Delay-Oriented Probabilistic Edge Caching Strategy in a Device-to-Device-Enabled IoT System
Xin Gao, Zhihong Qian, Xue Wang
Abstract
Device-to-device (D2D) communication and edge caching are important technologies in the field of the Internet of Things (IoT), which can offload some data resource requests to the network edge and avoid repeatedly requesting the same data from the data center. However, the existing caching strategies do not fully consider the interference environment of D2D communication and lack the scheme to optimize the user device caching strategy from the perspective of delay. In this article, the cumulative interference for D2D links and cellular links is derived using stochastic geometry. Based on a more comprehensive consideration of interference, this research derives the successful transmission probability of different data resource delivery strategies and models the transmission delays for user to obtain data through its own cache, D2D transmission, and macrocell base station (MBS) transmission. To support low-latency services, a probabilistic caching strategy optimization algorithm for delay minimization is designed and the caching strategy is obtained using an improved particle swarm optimization algorithm. The simulation results show that compared with the typical probabilistic caching strategies, the proposed strategy has a higher success rate of request offloading, and the data transmission delay is significantly reduced.