A Base Station Sleeping Strategy in Heterogeneous Cellular Networks Based on User Traffic Prediction

Abstract

Real-time traffic in a cellular network varies over time and often shows tidal patterns, such as the day/night traffic pattern. With this characteristic, we can reduce the energy consumption of a cellular network by consolidating workloads spreading over the entire network to fewer Base Stations (BSs). In this work, we propose a BS sleeping strategy for a two-tier Heterogeneous Cellular Network (HeCN) that consists of Macro Base Stations (MaBS) and Micro Base Stations (MiBS). We first use a Bidirectional Long Short-Term Memory (BLSTM) neural network to predict the future traffic of each user. Based on the predicted traffic, our proposed BS sleeping strategy switches user connections from underutilized MiBSs to other BSs, then switches off the idle MiBSs. The MaBSs are never switched off. All user connections have predefined Signal-to-Interference-plus-Noise Ratio thresholds, and we ensure that each user’s service quality, which is related to the user’s traffic demand rate, is not degraded when switching user connections. We demonstrate the effectiveness and superiority of our proposed strategy over four other baselines through extensive numerical simulations, where our proposed strategy substantially outperforms the four baselines in different scenarios.