Load migration mechanism in ultra-dense networks

As one of the main technologies in 5G networks, Ultra-dense networks (UDNs) can be used to improve the network coverage. The dense deployment of small cells in UDN hotspots generates an uneven traffic distribution. In this paper, we propose a novel mechanism in order to transfer the extra users from the small cells to the macrocells based on several load balancing approaches implemented within the small cells, which are formed based on the Radio-over-Fiber (RoF) system. To select the best overlapping zone and then the best candidate user to be handed-over between the access points of the small cells, a common zone approach, a worst zone approach and a mixed approach are proposed. With the objective of transferring the extra users to the macrocells, we suggest a transfer after approach, a transfer before approach and an active approach. The simulation results indicate that the proposed approaches succeed to balance the load among the access points and to migrate the required load from the overloaded small cells to the macrocells in selective way. In some cases, the balance improvement ratio can reach 97.94%. Moreover, the overall balance efficiency is increased by 51.32% compared to the case without transferring users to the macrocells.Peer reviewe