With the rapid development of wirelessly networked applications, components interact with each other more frequently. An essential challenge for security is to determine how one network entity can trust another. Most existing trust models evaluate trust, Choosing a single probability-based parameter based on probabilistic estimation. However malicious nodes can take advantage of this to either gain unfair trust values for themselves, or to degrade the trust values of other nodes. A multi-parameter trust framework for MANETs (MTFM) presented in this thesis employs multiple types of parameters for the evaluation of trust based on Grey theory. The proposed framework classifies the trust relationships into three types: direct, recommendation and indirect. The mobility of the MANET nodes in a trust grouping may cause large variances in the trust values. Simulations conducted in 6-node network demonstrate that MTFM can maintain consistent total trust values in the presence of various types of mobility. Misbehaviour detection strategies for the MTFM apply weight vector groups for multiple parameters, based on the design of an Analytic Hierarchy Process. This approach brings a significant benefit; the MTFM can not only detect misbehaviour, but can also detect the particular parameter use<:! in the strategy of the malicious node. A new combined prediction model provides a boundary to permit discrimination between normal and abnormal behaviour. And additional contribution is the issue of the trust group size. Although the resulting trust value can be resilient to unfair evaluation from malicious neighbour nodes, however, due to their limited recourses, having a lot of neighbours is not practical for wireless mobile nodes. Therefore. there needs to be a balance between maintaining trust values and the insensitivity of a subgroup to a changed trust value. Results and analysis are presented to indicate the preferred Size of a subgroup among a much larger network.
| Identifer | oai:union.ndltd.org:bl.uk/oai:ethos.bl.uk:601627 |
| Date | January 2013 |
| Creators | Guo, Ji |
| Publisher | Queen's University Belfast |
| Source Sets | Ethos UK |
| Detected Language | English |
| Type | Electronic Thesis or Dissertation |
Page generated in 1.1587 seconds