Contact duration aware cache refreshing for mobile opportunistic networks. Issue 4 (1st July 2016)

• Record Type: Journal Article
• Authors: Wang, Fang; Wang, Zhaocheng; Yang, Zhixing; Chen, Sheng
• Is Part Of: IET networks. Volume 5:Issue 4(2016)
• Page Start: 93
• Page End: 103
• Publication Date: 2016-07-01
• Subjects: cache storage -- mobile computing -- Markov processes -- tree data structures -- graph theory; mobile opportunistic networks -- mobile terminals -- data access latency reduction -- file update -- caching node mobility -- cached data freshness -- contact duration aware cache refreshing -- two‐state Markov chain -- tree structure -- refreshing efficiency improvement -- trace‐driven simulations; Computer network architectures -- Periodicals; Computer network protocols -- Periodicals; Information networks -- Periodicals; Telecommunication systems -- Periodicals; 004.605
• DOI: 10.1049/iet-net.2015.0086
• Languages: English
• ISSNs: 2047-4954

Abstract

Abstract : Caching is a practical data access approach in mobile opportunistic networks, which enables mobile terminals to obtain data from adjacent nodes and to reduces accessing latency, whereby the updating files should be delivered to the caching nodes on time when the source is refreshed in order to maintain the freshness of the cached data. Due to the node mobility and short communication range, the contact duration in mobile opportunistic networks is short and this limits the amount of data transmitted in one contact. However, current cache refreshing schemes often ignore this critical issue. In this study, the authors study the contact duration aware cache refreshing in mobile opportunistic networks. Specifically, they establish a two‐state Markov chain to describe the data delivered from one node to another. Based on the proposed model, they investigate a tree structure to refresh the cached data, where each node is responsible to refresh the data cached at its children and it may opportunistically rely on some relays outside the tree to improve the refreshing efficiency. Extensive trace‐driven simulations validate the effectiveness of the authors' scheme in terms of improving cache refreshing in challenging mobile environments.