Benchmarking of recommendation trust computation for trust/trustworthiness estimation in HDNs

Hesham El-Sayed, Heng Yu, Gokulnath Thandavarayan, Sharmi Sankar

    Research output: Contribution to journalArticlepeer-review


    In the recent years, Heterogeneous Distributed Networks (HDNs) is a predominant technology implemented to enable various application in different fields like transportation, medicine, war zone, etc. Due to its arbitrary self-organizing nature and temporary topologies in the spatial-temporal region, distributed systems are vulnerable with a few security issues and demands high security countermeasures. Unlike other static networks, the unique characteristics of HDNs demands cutting edge security policies. Numerous cryptographic techniques have been proposed by different researchers to address the security issues in HDNs. These techniques utilize too many resources, resulting in higher network overheads. This being classified under light weight security scheme, the Trust Management System (TMS) tends to be one of the most promising technology, featured with more efficiency in terms of availability, scalability and simplicity. It advocates both the node level validation and data level verification enhancing trust between the attributes. Further, it thwarts a wide range of security attacks by incorporating various statistical techniques and integrated security services. In this paper, we present a literature survey on different TMS that highlights reliable techniques adapted across the entire HDNs. We then comprehensively study the existing distributed trust computations and benchmark them in accordance to their effectiveness. Further, performance analysis is applied on the existing computation techniques and the benchmarked outcome delivered by Recommendation Trust Computations (RTC) is discussed. A Receiver Operating Characteristics (ROC) curve illustrates better accuracy for Recommendation Trust Computations (RTC), in comparison with Direct Trust Computations (DTC) and Hybrid Trust Computations (HTC). Finally, we propose the future directions for research and highlight reliable techniques to build an efficient TMS in HDNs.

    Original languageEnglish
    Pages (from-to)612-630
    Number of pages19
    JournalInternational Journal of Computers, Communications and Control
    Issue number5
    Publication statusPublished - 2017


    • Direct trust computations
    • Heterogeneous distributed networks (HDNs)
    • Hybrid trust computations (HTC)
    • Receiver operating characteristics
    • Recommendation trust computations (RTC)

    ASJC Scopus subject areas

    • Computer Science Applications
    • Computer Networks and Communications
    • Computational Theory and Mathematics


    Dive into the research topics of 'Benchmarking of recommendation trust computation for trust/trustworthiness estimation in HDNs'. Together they form a unique fingerprint.

    Cite this