Understanding complementary multi-layer collaborative heuristics for adaptive caching in heterogeneous mobile opportunistic networks

Huynh, Vu San Ha and Radenkovic, Milena (2018) Understanding complementary multi-layer collaborative heuristics for adaptive caching in heterogeneous mobile opportunistic networks. In: The 14th International Wireless Communications and Mobile Computing Conference (IWCMC 2018), 25-29 June 2018, Limassol, Cyprus. (In Press)

WarningThere is a more recent version of this item available.
[img]
Preview
PDF - Requires a PDF viewer such as GSview, Xpdf or Adobe Acrobat Reader
Download (455kB) | Preview

Abstract

Current research aims to deal with emerging challenges of the opportunistic discovery of content stored in remote mobile publishers and the delivery to the subscribers in heterogeneous mobile opportunistic networks. Innovative network and service architectures leverage in-network caching to improve transmission efficiency, reduce delay and handle disconnections. In this paper, we investigate the influences of multi-dimensional heuristics utilised by our adaptive collaborative caching framework CafRepCache on the performance of content dissemination and query in heterogeneous mobile opportunistic environments. We consider the complementary multi-layer heuristics that combine social driven, resources driven, ego network driven and content popularity driven analytics. We extensively evaluate the performance of each complementary heuristic and discuss the impact of each one on every layer of our caching framework across heterogeneous real-world mobility, connectivity traces and use YouTube dataset for different workload and content popularity patterns. We show that the multilayer heuristics enable CafRepCache to be responsive to dynamically changing network topology, congestion avoidance and varying patterns of content publishers/subscribers which balances the trade-off that achieves higher cache hit ratio, delivery success ratios while keeping lower delays and packet loss.

Item Type: Conference or Workshop Item (Paper)
Keywords: Opportunistic mobile networks, content discovery and retrieval, content caching, multi-layer heuristics and utilities.
Schools/Departments: University of Nottingham, UK > Faculty of Science > School of Computer Science
Depositing User: HUYNH, VU
Date Deposited: 10 Apr 2018 11:01
Last Modified: 25 Jun 2018 12:00
URI: http://eprints.nottingham.ac.uk/id/eprint/51047

Available Versions of this Item

Actions (Archive Staff Only)

Edit View Edit View