QoS可信的服务组合技术研究

Survey on service composition approach considering trustworthiness of QoS data

现有基于QoS的服务组合方法通常假定QoS数据可信,然而这种假设很难保证,虚假的服务对于服务组合实现的影响常常是致命的。针对该问题,提出一种考虑QoS可信的服务组合方法。分析了QoS来源,建立了相应的系统结构,划分了信任级别,根据社会交往方式量化了直接经验和间接经验在可信判断中的作用,并以此构建QoS的计算模型,设计了相应的算法。最后,通过仿真实验,将考虑可信的服务组合方法与未考虑可信的传统组合方法进行了比较,前者体现出明显的优越性,从而表明提出的方法在性能和执行效率上是可行的。

The existing QoS-based services composition approaches always assume that the QoS data coming from service providers and users are effective and trustworthy, but it is impossible in reality. The influence of deceiving services for service composition is generally fatal. So, this paper proposed a service composition optimize approach considering the trustworthiness of QoS data. It analysed the source of QoS data and proposed relevant system architecture, then proposed an algorithm to formulate the QoS dependability. According to the social manner, it gave an introduction of direct evidence and indirect evidence in QoS computation, differentiated the function of direct evidence and indirect evidence, and then proposed the model of QoS computation. It further gave corresponding algorithms. Finally, implemented a prototype system to evaluate the proposed approach through comprehensive experiments and achieved improved results both in usability and efficiency.