Web services are self-describing and self-contained modular applications based on the network. With the deepening of web service applications, service consumers have gradually increased their requirements for service functions and service quality. Aiming at how to select the optimal plan from a large number of execution plans with the same function and different QoS characteristics, this paper proposes a web service selection algorithm that supports QoS global optimization and dynamic replanning. The algorithm uses position matrix coding to represent all execution paths and replanning information of the service combination. By calculating the Hamming distance of the service quality between individuals, the quality of the service portfolio is improved. By specifying the total user time limit and implementing a good solution retention strategy, the problem of the impact of algorithm running time on service quality is solved. The experimental results show that the method proposed in this paper is effectively integrated into the development trend of QoS and close to the requester’s needs and can better meet user needs. This algorithm improves the user’s satisfaction with the returned service to a certain extent and improves the efficiency of service invocation.

中文翻译：

---

物联网中基于动态QoS的服务发现和选择

Web服务是基于网络的自我描述和独立的模块化应用程序。随着Web服务应用程序的不断深入，服务使用者逐渐增加了对服务功能和服务质量的要求。针对如何从功能相同，QoS特性不同的大量执行计划中选择最优计划，提出了一种支持QoS全局优化和动态重计划的Web服务选择算法。该算法使用位置矩阵编码来表示服务组合的所有执行路径和重新计划信息。通过计算个人之间服务质量的汉明距离，可以改善服务组合的质量。通过指定用户总时间限制并实施良好的解决方案保留策略，解决了算法运行时间对服务质量的影响问题。实验结果表明，本文提出的方法可以有效地集成到QoS的发展趋势中，并接近请求者的需求，可以更好地满足用户的需求。该算法在一定程度上提高了用户对返回服务的满意度，并提高了服务调用效率。