This dissertation was written as a part of the MSc in ICT Systems at the International Hellenic University.
Caching at the network edge is a recently proposed technique for the upcoming mobile generation 5G, to reduce the backhaul rates at the peak hours by prefetching popular contents and store them into memories at or near to the end users. However, we focus on a new revolutionary caching scheme named as coded caching that take advantage of the multicast medium of the mobile network to offer a considerable gain through information theory coding techniques. In this work, we analyze the performance of two dominant approaches. A comparative simulation-based study has been established of uncoded and coded caching under various levels of spatial locality of the user contents.
Our simulation results show that LFU (Least frequently used) uncoded caching scheme provides a better performance than coded caching schemes for real-life scenarios which were represented in our simulation as non-uniform content popularity. In addition, coded caching scheme still needs additional improvements regarding the supported number of users as well as the computational complexity imposed on users and server sides.
Collections
Show Collections