Reactive eviction algorithms for radio access networks

The double-digit growth of the number of mobile broadband subscribers has led to the exponential growth of the mobile traffic in the last several years. A major part of the mobile traffic is the duplication of the downloading the popular contents. A cache-enabled mobile network in which the popular contents are replicated at the radio access network (RAN) helps to reduce the transit traffic on the backhaul links and to improve the quality-of-experience of the end users. In this paper, we propose two reactive eviction algorithms called max-hit and min-transit with respect to two different caching objectives for a base-station in the RAN. Max-hit algorithm aims to maximize the number of cache hits at the base-station whereas min-transit algorithm minimizes the transit traffic at the backhaul link. We also develop an event-driven simulation to evaluate the performance of the proposed eviction algorithms. The complexity of our proposed algorithms is not higher than least-frequentlyused. However, depending on the objective of the content caching, our proposed algorithms show a better performance than the traditional eviction policies such as least-recently-used and leastfrequently- used.