Labeling Based Zero-forcing Beamforming for LTE-Advanced

For TDD, we take advantage of channel reciprocity and use UL-SRS for channel estimation to generate the DL BF weights. It may have better performance than FDD-LTE MIMO with limited feedback due to better CSI granularity, especially for multiple users. For the deployment of MU-MIMO and CoMP CB/JT in TDD-LTE, we needs solve three key problems: 1) to maximize the system sum rate of MU-MIMO and CoMP CB/JT; 2) to maximize the sum rate of selected paring users in the MU-MIMO or CoMP candidate user set; 3) to simplify the cross-layer resource allocation procedure between L1 and L2. If we use the traditional model, there will be a huge challenge for the scalable MIMO operation. This paper provides a general framework for the scalable system implementation with anticipated sub-optimal performance, but having low computing complexity based on the quantization to estimated channel using UL SRS signals. To reach the targets, we propose new codebook design criteria, new distributed algorithms for RRM and protocols for CoMP CB/JT.