Privacy-preserving range-free positioning of wireless devices

INV-23036

Background

Collaborative positioning can be a useful approach in situations where a person needs to exchange information with nearby users. This can be particularly helpful when using a GNSS receiver in an urban area with limited satellite visibility or when indoors without the necessary technology to determine their location. Collaborative positioning can address these issues, but it does involve the sharing of data between users, which can raise privacy concerns. To address these concerns, privacy-preserving network schemes can be adopted.

However, privacy-preserving network schemes have their own limitations. Although they provide a level of privacy by distorting sensitive data by adding a level of noise, while the noise level remains small, the sensitive data can be retrieved. This approach would also require a large number of users to take part in the collaborative network in order to be able to compute accurate positioning estimates.

Technology Overview

Researchers at Northeastern propose a collaborative network solution which uses homomorphic encryption methodology to eliminate the need to distort sensitive data and so reduce the number of users required to obtain valid estimated positions. Furthermore, the proposed approach introduces a second layer of encryption. The second layer of encryption is the layer where all the encrypted operations are performed and is seen as the encrypted layer of the first encrypted layer, or a ciphertext. All the users, who participate in the collaborative network encrypt their sensitive data to the first level ciphertext and that is then encrypted to the second level ciphertext. These users don’t have the authorization to decrypt any results, unless it is the primary user. This primary user, who participates in the collaborative calculation, has the private key for the second level ciphertext. When the primary user decrypts this second level ciphertext, which is within the second layer of encryption, the decrypted result is a first level ciphertext. So, the primary user decrypts from the second layer of encryption to the first layer of encryption. Now, with this first level ciphertext, the primary user is not capable to decrypt it, since it does not have the private key to the first layer of encryption. Only the user, whose estimates are unknown, has the private key of the first layer of encryption. This approach of having a collaborative network with two layers of encryption that do not share the same private key is unique and first of its kind.

Benefits

• It implements a proximity-based positioning algorithm that does not require computation of ranging among users, thus it is lightweight.
• The position of collaborative agents remains unknown to the user implementing NN positioning, thus preserving their privacy.
• Little communication exchange is required among users.

Applications

• Indoor localization where certain nodes have enhanced positioning capabilities and other nodes can simply estimate their position by averaging the positions of nearby users.
• Enhanced outdoors positioning where users might have poor GPS visibility but other users with better satellite visibility can help position the former.
• Positioning in crowd environments with large amounts of users in a given geographic area.

Opportunity

Seeking licensee and/or industry partner