Geometric optimization for joint localization and surveillance in distributed passive radar

This study addresses the geometric optimization of distributed passive radar systems, specifically the joint optimization of node positions and array orientations, to enhance both localization accuracy and surveillance coverage. A hybrid TDOA/AOA localization model is formulated, and its Cramér-Rao Lower Bound (CRLB) is analytically derived to quantify the impact of node geometry on localization accuracy. Under resource constraints, we introduce a node-key subarea allocation model (NKAM), which integrates spatial constraints between nodes and high-probability source regions (termed key subareas, KSs) to enable quantitative allocation strategy design. Based on this model, a multi-objective optimization framework is formulated to simultaneously maximize localization accuracy and surveillance coverage. Unlike traditional approaches that optimize only node positions, this work incorporates array orientations and NKAM-based assignments as decision variables, yielding a joint optimization problem for node geometry under NKAM-induced constraints. To address the non-convexity and complexity constraints of the problem, a multi-swarm relaxed-dominance-based multi-objective particle swarm optimization (MSRD-MOPSO) algorithm is proposed, efficiently generating a well-distributed Pareto-optimal front. Extensive simulations demonstrate the superiority of the proposed method over conventional deployment strategies, as well as its adaptability to diverse scenarios.