Research On Intent-based IoT Architecture And Algorithms

People think that the Internet of Things(IoT)is the result of the third revolution in technology.Its goal is to connect everything in the real world so that people can use more automated and smart services to make their lives easier.IoT has become one of the most talked about technologies as communication technology has grown quickly.IoT is now used in almost every industry and has a wide range of uses,such as smart homes,industrial automation,and smart farming.Due to the rapid development of the IoT,manufacturers are competing to design their own systems and products,resulting in a vertical development trend and strong device heterogeneity.It is difficult to control and reuse resources between systems,and devices lack semantic information,making it difficult to provide automated services to users.This article addresses these problems through research and analysis,with the main focus as follows:(1)This thesis proposes a universal device data model that describes how devices are put together on the inside based on how they are used.A device-to-device interaction model is designed based on this data model,which includes interaction modes between cloud and local devices as well as between local devices.In this interaction model,differences between directly connected devices and indirectly connected devices are also considered.This model provides the possibility for unified identification and control of heterogeneous devices between different systems.(2)This thesis tries to explain what IoT users want to do by using an intent model.After that,a modular IoT device ontology based on intent is made by taking into account the problems with existing ontologies in the IoT field.This ontology contains intent information,context information,and QoS information about device services,making it suitable for different application scenarios and providing a foundation for service matching.(3)This thesis proposes a hierarchical matching method based on the device ontology.The ontology’s method for calculating semantic similarity is improved,and the similarity between user intent and device services is calculated based on three factors:intent,context,and QoS.This helps choose the service that best fits the user’s needs.The simulation results show that the hierarchical matching algorithm can improve the accuracy of 8.7% compared with the traditional service matching algorithm.(4)This thesis designs and implements a simple demonstration system to automate the process from the user entering a request to the final device service call.This shows that the ideas in this article are possible.