This article presents an 8-element dual-polarized phased-array transceiver (TRX) front-end IC for millimeter-wave (mm-Wave) 5G new radio (NR). Power enhancement technologies for power amplifiers (PA) in mm-Wave 5G phased-array TRX are discussed. A four-stage wideband high-power class-AB PA with distributed-active-transformer (DAT) power combining and multi-stage second-harmonic traps is proposed, ensuring the mitigated amplitude-to-phase (AM-PM) distortions across wide carrier frequencies without degrading transmitting (TX) power, gain and efficiency. TX and receiving (RX) switching is achieved by a matching network co-designed on-chip T/R switch. In each TRX element, 6-bit 360° phase shifting and 6-bit 31.5-dB gain tuning are respectively achieved by the digital-controlled vector-modulated phase shifter (VMPS) and differential attenuator (ATT). Fabricated in 65-nm bulk complementary metal oxide semiconductor (CMOS), the proposed TRX demonstrates the measured peak TX/RX gains of 25.5/21.3 dB, covering the 24?29.5 GHz band. The measured peak TX OP1dB and power-added efficiency (PAE) are 20.8 dBm and 21.1%, respectively. The measured minimum RX NF is 4.1 dB. The TRX achieves an output power of 11.0?12.4 dBm and error vector magnitude (EVM) of 5% with 400-MHz 5G NR FR2 OFDM 64-QAM signals across 24?29.5 GHz, covering 3GPP 5G NR FR2 operating bands of n257, n258, and n261.

Recently, there has been increased attention toward 3D imaging using single-pixel single-photon detection (also known as temporal data) due to its potential advantages in terms of cost and power efficiency. However, to eliminate the symmetry blur in the reconstructed images, a fixed background is required. This paper proposes a fusion-data-based 3D imaging method that utilizes a single-pixel single-photon detector and millimeter-wave radar to capture temporal histograms of a scene from multiple perspectives. Subsequently, the 3D information can be reconstructed from the one-dimensional fusion temporal data by using an artificial neural network. Both the simulation and experimental results demonstrate that our fusion method effectively eliminates symmetry blur and improves the quality of the reconstructed images.

在车辆到基础设施（V2I）通信中，毫米波波束宽度窄、设备移动性高，有效的毫米波波束选择是一项关键而又富有挑战性的任务。针对波束搜索过程开销大的问题，文章提出了一种基于雷达和位置信息的毫米波波束选择算法。所提算法使用Top-k分类度量在激光雷达数据和波束跟踪通道数据集上训练神经网络结构，并且使用全局上下文网络（GCNet）模型预测最佳波束对。仿真结果表明，所提算法性能在分类识别准确率上有明显提升，与此同时，仅需搜索5个波束便能接近穷举搜索的性能，大大减少了波束搜索开销。