Intelligent Handover Triggering Mechanism in 5G Ultra-Dense Networks Via Clustering-Based Reinforcement Learning

Ultra-dense networks (UDNs) are considered as key 5G technologies. They provide mobile users a high transmission rate and efficient radio resource management. However, UDNs lead to the dense deployment of small base stations (BSs) that can cause stronger interference subsequently increase handover management complexity. At present, conventional triggering mechanism user equipment (UE) is only designed for macro mobility thus could result in negative effects such frequent handovers, ping-pong failures on process UE at UDNs. These degrade overall network performance. In addition, massive number BSs significantly maintenance system workload. To address these issues, this paper proposes an intelligent based Q-learning frameworks subtractive clustering techniques. The input metrics first converted state vectors by clustering, which improve efficiency effectiveness training process. Afterward, framework learns optimal policy from environment. trained Q table deployed trigger simulation results demonstrate proposed method ensure robustness improved 60%–90% compared approach with respect ping-ping rate, failure while maintaining other performance indicators (KPIs), is, relatively level throughput latency. through integration further average 20% terms all evaluated KPIs.