针对到达时间差被动定位任务,研究了分布式雷达的部署策略优化问题,以提升系统的定位和监视性能。现有研究大多仅关注节点位置的优化,而未充分考虑节点法线指向与系统监视性能间的耦合关系。文中结合实际应用场景,考虑了节点位置与法线指向的联合优化策略,构建了面向定位任务的单目标优化问题以及兼顾定位和监视任务的多目标优化问题。由于这些优化问题具有复杂耦合约束和非凸特性,解析解难以获得。文中提出一种区域约束多目标粒子群算法(RC-MOPSO),用以求解最优部署策略。该算法通过在粒子初始化和更新过程中引入约束区域,确保粒子在迭代过程中始终满足复杂耦合约束条件。仿真结果表明,所提方案实现了定位和监视性能的最优平衡,相较于随机部署方案表现出显著优势,同时对辐射源发射功率估计误差表现出较强的鲁棒性。

This paper addresses the optimization of deployment strategies for distributed radar systems in time difference of arrival localization tasks, aiming to enhance both localization and surveillance performance. Existing research typically focuses solely on optimizing radar node positions while neglecting the coupling between node orientation and surveillance performance. This paper introduces a joint optimization strategy for node placement and orientation, establishing a framework that addresses both single-objective localization optimization and multi-objective optimization, effectively balancing localization and surveillance tasks. The complex coupled constraints and non-convex nature of these optimization problems make analytical solutions difficult. Therefore, the paper also proposes a region-constrained multi-objective particle swarm optimization (RC-MOPSO) algorithm to find the optimal deployment strategy. This algorithm ensures that particles satisfy the complex constraints throughout the iterative process by introducing regional constraints during initialization and updates. Simulation results demonstrate that the proposed strategy achieves an optimal balance between localization and surveillance performance, significantly outperforming random deployment schemes, and exhibiting strong robustness to errors in radiation source power estimation.

TN957

国家自然科学基金资助项目(62201048, 62371046)