Symbol-Level Precoding Design for Integrated Sensing and Communication

The symbol-level precoding (SLP) design is considered for a multi-user multiple-input-multiple-output integrated sensing and communication (ISAC) system. This work exploits the SLP design to minimize the Cramer-Rae bound for radar sensing while satisfying the constant-modulus constraint, the similarity constraint and the communication quality of service requirements. However, because of the objective function and the constant-modulus constraint, the optimization problem considered in this work is non-convex. We exploit the Schur complement to transform this non-convex optimization problem into a semidefinite programming problem and use the convex optimization toolbox to solve it. Simulation results illustrate the advantage of the proposed design in the ISAC system.