A Transaction Model For The Semantic Web Of Things And Its Applications

As an emerging information technology of the 21^st century,the Internet of Things?IoT?has been playing a vital role in many fields of modern life.In the past,however,the lack of intelligence had once depressed the great potential of IoT.Hence,although people enjoyed many conveniences brought by IoT,there had been always great demands for the intelligent IoT.Recently,the advent of Semantic Web?SW?technology offered new insight into the intelligence of IoT.Not surprisingly,the Semantic Web of Things?SWoT?,a combination of IoT and SW,has become the mainstream of intelligent IoT research.Within this context,our research on SWoT comprises the following five parts:1?The Semantic Web-based model for transactions on SWoTWe proposed a model of transactions on SWo T based on SW technology.As a kind of intelligent IoT,SWoT is different from the general sensor networks in that it needs a special model to combine the Semantic Web and the IoT into one.In order to achieve this goal,by taking advantage of various intelligent technologies,we proposed a model for transactions on SWoT.The framework consists of top ontology,entity link layer,semantic label layer,service register center,transaction construction layer,transaction execution,and control layer.To satisfy the needs of intelligent IoT,the model first represents entity functions with Semantic Web services.Then it constructs the sequence of services automatically to meet the specific needs according to the analysis of the information.Also,it acts as a manager during the execution of a transaction and makes effective management and control to the entities.The model extends the IoT from sensor network to the real interconnection and also provides the underlying structure support for the intelligent IoT and the further study.We demonstrated the effectiveness and superiority of the model with a case study of the comprehensive rescue service for a third party logistics transportation accident.2?The top ontologies for the model of transactions on SWoTAccording to the model of transactions on SWoT,we proposed the top ontologies to meet the requirements of the model.As an inteligent Io T,SWoT not only includes domain ontology in various areas but also includes the ontologies to describe the transaction control.For describing the transaction control in SWoT,we proposed the top ontologies based on the concepts and relationships that may be involved in SWoT.The top ontologies include activity ontology,goal ontology,status ontology,role ontology,constraint ontology,space-time ontology and the corresponding domain ontology.In addition to their respective definitions,we also present a detailed description of the relationships between them.The proposed top ontologies provide the reasoning basis for the construction,execution and control of the transactions as the basic logic support of the SWoT transaction model.Experiments showed that in the third-party logistics transportation system,the top ontologies can effectively describe entities and support logical reasoning.3?The user oriented semi-automatic method for domain ontology constructionWe proposed a user oriented semi-automatic construction method for domain ontology.Information and concepts in domain ontology grow rapidly with the extensive development of IoT.It takes a very long cycle to build the domain ontolgy by domain experts.Furthermore,the built ontology may not meet the needs of end-users.On the other hand,as for the automatically generated ontology,its accuracy is low and user flexibility and scalability are poor.To address these issues,we developed a method for end-users to construct and expand the domain ontology according to their needs during the use of IoT.Our method only needs the users to input a descriptor set according to their requirements.Next,it automatically establishes a hierarchy structure based on descriptor set.With this method,the users can construct their own ontologies without the participation of experts.Besides,the method also applies to the expansion and collaborative development of ontologies.The validity of the method is verified by constructing the use case and the comparison test in our case study.4?The OWL-S based description for the dynamic entity servicesWe proposed a method to describe the entity services dynamically by extending OWL-S.In the proposed model of transactions on SWoT,Web Services have become an effective way to describe the entity functions.However,as a specification of information processing,Web Services cannot fully meet the needs of execution and control for the transactions on SWoT,for the effective perception and expression of physical space entities are missing.In order to handle this problem,a service description method is proposed based on OWL-S to add extended entity state description information.To ensure the validity of the transaction construction and execution,we use the QoS property as services selection inputs,as well as the extended information of entity status.Such description not only ensures the effectiveness at the information level,but also the effectiveness for the physical space entities.The method is compared with the Web service description method with the simulated test cases,which demonstrated the superiority of the method.5?The case study of the third-party logistics transportation systemWe implemented the model of transaction on SWoT in a case study of the third-party logistics transportation system.The SWoT transaction model is used to simulate the logistics transportation process.Service providers in the logistics system only need to care about goods delivery,receipt and costs.The storage time and place,selection of transport vehicles,transportation path determination and other issues should be decided by the IoT system.In the proposed model of transactions on SWoT,the reasoning is performed by the top ontologies,the entity functions are encapsulated and described as dynamic Web Services with extended OWL-S.Both the theoretical analysis and experimental studies validated that such a system greatly reduces the human involvement and interference.In summary,this thesis is centered around intellegent IoT.By combining IoT and SW,we proposed a model for transactions on SWoT.In particular,our work is focused on the description of the top ontologies,the construction for the domain ontology,and the formulation of dynamic entity services by extending OWL-S.Finally,we validated our model and methods through a case study of the third-party logistics transportation system.The experimental results proved that our method can construct and execute the user's transaction effectively.Compared with the traditional IoT models,our model is more automatic and intelligent in dealing with complex problems.Also,it is flexible enough to be easily extended in different fields with the corresponding domain ontologies to achieve their respective needs.