Filtering Effects of Media Merging in a Cache Hierarchy

In wireless networks, bandwidth is extremely scarce, especially for supporting on-demand media streaming. This study uses simulation to examine a caching hierarchy for wireless streaming video distribution, in combination with multicast stream merging. The main purpose is to gain insight into the filtering effects caused by caching and merging. Three metrics, namely entropy, request frequency vector, and inter-reference time vector, are used to characterize request traffic in a caching hierarchy. This paper shows how these metrics can assist with cache management decision-making.