Distributed optimal power and rate control in single-hop wireless interference networks

We study distributed optimal control in general single-hop wireless interference networks. We take into account the decoding errors due to channel outage by introducing “effective utility” associated with the average correctly received data rate, and formulate the effective network utility maximization (ENUM) w.r.t. power and rate control. The ENUM is a non-convex stochastic NUM. By introducing “average SINR”, we approximate the ENUMby a convex problem. The approximation error is asymptotically negligible. Borrowing the load-spillage characterization of the average SINR, we obtain a distributed stochastic algorithm to solve the approximated problem. Using stochastic approximation, we show that the proposed algorithm converges almost surely.