CSMA Using Statistical Physics Toward Throughput and Utility Optimal CSMA
In the recent past years, CSMA (Carrier Sense Multiple Access), which resolves contentions over wireless networks in a fully distributed fashion, has gained a lot of attentions since it has been proved in several papers that appropriate control of CSMA parameters provably guarantees optimality in terms of throughput and fairness. Most algorithms rely on the famous MCMC (Markov Chain Monte Carlo) technique, which enables one to find optimal CSMA parameters through iterative loops of simulation-and-update. However, such a simulation-based approach often becomes a major cause of slow convergence, being vulnerable to flow/topology changes. In this paper, the authors develop distributed iterative algorithms which produce approximate solutions with fast convergence.