Multi-domain Collaborative Awareness And Fusion Analysis Technology For Physical Objects Based On Internet

With the development of Internet of things,the IOT technology bring convenience to people,but also introduced a new security issues at the same time.Cross-domain penetration attacks are fatal to physical object in cyberspace,which could result in major safety accidents and loss of personal property.In order to comprehend the relevant information of the physical object and its distribution in cyberspace,it is necessary to perceive the physical objects in the cyberspace and collect the information as much as possible.At the same time,it's also necessary to combine the multi-domain information of the physical objects to process fusion correlation analysis,for the sake of the association between the physical objects.The main work of this paper includes:(1)Multi-domain Information Search Technology for Physical ObjectsThe Multi-domain information of physical objects can be classified into social domain information,cyber domain information and physical domain information.The method of information searching in each domain is different,and the multi-domain information search of physical objects is carried out by the characteristics of multidomain information of the physical object.In order to collect the physical domain information of the physical object,this paper mainly use the web crawler technique to collect the information.For the cyber domain information of physical objects,this paper collects based on the characteristics of network communication protocol.For the social domain information of physical object,this paper proposes a training model based on the deep neural network to collect the social domain information of the physical object.Finally,the effectiveness of the method by experiment.(2)Multi-domain Collaborative Awareness Model and Method Implementation of Physical ObjectIn order to complete the discovery of physical objects,we need to combine the multidomain information to further perceive and discover the physical objects.To complete the discovery of physical object,the need to combine multi-domain information to further perceive physical objects.In this paper,we propose a physical object sensing method based on multi-domain cooperative model.In physical domain,we use network specific protocol to perceive the physical object.In cyber domain,we use network topology and device fingerprint to detect physical object.In social domain,we mainly perceive related physical objects named entity to perceive the physical object.Finally,a model of multidomain cooperative sensing physical object based on multi-domain information is proposed and experiments are designed to verify.(3)Correlation Clustering Analysis Based on Multi-Domain Information of Physical ObjectsTo analyze the relationship between physical objects,in this paper,a calculation method of fusion of physical objects based on multi-domain information is proposed.We construct multi-domain association model of physical object based on the association of physical objects in social domain,information domain and physical domain.Then,the association map between the physical object nodes is established.Finally,the correlation between these nodes is calculated and analyzed based on the improved Markov clustering algorithm.(4)Build Prototype SystemAfter completing the above physical object information search,physical object multi-domain perception and physical object association calculation works,a physical object search and information display prototype System named NetThing is developed.The system uses B/S architecture,mainly including front web pages and background services,which provides the physical object location distribution query,specific information search and multi-domain information display function.