Detection of cyber machinery attacks
A method of cyber protection of a machine based on acquiring acoustic signals from a vicinity of the machine, while the machine is operative. The method includes analyzing the acquired acoustic signals to determine whether the machine or a controller of the machine is operating suspiciously and initiating a cyber measure on the controller of the machine, responsive to a determination based on the acquired acoustic signals that the machine or the controller is operating suspiciously.
1. A method of cyber protection of a machine, comprising:
acquiring acoustic signals from a vicinity of a machine, at a plurality of time points;
comparing, by a processor, the acoustic signals or parameters of the acoustic signals for different time points, to identify changes in the acoustic signals over time responsively to the acquired acoustic signals at the plurality of time points;
analyzing the identified changes to check whether the changes have one or more characteristics indicative of intentional changes; and
initiating, by the processor, a cyber alert, responsive to identifying in the checking, a characteristic indicative of intentional changes,
wherein analyzing the identified changes to check whether the changes have one or more characteristics indicative of intentional changes comprises identifying a plurality of changes occurring at time points having a regularity indicative of being non-random.
2. The method of claim 1 , wherein analyzing the identified changes to check whether the changes have one or more characteristics indicative of intentional changes comprises identifying abrupt changes followed by smooth operation.
3. The method of claim 1 , wherein analyzing the identified changes to check whether the changes have one or more characteristics indicative of intentional changes comprises identifying a first acoustic profile of the machine at a first time point, a second acoustic profile of the machine at a second time point, after the first time point, and a return to the first acoustic profile at a third time point after the second time point.
4. The method of claim 1 , wherein identifying the plurality of changes occurring at time points having a regularity indicative of being non-random comprises identifying a plurality of changes occurring at time points distanced from each other by equal intervals.
5. The method of claim 1 , wherein identifying the plurality of changes occurring at time points having a regularity indicative of being non-random comprises identifying a plurality of changes occurring at time points distanced from each other by round time intervals.
6. The method of claim 1 , wherein identifying the plurality of changes occurring at time points having a regularity indicative of being non-random comprises identifying a plurality of changes occurring at time points at round clock times.
7. The method of claim 1 , wherein comparing the acoustic signals or parameters of the acoustic signals for different time points comprises comparing the acoustic signals themselves.
8. The method of claim 1 , wherein comparing the acoustic signals or parameters of the acoustic signals for different time points comprises comparing frequency spectrums of the acoustic signals.
9. The method of claim 1 , wherein comparing the acoustic signals or parameters of the acoustic signals for different time points comprises comparing features extracted from the acoustic signals.