Utility fairness via association control in wireless mesh networks

We study the problem of utility fair bandwidth allocation via association control in wireless mesh networks (WMNs). We formulate an optimization problem that achieves an utility fairness objective and jointly considers user association and resource allocation in WMNs. As the problem is NP-hard, we propose an approximation algorithm whose solution is very close to the optimal one, which is demonstrated in our simulation results. Utility fairness is more general than other commonly used fairness objectives in resource management such as max-min fairness and proportional fairness. It is more flexible in controlling the tradeoff between resource utilization efficiency and user fairness. We introduce a user bandwidth boundary constraint to make the tradeoff more controllable and at the same time prevent extreme unfairness. We demonstrate through simulations how to control the tradeoff between efficiency and fairness to achieve the desired performance by tuning the control parameters.