Network Traffic Prediction with Reduced Power Consumption towards Green Cellular Networks

Abstract

The increased number of cellular network subscribers is giving rise to the network densification in next generation networks further increasing the greenhouse gas emission and the operational cost of network. Such issues have ignited a keen interest in the deployment of energy-efficient communication technologies rather than modifying the infrastructure of cellular networks. In cellular network largest portion of the power is consumed at the Base stations (BSs). Hence application of energy saving techniques at the BS will help reduce the power consumption of the cellular network further enhancing the energy efficiency (EE) of the network. As a result, BS sleep/wake-up techniques may significantly enhance cellular networks' energy efficiency. In the proposed work traffic and interference aware BS sleeping technique is proposed with an aim of reducing the power consumption of network while offering the desired Quality of Service (QoS) to the users. To implement the BS sleep modes in an efficient manner the prediction of network traffic load is carried out for future time slots. The Long Short term Memory model is used for prediction of network traffic load. Simulation results show that the proposed system provides significant reduction in power consumption as compared with the existing techniques while assuring the QoS requirements. With the proposed system the power saving is enhanced by approximately 2% when compared with the existing techniques. His proposed system will help in establishing green communication networks with reduced energy and power consumption.