagation causes severe ghost target issues, reducing radar reliability. To address this, we propose a trajectory-based ghost suppression method that integrates multi-target tracking with point cloud deep learning. Our approach consists of four key steps: (1) point cloud pre-segmentation, (2) inter-frame trajectory tracking, (3) trajectory feature aggregation, and (4) feature broadcasting, effectively combining spatiotemporal information with point-level features. Experiments on an indoor dataset demonstrate its superior performance compared to existing methods, achieving 93.5% accuracy and 98.2% AUROC. Ablation studies demonstrate the importance of each component, particularly the complementary benefits of pre-segmentation and trajectory processing.

Nowadays, Frequency-Modulated Continuous Wave (FMCW) radar systems are widely used. These radar systems usually suffer from undesired patterns include the interference from transmitter or LO, multiple reflections of static objects in the environment. Traditional baseband processing uses high-pass filters (HPF) to suppress these interferences; however, HPFs have both slow settling due to low cut-off frequency, and attenuation to real objects reflections at near frequency band. Some studies have tried to cancel the interference actively, however additional cancel errors are induced. In this paper, a novel active cancel methodology is proposed with pattern cancel DAC and remove the cancel error with feedforward compensation. It is verified on silicon to be effective to mitigate the interference and enhance the linearity of the FMCW baseband greatly.

This article presents a wideband digitally-controlled variable gain amplifier (VGA) with process voltage and temperature (PVT)-robust decibel-linear (dB-linear) characteristic, including two cascaded VGA cells, an exponential generator, a self-compensated control voltage generator and a self-adaptive current and voltage bias. The VGA cell is realized by a folded Gilbert-cell and it achieves a 2.4 GHz gain-independent bandwidth (BW). The exponential generator is realized by a parasitic bipolar transistor and it achieves a dB-linear gain control range of 59.7 dB with a gain error lower than 0.39 dB. A self-compensated control voltage generator is proposed to optimize the dB-linear gain response PVT robustness. The nonidealities that deteriorate the PVT robustness are analyzed and optimized. A self-adaptive current and voltage bias is proposed to alleviate the gain drift caused by the channel length modulation effect. Under PVT variations, the gain drift of the VGA is lower than 5.12 dB and its gain slope drift is lower than 3.99%. Based on the reconfigurability of the VGA’s dB-linear gain response, an automatic calibration flow is proposed to eliminate the gain response deviations between chips. Fabricated in a 40 nm standard CMOS process, the proposed VGA consumes 6.26 mW dc power under a 1.2-V supply voltage (excluding the output buffer) and occupies an active area of 0.036 mm2.

This article presents an ultra-wideband (UWB) low noise amplifier (LNA) with $g_{m}$ -assisted-feedback noise suppression technique for multi-band wireless communication. Compared with the traditional resistive feedback (RF) and source-follower feedback wideband LNAs, the proposed LNA blocks the noise current from the feedback resistor with a simple additional $g_{m}$ -cell, thus alleviating the $S_{11}$ -NF design tradeoff in the traditional structures. A source coupling transformer is also utilized to create $g_{m}$ -boosting and maintain a wideband flat gain. Fabricated in a 40-nm bulk CMOS process, the LNA prototype achieves a measured operating bandwidth (BW) of 1–21 GHz, measured NF of 1.95–3.1 dB, and $S_{21}$ of 20.6–22.4 dB, while consuming 21.6 mA from a 1.2-V power supply. It also supports UWB modulation signals with a data rate of 140 Gb/s. The proposed LNA reports the lowest NF among the recently reported bulk CMOS UWB LNAs.

Millimeter-wave (mmWave) radar has been widely applied in target detection. However, due to multipath and occlusion, radar often detects ghosts, especially in indoor environments. Existing solutions are mostly tailored to specific, simplified scenarios. To identify radar ghosts in diverse and complex indoor environments, we propose a data-driven approach. A thoughtful indoor radar ghost dataset is created with a multimodal data acquisition and automatic annotation system. And PairwiseNet, an end-to-end deep neural network adept at handling point-pair relationships within sparse point clouds, is proposed for radar ghost recognition. Multi-frame accumulation is also implemented in PairwiseNet. To further enhance PairwiseNet, an additional network incorporating grid maps and U-Net is developed for constructing environmental maps from sequential point clouds. This network is trained through cross-modal distillation, with a depth camera as the teacher. Finally, a series of experiments validate the effectiveness of the proposed method in identifying indoor radar ghosts and autonomously constructing environmental maps. The classification accuracy on the test set reaches 96.0%, accurately identifying ghosts in the vast majority of cases.

This article presents the design methodology of millimeter-wave wireless power transfer (WPT) system and the implementation of a 24-GHz beam-steerable multinode WPT system. We first propose a new figure of merit (FoM) to evaluate the cost efficiency and spatial coverage of WPT system. Considering loss in a practical system, we establish a mathematical model to optimize FoM and draw design curves to get some insights into tradeoff among frequency, area cost, and power efficiency at the target distance. To demonstrate the utility of the design methodology, we implement a 24-GHz WPT system consisting of an 8 $\times$ 8 phased-array transmitter with 56-dBm effective isotropic radiated power (EIRP) and 5 $\times$ 5 high-gain angle-deflection planar rectennas with LED loads. The spacing between the transmitter (TX) and receivers (RXs) is 1 m and the aperture size of TX antenna array and high-gain receiving antenna is 6 cm $\times$ 6 cm. Due to the beam-steering capability of TX, no calibration is required for the precise placement of individual rectennas. The measured results show that all LEDs can be lightened with maximum 5.7-dBm dc power and custom patterns can be displayed using such LED array by sequential beamforming of the TX.

Automatic fall detection plays a significant role in monitoring the health of senior citizens. In particular, millimeter-wave radar sensors are relevant for human pose recognition in an indoor environment due to their advantages of privacy protection, low hardware cost, and wide range of working conditions. However, low-quality point clouds from 4D radar diminish the reliability of fall detection. To improve the detection accuracy, conventional methods utilize more costly hardware. In this study, we propose a model that can provide high-quality three-dimensional point cloud images of the human body at a low cost. To improve the accuracy and effectiveness of fall detection, a system that extracts distribution features through small radar antenna arrays is developed. The proposed system achieved 99.1% and 98.9% accuracy on test datasets pertaining to new subjects and new environments, respectively.

Nowadays, Frequency-Modulated Continuous Wave (FMCW) radar systems are widely used in many domains, such as automotive and ADAS, industrial sensing, smart home applications and so on. However, these radar systems do suffer from undesired patterns such like interference from transmitter or LO, multiple reflections of static objects in the environments. Conventional systems employ high-pass filters (HPF) to remove the interference, those filters eliminate the interference and close objects blockage well, but they do attenuate all low frequency objects simultaneously no matter it is blocker signals or not. Also, such implementations tend to have smaller gain at the front end of the HPF, that cause bad noise and linearity balance. In this paper, a novel pattern cancel DAC system technique is proposed to mitigate the undesired pattern and enhance the signal processing capability of the FMCW systems greatly.