An Efficient Multicast Clustering Scheme for D2D Assisted Offloading in Cellular Networks

Device-to-device (D2D) multicast communication is considered as a promising technology to realize the innovative content-aware applications in the fifth generation (5G) networks, such as multimedia multicast service and group-based machine type communications (MTC) applications. Under such scenarios, mobile devices (MDs) that are interested in the same content can be clustered as a group and transmit cooperatively via multiple hops to achieve efficient content distribution. Such multi-hop multicast transmission helps to relief traffic pressure in cellular networks and improve the quality of service for cell edge users. However, it requires careful design of the clustering scheme to ensure delay performance and energy efficiency. In this paper, we propose a multicast clustering algorithm for multi-hop communications. As for each hop, we select a combination of nodes to act as cluster heads (CHs) and multicast the content to a group of next-hop users by jointly considering achievable multicast data rate, remaining operation time and end-to-end delay. In the case of overlapping clusters, we propose a cluster member (CM) reassignment criteria to minimize transmission redundancy. Simulation results show that the performance of our proposed clustering algorithm is competitive in terms of delay and robustness.