A Small Cell Outage Prediction Method Based on RNN Model

Small cells play an important bridging role in the nearly 5G network to connect the various IoT devices and internet of vehicles for low-latency and high-speed communication. The small cell outage diagnosis based on RNN model is proposed to improve the wireless network quality and customer experience due to the most small cells are deployed in the high-density population area. This research is used to diagnose the small cell obstacle events based on KPIs information and deep learning algorithm. The device KPIs are used as input factor in RNN deep learning model and the experiment result show the accuracy of validation dataset is to be 96%. The prediction result can repair the outage and adjust the parameters in advance according to the correlation between KPIs values change and outage events.