Enhancing Direct Position Determination in Distributed Base Station Systems Through Time-Varying Quantization Design

In distributed base station (DBS) systems, the traditional direct position determination (DPD) methods entail the transmission of raw data to the fusion center (FC), straining transmission bandwidth and hardware resources. One-bit analog-to-digital converters (ADCs) have emerged as a promising solution. However, the conventional zero-thresholding quantization (ZQ) scheme exhibits significant performance degradation. In response, we introduce a one-bit time-varying quantization (TQ) strategy tailored for DBS systems and develop a Rao-test-based method under the one-bit TQ scheme [TQ-one-bit Rao test (TQ-OBRT)], enabling simultaneous target detection and localization. To evaluate the performance of the proposed method, the one-bit Cramér–Rao lower bound (CRLB) is derived. Finally, we obtain the optimal quantization (OQ) scheme by minimizing the one-bit CRLB and introduce a simpler random quantization (RQ) scheme without iterative processing. Simulation results demonstrate the superior performance of the proposed OQ and RQ schemes compared with the ZQ scheme in terms of target detection and localization.

I. Introduction

Acommonly used approach for target detection and localization involves using a multiple BS system with widely distributed nodes, as it leads to significant enhancements in performance and resolution [1], [2], [3], [4], [5], [6], [7], [8], [9], [10]. In the context of DBS systems, the use of high-resolution ADC units for each receiver antenna has been a customary requirement [11]. However, the application of high-resolution ADCs along with multiple BSs leads to exponential growth in bandwidth and power requirements [12], [13], [14], [15], [16], [17]. Consequently, the demand for cost-effective and energy-efficient BS designs has escalated, aiming to ensure adherence to constraints related to bandwidth, power consumption, and hardware costs [18], [19], [20].