Codesign of Unimodular Waveform and Power Allocation for Polarimetric Radar

Joint design of waveforms and filters with adaptive power alloction has important applications in improving the target detection performance of polarization radar. The resulting problem is maximizing polarimetric radar target detection performance through unimodular waveform and filter design with power allocation, which is a non-convex and NP-hard problem. Existing methods mainly solve this problem by relaxing the objective function and obtaining filters with matrix inversion, which may introduce relaxation errors and high complexity. We construct a method based on unified product manifold to directly address this problem without matrix inversion. Specifically, a product manifold space is constructed to satisfy the constraints of unimodular waveform, filters and power. Then, an unconstrained problem is obtained by projecting the problem onto the unified product manifold. To solve this problem, a parallel conjugate gradient algorithm is derived. This algorithm can adaptively change the step size and fully explore the manifold space. Simulation results show that the proposed method can achieve better performance in shorter time than existing methods.