An Intelligent Energy Saving Strategy Recommendation Method of 5G Base Station Based on GBDT+FM

In order to find a better model of energy saving for 5G base stations to reduce energy consumption, this paper proposes an intelligent energy saving strategy recommendation method of 5G base stations based on GBDT+FM. Firstly, GBDT model and FM model is fused as GBDT+FM model to recommend energy strategy of 5G base stations. Then, in energy saving strategy recommendation layer, based on the GBDT+FM model, the identification model of the base station deployment scenario will be fused with the base station traffic prediction model to form the optimal AI energy saving strategy and it will be applied to the energy-saving system of 5G wireless communication unit (such as AAU). At the same time, the closed-loop feedback between energy-saving effects and energy-saving strategy recommendations can be considered to achieve a balance between energy-saving effects and quality of user experience for each base station, reduce overall energy consumption of base station without influencing quality of user experience, cut down OPEX ultimately. The evaluation result shows that the energy efficiency of base station main equipment based on GBDT+FM model proposed in this paper are all over 10–20 scores than the execution result of traditional method, the energy consumption reduction degree of CU/DU/AAU for 10 base stations based on GBDT+FM model proposed in this paper are all higher than 25%, higher than the execution result of a single model, and the power consumption of supporting equipment based on GBDT+FM model proposed in this paper are all 4–10 scores higher than the execution result of traditional method, which prove the validity of the proposed method in this paper.