Author

Gurpreet Kaur

Publication Date

Fall 2017

Degree Type

Master's Project

Department

Computer Science

Abstract

Advances in cellular network technology continue to develop to address increasing demands from the growing number of devices resulting from the Internet of Things, or IoT. IoT has brought forth countless new equipment competing for service on cellular networks. The latest in cellular technology is 5th Generation Cloud Radio Access Networks, or 5G C-RAN, which consists of an architectural design created specifically to meet novel and necessary requirements for better performance, reduced latency of service, and scalability. As part of this design is the inclusion of a virtual cache, there is a necessity for useful cache management schemes and protocols, which ultimately will provide users better performance on the cellular network. This paper explores a few different cache management schemes, and analyzes their performance in comparison to each other. They include a probability based scoring scheme for cache elements; a hierarchical, or tiered, approach aimed at separating the cache into different levels or sections; and enhancements to previously existing approaches including reverse random marking as well as a scheme based on an exponential decay model. These schemes aim to offer better hit ratios, reduced latency of request service, preferential treatment based on users’ service levels and mobility, and a reduction in network traffic compared to other traditional and classic caching mechanisms.

Available for download on Wednesday, December 12, 2018

Share

COinS