Property-driven Timestamps Encoding for Timeprints-based Tracing and Monitoring

Rehab Massoud, Hoang M. Le, Rolf Drechsler

In: 17th International Conference on FORMAL MODELING AND ANALYSIS OF TIMED SYSTEMS (FORMATS). International Conference on Formal Modelling and Analysis of Timed Systems (FORMATS-2019) August 27-29 Amsterdam Netherlands 2019.


Timeprints are temporal regularly-logged signatures, describ-ing a signal’s temporal behavior. They have been recently used in on-chip signals tracing and temporal properties checking. Timeprints are generated by aggregations of encoded timestamps marking where signal changes took place. This paper describes different timestamps encoding mechanisms, and shows how some system’s temporal properties can be used to create more efficient timestamps. The efficiency ofa timestamps-encoding is introduced in terms of the number of collisions in the timeprints-reconstruction solution space. We show how using property-based timestamps encoding reduces the number of such colli-sions, leading to better chances capturing unexpected behaviors.


German Research Center for Artificial Intelligence
Deutsches Forschungszentrum für Künstliche Intelligenz