Clustering Optimization for Out-of-Band D2D Communications

Abstract

Significant increase in multimedia traffic challenges 5G networks in terms of capacity and correspondent QoS parameters. Device-to-device communication paradigm has already become an integral part of 3GPP standards; nevertheless it has not yet been widely deployed due to many different reasons. D2D is expected to leverage implementation of many qualitatively new services and to efficiently accomplish it D2D devices are supposed to form clusters. Due to practical limitations, current D2D implementations are mostly out-of-band and use Wi-Fi Direct. In this paper, we propose a novel model for throughput optimization in out-of-band D2D clusters. We delivered numerical results for different typical cluster member distributions and revealed key functional dependencies. Further, for the first time we compare clustering algorithms for out-of-band D2D and identify effective clustering algorithm that increases network resource utilization rate.