School of Computer Science, Inner Mongolia University, Hohhot 010021, China.
Inner Mongolia A.R. Key Laboratory of Wireless Networking and Mobile Computing, Hohhot 010021, China.
Sensors (Basel). 2018 Jul 7;18(7):2190. doi: 10.3390/s18072190.
The Internet of things (IoT) technology is developing rapidly, and the IoT services are penetrating broadly into every aspect of people’s lives. As the large amount of services grows dramatically, how to discover and select the best services dynamically to satisfy the actual needs of users in the IoT service set, the elements of which have the same function, is an unavoidable issue. Therefore, for the robustness of the IoT system, evaluating the quality level of the IoT service to provide a reference for the users choosing the most appropriate service has become a hot topic. Most of the current methods just use some static data to evaluate the quality of the service and ignore the dynamic changing trend of the service performance. In this paper, an estimation mechanism for the quality level of the IoT service based on fuzzy logic is conducted to grade the quality of the service. Specifically, the comprehensive factors are taken into account according to the defined level changing rules and the effect of the service in the previous execution process, so that it can provide users with an effective reference. Experiments are carried out by using a simulated service set. It is shown that the proposed algorithm can estimate the quality level of the service more comprehensively and reasonably, which is evidently superior to the other two common methods, i.e., the estimating method by a Randomization Test (RT) and the estimating method by a Single Test in Steps (STS).
物联网（IoT）技术发展迅速，物联网服务广泛渗透到人们生活的方方面面。随着服务数量的急剧增长，如何在具有相同功能的物联网服务集中动态发现和选择最佳服务来满足用户的实际需求，已成为一个不可避免的问题。因此，为了提高物联网系统的健壮性，评估物联网服务的质量水平，为用户选择最合适的服务提供参考，已成为一个热门话题。目前大多数方法仅使用一些静态数据来评估服务质量，而忽略了服务性能的动态变化趋势。在本文中，基于模糊逻辑的物联网服务质量水平估计机制被提出，以对服务质量进行分级。具体来说，根据定义的级别变化规则和服务在前一执行过程中的效果，综合考虑各种因素，从而为用户提供有效的参考。通过模拟服务集进行实验，结果表明，所提出的算法可以更全面、更合理地估计服务的质量水平，明显优于其他两种常用方法，即随机测试（RT）估计方法和分步单测试（STS）估计方法。
