This work presents an IEEE 802.15.4z-compliant UWB SoC with a TX spillover cancellation enabling shared-antenna fullduplex operation. The proposed PVT-robust FIR filter based SIC achieves 27 dB average rejection $(\sigma=1.4 \text{dB})$, improving the RX NF from 21 to 4 dB. A reconfigurable TIA supports both of low-power ranging and wide bandwidth radar modes using a feed-forward architecture. Child presence detection is demonstrated in a real vehicle using a certified infant dummy and human subjects, including worst-case legroom scenarios.

The latest wireless local area networks (WLANs) including IEEE 802.11be have become increasingly sensitive to synchronization errors, as per stringent requirements in pursuit of high-end performance. In particular, stemming from common clock mismatch between transceivers, carrier and sampling frequency offsets (CFO and SFO) jointly induce time/frequency-wise phase rotations whose detrimental impact grows with wider channel bandwidth. This paper presents a comprehensive study into the coupled CFO and SFO impairments in emerging WLANs, with up to 320 MHz wideband operation. For better understanding of their joint effect, we introduce a useful concept of effective CFO (eCFO) referring to individual frequency drift per subcarrier. Our in-depth analysis and scrutiny into existing receiver designs reveal that wideband WLANs cannot stick to conventional synchronization mechanisms. To tackle this situation, we propose enhanced approaches for initial estimation and compensation of both frequency offsets, exploiting the frequency-dependent nature of eCFO. Extensive hardware-benchmarked simulations demonstrate the effectiveness and robustness of our strategies in coping with a variety of frame reception across wideband scenarios. Ultimately, the proposed receiver design achieves near optimal end-to-end performance by mitigating the impairments, even under the most challenging operation regime.

This paper presents 60-and-77GHz FMCW radar RFICs with 4T/4R MIMO and cascading capability in 28nm CMOS. The RFICs demonstrate excellent RF/analog/digital performance, including $>14.8 \text{dBm}$ TX $P_{\text{SAT }}$, 7-bit phase resolution, and $>36 \text{MHz}$ RX IF BW, with a compact $4.5 \times 4.5 \text{mm}^{2}$ chip size and $<2.1 \mathrm{W}$ power consumption. The RFICs support multiple radar modes and are validated through OTA tests, demonstrating their suitability for various radar applications.

A reconfigurable UWB receiver supporting the IEEE 802.15.4ab narrowband-assisted multi-millisecond (NBA-MMS) ranging scheme is presented. To minimize the area overhead of NB support, a current-amplification-based capacitance-boosting (CBST) technique is proposed, achieving 49% baseband area reduction with only 0.025 mm2 for both I and Q paths. In addition, a PVT-robust RSSI based on a mismatch-cancellation amplifier (MCA) enables reliable NB-to-UWB gain setting, achieving ± 0.5 dB nominal accuracy and within ± 1.9 dB across PVT variations. The receiver supports both NB and UWB operation while consuming 78.3 mW and occupying 0.52 mm2.

This article presents a radar-enabled impulse radio ultrawideband (IR-UWB) system by adopting a self-interference-resistant RF transceiver. It demonstrates that child presence detection (CPD) using IR-UWB radar is feasible through the implementation of a radar system that includes physical (PHY), media access control (MAC), and application (APP) processors. The RF transceiver shows high receiver (RX) sensitivity and high transmitter (TX) peak power with low power. In addition, a radar-dedicated TX is used to improve the TX signal isolation to the RX. To achieve the high RX chain gain and low noise figure (NF) of RX, an RX front end (RX-FE) low-noise amplifier (LNA) and voltage-to-current (V2I) with a 50% duty cycle local oscillator (LO) is implemented. Furthermore, a notch pulse shaping filter (PSF) technique is proposed to meet spectrum emission mask (SEM) specifications with low power consumption. An LNA with wide gain ranges is also proposed to improve the sensitivity and support various applications related to radar and positioning applications. The UWB transceiver achieves -96.5-dBm sensitivity at BPRF 6.81 Mb/s and 14-dBm TX peak power and supports radar operation with UWB channels 5 and 9.

