Delay-aware User Association for Caching Enabled Small Cell Networks

Abstract

Capable of reducing the content-user distance, content caching is shown to increase the hit ratio, relieve the backhaul pressure and improve the overall system throughput in caching enabled small cell networks (CSCN). In this work, however, we show that the performance of caching will be significantly degraded in terms of the local delay in dense CSCN. Especially, local delay, which is defined as the number of time slots for one successful transmission, first decreases and then exponentially increases with the growing deployment density of small cell base stations (SBSs) when the number of users is large. The reason is that transmission failures are more likely to occur supposing that content-oriented user association is applied and over-deployed SBSs are located closer to the intended user than the associated SBSs, which results in overwhelming near-field interference. To reduce the local delay, we design and optimize a delay-aware user association strategy, in which a distance threshold is used to determine whether to obtain content through content caching or backhaul resources. Through the optimization of distance threshold, the near-field interference could be effectively alleviated. Numerical results show that the local delay performance is improved under delay-aware user association strategy, especially for high signal-to-interference ratio (SIR) threshold.