Handling of Past and Future with Phenesthe+
Manolis Pitsikalis, Alexei Lisitsa, Patrick Totzke
Abstract
Open-access reader
Manolis Pitsikalis, Alexei Lisitsa, Patrick Totzke
Abstract
Open-access reader
Writing temporal logic formulae for properties that combine instantaneous events with overlapping temporal phenomena of some duration is difficult in classical temporal logics. To address this issue, in previous work we introduced a new temporal logic with intuitive temporal modalities specifically tailored for the representation of both instantaneous and durative phenomena. We also provided an implementation of a complex event processing system, Phenesthe, based on this logic, that has been applied and tested on a real maritime surveillance scenario. In this work, we extend our temporal logic with two extra modalities to increase its expressive power for handling future formulae. We compare the expressive power of different fragments of our logic with Linear Temporal Logic and dyadic first-order logic. Furthermore, we define correctness criteria for stream processors that use our language. Last but not least, we evaluate empirically the performance of Phenesthe+, our extended implementation, and show that the increased expressive power does not affect efficiency significantly.
A significance statement is not available in the OpenAlex record.
A contribution statement is not available in the OpenAlex record.
Method details are not available in the OpenAlex metadata.
Findings are not separately available in the OpenAlex metadata.
Limitations are not available in the OpenAlex metadata.
Application details are not available in the OpenAlex metadata.
Writing temporal logic formulae for properties that combine instantaneous events with overlapping temporal phenomena of some duration is difficult in classical temporal logics. To address this issue, in previous work we introduced a new temporal logic with intuitive temporal modalities specifically tailored for the representation of both instantaneous and durative phenomena. We also provided an implementation of a complex event processing system, Phenesthe, based on this logic, that has been applied and tested on a real maritime surveillance scenario. In this work, we extend our temporal logic with two extra modalities to increase its expressive power for handling future formulae. We compare the expressive power of different fragments of our logic with Linear Temporal Logic and dyadic first-order logic. Furthermore, we define correctness criteria for stream processors that use our language. Last but not least, we evaluate empirically the performance of Phenesthe+, our extended implementation, and show that the increased expressive power does not affect efficiency significantly.
Key concepts: Temporal logic, Temporal logic of actions, Linear temporal logic, Interval temporal logic, Correctness, Expressive power, Computer science, Multimodal logic