Joint Transmission with Limited Backhaul Connectivity

Downlink beamforming techniques with low signaling overhead are proposed for joint processing coordinated (JP) multi-point transmission. The objective is to maximize the weighted sum rate within joint transmission clusters. As the considered weighted sum rate maximization is a nonconvex problem, successive convex approximation techniques, based on weighted mean-squared error minimization, are applied to devise algorithms with tractable computational complexity. Decentralized algorithms are proposed to enable JP even with limited backhaul connectivity. These algorithms rely provide a variety of alternatives for signaling overhead, computational complexity and convergence behavior. Time division duplexing is exploited to design transceiver training techniques for two scenarios: stream specific estimation and direct estimation. In the stream specific estimation, the base station and user equipment estimate all of the stream specific precoded pilots individually and construct the transmit/receive covariance matrices based on these pilot estimates. With the direct estimation, only the intended transmission is separately estimated and the covariance matrices constructed directly from the aggregate system-wide pilots. The proposed training schemes incorporate bi-directional beamformer signaling to improve the convergence behavior. This scheme exploits the time division duplexing frame structure and is shown to improve the training latency of the iterative transceiver design. The impact of feedback/backhaul signaling quantization is considered, in order to further reduce the signaling overhead. Also, user admission is being considered for time-correlated channels. The enhanced transceiver convergence rate enables periodic beamformer reinitialization, which greatly improves the achieved system performance in dense networks.