Mobile and Ubiquitous Systems: Computing, Networking, and Services. 7th International ICST Conference, MobiQuitous 2010, Sydeny, Australia, December 6-9, 2010, Revised Selected Papers

Research Article

A Quantitative Comparison of Communication Paradigms for MANETs

Download120 downloads
  • @INPROCEEDINGS{10.1007/978-3-642-29154-8_22,
        author={Justin Collins and Rajive Bagrodia},
        title={A Quantitative Comparison of Communication Paradigms for MANETs},
        proceedings={Mobile and Ubiquitous Systems: Computing, Networking, and Services. 7th International ICST Conference, MobiQuitous 2010, Sydeny, Australia, December 6-9, 2010, Revised Selected Papers},
        proceedings_a={MOBIQUITOUS},
        year={2012},
        month={10},
        keywords={},
        doi={10.1007/978-3-642-29154-8_22}
    }
    
  • Justin Collins
    Rajive Bagrodia
    Year: 2012
    A Quantitative Comparison of Communication Paradigms for MANETs
    MOBIQUITOUS
    Springer
    DOI: 10.1007/978-3-642-29154-8_22
Justin Collins1,*, Rajive Bagrodia1,*
  • 1: University of California
*Contact email: collins@cs.ucla.edu, rajive@cs.ucla.edu

Abstract

Mobile ad hoc networks (MANET) present a challenging area for application development. The combination of mobile nodes and wireless communication can create highly dynamic networks with frequent disconnections and unpredictable availability. Several language paradigms have been applied to MANETs, but there has been no quantitative comparison of alternative approaches. This paper presents the first quantitative evaluation of three common communication paradigms (publish/subscribe, RPC, and tuple spaces) compared within realistic MANET environments using real applications. We investigate the application-level performance of the paradigms and present a summary of their relative strengths and weaknesses. We also demonstrate the impact of wireless and mobility on application-level metrics, the most dramatic being delivery rates dropping to nearly 25% and round trip times increasing up to 2000% in a mobile scenario.