Home
This Title All WIREs
WIREs RSS Feed
How to cite this WIREs title:
WIREs Comp Stat

Alarm activation, pattern discovery, and anomaly detection in sensor networks

Full article on Wiley Online Library:   HTML PDF

Can't access this content? Tell your librarian.

Abstract This article describes some recent sensor network research. We have looked in more detail at three data analysis issues. First, we have considered the general issue of what constitutes normal activity and what constitutes anomalous activity. Detecting anomalies requires some measure of ‘normal’ to determine if there is an anomaly. While approaches such as Bayesian networks work well on large and diverse networks, a simpler approach may be useful for smaller networks. We describe an approach for keeping an ongoing record of sensor activation events and turning on an alarm when the frequency of events exceeds a set threshold. The approach is devised for efficiency of computation and storage. We considered groups of days such as seasons, days of the week, and work days to produce several sets of average activation levels. Second, we modeled different time intervals as independent Poisson processes and used this model to compile average data to look for and display anomalies at different sensors and different times. Using a sensor data set, we studied groups of days such as seasons, days of the week, and work days to produce several sets of average activation levels. Third, we consider possible paths that could be caused by an intruder activating several sensors in a spatial and temporal neighborhood. We are developing a simple tool to calculate and display such paths and their probabilities. WIREs Comput Stat 2012, 4:565–570. doi: 10.1002/wics.1233 This article is categorized under: Algorithms and Computational Methods > Computer Graphics Statistical Learning and Exploratory Methods of the Data Sciences > Pattern Recognition

Spatial locations of 15 out of 26 sensors from the test data set; the other 11 were omitted for visual clarity.

[ Normal View | Magnified View ]

A graph of possible intruder paths based on activations of sensors.

[ Normal View | Magnified View ]

Anomaly detection graph for October 11, low significance threshold.

[ Normal View | Magnified View ]

Anomaly detection graph for October 11, medium significance threshold.

[ Normal View | Magnified View ]

Anomaly detection graph for October 11, high significance threshold.

[ Normal View | Magnified View ]

Average activation level for each sensor over all Spring (March–May 2010). Dark green is the lowest activation (none), white is the highest activation, lighter colors denote higher activation.

[ Normal View | Magnified View ]

Average activation level for each sensor over Winter (December 2009–March 2010). Dark green is the lowest activation (none), white is the highest activation, lighter colors denote higher activation.

[ Normal View | Magnified View ]

Average activation level for each sensor over Autumn (September–December 2009). Dark green is the lowest activation (none), white is the highest activation, lighter colors denote higher activation.

[ Normal View | Magnified View ]

Average activation level for each sensor over Summer (July–September 2009). Dark green is the lowest activation (none), white is the highest activation, lighter colors denote higher activation.

[ Normal View | Magnified View ]

Average activation level for each sensor over all days, July 2009–May 2010. Dark green is the lowest activation (none), white is the highest activation, lighter colors denote higher activation.

[ Normal View | Magnified View ]

Activations from midnight (left side of graph) to 11:59 pm (right side of graph) for sensors 2 (bottom of graph) to 26 (top of graph) on August 26, 2009.

[ Normal View | Magnified View ]

Activations from midnight (left side of graph) to 11:59 pm (right side of graph) for sensors 1 (bottom of graph) to 21 (top of graph) on July 11, 2009.

[ Normal View | Magnified View ]

Browse by Topic

Algorithms and Computational Methods > Computer Graphics
Statistical Learning and Exploratory Methods of the Data Sciences > Pattern Recognition

Access to this WIREs title is by subscription only.

Recommend to Your
Librarian Now!

The latest WIREs articles in your inbox

Sign Up for Article Alerts