The aggregate throughput in random wireless networks with successive interference cancellation

The feasibility of successive interference cancellation (SIC) depends on the received power ordering from different users, which, in turn, depends on the fading distribution, path loss function and network geometry. Using a framework based on stochastic geometry, this paper studies the aggregate throughput in d-dimensional random wireless networks with SIC capability. We consider networks with arbitrary fading distribution, power-law path loss; the network geometry is governed by a non-uniform Poisson point process (PPP). Our results demonstrate how the performance of SIC changes as a function of the network geometry, fading distribution, and the path loss law. An important observation is that, in interference-limited networks, lower per-user information rate always results in higher aggregate throughput, while in noisy networks, there exists a positive optimal per-user rate at which the aggregate throughput is maximized.