Rohit Mital, SGT, Joseph Coughlin, SGT, Mike Canaday, SGT
Keywords: Big Data, Analytics, Sensor,
Abstract:
A goal of big data analytics is to help leaders make informed and rapid decisions by analyzing large volumes of complex data, as well as other forms of data that may be untapped by conventional analyses, and presenting it in a form that facilitates decision making. Big data analytics is the process of examining large data sets containing a variety of data types to uncover hidden patterns, unknown correlations, and other useful information.
Sensors typically record significant amounts of data but it is often not exploited except in special cases and after historically large amounts of analysis time. Big data analytics provides a mechanism to routinely monitor these data sets while also providing insight into anomalous events, such as are encountered in large sensor systems such as those in the space surveillance network.
In this study, we simulate recorded data from a notional radar or optical sensor and use big data technologies and the analytics to process the data to analyze and predict sensor performance. This study focuses on data products that would commonly be analyzed at a site and how big data technologies can be used to detect anomalies.
This study shows how the ability to rapidly drill down into the data enables an analyst or decision maker to assess potential system anomalies. This study shows how current technologies and predictive analytical techniques can be used to view the data, detect and explain anomalies, and predict preventative maintenance actions in a timely manner.
Date of Conference: September 15-18, 2015
Track: Poster