Enhanced Entropy‐Based Resource Searching in Unstructured P2P Networks. Issue 2 (1st April 2015)

• Record Type: Journal Article
• Authors: Gong, Weihua; Jin, Rong; Yang, Lianghuai; Huang, Decai
• Is Part Of: Chinese journal of electronics. Volume 24:Issue 2(2015)
• Page Start: 229
• Page End: 235
• Publication Date: 2015-04-01
• Subjects: peer‐to‐peer computing -- telecommunication network routing -- telecommunication network topology -- trees (mathematics); enhanced entropy‐based resource searching -- unstructured P2P networks -- P2P overlay topology -- data representation -- query routing tree -- QRT -- optimized routing scheme -- query conditions -- optimal routing tree; Electronics -- Periodicals; Electronics -- China -- Periodicals; Electronics; China; Periodicals; 621.38105
• DOI: 10.1049/cje.2015.04.002
• Languages: English
• ISSNs: 1022-4653

Abstract

Abstract : How to find desired resources efficiently and accurately is one fundamental challenge of any unstructured P2P networks, which is mainly involved some difficulties in the P2P overlay topology, data representationin peers and routing mechanism. In this paper, weaddress the issue of resilient routing in unstructured P2Pnetworks. An efficient algorithm called Query routing tree(QRT) based on maximum mutual information is proposedto improve the performance of resource searching, whichhas tightly associated the resource contents of peers withthe logical links in P2P network that makes the query messagesforwarded more effectively in similar peers and canhit more target resources faster. Additionally, we presentan optimized routing scheme with the query conditionstaken into account, to obtain the optimal routing tree withthe minimal information gain from the candidate tree setso as to adapt to different query types more flexibly. Thesimulation results show the proposed QRT can reduce thesearch cost more effectively and maintain higher targetshit rate than existing typical algorithms such as Flooding, k‐RW and APS. Finally, our optimized scheme is alsoproved to conduct high searching performance with nicerself‐adaptability and scalability in unstructured P2P networks.