Aug 2026· PLoS ONE· Vol 21· 0 citations· 39 references
Medicine
TL;DR
A proactive mitigation framework that applies the unified Autoregressive Recurrent Neural Network (AR-RNN) that significantly improves network reliability, reducing the network outage probability by up to 50% compared to standard reactive handover procedures.
Abstract
Developments in 5G New Radio (NR) aim to provide the best possible user experience, but barred and resource-depleted cells in dense small cell networks can cause service interruptions and increase handover latency. To address these challenges, we propose a proactive mitigation framework that applies the unified Autoregressive Recurrent Neural Network (AR-RNN). This framework simultaneously predicts a user’s future cell trajectory and their resource requirements by learning from historical data. By integrating these predictive forecasts with standard network-monitoring alerts (e.g., from an Intrusion Detection System (IDS), a Network Management System (NMS), or Operations, Administration, and Maintenance (OAM) systems), the system preemptively reroutes users to avoid flagged or unavailable cells. Applying the algorithm to both Homogeneous and Heterogeneous networks, our simulations demonstrate that this proactive approach yields substantial and quantifiable improvements in network performance across both topologies. The AR-RNN model achieves a next-cell prediction accuracy of up to 95.8%, establishing high dependability. This accuracy directly translates to enhanced Quality of Service (QoS) by reducing handover latency even to as low as nearly 5 ms. Furthermore, the framework significantly improves network reliability, reducing the network outage probability by up to 50% compared to standard reactive handover procedures. These results demonstrate a concrete and effective method for creating a more seamless and efficient telecommunications experience in dense 5G environments.
A comprehensive survey of AI-enabled mobility management strategies for 5G, Beyond 5G, and upcoming 6G networks, with particular attention to HO optimization and load balancing is presented.
H. Asif, Abdulraqeb Alhammadi, N. Tarhuni et al.· Future Internet· 0 citations
Recently, the development and deployment of intelligent controllers for radio access networks (RAN) has attracted significant attention from network operators and international telecommunications organizations, driven by rapid advances in artificial intelligence. Mobility management plays a fundamental role in ensuring seamless connectivity and service quality in 5G RAN. In fact, optimal control in 5G RAN is highly challenging due to its complex, dynamic, and distributed environment. Many approaches have been proposed to address this problem, particularly those based on deep reinforcement learning (DRL). However, contrary to the dense reward assumption in many DRL-based studies, mobility feedback in practical RAN environments is characteristically sparse and delayed. In this paper, we propose WHO (World Model for Handover Optimization), a novel method designed to bridge the gap between sparse feedback and efficient learning in 5G networks. WHO utilizes a world model to convert event-driven rewards into dense predictive signals, facilitating robust multi-agent optimization. Field experiments involving 13 base stations and 39 cells show that the proposed method significantly improves handover performance and network stability compared to conventional DRL approaches, achieving 19–40% higher prediction precision and up to 32% improvement in key performance indicators (KPIs).
Uyen Thi Thu Truong, Doan Van Nguyen, Do Ngoc Tuan et al.· International Conference on...· 0 citations
A Hybrid deep double-Q networks (DDQN)–bidirectional long short-term memory (Bi-LSTM) Framework that integrates bi-directional mobility prediction and DRL-based adaptive decision-making is introduced that highlights the effectiveness of integrating predictive intelligence with reinforcement learning for reliable mobility management in 5G-Advanced and emerging 6G networks.
The deployment of Ultra-Dense Networks (UDNs) in 5G systems is to meet the growing demand for high data rates and massive connectivity. However, the dense deployment of small cells increases handover frequency, leading to challenges such as handover failures (HOF), unnecessary handovers, and the ping-pong effect, leading to degradeuser Quality of Service (QoS). This paper proposes a velocity-aware adaptive handover control approach for efficient mobility management in 5G ultra-dense networks. The proposed approach dynamically adjusts Handover Control Parameters (HCPs) called Time-to-Trigger (TTT) and Handover Margin (HOM) on the real-time velocity of User Equipment (UE) and signal conditions. The system is modeled as a two-tier heterogeneous network consisting of a macrocell overlaid with multiple small cells, and performance is evaluated using the Cost 231-Hata propagation model. The findings demonstrate that the proposed algorithm significantly reduces the total number of handovers, mitigates the ping-pong effect, and lowers handover failure rates compared to conventional static schemes. The results confirm that velocity-aware adaptive control enhances network reliability, reduces signaling overhead, and improves overall mobility performance in 5G ultra-dense environments.
Halah Hassen Aldumaini, Hanadi Esmeail Yahya, Oloof Ameen Mohmmed et al.· 2026 6th International Confe...· 0 citations
Frequent handovers remain a challenge in vehicular fifth-generation (5G) networks, especially in dense urban areas with small cells and intersections. Conventional handover decisions mainly follow the current radio condition and may select a target that is not stable along the vehicle's future route. This paper proposes a mobility- and radio-aware handover control method for 5G vehicular networks. The method keeps the default received-signal trigger as the initial detector, predicts a short future route from lane position, heading continuity, road topology, and transition-support information, and validates the candidate using a historical signal-to-interference-plus-noise ratio (SINR) grid map. It suppresses candidates that indicate a ping-pong return to the serving next-generation NodeB (gNB) or a short stay before another gNB becomes preferable, while a current SINR-based rescue rule avoids keeping the user equipment (UE) on a weak serving link. In a real-map Innsbruck urban microcell scenario with 85 vehicular UEs and 24 gNBs, the proposed method reduces total handovers by 28.9%, ping-pong handovers by 55.9%, and short-stay handovers by 69.0% compared with the default policy. The post-simulation reference signal received power (RSRP) evaluation also remains strong, with 99.46% of the proposed-method samples in the Excellent category and 0.54% in the Good category. These results show that mobility prediction with future SINR-map validation improves vehicular handover stability while preserving serving-link quality.
Heriansyah, G. Kotsis· International Conference on...· 0 citations
Ultra-dense 5G networks require advanced traffic steering to maintain performance and balance load amid growing user and base station (gNB) densities. Traditional heuristics such as nearest-base-station and Signal-to-Interference-plus-Noise Ratio (SINR)-based selection provide simple solutions but struggle to adapt to dynamic user mobility, diverse traffic, and fluctuating radio conditions at the mobility-control level, often leading to inefficient handovers and degraded network quality. We propose a deep reinforcement learning (DRL) framework to dynamically tune a global handover hysteresis margin that governs handover triggering decisions, optimizing handover success, reducing failures, and enhancing throughput and fairness. Implemented in Python using Stable Baselines3 and NumPy, our custom simulation environment models key mobility-related 5G dynamics at a high level, including user mobility, pathloss-based signal degradation, and interference. We evaluate DRL agents-Deep Q-Network (DQN) and Proximal Policy Optimization (PPO)-against heuristic and hysteresis-based baselines. Results show that DRL-based hysteresis optimization provides strong and robust performance under the considered ultra-dense mobility conditions in handover success rate, average SINR, throughput, and fairness, with PPO demonstrating the most consistent behavior across configurations. This work offers a reproducible simulation framework for further research into adaptive mobility management.
Damianos Diasakos, V. Kokkinos, C. Bouras et al.· International Conference on...· 0 citations