Distributed Energy Resources Optimization for Wireless Base Stations using Power Amplifier Switching

With the increased deployment of cellular networks due to advanced transmission techniques, overall energy consumption of the telecommunication sector across the globe has grown. Numerous challenges for operators are springing up including the escalating number of base station (BS) sites and also the increasing operation cost as well as environmental impacts due to the use of sources such as fossil fuel. Therefore, to achieve optimization in power consumption of cellular network, global network operators are deploying different energy-saving strategies. In this paper, a simple software-based effective power amplifier switching algorithm requiring no change of infrastructure is proposed to save a significant amount of energy at the base station using a real-time traffic scenario. The proposed algorithm uses the Transmit Channel Allocation method to switch the power amplifier ON/OFF at the transceivers of BS sites automatically based on the continuous traffic analysis. To validate the proposed algorithm, a model consisting of 5 BSs with a solar system is considered, and the power amplifier switching algorithm is run via Matlab at these BSs to record the hourly power consumption over the year. The resulting optimized BSs power load represents a 17.31% reduction in average power consumption over the course of one year. A battery storage model using real-world hourly irradiance and temperature for 5 BSs was used to analyze the grid dependency of the model for both optimized and unoptimized annual load. Grid power utilization for both cases is compared to conclude that the proposed algorithm improved the overall power efficiency and lowered the cost of operation of the cellular network by reducing the average power consumption of the BSs. The hybrid solar power BS model using an optimum battery storage capacity became more grid independent with a 72.14% reduction in grid dependency as an impact of the optimization achieved using the proposed PA switching algorithm.