ON THE ENERGY EFFICIENCY OF CLUSTERING ALGORITHMS FOR RELAY ASSISTED DEVICE-TO-DEVICE COMMUNICATION

The energy efficiency of data transmission in Device-to-Device (D2D) communication extends battery lives necessary for service continuity. One communication method in D2D is clustering where one User Equipment (UE) relays data between cluster members and the base station (NodeB). This paper evaluates the energy efficiency of three D2D clustering algorithms: Round Robin, where the cluster head selection is rotational among cluster members, Minimum Distance, where the UE closest to the NodeB is selected cluster head, and Minimum Remaining Power where the UE with the highest battery level is the cluster head. A simulation is developed considering various scenarios where UEs are at different distances from the NodeB, and either stationary or moving at certain speeds. Results reveal that the Minimum Distance is the most energy efficient algorithm when UEs are nearby the NodeB and moving. It also outperforms other algorithms as long as UEs are on the move; so that selection of cluster head is rotated among cluster members. Results also show that maximum energy savings due to clustering are up to 47.4% compared to traditional LTE.