Spatial Multiplexing in Random Wireless Networks

We consider a network of transmitters, each with a receiver at a fixed distance, and locations drawn independently according to a homogeneous Poisson Point Process (PPP). The transmitters and the receivers are equipped with multiple antennas. Under a channel model that includes Rayleigh fading and path-loss, and an outage model for packet successes, we examine the performance of various spatial multiplexing techniques, namely zero-forcing (ZF), ZF with successive interference cancellation (ZF-SIC or VBLAST) and DBLAST. In each case, we determine the number of streams that maximizes the transmission capacity, defined as the maximum network throughput per unit area such that a constraint on the outage probability is satisfied. Numerical results showcase the benefit of DBLAST over ZF and VBLAST in terms of the transmission capacity. In all cases, the transmission capacity scales linearly in the number of antennas.