Distributed Precoding for Worst-Case MSE Optimization in Multi-Cell Downlink Systems

To mitigate the inter-cell interference in multi-cell downlink systems, robust precoder design is important for multi-cell cooperation as the channel state information (CSI) available at the base station is imperfect. In this letter, we investigate the worst-case performance optimization problem with bounded CSI error. According to the leakage-based schemes, the sum mean-square-error (MSE) minimization problem is reformulated as a series of parallel sub-problems that can be recast as a semidefinite programming (SDP) problem, and solved in a distributed manner. Moreover, a modified cost function is proposed to get a distributed solution for the Min-Max MSE optimization problem. In the modified MSE, inter-cell leakage is introduced to replace inter-cell interference in the traditional MSE. Simulation results indicate that the inter-cell leakage is comparable to the inter-cell interference for each user. Moreover, simulation results demonstrate the robustness of the distributed precoder designs through the bit error rate (BER) performance.