You are now in the main content area

In-network Cache

Caching in 5G

In this project, we present a cache placement scheme to deal with bandwidth constraint of backhaul links in Small Cell Network for 5G wireless network. We formulated the cache placement problem as a graph matching problem and presented an optimal file-helper matching algorithm. We defined stability criterion for the matching and found that our matching solution is stable in the sense that every helper finds at least one file to cache given that no file exceed minimum cache size. We achieved a unique placement of a file within a cluster of helpers to increase the number of files cached within a cluster. Further, our experimental evaluation demonstrates that our algorithm increases local and neighbor hit ratios as compared to a random placement, which in turn significantly decreases the traffic that goes over the backhaul bottleneck

Caching in Information Centric Network (ICN)

Information Centric Network (ICN) is based on a long-term vision of the network design to make the access and distribution of content easy using publish-subscription model. The four building blocks of ICN are: content naming, request forwarding, in-network caching, and network control. The content name is fundamental to ICN and it should be persistent, globally unique, secure, location independent and user friendly. In ICN, content is identified by a URL-like name. The content routers (CRs) store and propagate the reachability information of content through routing protocol. A user generates Interest packet to identify its interest in receiving a content. All CRs along the path from the content provider to the user cache the content. The ICN provides native multicast since CRs along the path to content can aggregate Interest packets if they arrive within a time window. In-network caching increases the capacity of network processing Interests and reduces content latency. Content streaming and deployment of ICN at the network edge are two practical use cases. The ICN is capable of caching and invoking congestion control optimized for different content type, which enables more effective design of dynamic adaptive streaming (DAS). The NFV can be used to create NS for the ICN that allows cohabitation of ICN and IP in the same 5G edge network. The ongoing work on the convergence of two major ICN architecture, NDN and CCNx has developed their deployment scenarios. In this project, we are investigating cache management in ICN and cache security threats.

Team
  • Farshad Daliri
  • Lubna Badri
Partner
  • Dr. Alagan Anpalagan
  • Dr. Jamshid Abouei

Selected Publications

  • S. Sadeghpour, M. Jaseemuddin, and S. Kharbish, “Cluster-based Cache Placement in 5G Networks”, 20th IEEE Int. Conf. on High Performance Computing and Communications (HPCC), June 2018.
  • L. Mohammed, M. Jaseemuddin, and A. Anpalagan, “Fuzzy Soft-Set Based Approach for Femto Caching in Wireless Networks”, 20th IEEE Int Conf. on. High Performance Computing and Communications (HPCC), June 2018.