A Block Coordinate Descent Approach to Design Constant Modulus Waveform for MIMO Radar with Low-Resolution DACs

Abstract

In recent years, the problem of waveform design for multiple-input multiple-output (MIMO) radar has drawn extensive attention. Noted that the most of existing methods are devoted to systems with infinite-resolution digital-to-analog converters (DACs), which increase the hardware-cost and circuit power consumption. In this paper, the problem of constant modulus (CM) waveform design for MIMO radar which deploys low-resolution phase-only DACs is investigated. To achieve better radiation performance, the integrated sidelobe to mainlobe ratio (ISMR) is considered as the optimization metric. However, the optimization problem is difficult to solve due to the quadratic fractional function and discrete constraints. Towards this end, we employ the Dinkelbach algorithm to transform the quadratic fractional problem into a parametric subtractive-form problem. Leveraging on the block coordinate descent (BCD) method, the original multi-dimensional problem is decomposed into multiple smaller dimensional problems. We sequentially update each variable block while fixing the remaining blocks to solve each subproblem. Finally, we analyze the performance of the proposed algorithm in terms of convergence and beampattern of the wave-form, in comparison with other algorithms through numerical simulations. The results show that the proposed method can effectively achieve a low ISMR.