| As the development of MEMS, mobile computing, and wireless communications technology, People-centric Sensing Network has a large space to develop and a big application prospect as a new type of information transmission systems. However, due to the features, such as dynamics and vulnerability, many traditional privacy-preserving measures are no longer applicable. Therefore, designing a privacy-reserving scheme for People-centric Sensing Network is an important and challenge work. Considering the shortage of the current schemes, we investigate the privacy preservation technology for data transmission, data aggregation, and data query in People-centric Sensing Network in this dissertation. The main contributions of this dissertation are as follows:Firstly, in order to protect the privacy of sensing data transmission, we present the design of PPSense. In this scheme, we design an Attribute Based Encryption which based on the base station to guarantee the effectiveness of encryption algorithm. Then, we separate the users’ identify and the data based on anonymous technique. At last, we verify the validity of this scheme.Secondly, in order to protect the privacy of sensing data fusion, we present the design of PDA. In this scheme, homomorphic encryption is used to preprocess the sensing data in sensor node. Moreover, the aggregation node realizes the data fusion operation by the homomorphism. Then, we use the secret sharing to complete the reliability design. At last, we verify the validity of this scheme.Finally, in order to protect the privacy of sensing data querying, we present the design of SPS. In this scheme, we based on SensFlow algorithm. At first, we use Analytic Hierarchy Process to obtain the appropriate sensitivity threshold value for the user’s requirement. Then, we get the best obfuscation space division method using statistical learning theory. The experimental results show that SPS can defend the semantic location attack under a reasonable cost. |
PDF Full Text Request