Anonymity-based Privacy-preserving Data Reporting for Participatory Sensing ABSTRACT

Anonymity-based Privacy-preserving Data

Reporting for Participatory Sensing

Abstract:

In this paper, we propose an efficient anonymous data reporting protocol for participatory sensing, which provides strong privacy protection, data accuracy and generality. The protocol consists of two stages, namely slot reservation and message submission. In the slot reservation stage, a group of N participants cooperate to assign each member a message slot in a vector which is essentially a message submission schedule, in such a manner that each participant’s slot is oblivious to other members and the application server. In the message submission stage, each participant transmits an encoded data to the application server based on the slot information known only to herself, in such a way that the application server cannot link a data to a specific participant. With such a data reporting protocol, the link between the data and the participants is broken, and as a result, participant’s privacy is protected. We conduct theoretical analysis of the correctness and anonymity of our protocol, as well as experiments to demonstrate the efficiency in small-scale applications with periodic data sampling.

Existing System:

v In existing methods, if the application server colludes with a global eavesdropper who can monitor the traffic across the network, it can link each data with its contributor.

v Privacy in participatory sensing systems has been addressed by many works, instead of sending an accurate location to the server, spatial cloaking is employed to calculate an anonymity set.

v As a result, the user is protected. The work in also follows this idea of generalization, in which k pieces of data are combined together before sending to the server, with the intension of adding enough “confusion” in the data to make it difficult to obtain exact times and locations for the individual data.

Proposed system:

v Privacy in participatory sensing systems has been addressed by many works, instead of sending an accurate location to the server, spatial cloaking is employed to calculate an anonymity set.

v  As a result, the user is protected. The work in  also follows this idea of generalization, in which k pieces of data are combined together before sending to the server, with the intension of adding enough “confusion” in the data to make it difficult to obtain exact times and locations for the individual data.

v  The most efficient anonymous message protocol so far is proposed in , where a ((N, N)-SS) secret sharing protocol is employed for both slot reservation and data submission.

v  The computation is efficient, as it doesn’t use public key encryption. But the communication cost is very heavy.

Problem statement:

Privacy protection is an important issue in participatory sensing. We propose an anonymous data reporting protocol for participatory applications to protect user privacy. The intuition behind the protocol is that, if the data itself does not contain identification information, and we can break the link between the data and the participant that reports the data, the user’s privacy can be protected. The anonymous data reporting protocol is divided into two stages, a slot reservation stage and a data submission stage.

Implementation of modules:

                                         System architecture:

In this Anonymity-based Privacy-preserving Data Reporting for Participatory Sensing five  modules such as given below,

1. Participant module.

2. Application Server module.

3. End Users module.

4. Privacy.

5. Trusted certification authority.

Participant Module:

 particiaptor can submit online reports using public/private keys, that may be related to personal ,group,community.participator can have limitation for submitting records into the server.

Participator can perform the following works

v .participator registeration.

v participator  login.

v view profile.

v slot reservation.

v Data submission.

Application Server module:

Application server is nothing but admin after admin login, he will collect the information from the server and analyze  the reports and prepare statistics, after that view the end user requests.

Steps:

v server login.

v view participator online reports collect the statistics.

v view end user details.

Trusted Certification authority:

 We assume that each participant Pi has a private/public key pair (xi, yi), and each participant knows the public keys of all the others. Thus, all participants can identify each other by public keys. In practice, the participants apply their key pairs from a trusted certification authority whose job is to associate each individual with her public key. Or alternatively, each participant can generate the key pair herself and publish the public key at the application server, if she does not trust any authority.

Steps:

v login.

v view user requests.

v accept user slot requests and provide pair of public/private keys.

End User module:

User can view the online reports and give the feed backs according to the type of posts.

Steps:

v user registeration.

v user login.

v search

v give feed backs and suggestions.

Participatory Privacy sensing:

participating in a participatory sensing task, especially a community-scale task, could result in private information leakage. Some tasks require users to submit data containing sensitive information, for example, disease symptoms. Some applications don’t directly use sensitive data, but still result in privacy leakage. For example, in a power consumption monitoring application , from temporally fine-grained energy consumption reports submitted by users, household activities can be inferred easily. In addition, data in participatory sensing application are usually geo- and timetagged. From multiple data reported by a participant, an adversary can derive much sensitive information. Thus, users are reluctant to contribute to the sensing campaigns, if their privacy cannot be protected.

A variety of methods have been proposed to protect the privacy of each participant for participatory sensing applications. A naive mechanism to protect the privacy is to use pseudonyms. However, as demonstrated in , the use of pseudonyms does not necessarily guarantee privacy. Some privacy protection methods employ generalization or perturbation, both of which intend to allow the application server to determine community trends without revealing individual data, by deliberately reducing the accuracy or precision of the sensed data.

Conclusion:

 Privacy protection is an important issue in participatory sensing. We propose an anonymous data reporting protocol for participatory applications to protect user privacy. The intuition behind the protocol is that, if the data itself does not contain identification information, and we can break the link between the data and the participant that reports the data, the user’s privacy can be protected. The anonymous data reporting protocol is divided into two stages, a slot reservation stage and a data submission stage. We propose an anonymous slot reservation scheme based on public key encryption and message shuffle, and a data submission scheme based on efficient XOR operation. The theoretical analysis verifies the correctness and the anonymity of the protocol. The experiments demonstrate that, for small-scale applications with only tens of participants where data is collected in a periodic manner, the proposed protocol is efficient and applicable.