Abstract: Trust chain is one of the key technologies in trusted computing, and how to express it and assess it is a current hot research focus. According to the complex factors which influence the trust assessment and the uncertainty and dynamism of trust relation in trust computing environment, a trust computing assessment method was proposed based on fuzzy theory. This method established level five space for the trust particle, the historical measurement record and time attenuation factor were introduced to construct the direct trust, and a fuzzy evaluation method of indirect trust was also given. By defining similar degree function with the improved Einstein operator, the trust entity fuzzy reasoning and evaluation process of trust chain was given. This method combined fuzzy reasoning with trust transferring, and it could assess the trust of entity in trust chain fully. The simulation experiments show that, the proposed method has a better ability to resist the malicious assessment and has better credibility and reliability in the results of the assessment as compared with other similar methods. It is a new method for the trust assessment in trusted computing.

Key words: trusted computing, trust assessment, fuzzy theory, trust chain, measurement

摘要： 信任链是可信计算的关键技术之一，如何对其表达和评估是研究热点。针对可信计算环境中影响信任评价的各种复杂因素以及信任关系的不确定性和动态性，提出一种基于模糊理论的可信计算信任评估方法。引入历史度量记录和时间衰减因子以构建直接信任度，建立粒度为5的信任等级空间，并给出间接信任度的模糊评判方法。运用改进的Einstein算子定义相似度函数，在此基础上给出信任链上实体的信任模糊推理和评估过程。方法把模糊推理和信任传递相结合，可以全面地评估信任链上实体的信任度。模拟实验表明，与同类方法相比较，所提方法对恶意评估具有较好的抵抗能力，且评估结果的可信性和可靠性更好，是一种可信计算信任评估的新方法。

关键词: 可信计算, 信任评估, 模糊理论, 信任链, 度量