Blog posts



Fairness in Wireless Networks

10 minute read


Wireless Networks (WNs) are characterized for sharing a medium with limited resources (i.e., the air), which may lead to coexistence issues that harm the overall performance of the overlapping devices. Due to the possibility of choosing different configurations (e.g., range of channels, carrier sense threshold, transmit power), we often find an unbalanced share of the resources between WNs. Such imbalance may be solved by the application of fairness policies. However, a fair solution does not always entail a maximization of the aggregate performance. In general, fairness is achieved by a central unit that decides the configuration of the wireless devices. However, we aim to extend this concept to collaborative approaches that share the same fairness goal that boosts the overall performance.

In this document, we aim to shed some light on the fairness problem in WNs, as well as on the main considerations to be done for a proper resource allocation. We also aim to bound the meaning of \emph{the cost of fairness}. Note, as well, that we focus on configurations in which the utilities of players are known. However, there are many other situations in which players show a selfish behavior and do not truthfully reveal their utilities. Such situations lead to what is commonly known as \emph{the price of anarchy}.

Disclaimer: most of the material in this document has been retrieved from Bertsimas, D., Farias, V. F., & Trichakis, N. (2011). The price of fairness. Operations research, 59(1), 17-31.