Slick algorithm helps spot tech trouble in everything from networks to satellites

Outlier Detection Algorithm could let operators keep a system performing even while it's failing

Researchers say they have developed an algorithm that can be run against a variety of systems including cars, power plants, satellites and networks that would let users spot and fix problems faster.

In theory the technique, known as the Intelligent Outlier Detection Algorithm, or IODA could let operators keep a system performing even while it's failing, Andrew Weekley, a software engineer at the National Center for Atmospheric Research (NCAR) who led the algorithm development effort stated in a release. "When a system starts to fail, it's absolutely critical to be able to control it as long as possible. That can make the difference between disaster or not." 

Layer 8 Extra:

15 genius algorithms that aren't boring

12 mad science projects that could shake the world 

According to researchers, if sensors malfunction and begin transmitting bad data, computers programmed with IODA could spot the problem and isolate bad data. The algorithm, researchers say, is an expert system that uses statistics, graph theory, image processing and decision trees, can be applied in cases where the correct assessment of data is critical, the incoming data are too numerous for a human to easily review, or the consequences of a sensor failure would be significant, researchers stated. IODA typically works by evaluating data from sensor systems and treating them as "images" that mimics the way a person might look at a plot of data points to spot an inconsistency, researchers said. 

To find bad data, complex operations typically rely on multiple sensors, as well as algorithms that characterize specific relationships among the data being collected, and identify failures when the data unexpectedly change, researchers stated.

 A drawback in most of those methods is they are designed for a particular type of time series and can fail catastrophically when applied to different types of data, especially in situations where there are numerous and sometimes subtle errors, experts said.

 The algorithm consists of several thousand lines of a technical computing language known as MATLAB. Developers may expand and translate it into a computer programming language such as C so it can be used for commercial purposes, researchers said. 

This is not the first use of Outlier algorithms.  Other examples of its use can be found in data mining applications for example.

Layer 8 in a box

Check out these other hot stories:

IBM, researchers get 24M DOE supercomputer hours to develop controversial lithium air battery

NASA test drives all-composite prototype spacecraft

NASA Mars Rover Spirit ends one mission to start another

How will NASA defend Earth against killer asteroids and comets?

NASA Mars rover Opportunity grinds "cool" rock

Military wants protection from laser weapons

NASA wants your Mars photo target ideas

Brain drain; admin failures threaten FCC role

FBI flashes Big Apple with criminals on Times Square billboard

DARPA wants crowdsourcing applications to fly small satellites

US preps cyber outfit to protect national electric grid

Open source tool, algorithm aim to slash software errors

Join the Network World communities on Facebook and LinkedIn to comment on topics that are top of mind.

Copyright © 2010 IDG Communications, Inc.

SD-WAN buyers guide: Key questions to ask vendors (and yourself)