Joint Precoder, Reflection Coefficients, and Equalizer Design for IRS-Assisted MIMO Systems

The incorporation of intelligent reflecting surface (IRS) into wireless communication systems can extend the coverage and enhance data transmission rate. This paper studies joint transceiver IRS designs in IRS-assisted multi-input multi-output (MIMO) under both perfect channel state information (CSI) imperfect CSI. Specifically, transmit precoder, reflection coefficients at IRS, receive equalizer are jointly optimized to minimize detection mean square error (MSE), subject power constraint modulus constraints for coefficients. design problems, non-convex challenging, tackled framework alternating optimization. For with CSI, we successively optimize given precoder present closed-form optimal angle one coefficient others. robust first average MSE over uncertainties by using a generalized statistical CSI model. Then, averaged is approximated more tractable upper bound. Subsequently, problem elaborately transformed form similar Numerical results demonstrate effectiveness proposed as compared various benchmark schemes.