Hier finden Sie wissenschaftliche Publikationen aus den Fraunhofer-Instituten.

Increasing the efficiency of syndrome coding for PUFs with helper data compression

: Hiller, Matthias


Preas, Kathy ; European Design Automation Association -EDAA-; IEEE Computer Society, Technical Council Test Technology:
Design, Automation and Test in Europe Conference and Exhibition, DATE 2014. Vol.1 : 24 - 28 March 2014, Dresden, Germany
Piscataway, NJ: IEEE, 2014
ISBN: 978-1-4799-3297-9
ISBN: 978-3-9815370-2-4
Design, Automation and Test in Europe Conference & Exhibition (DATE) <17, 2014, Dresden>
Fraunhofer AISEC ()

Physical Unclonable Functions (PUFs) provide secure cryptographic keys for resource constrained embedded systems without secure storage. A PUF measures internal manufacturing variations to create a unique, but noisy secret inside a device. Syndrome coding schemes create and store helper data about the structure of a specific PUF to correct errors within subsequent PUF measurements and generate a reliable key. This helper data can contain redundancy. We analyze existing schemes and show that data compression can be applied to decrease the size of the helper data of existing implementations. We introduce compressed Differential Sequence Coding (DSC), which is the most efficient syndrome coding scheme known to date for a popular reference scenario. Adding helper data compression to the DSC algorithm leads to an overall decrease of 68% in helper data size compared to other algorithms in a reference scenario. This is achieved without increasing the number of PUF bits and a minimal increase in logic size.