Overhead-Performance Tradeoff for MISO Broadcast Channels With Zero-Forcing Precoder

摘要

By applying a rate-distortion approach, we investigate the relationship between the amount of overhead and the system performance from an information-theoretic point of view. Though the purpose of rate-distortion theory is to find a lower bound on lossy source coding problems, the concepts of distortion and rate can be extended to communication performance measures and the quality of feedback overhead, respectively. In this article, we study the overhead-performance tradeoff for a downlink MU-MISO channel system with limited feedback. The required feedback bits for characterizing the channel state information is represented as a function of tolerable rate loss. The proposed method to derive this tradeoff is valid for any number of transmit antennas and users, and can help in designing practical systems where the impact of channel feedback overhead cannot be neglected. Comparing the obtained rate-distortion curves with vector quantization, it can be concluded that, for each user, by feeding back different numbers of bits for different channel power gains, the optimal performance can be achieved.