Computational design of optimal waveforms for MIMO radar via multi-dimensional iterative spectral approximation

We propose an approach to design waveform sets with good auto- and cross-correlation properties. The designed waveform sets can be used for multiple-input multiple-output radars to minimize the cross-interference, and consequently improve the radar performance. Firstly, we transform the original correlation optimization problem into a spectral approximation problem. Secondly, we define the objective function based on the square error between the designed waveform matrix and the ideal one. Finally, the problem is solved by using an algorithm based on alternating projection and phase retrieval. We further improve the proposed method, and derive an extended version to design waveforms with sparse spectrum and good correlation property for combating electronic jamming. The algorithms are computationally efficient because their main steps are based on fast Fourier transform. The effectiveness of the proposed approach is demonstrated by numerical simulations.