摘要
文章提出一种基于XGBoost算法的自适应网络切换方法,优化工业物联网(Industrial Internet of Things,IIoT)环境中Wi-Fi与5G网络的切换效率。通过XGBoost模型深度学习历史网络性能数据和环境参数,智能预测最优网络切换时机和目标网络类型。该方法实现了动态网络选择,并结合动态缓存系统利用历史数据优化决策,提高了切换效率和响应速度。引入的回滚检查机制确保在网络性能下降或切换失败时能够迅速恢复到稳定状态,保障通信质量。实验评估表明,该方法在切换成功率、平均延迟和系统开销方面表现优异,为提高IIoT设备的通信性能提供了有效解决方案。
This paper proposes an adaptive network switching method based on the XGBoost algorithm to optimize the switching efficiency between Wi-Fi and 5G networks in Industrial Internet of Things(IIoT)environments.By deeply learning historical network performance data and environmental parameters,the XGBoost model intelligently predicts the optimal switching timing and target network type.This method enables dynamic network selection and integrates a dynamic caching system to utilize historical data for decision optimization,enhancing switching efficiency and response speed.A rollback check mechanism ensures quick recovery to a stable state in case of network performance degradation or switching failure,maintaining communication quality.Experimental evaluations demonstrate that this method excels in switching success rate,average latency,and system overhead,providing an effective solution for improving the communication performance of IIoT devices.
作者
陈文海
CHEN Wenhai(Guangdong Southern Planning&Designing Institute of Telecom Consultation Co.,Ltd.,Shenzhen 518038,China)
出处
《通信电源技术》
2024年第20期17-19,共3页
Telecom Power Technology
作者简介
陈文海(1986-),男,广东梅州人,本科,工程师,主要研究方向为5G通信技术。
