An auxiliary caching strategy for mobile cache networks

With the explosive growth of wireless data transmission, wireless content sharing is facing significant challenges such as a surge in data transmission scale, resource shortages, and a volatile communication environment. Consequently, dynamic caching technology based on D2D communication has emerged. Cache network is a distributed network architecture where network nodes (such as user devices or base stations) have caching capabilities and can store part of the content to reduce content transmission delay and network traffic. In cache networks, considering the short distance characteristics of direct device communication and the self offloading advantage of user caching, a cache mechanism is proposed. The paper analyzes the file offload capability and packet loss rate of the cache network. First, this paper analyzes network interference. Considering the randomness of content caching and requests, it derives a closed-form expression of the successful transmission probability based on random geometric theory. Secondly, considering the randomness of content requests and the limited cache capacity, a user cache queue model is established to analyze the transmission process of request files, and an M/G/1 queuing service model is established. The files requested by users can be obtained through two offloading methods: self offloading and device direct connection (D2D) communication offloading. The queue state of the files is constructed as a Markov chain, and the file queue state under different transmission modes is analyzed. Key performance parameter expressions such as successful offloading probability are obtained. Simulation shows that in networks with D2D assisted caching, in the actual situation of limited cache space, the higher the popularity index, the easier it is for the requested content to be satisfied by its own cache, the smaller the cache queue length, and the lower the packet loss rate. Auxiliary caching strategies can leverage the advantages of their own caching and D2D communication, effectively reducing the mean delay for users to obtain request files.