Cyber-physical systems (CPSs) have become more complex, more sophisticated, and more intelligent. In addition to this complexity, they have also been exposed to some important disturbances due to unintentional and intentional events since the number of cyber attacks has increased, and their behaviors have become more sophisticated. The openness, virtualization, and ubiquitous access traits of the combination of CPS and cloud computing may cause coupling problems. When malicious users or attackers simultaneously request the same physical nodes, it may lead to a failure of services as well as a security threat to the system. In this article, we design a low-coupling system based on the edge computing platform to counter coupling problems. The edge computing platform acts as a middleware platform and provides the scheduling method. Based on the edge computing platform and artificial intelligence technology, we design two buffer queues to reduce the coupling degree of the system in parallel. Moreover, we improve the Kuhn-Munkres algorithm to obtain the maximum matching between users' requests and resources to achieve optimal resource distribution. The experimental results indicate that the proposed edge-based scheme can effectively counter the coupling problem for CPSs.

中文翻译：

---

基于智能边缘计算的计算机物理系统中耦合问题的解决方法

网络物理系统（CPS）变得更加复杂，更加复杂和智能。除了这种复杂性之外，由于网络攻击的数量增加了，并且行为变得更加复杂，它们也由于一些无意和有意的事件而受到了一些重要的干扰。CPS和云计算的组合的开放性，虚拟化和无处不在的访问特性可能会导致耦合问题。当恶意用户或攻击者同时请求相同的物理节点时，可能导致服务失败以及对系统的安全威胁。在本文中，我们设计了一个基于边缘计算平台的低耦合系统来解决耦合问题。边缘计算平台充当中间件平台并提供调度方法。基于边缘计算平台和人工智能技术，我们设计了两个缓冲区队列以降低系统的并行耦合度。此外，我们改进了Kuhn-Munkres算法，以获得用户请求和资源之间的最大匹配，以实现最佳的资源分配。实验结果表明，提出的基于边缘的方案可以有效地解决CPS的耦合问题。