“Developing caching techniques using Collaborative Filtering methods in 5G Networks” by Farnaz Hassanzadeh

Loading Map....

Date/Time
Date(s) - 14/12/2017
4:00 pm - 5:30 pm

Location
IBM Lab

Categories


Abstract
As a solution for decreasing end-to-end latency, we propose collaborative filtering (CF) on mobile edge computing (MEC) based caching method . CF is a mechanism in which
automatic predictions will be generated based on users preferences by collecting similar information from other users and MEC provides a situation in which the applications are
deployed and ran at the edge of a mobile network, which is close to users. It can support low latency schemes by providing the shortest route between the applications. Inserting
the applications nearby the sources improves the performance of the network by processing large amounts of data locally.
To improve the end to end latency in 5G mobile networks, we use caching method. The experiences show that most of the request from the users are repeating frequently. If there would be a method to prevent transmission of the same request over and over, the transmission delay will decrease. To reach this issue, caches are used to store the contents of the requests and whenever there was a need for a repetitive request, the content can be easily prefetch from cache without any delay. Here, the most important problem is locating the caches in such a way to get the maximum efficiency. Moreover, due to limitations in the capacity of the caches, it should be effectively decided what to be stored and released in the caches.

CF caching,pdf