Unlike received signal-strength-indicator (RSSI)-based localization methodology of Wi-Fi and BLE, which have a distance error of several meters or more [1], impulse-radio ultra-wideband (IR-UWB) using narrow pulse width enables accurate distance information within a few centimeters [2]. Additionally, IR-UWB has demonstrated stand-alone radar capabilities [3], enabling expanded radar applications such as child presence detection, augmented reality, and motion sensing. With the expansion of UWB applications, multiple UWB ICs are widely used in various areas such as IoT devices, mobile devices, and automobiles. Consequently, there is growing concern in the market about the potential for these devices to interfere with each other's communication. Furthermore, the channel environment is getting worse as the 6GHz band is becoming increasingly crowded due to both Wi-Fi and cellular New Radio (NR) bands. With the UWB solution of previous works [3], [4] only supporting two channels (ch5 and ch9), there are limitations in handling multiple users effectively. To address these issues, UWB technology needs to support multiple channels to make better use of the channel environment. In addition, to minimize the number of antennas, previous works [3]–[5] have adopted a cost-effective structure that shares TX/RX antennas using RF switches. In [5], external RF switches based on the silicon-on-insulator (SOI) process are used to share TX and RX antennas; however, the use of external RF switches and the system-in-package (SiP) incur additional costs, which can be challenging for mass production. In [3], [4], the RF switches are fully integrated using the CMOS process, but the high insertion loss of the internal RF switch increases RX noise figure (NF), leading to degraded sensitivity.

Ever-growing applications, such as 5G communication, deep learning, advanced driver-assistance systems (ADAS), and extended reality (XR), have fueled demand for increased computing power and per-pin interface bandwidth. Recently, four-level pulse-amplitude modulation (PAM4) has been adopted as a solution [1-3]: the throughput is doubled without increasing the baud (Nyquist) rate. Compared to a conventional non-return-to-zero (NRZ) signaling, PAM4 requires more design effort: varying from the precise design of I/O circuits to the off-chip characterization. This is in part due to SNR degradation and an increased switching jitter (SWJ). For a $1^{\text{st}}$ -order low-pass filter with a Nyquist-frequency cutoff, SWJ is 35% for the middle eye and 51.2% for the top and bottom eyes [4]. Maximum-transition-avoidance (MTA) encoding [3] can be used to reduce SWJ, but at the cost of additional encoder/decoder hardware and an auxiliary channel to compensate for data loss.

This paper presents a radar-enabled IR-UWB system by adopting a self-interference resistant RF transceiver. It shows that child presence detection using IR-UWB radar is possible by implementing a radar system including PHY, MAC, and APP processor. RF transceiver shows high receiver sensitivity and transmitter peak power with low power. Furthermore, a radar dedicated transmitter is used to improve the transmitter signal isolation to the receiver. Proposed notch pulse shaping filter is adopted to satisfy the spectrum emission mask specification. Additionally, a low noise amplifier with wide gain ranges is proposed to improve the sensitivity and support various applications such as radar and positioning detection application. The UWB transceiver achieves -96.5 dBm sensitivity at BPRF 6.81 Mb/s and 14 dBm transmitter peak power and supports the UWB channel 5 and 9.

Givens rotation based channel state information (CSI) feedback has been adopted as a limited feedback technique for beamforming in the wireless local area networks (WLANs). On the other hand, the long term evolution (LTE) systems utilize the predefined codebook for the CSI feedback. In this paper, we propose a dual CSI feedback technique for next generation WLANs, which combines the codebook and Givens rotation to derive the benefits of both techniques. Machine learning (ML) technique is adopted for the improved codebook design. The extensive simulation is carried out via IEEE 802.11be link-level simulator to verify the performance of the proposed scheme. It shows that the proposed scheme can reduce feedback overhead by more than 50% compared to the scheme adopted in current WLANs, and enhances the throughput significantly.

