Search

The sharing economy has greatly promoted the rapid development and application of spatial crowdsourcing. Although privacy-preserving spatial task matching as an indispensable part has been extensively explored, existing schemes cannot be deployed into the practical environment due to drawbacks in the one-side location protection, the matching efficiency, and the dynamic updates. In this study, we propose a novel Secure and Efficient Spatial Task Matching framework (SESTM) with utilizing multi-user searchable encryption and secure index technique, which enables to preserve the location privacy of requesters and workers while achieving efficient task allocation and good user scalability. Specifically, requesters firstly transform and encrypt their task locations before being outsourced, and we secondly design a secure and dynamic tree-based index SD-Tree for SC-server to merge these uploaded encrypted data without knowing their underlying content. Finally, SESTM provides efficient task matching services for multiple workers based on encrypted queries. Furthermore, SD-Tree also provides fast delete and insert operations under logarithmic time to reduce the dynamic update overhead for real SC services. Extensive theoretical analysis and performance evaluation demonstrate the practicality of our method.

With the development of global positioning systems, communication technologies, the Internet and mobile terminals, location-based services (LBS) have become increasingly widely used. LBS brings great convenience to people's daily life and social activities. At the same time, people pay more attention to the problem of sensitive information leakage when using LBS. Users need to provide their accurate location information when sending queries to the LBS server. User location privacy is extremely vulnerable to serious threats during the exchange of services, and these location-related queries may cause serious privacy issues. In this paper, considering that the opponent may use auxiliary information such as data analysis and crawlers to determine the approximate location, we propose an effective location privacy-preserving k-anonymity scheme (PPKS) to generate virtual locations based on the probability density function in mathematical statistics to realize the k-anonymity of users with privacy zone awareness in LBS. We first determine the origin of the region and then calculate the probability density function of the X-axis and the Y-axis. Subsequently, the k-1 positions are distributed to the X-axis and Y-axis in proportion, respectively. Finally, according to the proportions, the k-1 virtual positions are generated. The results of security analysis and experimental evaluation show that the solution can significantly improve the level of privacy and anonymity.

Trajectory community recommendation (TCR) is a location-based social network (LBSN) service whereby LBSN server recommends a user a community in which the trajectories have similar movement patterns with the user’s trajectory. Due to privacy concerns, the trajectory should be protected. However, the data availability of traditional privacy-preserving schemes is limited, and previous differential privacy (DP) methods cannot achieve high data utility in TCR and rely on a fully trusted third party. In this paper, we propose a DP-based trajectory community recommendation (DPTCR) scheme to perform effective TCR service while protecting trajectory privacy by the user himself. First, DPTCR transits the actual trajectory’s locations into noisy feature locations based on private semantic expectation method, which ensures the semantic similarity between noisy locations and the actual locations. Second, DPTCR uses a private geographical distance method to construct a noisy trajectory that has the smallest geographical distance with the actual trajectory. Finally, DPTCR uses a semantic-geographical distance model to cluster a community in which the trajectories have high similarity with the constructed noisy trajectory. Security analysis proves that our DPTCR scheme achieves ϵ -DP, and experimental results show that our scheme achieves high efficiency and data utility.

Mobile cloud computing has received a particular attention in recent years thanks to the expansion and easy accessibility of smartphones with enhanced capabilities, pushing for data owners to outsource more data to the cloud in order to serve the increasing number of mobile users. To protect data privacy, data have to be encrypted prior to outsourcing. Nevertheless, mobile users should still be able to query data, get the results corresponding to their requests and verify the integrity of the query results, all within a query privacy-preserving and efficient fashion. We focus in this work on the secure range query allowing to search for data belonging to a specific range. At the difference of previous works where data privacy and query results verification require separate structures and thus more overhead, our work achieves both data privacy and query results completeness verification using the encrypted data index and the vector neighbor chain (VNC). As a response to the intrinsic limitations of mobile devices, we design a lightweight protocol based on linear algebra operations for both 1-dimensional and multi-dimensional data. Practical experiments analysis on a real-world U.S. Census demographic dataset shows that our scheme guarantees the privacy-preserving property and is efficient in terms of computation and communication overhead.

Mobile cloud data storage allows mobile users to outsource personal and corporate data to the cloud for flexibility and money savings. However, data outsourcing introduces high confidentiality and privacy risks. To solve the above issues in mobile cloud data storage, attribute-based encryption has been proposed thanks to the flexibility in encryption policy and the fine grained access control. Existing multi-authority attribute-based encryption schemes however still require a trusted central authority to publish system parameters and to generate user secret keys. They give to the trusted central authority enough privileges to access the plaintext information meant for the user, a problem referred to as key escrow issue. They moreover require high computational and communicational overhead on the different operating entities. We propose in this paper an efficient decentralized multi-authority attribute-based scheme for mobile cloud data storage. Our scheme solves the key escrow problem by removing the central authority, without making use of any global user identity. We demonstrate that our scheme is flexible and exhibits less communication and computation overhead on the data user side, both in an online and offline fashion, as well as prove its security under the decisional bilinear diffie-hellman assumption.