Distributed cross-layer resource allocation in wireless ad hoc networks

In this paper, based on elastic link capacity which a function of transmission power, we construct a joint optimization formulation of congestion control and power control to maximize the network utility with multiple alternative subcarrier paths. The problem is a non-convex and non-separable optimization problem due to the multiple route nature of the flows. Furthermore, by integrating congestion control and power control, a distributed coordinative algorithm is proposed to maximize the aggregate source utility and increase end-to-end throughput. We obtain a decoupled and dual-decomposable convex formulation by applying an appropriate transformation and introducing some new variables and small disturbance quadratic terms. This is accomplished by the interaction and coordination between the congestion control sub-algorithm and the power control sub-algorithm through the congestion prices. Simulation results demonstrate that our distributed algorithm converges much faster and greatly increases network utilization and throughput.