Network Control and Rate Optimization for Multiuser MIMO Communications

We present a network control and rate optimization solution for multiuser multiple input multiple output (MU-MIMO) communications in wireless networks. This solution is decentralized and includes scheduling and routing of the MU-MIMO communication links that adapt to dynamic channel, interference, and traffic conditions. We start with analyzing the ergodic sum rates of MIMO multiple access channel (MAC) and interference channel (IC) configurations by considering the error, and overhead effects due to channel estimation (training) and quantization (feedback). By taking the practical considerations such as channel estimation, quantization error and in-network interference into account, we show the rate gain with an increasing number of antennas compared with single-input single-output (SISO) systems. Next, we present a distributed channel access protocol to select and activate MU-MIMO configurations with the maximum achievable sum rates using local information on channel, interference, and traffic conditions. Then, we extend this scheduling algorithm to routing via a cross-layer solution based on a decentralized version of the backpressure algorithm. After accounting for the control message overhead, we show that the proposed MU-MIMO scheduling and routing solution improves the stable throughput over the minimum distance routing based on frequency of encounters and single user MIMO communications in a mobile ad hoc network (MANET) setting.