This paper considers the waveform design for colocated MIMO radar to improve target detectability embedded in signal-dependent interferences, and maximize the SINR (signal-to-interference-plus-noise ratio). Due to constant modulus constraint, the problem is indeed non-convex and NP-hard. To address the challenge of this intractable problem, we propose a novel optimization framework for solving the problem based on convex relaxation. In the proposed algorithm, the original optimization problem is reformulated heuristically by introducing an auxiliary variable. Then it can be relaxed to a convex optimization problem and efficiently solved by alternative optimizing. Numerical results show that the proposed algorithm improves the beampattern performance significantly by generating much deeper notches at the interference directions, then obtains higher SINR over the state-of-the-art method.

本文考虑了协同定位 MIMO 雷达的波形设计，以提高嵌入在信号相关干扰中的目标可检测性，并最大化 SINR（信号干扰加噪声比）。由于恒模约束，问题确实是非凸和NP-hard的。为了解决这个棘手问题的挑战，我们提出了一种新的优化框架来解决基于凸松弛的问题。在所提出的算法中，通过引入辅助变量，启发式地重新表述了原始优化问题。然后它可以放松到一个凸优化问题，并通过替代优化有效地解决。数值结果表明，所提出的算法通过在干扰方向产生更深的陷波，显着提高了波束图的性能，