Distributed Radar Target Detection with Doppler Channel Maximum Quantization

In distributed radar, a local radar site may perform coherent accumulation for a better signal-to-noise (SNR) output and it is often unnecessary to transmit statistics in all Doppler channels to a fusion center (FC), if the probability that more than one target is present in a range cell is low. In this paper, we propose a low-communication-cost algorithm for distributed detection with locally coherent multi-pulse accumulation. Under the assumption that no more than one target is present in a range cell, we take the maximum of the test statistics over all Doppler channels to represent the presence of a target and then quantize the statistic maximum with the least squares quantization (LSQ) algorithm for a low communication bandwidth requirement. An optimal fusion rule based on the Neyman-Pearson (NP) criterion is formulated to make a global decision under a constant false alarm rate. Numerical results indicate that compared with traditional algorithm, the proposed algorithm with 3-bit LSQ quantization results in less than 0.5 dB SNR loss whereas the communication bandwidth is reduced by 99.71%.