MU-MIMO downlink scheduling based on users' correlation and fairness

A scheduling algorithm based on weighted users' correlation and fairness (WUCFS) is proposed for MU-MIMO downlink broadcast channels (BC), to address the problem that traditional fairness scheduling algorithms cannot handle, i.e., they cannot accurately compute the achievable rate for each user in their iterative user selection process, thus creating an undesirable tradeoff between user fairness and system sum-rate. The proposed WUCFS is used to activate mobile subscribers by evaluating the correlation between the candidate and the selected users as well as those who may be scheduled for data transmission. Weighted inverse-correlation maximization is then employed to determine which users to be scheduled for data transmission. By estimating potential co-channel interference (CCI) a priori, our algorithm can accurately estimate the actual transmission rate of each mobile user, and fairly activate a set of users with small mutual interference. Compared to existing schemes, the proposed algorithm is shown to improve fairness among users while ensuring high system sum-rate.