Energy-efficient resource allocation in wireless LANs under non-linear capacity constraints

In this paper, we present an approach and an algorithm aimed at minimising the energy consumption of enterprise wireless LANs during periods of low user activity. We act on two network management aspects: powering off some access points (APs), and choosing the level of transmission power of each AP. An efficient technique to allocate the user terminals to the various APs is the key to achieve this goal. The approach has been shaped as an integer programming model with nonlinear constraints, which comes from a general but accurate characterisation of the WLAN system. To solve this model to optimality we present an algorithm named Benders' Decomposition-based Algorithm (BDA). The computational results in realistic scenarios proved the possibility to obtain remarkable power savings.