Price coordination method for congestion control in ad Hoc networks

In this paper, the congestion control problem in Ad Hoc networks is represented as the nonlinear optimization problem, which is the network utility maximization with the capacity constraints based on the link's interference set describing the contention relationship between flows. Based on dual decomposition and gradient projection method, Price Coordination Approach (PCA) is proposed to solve this utility maximization problem. Also, Neighbor Set Approximation method (NSA) and Queue Size Estimation method (QSE) are proposed to implement PCA in NS2 simulator. The MATLAB-based simulation results illustrate that PCA can approach the globally optimal solution. The NS2-based simulation results demonstrated that PCA outperforms default TCP and ATCP to achieve the efficient and fair resource allocation in Ad Hoc networks.