Port-logistics-oriented Complex Event Processing Of Internet Of Things

With the rapid development of international trade, port logistics plays an increasingly important role in international logistics system because of its advantages such as high safety performance, convenience, low operating costs, high degree of standardization, being able to effectively reduce the complexity of the operating processes, etc. However, while modern port logistics continuously develops, some of the problems are also increasingly prominent, such as the artificial or semi-artificial acquisition and transmission of port logistics information, lack of transparency in port logistics process, the security issues of port logistics. Information level of port logistics has become a key factor restricting the development of port logistics. The rise of the Internet of Things in recent years is an effective way to solve the above problems, and the technology of middleware of Internet of things and the complex event processing are the keys while applying Internets of things to port logistics and giving full play to its advantages.Firstly on the basis of the reference to existing middleware systems and related literatures and considering the characteristics of port logistics, the overall middleware architecture model of a three-tiers and three-modules were designed and given, and detailed design of each module were also given. The port business demands of middleware backend applications were fully taken into account in this design, and the core functions of middleware device management, complex event processing, and multi-source information fusion were highlighted. This design had a certain reference significance to the middleware design in the field of port logistics.Then considering the characteristics of port logistics and complex event processing theory, IoT based port logistics services and complex events involved in port logistics were analyzed and summarized, and based on Rete algorithm, complex events including container aggregate event, container truck route event, container entry/exit event and container loading/unloading were studied. Considering the disadvantage of not supporting timing operation in Rete algorithm, Allen’s time operators were introduced and expanded. After that, simulation experiment comparisons between complex event detection methods based on Rete algorithm and Petri net were made. Experimental results showed that Rete algorithm was applicable to complex event processing, and Rete algorithm was superior to Petri net in the aspects of accuracy, stability and detection efficiency. For large quantities of complex event detection (complex events over1000), a single complex event detection took an average of no more than0.024s, and the accuracy rate was up to100%.Finally, the core function of complex event processing of middleware data processing and fusion-integration module was accomplished using Rete algorithm. This module was able to achieve a variety of complex events from a large number of data streams to achieve the desired effect.