Since biometric data contains mathematical descriptions of certain characteristics of body features such as fingerprints, iris or facial details, and vein patterns, it is considered highly critical and must be protected in a particular manner - not only to avert data breaches but especially to prevent cybercriminals from stealing entire sets of biometric data.
The process can be used wherever sensitive data needs to be protected more strongly – also outside of biometric systems. For example, with the split-template method, tokens can be protected even better against unauthorized access.
To clarify the process and the advantages of the split-template method, we need to take a detailed look at previous methods:
Potential security gaps in biometric access controls
In high-security biometric access controls, biometric characteristics such as fingerprints, veins, or iris features are recorded by a corresponding sensor system and compared to biometric features already stored in the system. If these results show sufficient similarities, a so-called "match" is found. Biometric data used for comparison belongs to the person who presented the corresponding biometric feature. Since the system recognizes the person, who previously generated the biometric data set used for comparison, the person is thereby identified.
Cybercriminals can steal or manipulate this biometric data. For example, if fingerprint data is published online, any person with the appropriate knowledge can create a mock-up for a so-called "presentation attack" and thus defraud biometric security systems.
For the person, whose fingerprint was exposed, this means that their it can never again be used in the biometric system. Since the biometric feature is now known in its pure form, a replica can be generated at any time. In addition, digital fingerprints are sensitive, personal data that must be specially protected according to the EU-GDPR. They may be exempt from permanent storage altogether. For this reason, data protection law prohibits the central storage of sensitive biometric data. This in turn may mean that biometric systems are not allowed to be used in various applications.
Difficulties of previous solutions
To circumvent these issues, there currently is one procedure in place: for later comparison, biometric data is stored on servers that belong to a highly secure and non-vulnerable IT infrastructure.
This works quite well in practice. However, due to increasingly complex IT procedures, it is becoming more and more difficult and cost-intensive to maintain this level of highly secure infrastructure.
In addition, and as mentioned above, centralized storage of personal data, such as biometric data, is a fundamental data protection issue.
Is mobile data storage a GDPR-friendly alternative?
To comply with the GDPR, biometric data must be stored exclusively on mobile media such as RFID cards or mobile devices. This way, users can access their personal data at any time and also delete it if in doubt.
By holding the card or mobile device up to a corresponding reader, the data is recorded. The system then compares the data and immediately deletes the used data. This way, biometric data is only in the system at the time of use and is not persistently stored there. In order for this method to comply with the EU-GDPR, additional prerequisites need to be met.
EU-GDPR (Article 9)
- Consent: the data subject has given explicit consent to the processing of biometric data for one or more specified purposes, except where Union or Member State law provides that the prohibition may not be lifted by the data subject's consent.
- Processing is necessary for the controller or the data subject to carry out the obligations and exercising specific rights in the field of employment, social security, and social protection law. This is to be authorised by Union or Member State law or a collective agreement pursuant to Member State law providing for appropriate safeguards for the fundamental rights and the interests of the data subject.
This environment meets data protection requirements - but in terms of IT security, the data is not secured or monitored. Consequently, both methods can only be used to a limited extent for highly secure data storage of critical data records.
Why is the split-template process better?
The split-template method uses the best of the methods described, taking it to the next level to make biometric data unrecognizable:
- First, critical data blocks are split into two or more data portions.
- The individual portions are then stored on different media and/or in different locations.
- Storage locations can be data carriers such as the RFID card AND the server within the IT infrastructure.