Sequence Set Design with Min-Max Criterion Using Majorization-Minimization

This paper investigates the problem of designing a sequence set with both good auto- and cross-correlation properties, which is required in lots of communication systems and active sensing, e.g., code-division-multiple-access (CDMA) cellular systems and multiple-input-multiple-output (MIMO) radar systems. In order to achieve uniform correlation sidelobes of the designed sequence, we design the sequence set construction algorithm using the min-max criterion. But the sequence set design problem at hand is highly nonlinear and non-convex. To address the optimization problem arising from the sequence set design, we propose a lightweight and efficient iterative algorithm based on majorization-minimization and block coordinate descent methods. Numerical results demonstrate the performance of the proposed method.