Staggered-sampling technique for detecting sensor anomalies in a dynamic univariate time-series signal

A non-transitory computer-readable storage medium storing instructions that when executed by a computer cause the computer to perform a method for detecting sensor anomalies in a univariate time-series signal, wherein during a surveillance mode, the method comprises:

• receiving the univariate time-series signal from a sensor in a monitored system;
• performing a staggered-sampling operation on the univariate time-series signal to produce N sub-sampled time-series signals, wherein the staggered-sampling operation allocates consecutive samples from the univariate time-series signal to the N sub-sampled time-series signals in a round-robin ordering;
• using a trained inferential model to generate estimated values for the N sub-sampled time-series signals based on cross-correlations with other signals in the N sub-sampled time-series signals;
• performing an anomaly detection operation to detect incipient sensor anomalies in the univariate time-series signal based on differences between actual values and the estimated values for the N sub-sampled time-series signals;
• and
• when an incipient sensor anomaly is detected, generating a notification.

A system that detects sensor anomalies in a univariate time-series signal, comprising:

• at least one processor and at least one associated memory;
• and
• an anomaly detection mechanism that executes on the at least one processor, wherein during operation, the anomaly detection mechanism:
• receives the univariate time-series signal from a sensor in a monitored system;
• performs a staggered-sampling operation on the univariate time-series signal to produce N sub-sampled time-series signals, wherein the staggered-sampling operation allocates consecutive samples from the univariate time-series signal to the N sub-sampled time-series signals in a round-robin ordering;
• uses a trained inferential model to generate estimated values for the N sub-sampled time-series signals based on cross-correlations with other signals in the N sub-sampled time-series signals;
• performs an anomaly detection operation to detect incipient sensor anomalies in the univariate time-series signal based on differences between actual values and the estimated values for the N sub-sampled time-series signals;
• and
• when an incipient sensor anomaly is detected, generates a notification.