[2412.17565] Evaluation of Bio-Inspired Models under Different Learning Settings For Energy Efficiency in Network Traffic Prediction

PDF display of the paper entitled Evaluating the Biomedical Models under various educational settings for energy effectiveness in the network movement, by Theodoros Tsiolakis and 3 other authors

PDF HTML (experimental) view

a summary:The cellular traffic prediction is an important task that enables the network operators to customize resources efficiently and treat abnormal cases in sophisticated environments quickly. Si growth of data collected from basic stations represents major challenges in processing and analysis. While ML) appeared as strong tools for dealing with these large data collections and providing precise predictions, their environmental effect, especially with regard to energy consumption, is often ignored in favor of their predictive capabilities. This study is examining the capabilities of two vital -inspired models: nerve networks (SNNS) and the tank calculation through the Echoae (ESNS) networks to predict cellular movement. The evaluation focuses on both its alert performance and energy efficiency. These models are implemented in both central settings and handrails to analyze their effectiveness and energy consumption in decentralized systems. In addition, we compare the vital -inspired models with traditional structures, such as the todient nerve networks (CNNS) and multi -layer layers (MLPS), to provide a comprehensive evaluation. Using the data collected from three varied sites in Barcelona, Spain, we study the differentials between predictive accuracy and energy requirements through these methods. The results indicate that the models inspired by vital, such as SNNS and ESNS, can achieve significant energy savings while maintaining predictive accuracy similar to the traditional structure. Moreover, unified applications were tested to assess energy efficiency in decentralized environments compared to central systems, especially in conjunction with vital inspired models. These results provide valuable visions about the potential of vital inspired models to predict the sustainable cellular movement and maintain privacy.