Mobility-Aware Handover Optimization Using Adaptive Time-To-Trigger Mechanisms in 5G Networks

Abstract

In 5G networks, handover management is critical for ensuring seamless mobility, low latency, and high-quality user experiences. However, traditional handover mechanisms suffer from frequent handover failures and ping-pong effects, especially when users move at high speeds or across densely deployed small cells. This paper proposes a mobility-prediction-based handover approach with an adaptive Time-To-Trigger (TTT) mechanism that adjusts dynamically to user speed and mobility patterns. The system comprises three key components: a Mobility Prediction Module utilizing recurrent neural networks (RNNs) to analyze historical movement data and real-time parameters including signal strength and user speed; Dynamic TTT Adaptation that reduces TTT for high-speed users to enable faster handovers while increasing TTT for slow-moving users to prevent ping-pong effects; and a Handover Decision Algorithm that integrates predicted mobility with real-time signal quality measurements. Simulation results demonstrate that the approach improves handover success rates, reduces ping-pong effects and enhances overall network performance. Additionally, the adaptive framework contributes to better resource utilization and overall network efficiency, The proposed system achieved 94.2% success rate, 67% fewer ping-pong events, 42ms average delay, 15.8% throughput gain, and 89.4\% prediction accuracy with low computational cost.