Optimal and suboptimal routing based on partial CSI in wireless ad-hoc networks

In this paper, we analyze and optimize the performance of distributed routing schemes in multihop wireless ad-hoc networks. We assume that the nodes are distributed according to a Poisson-Point-Process (PPP) and consider routing schemes that select the next relay based on geographical locations and local knowledge of the channel state (CSI). At the first stage we define the optimization problem and give an expression for the optimal routing metric in a network that achieves the ergodic rate density (ERD) in each link. We then use a recently published bound on the ERD, and present a low complexity, suboptimal routing metric. Numerical results demonstrate that the low complexity scheme performs nearly as good as the optimal scheme.