MISL-based colocated MIMO radar waveform coefficients matrix design for beampattern matching

In this paper we focus on waveform coefficient matrix (WCEM) design to perform matched beampattern design with lower sidelobe level for generalized MIMO configurations. First, the relationship between transmit waveforms and beam-pattern, and the limitation of conventional waveform covariance matrix (WCVM) design are briefly discussed. Then a power approximation method (PAM) by minimizing both integrated sidelobe level (ISL) and mean square error (MSE) for beampattern formation is proposed. Based on the proposed method, two optimization algorithms are designed according to the different types of WCEM. The first algorithm is to solve a semi-definite programming (SDP), thus convex toolbox is directly used, while the second algorithm is to solve a nonconvex quadratically constrained quadratic programming (QCQP), then randomized semi-definite relaxation (SDR) is proposed to determine feasible solutions close to the solution to the original problem. Finally, numerical comparisons with classical methods demonstrate that the proposed algorithms can obtain well-matched beampatterns with lower sidelobe levels.