Pre-Connect Handover Management for 5G Networks Using Multi-Agent Deep Q-Networks
Why this work is in the frame
A frame that forgets how it found something cannot be audited. These are the routes that admitted this work.
Bibliographic record
Abstract
Effective handover management is crucial for maintaining seamless connectivity in wireless networks, and it becomes increasingly challenging in fifth-generation (5G) networks due to strict Quality of Service (QoS) and Quality of Experience (QoE) requirements. This paper addresses the challenge of ensuring reliable and low-latency handovers in high-mobility scenarios for 5G networks by introducing a novel pre-connect handover (PHO) mechanism enhanced with Deep Reinforcement Learning (DRL). The proposed approach leverages Deep Q-Networks (DQN), a model-free DRL algorithm, to proactively select the best target cell based on prediction for handover. DQN makes predictive decisions based on Reference Signal Received Quality (RSRQ) values and their rate of change among the candidate cells that a user equipment (UE) can receive signals from simultaneously. To further reduce handover latency and improve reliability, the mechanism incorporates packet buffering at the target cell before handover execution. The DQN-assisted PHO solution is implemented and evaluated using Network Simulator 3 (NS-3) integrated with NS3-Gym, focusing on real-time online prediction for high-speed mobility scenarios. Furthermore, this paper explores the feasibility of extending the approach to Multi-Agent DRL (MADRL), where agents manage handovers independently. Experimental results demonstrate that the proposed DQN-assisted PHO significantly improves handover success rates by triggering the handover process 800–900 ms earlier, and the MADRL can also achieve up to 100% success rate for specific scenarios. These findings highlight the potential of DRL-based techniques for enhancing handover reliability and performance in 5G and beyond wireless networks.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
Full frame distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.000 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.001 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.000 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from it