Department: Electrical & Computer Engineering
Faculty Advisor(s): Sujit Dey

Primary Student
Name: Hasti Ahlehagh
Email: hahlehag@ucsd.edu
Phone: 858-534-7883
Grad Year: 2013

In this poster we introduce distributed caching of videos at the base-stations of the Radio Access Network (RAN) as a way to reduce the need to bring requested videos from Internet CDNs, thereby addressing backhaul bottlenecks within the RAN by reducing backhaul transmissions, improving video quality of experience - delay and video stalling - and increasing overall network capacity to support more simultaneous video requests. In addition, we investigate supplementing these RAN caches with a hierarchical caching scheme, where the gateways in the Core Network (CN) also have video caches. The hierarchical caching approach further improves video capacity when there is a bottleneck either in the CN or gateway to the Internet by enabling multiple base-stations to share caches at higher levels of the hierarchy. Furthermore, using hierarchical caching can better accommodate mobility. Unlike Internet CDNs that can store millions of videos in a relatively few, large-sized caches, our proposed caching architecture consists of a very large number of micro-caches, with each micro-cache being able to store only thousands of videos, and hence they may not have high cache hit ratios. To address this challenge we propose new caching policies based on the User Preference Profile (UPP) of active users in the cell: Reactive UPP (R-UPP) and Proactive UPP (P-UPP). We further extend these caching policies for hierarchical caching of videos in the CN and evaluate their performance relative to traditional caching schemes, namely Most-Popular-Videos (MPV) and Least-Recently-Used (LRU). Further, we propose a video scheduling method that allocates the RAN and CN backhaul resources so as to reduce video latency and increase network capacity. We develop a simulation framework using MATLAB to study the performance of RAN-only and hierarchical caching. Our simulation results show that the RAN micro-caches with the proposed UPP-based caching policies, along with the proposed scheduling approach, can improve the probability that video requests can meet initial delay requirements by almost 60%, and network capacity by up to 100%. The results also show that UPP based policies can enhance network capacity by up to 30% compared to conventional caching policies. Furthermore, our simulation results show that using hierarchical caching can enhance cache hit ratio by 24% and network capacity by up to 45% compared to RAN-only caching in networks where the CN is the capacity bottleneck. For the specific settings with high mobility, hierarchical caching can lead to a 47% capacity improvement compared with the RAN-only caching.

« Back to Posters or Search Results