Detecting motion-sensor data anomalies
AssigneeInternational Business Machines Corporation
InventorsThiago Bianchi, John Donald Vasquez, Joao Souto Maior
Life of this patent
Estimating the term…
Abstract
Detecting motion-sensor data anomalies by registering an event associated with received motion-detector sensor data at a current timestamp, calculating, a timestamp delta between a previous timestamp and the current timestamp, and counting transitions between a set of timestamp deltas using a finite state machine. The anomaly detection also includes using a Markov chain transition matrix, generated from the finite state machine, to determine that, a probability of transitioning from a current state to a next state associated with the event, is less than a predetermined value, and registering an anomaly associated with the sensor and the event.
Claims20
A computer implemented method for processing sensor data, the computer implemented method comprising:
- receiving, by one or more computer processors, motion-detection data from a sensor;
- registering, by one or more computer processors, a current event associated with the motion-detection data at a first timestamp;
- incrementing, by one or more computer processors, a cumulative count associated with the sensor;
- calculating, by one or more computer processors, a difference between a second timestamp and the first timestamp, wherein the difference is a timestamp delta;
- generating, by one or more computer processors, a finite state machine to count transitions between a set of timestamp deltas;
- generating, by one or more computer processors using the finite state machine, a Markov chain transition matrix;
- determining, by one or more computer processors using the Markov chain transition matrix, a probability of transitioning from a current state to a next state associated with the event;
- determining, by one or more computer processors, that the probability of transitioning from a current state to the next state associated with the event is less than a predetermined value;
- and
- registering, by one or more computer processors, an anomaly associated with the sensor and the event.
A computer program product for processing sensor data, the computer program product comprising one or more computer readable storage devices and stored program instructions on the one or more computer readable storage devices, the stored program instructions comprising:
- programmed instructions for receiving motion-detection data from a sensor;
- programmed instructions for registering a current event associated with the motion-detection data at a first timestamp;
- programmed instructions for incrementing a cumulative count associated with the sensor;
- programmed instructions for calculating a difference between a second timestamp and the first timestamp, wherein the difference is a timestamp delta;
- programmed instructions for generating a finite state machine to count transitions between a set of timestamp deltas;
- programmed instructions for generating, using the finite state machine a Markov chain transition matrix;
- programmed instructions for determining, using the Markov chain transition matrix, a probability of transitioning from a current state to a next state associated with the event;
- programmed instructions for determining that the probability of transitioning from a current state to the next state associated with the event, is less than a predetermined value;
- and
- programmed instructions for registering an anomaly associated with the sensor and the event.
A computer system for processing sensor data, the computer system comprising:
- one or more computer processors;
- one or more computer readable storage devices;
- stored program instructions on the one or more computer readable storage devices for execution by the at least on computer processor, the stored program instructions comprising:
- programmed instructions for receiving motion-detection data from a sensor;
- programmed instructions for registering a current event associated with the motion-detection data at a first timestamp;
- programmed instructions for incrementing a cumulative count associated with the sensor;
- programmed instructions for calculating a difference between a second timestamp and the first timestamp, wherein the difference is a timestamp delta;
- programmed instructions for generating a finite state machine to count transitions between a set of timestamp deltas;
- programmed instructions for generating, using the finite state machine a Markov chain transition matrix;
- programmed instructions for determining, using the Markov chain transition matrix, a probability of transitioning from a current state to a next state associated with the event;
- programmed instructions for determining that the probability of transitioning from a current state to the next state associated with the event, is less than a predetermined value;
- and
- programmed instructions for registering an anomaly associated with the sensor and the event.
Dependencies are inferred from the claim text.