A Cascading Bandit Approach To Efficient Mobility Management In Ultra-Dense Networks

Efficient mobility management is a key problem in modern wireless networks with high node density. In this paper, we propose an online learning approach for mobility management in ultra-dense networks, based on the cascading multi-armed bandits model. The proposed Cost-aware Cascading Bandit Neighbor Cell List (CCB-NCL) mobility protocol relies on the active neighbor cell list to assist the user equipment to explore the base station selection sequentially. Simulation results show that the proposed algorithm reduces the handover latency with lower dropped call rate, hence it is a better fit to efficient mobility management.