Collaborative multi-target detection in radar sensor networks

In many military and civilian applications, estimating the number of targets in a region of interest plays a primary role in performing important tasks such as target localization, classification, recognition, tracking, etc. Such an estimation problem is however very challenging since the number of targets is time-varying, targets' states are fluctuating, and various kinds of targets might appear in the field of interest. In this paper, we develop a framework for estimating the number of targets in a sensing area using Radar Sensor Networks (RSN): (1) the multi-target detection problem is formulated; (2) signals, interference (e.g., clutter, jamming, and interference between radars), and noise at radar sensors are modeled; and (3) a Maximum Likelihood Multi-Target Detection (ML-MTD) algorithm is proposed to combine received measurements and estimate the number of targets present in the sensing area. We evaluate multi-target detection performance using RSN in terms of the probability of miss-detection P(MD) and the root mean square error (RMSE). Simulation results show that multi-target detection performance of the RSN is much better than that of single radar systems.