An Energy-Efficient Intelligence Sharing Scheme in Intelligence Networking-Empowered Edge Computing

Advanced artificial intelligence (AI) and multi-access edge computing (MEC) technologies facilitate the development of edge intelligence, which enables the intelligence learned from remote cloud to network edge. To realize automatic decision-making, the accuracy of AI models and training efficiency are crucial for edge intelligence. However, the data volume collected by each network edge node from various sensors is limited, which may cause the over-fitting of AI models. To improve the accuracy of AI models and training efficiency for edge intelligence, intelligence networking-empowered edge computing (INEEC) is emerging as a promising solution, which enables each network edge node to improve its AI models quickly and economically with the help of other network edge nodes’ sharing of their learned intelligence. Sharing intelligence among network edge nodes efficiently is essential for INEEC. Thus in this paper, we study the intelligence sharing scheme, which aims to maximize the system energy efficiency while ensuring the latency tolerance via jointly optimizing intelligence requesting strategy, transmission power control and computation resource allocation. The system energy efficiency is defined as the ratio of model performance to energy consumption. To solve this complex optimization problem, a hybrid algorithm that integrates GA and PSO is proposed to make the optimal intelligence sharing decision. Finally, the convergence and superiority of the proposed scheme in terms of intelligence sharing efficiency are evaluated through extensive simulation experiments.