Design of multiple unimodular waveforms with low auto- and cross-correlations for radar via majorization-minimization

We develop a new efficient method for designing unimodular waveforms with good auto- and cross-correlation properties for multiple-input multiple-output (MIMO) radar. Our waveform design scheme is conducted based on minimization of the integrated sidelobe level of designed waveforms, which is formulated as a quartic non-convex optimization problem. We start from simplifying the quartic optimization problem and then transform it into a quadratic form. By means of the majorization-minimization technique that seeks to find the solution of a corresponding quadratic optimization problem, we resolve the design of waveforms for MIMO radar. Corresponding algorithms that enable good correlations of the designed waveforms and meanwhile show faster convergence as compared to their counterparts are proposed and then tested.