SSIM-based Weak Zero-forcing Precoder Design for Multiuser MIMO System

The optimization of zero-forcing precoders is studied for the structural similarity (SSIM) index in this paper. When the input signal is an image for the multi-user multi-input multi-output (MIMO) downlink communication channels. the traditional precoders design with maximizing signal to noise ratio(SNR) or mean squared error(MSE) may not retain the structural information of transmitted signal. In this paper, we consider the problem of linear precoding design with SSIM index under the weak zero-forcing condition. A general solution for zero-forcing precoder is provided to guarantee the quality of the received image, under total power constraint or per-antenna power constraint. The precoder design with SSIM is a non-convex problem. We use the polynomial fitting method to convert it into convex problem, and then use the convex optimization toolbox to solve it. Experiment results have shown the effectiveness and advantages of the proposed precoder.