Caching contents at the edge of mobile networks is an efficient mechanism that can alleviate the backhaul links load and reduce the transmission delay. For this purpose, choosing an adequate caching strategy becomes an important issue. Recently, the tremendous growth of Mobile Edge Computing (MEC) empowers the edge network nodes with more computation capabilities and storage capabilities, allowing the execution of resource-intensive tasks within the mobile network edges such as running artificial intelligence (AI) algorithms. Exploiting users context information intelligently makes it possible to design an intelligent context-aware mobile edge caching. To maximize the caching performance, the suitable methodology is to consider both context awareness and intelligence so that the caching strategy is aware of the environment while caching the appropriate content by making the right decision. Inspired by the success of reinforcement learning (RL) that uses agents to deal with decision making problems, we present a modified reinforcement learning (mRL) to cache contents in the network edges. Our proposed solution aims to maximize the cache hit rate and requires a multi awareness of the influencing factors on cache performance. The modified RL differs from other RL algorithms in the learning rate that uses the method of stochastic gradient decent (SGD) beside taking advantage of learning using the optimal caching decision obtained from fuzzy rules.
Modified reinforcement learning based- caching system for mobile edge computing
