Showing papers by "Bernard Cousin published in 2013"

Joint power-delay minimization in green wireless access networks

Abstract: Growing energy demands, the increasing depletion of traditional energy resources, together with the recent surge in mobile internet traffic, all call for green solutions to address the challenge of energy-efficient wireless access networks. In this paper, we consider possible power saving by reducing the number of active BSs and adjusting the transmit power of those that remain active while maintaining a satisfying service for all users in the network. We thus introduce a joint optimization problem that minimizes the network power consumption of the network and the sum of network user transmission delays. Our formulation allows us to investigate the tradeoff between power and delay by tuning the respective weighting factors. Moreover, to reduce the computational complexity of the optimal solution of our non-linear optimization problem, we convert it into a Mixed Integer Linear Programming (MILP) problem. We provide extensive simulations for various decision preferences such as power minimization, delay minimization and joint minimization of power and delay. The results we present show that we obtain power savings of up to 16% compared to legacy network models.

Power-Delay Tradeoffs in Green Wireless Access Networks

Abstract: Targeting energy efficiency while meeting user Quality of Service (QoS) is one of the most challenging problems in green wireless networks. In this paper, we propose an optimization model based on finding a tradeoff between reducing the number of active radio cells and increasing the transmit power of base stations (BSs) to better serve all users in the system. The main contribution of the paper is the formulation of a multiobjective optimization problem that jointly minimizes the network power consumption and the sum of the network user transmission delay. Our proposed problem is solved using an exhaustive search algorithm to obtain the optimal solution. Solving the optimization problem at hand is very challenging due to the high computational complexity of the exhaustive search. Therefore, we run simulations in a small network to give insights into the optimal solution. Specifically, we study different cases by tuning the respective weights of the power and delay costs. This is a distinctive and important feature of our model allowing it to reflect various decision preferences. Regarding these preferences and under various spatial distribution of users, results show that our solution allows the optimal network configuration to be selected in terms of power consumption while guaranteeing minimal delay for all users in the network.