Detection and Localization of One-Bit Signal in Multiple Distributed Subarray Systems.

Lihua Ni,Di Zhang,Yimao Sun,Ning Liu,Jing Liang,Qun Wan
DOI: https://doi.org/10.1109/tsp.2023.3300634
IF: 4.875
2023-01-01
IEEE Transactions on Signal Processing
Abstract:A multiple distributed subarray (MDS) system usually transmits the raw data to the fusion center (FC), where the target detection and localization can be processed. Transmitting raw data requires large bandwidth and is power-consuming, so that technologies such as one-bit analog-to-digital converters (ADCs) have been applied to reduce the volume of data. This paper focuses on the detection and localization problems in a MDS system with one-bit ADCs. A Rao test-based method for one-bit data, called one-bit Rao test (OBRT), is proposed, which handles the detection and localization in one framework. Since the quantization is significant to the performance, an optimal one-bit quantizer is designed for the proposed detector and location estimator. To remove the performance loss caused by one-bit quantization, we analyze the asymptotic detection performance of the proposed method compared with the infinite-bit (analog measurements) generalized likelihood ratio test (IBGLRT)-based method. Furthermore, we provide a closed-form Cramér-Rao lower bound (CRLB) for one-bit direct position determination (DPD) as a benchmark to evaluate the localization performance of the proposed OBRT. The theoretical analysis and simulation results show that 1) the detection performance loss of the proposed method can be removed by increasing 0.57 times samples or 1.5 times nodes; 2) compared to the IBGLRT, the localization gap of the OBRT is $\boldsymbol{\pi/2}$ in low-SNR scenarios.
What problem does this paper attempt to address?