060 - Novel privacy preserving PNT processing techniques
Up to now, cloud-based processing of positioning data (e.g. samples, raw measurements, anonymised Position, Velocity and Time) has always been susceptible to data breaches and raised concerns about privacy protection. Location-based services can fail in their path to wide-spread adoption, because they are perceived as invasive by the users and fail to properly address their risks. Similar issues appear for patient monitoring, and very recently, the COVID-19 pandemic revealed the necessity of “contact tracing” without breaking confidentiality barriers of the user. The lack of technologies allowing the processing of PNT data without impairing the user privacy has led most of the countries to avoid using PNT in “tracking apps”.
By shifting the privacy control on the user side and also decentralise processing with the involvement of the user, trusted services can be built without having to rely on the security of the infrastructure. One can bear the full responsibility for its data if, for example, if allowed to act in anonymity or to encrypt its information before it is shared for processing. Different techniques can be envisioned to avoid disclosing sensitive data. For example, homomorphic encryption allows the manipulation and processing of encrypted data sets without the need of revealing any of the original data. This enables third parties such as cloud service providers to apply various functions over the user’s data while not infringing upon any privacy barriers. The user maintains all its information and encryption keys private, and computationally intensive operations are blindly performed by dedicated servers.
In the context of tracking applications, these encryption schemes can be applied over sensitive PNT data. One can process trajectory data and derive relevant events, such as intersections, and still maintain confidentiality. Moreover, to prevent security breaches, multi resolution sharing allows the user to control the granularity of information shared to different users. Other crypto technologies such as blockchain can enable a community data sharing protocol robust against tampering from malevolent parties. The distributed ledger can offer protection of data without the need of a central authority and can allow users to access it without revealing their identity.
Transferring the aforementioned processing techniques to the GNSS field will open the door for a new range of location-based-services (e.g. non-disclosing trajectory operations) and will enable highly-secure distributed GNSS architectures (e.g. cloud correlators) which will minimize the information leakage. Furthermore, applications such as healthcare analysis (e.g. disease tracking, identification of risk factors, etc) or marketing services could experience a leap by including PNT information to the plethora of big data.
Many applications could be considered, such as:
- healthcare, e.g. contact tracing for establishing risk of infection and minimizing the spread;
- incident awareness for users in areas affected by natural disasters or in an emergency state;
- automatic tolling for passenger cars;
- enforcing regulations on the airspace used by autonomous drones;
- distributed secure processing of GNSS signals.
The main objective of the activity is therefore to define and develop a proof-of-concept of an end-to-end privacy-preserving positioning solution.
The tasks to be performed include:
- investigating novel architectures, algorithms and techniques for processing GNSS data (signal, measurements, etc.) based on advanced encryption schemes and data-sharing protocols, including e.g. homomorphic cryptography, blockchain;
- developing a proof-of-concept and validating it through simulations and laboratory tests.
The main results of the activity will provide:
- breadboard demonstrator of the innovative privacy preserving concepts;
- test data, including simulation and laboratory test results and benchmark with standard PNT processing solutions in various environments;
- roadmap towards development of industrial products, for instance through initiatives in NAVISP Element 2.