| Abstract |
Providing high-quality services to citizens is the aim of smart cities. For this requirement to be met, several applications and networks are implemented, which consume a large amount of energy. As a result, the research proposal is to create a proactive strategy for optimizing energy consumption in cellular networks. Therefore, this chapter summarizes recent techniques for predicting cellular traffic. Next, we introduce a new model that preserves energy. The model uses clustering based on modified k-means of traffic based on its behavior. Next, the clustered time series data is used with the KNN regression algorithm to determine the traffic pattern. Utilizing the predicted cell tower traffic with the proposed energy-saving model, additional devices are powered on and off in order to reduce energy requirements. The proposed proactive model is validated using a flow of data over 4G LTE networks. Additionally, model implementation is carried out using the Google Colab service. The proposed model provides acceptable accuracy with limited data. According to the experimental analysis, the model is able to preserve between 20 and 28% of the energy in the cell towers when compared to the traditional approach. © 2025 selection and editorial matter, Vivek Kumar Singh, Anil Kumar Sagar, Parma Nand, Rani Astya and Omprakash Kaiwartya. All rights reserved. |
