一种结合机器学习的移动边缘计算的切换预测方法

随着移动通信的发展，减少通信延时成为关键性问题，因此，提出了一种使用机器学习方法的移动边缘计算(MEC)移动性管理。移动性决策基于参考信号接收功率(RSRP)值和不确定性预测器。使用神经网络建立预测器，输出不同相邻单元的RSRP平均值和标准偏差，推导了切换概率的封闭表达式。基于这些可能性，MEC服务器能够提前缓存用户服务，以便将切换期间的中断降至最低。实验结果表明，提出的方法能够满足实际需求。

A Handoff Prediction Method for Mobile Edge Computing Based on Machine Learning

With the development of mobile communication, reducing communication delay becomes a key issue, therefore, this paper proposes a mobile edge computing (MEC) mobility management using machine learning method. Mobility decisions are based on Reference Signal Received Power (RSRP) values and uncertainty predictors. Use a neural network to build a predictor that outputs the RSRP mean and standard deviation for different neighboring cells. A closed expression for the switching probability is derived. With these possibilities, the MEC server is able to cache user services ahead of time in order to minimize disruption during handovers. The experimental results show that the proposed method can meet the actual needs.