Intelligent Load Balancing Based on Traffic Behavior in Wireless Systems | ||
| Port-Said Engineering Research Journal | ||
| Articles in Press, Accepted Manuscript, Available Online from 02 October 2025 | ||
| Document Type: Original Article | ||
| DOI: 10.21608/pserj.2025.409959.1429 | ||
| Author | ||
| Mona Nashaat* | ||
| Port Said University | ||
| Abstract | ||
| As the adoption of smart devices and bandwidth-intensive applications continues to grow, wireless networks face increasing pressure to deliver consistent, high-performance broadband services. Ensuring efficient traffic management and maintaining communication reliability are critical challenges, particularly in next-generation wireless infrastructures such as 5G. One of the key difficulties lies in the uneven distribution of data traffic across network zones, which can lead to congestion and degraded Quality of Service (QoS). This paper introduces AutoBalancer, a predictive framework designed to enhance load distribution in wireless networks by forecasting traffic levels and optimizing handover control parameters. The framework employs an autoencoder-based model to accurately predict traffic conditions across network segments and recommends adaptive adjustments to Cell Individual Offset (CIO) values to facilitate smooth handovers while maintaining network balance. Experimental evaluations show that the proposed model can exceed the performance of state-of-the-art models in both traffic prediction accuracy and computation time. Additionally, AutoBalancer has the capability to substantially improve load balancing efficiency, enhance overall cell throughput, and increase the load fairness index. This evaluation relies on real 5G mobile network data from a high-traffic urban area in Egypt and demonstrates that the proposed model can assist 5G networks in fulfilling the rising demand for high-speed data services. | ||
| Keywords | ||
| 5G Networks; Autoencoder; Load Balancing; Network Traffic Prediction | ||
|
Statistics Article View: 17 |
||
