LPI-Based Robust Node Selection and Power Allocation Scheme for Multiple Targets Tracking in Colocated MIMO Radar Network

• Published in: IEEE transactions on aerospace and electronic systems Vol. 61; no. 4; pp. 10394 - 10409
• Main Authors: Ding, Lintao, Long, Fei, Shi, Chenguang, Wang, Jian, Zhou, Jianjiang
• Format: Journal Article
• Language: English
• Published: New York IEEE 01.08.2025; The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
• Subjects: Accuracy; Colocated multiple-input multiple-output (C-MIMO) radar network; Constraints; convex optimization; Convexity; Error reduction; Interception; low probability of intercept (LPI); Measurement; Multiple target tracking; multiple targets tracking (MTT); Optimization; Quality of service; Radar; Radar cross-sections; Radar networks; Radar tracking; Resource management; Resource utilization; robust node selection and power allocation (RNS-PA); Robustness; Robustness (mathematics); Target tracking; Tracking
• ISSN: 0018-9251; 1557-9603
• DOI: 10.1109/TAES.2025.3560607

In this study, a low probability of intercept (LPI)-based robust node selection and power allocation (RNS-PA) scheme is proposed for multiple targets tracking (MTT) within a colocated multiple-input multiple-output (C-MIMO) radar network. The core of the RNS-PA scheme is to adaptively coordinate the radar-to-target assignment and transmit power to improve the LPI performance of the radar network, subject to the MTT accuracy requirements and system resource budgets. Unlike the existing resource-aware design algorithms that prioritize total resource consumption as the optimization objective (commonly resulting in decreased resource utilization and suboptimal solutions), we employ the closed-form expression for the worst-case probability of the C-MIMO radar network being intercepted by hostile targets as the optimization criterion. It is shown that the RNS-PA scheme is formulated as a nonlinear and nonconvex optimization problem. By evaluating whether all targets can satisfy the predefined tracking accuracy demands, the original problem is transformed into two subproblems: first, accelerating the reduction of MTT error under a given interception probability threshold (when the original problem is infeasible), and second, minimizing the worst-case interception probability subject to the MTT accuracy threshold constraints (when the original problem is feasible). We verify that the interception probability constraint forms a convex set and then propose a fast, comprehensive solution method to tackle the optimization problem efficiently. Simulation results verify the robustness and effectiveness of the proposed RNS-PA scheme.