Font Size: a A A

Research On Some Key Issues Of Security For Privacy Preservation In Internet Of Things

Posted on:2015-01-18Degree:DoctorType:Dissertation
Country:ChinaCandidate:Q M LinFull Text:PDF
GTID:1228330467474593Subject:Information networks
Abstract/Summary:PDF Full Text Request
The internet of things has been paid more and more attention by the government, academe andindustry all over the world. The core concept of IoT is perception, control, transmission andintelligence. IoT can construct an integrated network system over the wireless sensor networks,Internet and mobile communication network. Thereby IoT can achieve a collaborative relationshipbetween objects and persons. The data that collected from users may have some sensitiveinformation on their daily activities(for instance, environments, location, habits of rest and work,and so on), and the information is regarded as privacy by some IoT users. And hence privacypreserving and security protection become increasingly vital in applications of IoT. However,subject to the network structure, terminal equipment, communication methods and applicationscenarios, privacy and security issues in IoT cannot be directly resolved through the existingInternet security techniques. Therefore, to address this need, it becomes necessary to deeply studyprivacy preserving issues in IoT.First of all, we make a brief description on IoT and its security protection and privacypreservation. Then, the states of IoT privacy security research and application home and abroad aregiven in this dissertation. Later, we summarize the existing privacy security issues over the IoT.Finally, we deeply study four key issues which are secure key predistribution, mutual authenticationand key agreement, digital authentication and steganography. The main jobs of the dissertation areshown as follows.(1) With the widespread application of internet of things, security in wireless sensor networks isa hot research field which is essential for privacy preserving. Key management plays an vital role inachieving security in wireless sensor networks. In this dissertation, an elliptic curve based keypredistribution approach for wireless sensor networks is offered. In the proposed approach, a seedkey, which is a distinct point in an elliptic curve, is assigned to each sensor node prior to itsdeployment. The private key ring for each node is generated via the point doubling mathematicaloperation over the seed key. If two sensor nodes share a common private key, then a secure link isestablished between these two sensor nodes. By properly selecting the size of key ring, theprobability of two sensor nodes sharing the same private key could be raised. Simulations indicatethat the performance is better for the proposed approach compared with the other approach in termsof connectivity and resilience against node capture. (2) Authentication and key agreement can serve as a core security guard in the radio frequencyidentification(RFID) system, as is the basis for the implementation of access control. A novel RFIDmutual authentication and key agreement protocol is proposed, which can effectively resolve theRFID system privacy protection issues. The main features of the protocol are safety, low storagecosts, low communication costs, novelty, easy to implement and so on. The protocol can effectivelyresist server compromise attacks, the man-in-the-middle attacks, offline guessing attacks andimperceptible online guessing attacks, while providing mutual authentication, strong secure sessionkey, untraceability and forward security. All in all, the protocol is very suitable for the application inRFID systems which require privacy protection.(3) A privacy secure communication mechanism is necessary in the applications of wirelessmultimedia sensor networks, which is more vulnerable to privacy security attacks due to thepresence of multimedia data. Additionally, given the limited technological resources (in term ofenergy, computation, bandwidth, and storage) of sensor nodes, security and privacy policies have tobe combined with energy-aware algorithms and distributed processing of multimedia contents inwireless multimedia sensor networks. To solve these problems, an energy efficient distributedsteganography scheme, which combines steganography technique with the concept of distributedcomputing, is proposed for privacy secure communication in wireless multimedia sensor networks.The simulation results show that the proposed method can achieve considerable energy efficiencywhile assuring the privacy communication security simultaneously.(4) As one of the most successful techniques of authenticaiton, face recognition has recentlyreceived significant attention, especially during the past few years. In order to construct anautonomous and robust security system in IoT, the auther explores the application of facerecognition technique in wireless multimedia sensor networks. A facial components basedrecognition mechanism is firstly applied to ensure the recognition accuracy. Secondly, in order toaddress the problem of resource constraints, a distributed scheme based on K-d tree is deployed forboth the face image transmission and retrieval. According to the simulation results, the proposedmethod is capable of achieving considerable energy efficiency, while assuring the recognitionaccuracy. Although the face recognition system gains flexibility by integrating into a wirelessmultimedia sensor network, unprotected facial feature directly expose to the malicious personsduring face recognition. As a result, users’ privacy and security can not be guaranteed. A facialfeature steganography based on wavelet transform technique described earlier is proposed. It embeds encrypted facial feature into cover image, as makes criminals can not distinguish or toextract useful facial feature when the hidden image is even intercepted. The purpose of privacyprotection is achieved, but facial feature steganography also faces the threats of replay attacks. Andhence, a facial feature steganography based authentication and key agreement protocol is proposedso as to offer effective protection of system authentication and privacy security.
Keywords/Search Tags:Security in Internet of Things, Privacy Preserving, Wireless Sensor Networks, RadioFrequency Identification, Elliptic Curve Cryptography, Authentication and Key Agreement, Steganography, Face Recognition
PDF Full Text Request
Related items