Optimizing energy efficiency for edge caching for IIoT networks

• Main Author: Ng, Jovian Nursan
• Other Authors: A S Madhukumar
• Format: Final Year Project
• Language: English
• Published: Nanyang Technological University 2024
• Subjects: Computer and Information Science; IIoT
• Online Access: https://hdl.handle.net/10356/175259
• Institution: Nanyang Technological University

The Internet of Things (IoT) is a dynamic global network infrastructure, where devices and sensors are interconnected, enabling information sharing and collective decision-making capabilities between the devices in the network [1]. IIoT focuses on improving industrial process productivity by interconnecting industrial devices such as AMRs and sensors to infrastructures such as BS and cloud servers. This facilitates the data exchange and analysis between devices within the IIoT network, as shown in Figure 1 below. It requires a significant amount of energy consumption to maintain the transmissions within the IIoT network. As such, this project aims to propose a framework, mainly machine learning models, to optimize the energy EE of the system with the aim of minimizing the cost of maintaining the system. The models will be evaluated using the benchmark method as a baseline. Evaluation criteria include prediction accuracy and time taken compared to the benchmark method. This evaluation aims to predict optimal parameters for achieving optimal EE. The proposed machine learning models are shown to be able to predict around 800 times faster while only losing around 3-4% in terms of prediction accuracy compared to the benchmark method.