Project results

This project considers the problem of minimizing the energy consumption of wireless access networks through switching on and off and adjusting the transmitted power of wireless network devices according to the realistic traffic patterns.  We propose an optimization approach based on development of integer linear programming (ILP) models that minimize energy consumption of whole network while ensuring area coverage and enough capacity for guaranteeing quality of service. Proposed mathematical models capture system characteristics considering different management constraints that can be based on traffic requirements and application scenarios.


Illustrative movie visualizes results of optimization for wireless local area network (WLAN) consisted of 61 access points (APs) allocated on 900 m x 1200 m service area. On or off state of APs including level of transmitted power is changing among different time periods according to traffic pattern (user activity).


Energy minimization problems are solved to the optimum or with a gap to the optimum of less than 3% on a set of network instances which sizes and number of network elements corresponds to the real ones. Obtained results show that remarkable energy savings, up to 50% can be yielded. This savings can be accomplished through implementation of dynamic management of network resources in accordance with traffic load. To cope with the problem of high computational time characteristic for some ILP models, we have developed own heuristic algorithms based on greedy methods and local search. Although heuristics results have been up to 10% higher in comparison to the ones obtained for ILP models, each of heuristic algorithms ensures minimization of network energy consumption in reasonable amount of time. This makes heuristics algorithms applicable for practical implementation in real network management systems. It is important to emphasize that developed mathematical models and heuristic algorithms have general structure what makes them suitable for implementation, with rather small adaptations, on wireless networks of different access technologies. For networks of different sizes and wireless access technologies, this enables estimation of possible energy savings through implementation of energy efficient network management.