Research And Implementation Of Privacy Protection For IoT In The Context Of Smart Home

With the development of wireless communication and sensor technology,the emergence and development of the Internet of things with the meaning of the times has been promoted.The Internet of things has become a hot topic in the field of IT,which has been widely concerned and developed rapidly in recent years.Among them,the smart home as an important branch of the Internet of things,to bring great convenience to the people at the same time,privacy protection is also facing enormous challenges.First of all,in the perception layer,through a method of data encryption can guarantee the security of data fusion,but attacks cannot decrypt the ciphertext transmission case through statistical analysis to obtain the user's privacy;secondly,when the data fusion to the sink node,mapping the traditional access control can only achieve a single authority and data source that cannot be associated reasoning,an attacker can get the edge information,infer the privacy of data;finally,when the data transmission through the Internet to Home Furnishing intelligent service provider,if cloud security cannot be guaranteed,there are still risks to user privacy disclosure.Aiming at the above problems,this paper based on the existing work,the construction of user privacy protection framework for intelligent Home Furnishing environment,focusing on data attacks against the perception layer method,reasoning method of access control and data privacy protection scheme based on homomorphic encryption based on.The main results are as follows:1.In this paper,we propose a privacy preserving method based on supervised learning in the smart home environment.This method mainly deals with the behavior of the attack on the data sensing layer.By allowing the sensor nodes of the sensing layer to automatically add noise to protect the privacy of users,so as not to be stolen by the bypass attack.The method can generate the simulation parameters through the analysis of large data,and then add the noise to the perceptual nodes adaptively by learning the user habits.Compared with other existing algorithms,this algorithm has the advantages of good privacy protection and high efficiency.2.In the network layer,an ontology based access control method is proposed.The user sets the privacy according to the privacy attribute,but if sufficient non privacy data is available,the user's privacy data is inferred.Ontology based access control method,according to the privacy permissions set by the user,according to the requestor context request information ontology reasoning,set may leak privacy incidents,so as to judge whether the request leads to leakage of user privacy.Compared with the existing access control model,the access control method can effectively enhance the privacy protection effect and avoid the privacy leakage caused by associated data inference.3.A privacy preserving data protection method based on homomorphic encryption is proposed.The method can guarantee the security of data in the transmission process and smart home cloud server.The use of homomorphic encryption algorithm for data encryption functions and privacy protection function to ensure the security of data transmission and cloud computing environment.