Optimal backhaul resource management in wireless-optical converged networks

Abstract

The introduction of the new 4G technologies promises to satisfy the increasing demands of the end-users for bandwidth consuming applications. However, the high data rates provided by 4G networks at the air interface raise the need for more efficient management of the backhaul resources. In the current work, the authors study the problem of the efficient management of the backhaul resources at the side of the base station. Specifically, a novel scheme is proposed that, initially, predicts the forthcoming demand using artificial neural networks and, then, based on the prediction results, it proactively requests the commitment of the appropriate resources using linear optimisation techniques. The experimental results show that the proposed scheme can efficiently and cost-effectively manage the backhaul resources, outperforming the traditional flat commitment approaches.