中文摘要：

多输入多输出雷达的波形优化基于初始参数估计值，通常对参数估计误差较敏感。稳健波形设计通过将参数不确定性显式包含进优化问题以减轻敏感性。基于克拉美-罗界，在参数不确定凸集上考虑改善最差参数估计性能的稳健波形优化问题。提出一种优化波形相关阵的迭代算法以改善最差参数估计性能。迭代中的每步都可通过凸松弛求解。仿真表明，相对于不相关波形，所提方法可显著提高最差参数估计性能。

英文摘要：

Waveform optimization of MIMO radar, which depends on the initial parameters estimation, is often sensitive to error estimation of parameters. Robust waveform design attempts to alleviate this kind of sensitivity with incorporating a parameter uncertainty model in the optimization problem explic- itly. The robust waveform optimization is considered to improve the worst-case performance of parame- ter estimation over a convex uncertainty model, which is based on the Cramer-Rao bound （CRB）. An iterative algorithm is proposed to optimize the waveform covariance matrix （WCM） so that the worst- case performance can be improved. Each iteration step in the proposed algorithm could be solved with a convex relaxation. Numerical results show that compared to uncorrelated waveforms, the worst-case performance can be improved considerably by the proposed method.