Energy Efficient Channel Assignment with Switching Optimization in Multi-radio Wireless Networks

In multi-radio wireless networks, it is vital to efficiently utilize the resource of non-overlapping channels, and thus the problem of channel assignment has been widely studied. Most proposed channel assignment approaches focus on reducing interference or maximizing throughput, assuming that all the radios on each node are keeping active. However, it is a significant waste of energy when network traffic is low, since the network is designed for peak traffic demands but does not serve peak traffic demands all the time. Even if the number of active radios is dynamically adaptive to time-varying network traffic, some radios will inevitably switch their channels, which significantly increase network delay. Therefore, we investigate the energy efficient channel assignment problem with switching optimization in this paper. To address this problem, we first define the optimization objective to minimize the number of active radios and channel switchings, and then propose several constraints, such as channel switching, radio amount, transmission existence, link rate, link interference, traffic demand and routing constraints. On this basis, we present an MILP model called EECA-SO, and evaluate its performance through numerical analysis. The results show that the proposed EECA-SO model can not only achieve energy saving effectively, but also obtain the trade off between energy efficiency and switching optimization.