Distributed file allocation using matching game in mobile fog-caching service network

Abstract

The fog-caching enabled radio access network has been identified as an effective solution to reduce content down-loading latency of mobile users (MUs). The quality of service (QoS) can be dramatically enhanced by caching popular files into the storages of femto-cell access points (FAPs) adjacent to MUs. However, file allocation is challenging when the mobility of users is considered. In this scenario, the serving users of each FAP are not fixed, and they vary from time to time. In order to address this, we design distributed algorithms, using two separate matching games, with the aim to minimize the system file downloading latency. To be specific, we first allocate files to each FAP and then associate an MU with a proper FAP, considering the mobility of users. At last, numerical results are presented to demonstrate the effectiveness of the proposed algorithms. It is verified that the proposed algorithms outperform the benchmark in terms of achieving a lower transmission delay.