Energy and Resource Efficiency by User Traffic Prediction and Classification in Cellular Networks

There is a lack of research on the analysis per-user traffic in cellular networks, for deriving and following traffic-aware network management. In fact, legacy design approach, which resource provisioning operation control are performed based cell-aggregated scenarios, not so energy- cost-efficient need to be substituted with user-centric predictive mobile proactive Here, we shed light this problem by designing prediction tools that utilize standard machine learning (ML) tools, including long short-term memory (LSTM) autoregressive integrated moving average (ARIMA) top data. We present an expansive empirical evaluation designed solutions over real dataset. Within analysis, impact different parameters, such as time granularity, length future predictions, feature selection investigated. As potential application these solutions, ML-powered Discontinuous reception (DRX) scheme energy saving. Towards end, leverage derived ML models dynamic DRX parameter adaptation user traffic. The performance results demonstrate superiority LSTM ARIMA general, especially when training series high enough, it augmented wisely -selected set features. Furthermore, show parameters online provides much more energy-saving at low latency cost comparison cell-wide adaptation.