Jamming Power Allocation for MIMO Radar Localization Aided by Reconfigurable Intelligent Surface

As a promising technology, reconfigurable intelligent surface (RIS) can customize the wireless channel through the adjustmentof its phase, and strengthen the localization performance and anti-jamming capability of multiple input multiple output (MIMO) radar. Thus, it is imperative to investigate a jamming strategy against the RIS-aided MIMO radar localization system. In this paper, we consider a jamming power allocation problem for RIS-aided MIMO radar based on cramér-rao bound (CRB) criteria. To this end, we first derive the CRB regarding target location estimation of RIS-aided MIMO radar in the presence of jammers. Subsequently, we formulate the jamming power allocation problem to maximize the minimum of the average CRB of RIS-aided MIMO radar via jointly optimizing the jamming power and RIS phase under the power budget and peak power constraints. Due to the variable coupling and high non-convexity of CRB, the optimization problem is extremely challenging to tackle. To solve the non-convex problem, we develop an alternating iterative algorithm with respect to the jamming power allocation and RIS phase design optimization subproblems. Specifically, we first transform the jamming power allocation subproblem into the form of linear programming for solution. Then, the RIS phase design subproblem is relaxed by the arithmetic-harmonic mean inequality to circumvent the complex matrix inversion, and converted into a fractional programming problem that can be solved through quadratic transformation approach. Owing to a degeneration of the solution induced by the inequality relaxation, we further propose a tightening-relaxation strategy to improve the solution quality, and re-optimize the RIS phase based on the previously derived solution. The ultimate solution can be achieved by alternately optimizing the jamming power and the re-optimized RIS phase. Simulation results are provided to demonstrate the effectiveness of the proposed algorithm.