This article presents a fully integrated ultrawideband (UWB) system-on-chip (SoC) that complies with IEEE 802.15.4/4z standard, providing precise positioning and secure communication capabilities. The proposed UWB SoC includes a radio frequency (RF) transceiver, a modem, a microcontroller unit (MCU), an eFlash, a power management unit (PMU), clock generation IPs, and peripheral blocks. The RF transceiver integrates RF switches and comprises two transmitters and three receivers. The proposed nonuniform quantization digital-to-analog converter (NQ-DAC) is used in the transmitter to enhance the quality of pulse sidelobe, and a receiver I/Q mismatch estimation technique using a double-sideband (DSB) transmitter as a reference is proposed. The UWB SoC operates in the frequencies at 6.5 GHz (Ch5) and 8 GHz (Ch9), achieving a location estimation precision of a few centimeters. The prototype UWB SoC is fabricated in a CMOS 28-nm process, and the experimental results under base pulse repetition frequencies (BPRFs) 6.81 Mbps show a receiver sensitivity of −96.1 dBm and a transmitter output peak power of 14.25 dBm, achieving the state-of-the-art performance. The prototype UWB SoC demonstrates the power consumption of 198.6 mW during transmitter operation and 309.7, 442.5, and 528.9 mW for 1-RX, 2-RX, and 3-RX during packet reception. The proposed UWB SoC architecture and techniques provide a practical and efficient solution for IEEE 802.15.4/4z compliant UWB systems.

Based on robust 28-nm embedded flash (eFlash) process, IoT One-chip for high-speed and low power applications which MCU-chip (10Mb eFlash) and connectivity-chip (BLE/Zigbee) are integrated for the first time. By introducing new devices on 28-nm low-power eFlash process, high-speed ( random read), ultra-low power sleep mode current, 10/13mA RF current at Tx/Rx mode) and robust reliability (-40 ~ 125°C stable operation, 100K cycle endurance, 150C/RT retention up to 200K hours) are achieved. LDD-first IO transistor with low Vth (~0.5V) for low-Vdd (~1.0V) operation [1] and ultra-low leakage (ULL) SRAM bit-cell (0.1x vs. normal) supporting low sleep mode chip current are applied to extend battery life-time. Stable endurance and high (/low)-temperature retention after cycling stress are achieved by robust split-gate type eFlash cell.

Bluetooth and Wi-Fi are the most widely used wireless technologies because they use unlicensed spectrum and are widely deployed on the latest mobile devices. For seamless Wi-Fi connectivity in mobile environments, the mobile device should maintain the information of adjacent access points (APs) through the scanning procedure, which often consumes a significant amount of energy and time. In this paper, we develop SplitScan that enables mobile devices to share Wi-Fi scanning information with adjacent stations (STAs) via Bluetooth packet exchange. We evaluate its performance through experiment with a testbed implementation as well as extensive simulation. The results show that SplitScan saves vonsiderable energy and time during the Wi-Fi scanning process.

In this paper, we consider a multi-user (MU) radio-frequency (RF) beam training scenario with user selection. We propose a new beam training scheme that alleviates the latency issue in the conventional IEEE 802.11ad by adopting a sequential downlink-downlink transmit sector sweep combination. Then, we analyze the average rate performance of several RF beam training schemes in two different asymptotic scenarios. In addition, we characterize the performance gain of the proposed method over other schemes. Our analytic results confirm that the proposed method achieves the average rate performance of the optimal fullsearch method in the asymptotic region with much reduced training overhead. It is shown from the simulation results that the proposed scheme outperforms the conventional beam training schemes and achieves a 35% performance gain over the optimal full search scheme when considering the beam training overhead in practical MU millimeter-wave channel environments. We also confirm that our analytical results match well with the numerical results.