Executive Summary : | The growth of intensive applications has increased computational requirements and energy consumption in ecosystems like smartphones and IoT. Mobile edge computing (MEC) offers computing services for mobile applications, but traditional terrestrial infrastructure-based MEC networks are not suitable in remote regions or natural disaster areas. A new paradigm of UAV-enabled aerial MEC is proposed, integrating UAV communications and MEC for next-generation wireless networks. This project proposes an intelligent mechanism based on Deep Reinforcement Learning (DRL) for optimized computation offloading in dynamic UAV-Enabled MEC environments for next-generation networks, aiming to minimize latency, energy consumption, and task failure rates in real-world environments. The project aims to set up a UAV-enabled MEC system for next-generation networks 5G and Beyond, develop machine/deep learning models for intelligent computation offloading, and perform performance analysis and verification using 5G test-beds setup in India. The methodology includes literature review, setting up the MEC environment, implementing AI models, integrating AI models with communication environment, and evaluating and validating the system. The goal is to optimize computation offloading, enhance performance, and improve real-time application service performance. |