Research Of Key Technologies Of Credentials In Automated Trust Negotiation

Automated trust negotiation is an effective way to achieve the sharing of the P2P network resource across security domains and to solve the issues regarding accessing security. The credential module is an underlying sub-module of the automated trust negotiation system, it is responsible for the credential storing and credential chain discovery. The design of the credential module has a direct impact on the establishment of a trust relationship between or among the negotiating parties. The credential module consists mainly of credential storage strategy and credential chain discovery algorithm. The existent distributed credential storage strategies have the problem of poor fault-tolerant. The thesis pays much attention on design a good fault-tolerant storage strategy which can be used by trust credentials and designing an efficient search algorithm based on the strategy of the trust credential.In order to solve the poor fault-tolerant questions of the existing distributed storage strategy, an improved storage strategy based on resource redundancy theory is proposed on the basis of the analyses of existing credential storage strategy. The strategy employs multiple backup trust credential, when the network node failure problem occurs, timely restoring missing trust credential, credential chain of trust established to avoid defeat.On the basis of the improved distributed storage strategy, a search algorithm regarding trust credential chain is proposed by this paper. The algorithm uses greedy search algorithm and two-way searching algorithm, which increases the efficiency of the algorithm by limit the size of trust credential graph during the process of credential chain discovery.The result of the credential module compared with other credentials modules shows that the credential module has a good fault-tolerance and load balance, it can minimize the impact of the failure node against the system. and the credential chain discovery algorithm has high efficiency compared to other algorithms, .