A Joint Allocation Method for RF Stealth Resources in Distributed Networked Radar Multi-Target Tracking Based on Penalized SCA

In multiple target tracking scenarios, distributed networked radars face the challenge of maintaining tracking accuracy while minimizing radiation resource consumption to enhance radio frequency (RF) stealth. To address this issue, this paper proposes a joint allocation method for radar transmit power and dwell time. First, the Bayesian Cramér-Rao lower bound (BCRLB) is derived, which incorporates node selection variables and physical radiation parameters. However, the dimensional differences between position and velocity often cause optimization weight distortion. To overcome this, a position selection matrix is introduced to construct a dimensionally unified error lower bound, thereby eliminating magnitude interference. Subsequently, a nonlinear resource allocation problem is formulated to minimize the weighted sum of normalized transmit power and dwell time. This formulation is subject to two constraints: a preset tracking accuracy threshold and an upper limit on the total system radiation resources. Consequently, the resulting optimization model contains non-convex constraints. To handle this non-convexity, slack variables are introduced to convert non-convex constraints into penalty terms. This approach expands the iterative feasible region and prevents the infeasibility issues commonly encountered in the traditional sequential convex approximation (SCA) algorithm. Furthermore, the penalty factors are updated dynamically, and the convex subproblems are solved using the interior-point method. Simulation results demonstrate that, compared with the uniform resource allocation algorithm, the proposed method can adaptively allocate tracking resources on demand. It effectively reduces resource consumption and enhances the RF stealth capability of distributed networked radars.