The Trust Management Researches About Trust Quantification-indicate And Credential Fair-exchange

In real life, trust is a basic requirement among people engaged in trading activities. In such transactions, people tend to be face to face, or is based on the historical basis of the once face-to-face interaction experience. With the rapid development of computer-related technology and the increasing popularity of the online community network applications, people in the network virtual society will no longer need the face-to-face. Therefore, it is necessary to build a safe and reliable trust mechanism of the virtual network society, which can help people to quickly establish a trust relationship when online transactions and other activities, and meet the trust management requirements of network applications on the entity identity and its related social attributes.In order to carry out the effective management on the trust relationship among the network entities, with the trust management researches on the trust relationship in the network environment, this paper divides the trust management into the subjective trust management based on the reputation and the objective trust management based on the credentials. For the quantification-indicate of user trust attributes and the fair-exchange of user trust credentials, and focus on the soft-division method of trust-attribute-interval, the trust management model of multi-dimensional attribute, and the fair-exchange improved model of trust credentials, our research work carried out from three aspects as follows.1)A soft-divided method for trust range based on cloud theory was proposed. For the question about the quantification-indicate of user trust attributes, different from the hard-division method based on the equidistant intervals of trust range, this paper proposed a soft-divided method for trust range based on cloud theory(TRSDM). With the collected data through the the conversion of atomic-concept, the leap of qualitative-concept, and the formation of interval-concept-table, the TRSDM method defines a detail process of the trust range soft-divide. The TRSDM has achieved the convertion from the quantitative values of the trust-attribute to the qualitative concept of the trust relationship, and it implements the soft-division of trust-attribute-range based on the qualitative concept of cloud theory using the interval-concept table. The simulation experiments and results analysis shows that the TRSDM method can effectively solve the distribution characteristics of the data collected, the uncertainty of qualitative concept, and the timeliness of the data collected.2)A trust model based on the trust-degree concept table of multi-dimensional trust-attribute was proposed. For the question about the quantification-indicate of user trust attributes continuted, different from the indication method which merges the trust-degrees of the multi-dimensional trust-attribute, based on the TRSDM method, this paper proposed a trust model based on the trust-degree concept table of multi-dimensional trust-attribute(TCTTM). With the formation of interval-concept-table, the formaiton and update of trust-degree-concept-table, the determination of membership-concept, the evalution of trust-degree, the TCTTM model defines a detail process of the combinatorial indication of the multi-dimensional trust-attribute. The TCTTM model has revealed the intrinsic correlation among multi-dimensional trust-attribute, and it implements the indication of multi-demensional trust-attribute based on the trust-degree concept of cloud theory using the trust-degree table. The simulation experiments and results analysis shows that the TCTTM model is an effective solution to how to use more trust-degree qualitative concept to evalute entity’s trust-degree.3)A four-step fair-exchange protocol and a credential exchange model improved were proposed. For the question about the fair-exchange of user trust credential, firstly this paper proved that the non-repudiation and fair-exchange protocol based on the third-party at least run four-step. Then this paper propose a common model of convertible authenticated encryption schemes, propose a four-step and fair-exchange protocol(FFEP) based on the convertible authenticated encryption scheme using the common model. Formal analysis shows that the FFEP protocol has non-repudiation features of both sender and receiver. Lastly for the question about the credential fair-exchange in trust development stage of automated trust negotiation HBTN model, this paper propose a fair-exchange model improved based on the FFEP protocol, to prevent the unfair exchange of the credential in automated trust negotiation process.