US 2008/0077355 A1Published application

Methods and systems for statistical based monitoring and power management

AssigneesInternational Business Machines Corporation, Mayo Foundation for Medical Education and Research

InventorsDavid C. Casler, Joan L. Mitchell, Piet C. De Groen

Life of this patent

Estimating the term…

Loading patent…

Abstract

Embodiments of the present invention provide devices and techniques for detecting anomalies in monitored environments. In one embodiment of the invention, data may be acquired by monitoring an environmental variable. Based on the acquired data, statistics may be calculated which model the behavior of the environmental variable. Based on the calculated statistics, a long-term behavior and a short-term behavior of the statistics may also be calculated. A difference between the long-term statistics and the short-term statistics may then be calculated. If the difference between the long-term statistical behavior and short-term statistical behavior exceeds a dynamic or predefined threshold, an action may be taken.

Claims20

Dependent claims

A computer-implemented method of detecting an anomaly in a signal, comprising:

calculating a first plurality of values based on a set of statistics which model the signal, wherein the first plurality of values is calculated over a first period of time;
calculating a second plurality of values based on the set of statistics which model the signal, wherein the second plurality of values is calculated over a second period of time;
calculating a third plurality of values, wherein the third plurality of values is calculated using the first plurality of values and the second plurality of values, and wherein the third plurality of values is calculated based on a least an autocorrelation of the signal;
calculating a difference between at least one of the third plurality of values and a corresponding long-term value, wherein the long-term value is calculated over a third period of time, and wherein the third period of time is greater than the first and second periods of time;
comparing the difference to a threshold wherein the difference exceeding the threshold is indicative of the anomaly in the signal;
if the difference exceeds the threshold, performing an action wherein performing the action requires a relative increase in power consumption;
and
transmitting an alert signal or a portion of the signal indicating when the difference exceeds the threshold.

5 more elements

An apparatus for detecting an anomaly in a signal, comprising:

a sensor;
and
a signal processing unit coupled with the sensor, wherein the signal processing unit performs operations comprising:
calculating a first plurality of values based on a set of statistics which model the signal, wherein the first plurality of values is calculated over a first period of time;
calculating a second plurality of values based on the set of statistics which model the signal, wherein the second plurality of values is calculated over a second period of time;
calculating a third plurality of values, wherein the third plurality of values is calculated using first plurality of values and the second plurality of values and wherein the third plurality of values is calculated based on at least an autocorrelation of the signal;
calculating a difference between at least one of the third plurality of values and a corresponding long-term value, wherein the long-term value is calculated over a third period of time, and wherein the third period of time is greater than the first and second periods of time;
comparing the difference to a threshold, wherein the difference exceeding the threshold is indicative of the anomaly in the signal;
if the difference exceeds the threshold, performing an action wherein performing the action requires an increase in power consumption;
and
transmitting an alert signal or a portion of the signal indicating when the difference exceeds the threshold.

8 more elements

Dependencies are inferred from the claim text.