Energy-efficient dynamic load distribution for heterogeneous access networks

Abstract

In heterogeneous wireless access networks, it is important to achieve high variability of access resources. Bandwidth Aggregation (BAG) is considered to be a promising solution for this objective. However, power consumption has not been adequately addressed, although prolonging the battery life time of mobile devices with low power consumption has received attention lately. In this paper, we propose a novel dynamic energy-efficient load distribution algorithm for heterogeneous wireless networks. For the optimal load distribution, which is challenging combinatorial problem, we propose simple greedy load distribution algorithms that are inspired by the mathematical background of the Lagrangian algorithm. Furthermore, we propose another heuristic algorithm that extends the proposed algorithm when the retransmission scheme is hired for packet loss (e.g, TCP). Finally, an extensive Network Simulator 2 (NS2) simulation under various practical configurations demonstrates that the proposed algorithm could converge to the optimal distribution point with low complexity. Furthermore, it gives significantly increased the battery life time of mobile device with guarantee required Quality of Service (QoS) constraints.