MIMO Radar Transmit Beampattern Matching Based on Block Successive Upper-bound Minimization

Waveform design with considering discrete phase constraint at the design stage tends to be pertinent in the emerging radar designs, especially since the digital to analogue converters are limited in the resolution. However, this constraint confines the degree of freedom to be only the waveform phase, which should be selected from a limited alphabet. In this paper, we aim to approximate a desired beampattern closely by designing the transmit waveform while considering the discrete phase constraint at the design stage. To this end, we consider a novel
$\ell_{p}$
-norm metric to achieve quasi-equiripple beampattern, which reduces the interference from the undesired directions. This problem leads to a NP-hard and non-convex optimization problem, where to efficiently solve it, we utilize the BSUM algorithm which successively optimizes the objective function by optimizing a certain upper bound of the original objective in a coordinate wise manner. In the numerical results, we show the performance of the proposed method and compare it with the state-of-the art.