TerraEchos

Streaming data technology supports covert intelligence and surveillance sensor systems

Published on 27-Dec-2011

Validated on 12 Nov 2013

"We are not only capturing and reducing a tremendous amount of digital acoustic data, which is exactly what InfoSphere Streams does very well, but we are also running intensive computational statistical analysis—and we are doing all of this in one-fourteenth of a second." - Dr. Alex Philp, Founder and CTO, TerraEchos, Inc.

Customer:
TerraEchos

Industry:
Professional Services

Deployment country:
United States

Solution:
Big Data, Information Infrastructure, Information Integration, Security: Governance, Risk and Compliance

Overview

A U.S. Department of Energy National Lab turned to IBM Business Partner, TerraEchos, to implement an advanced, covert security and surveillance system, based on the TerraEchos Adelos S4 System and IBM technology. Because the solution captures and transmits real-time, streaming acoustical data from around the lab premises, security staff has unprecedented insight into any event and can “hear” what is going on—even when the disturbance is miles away. 

Business need:
A U.S. Department of Energy National Lab needed a solution to detect, classify, locate and track potential threats to secure its perimeters and border areas.

Solution:
IBM Business Partner, TerraEchos, implements an advanced security and covert surveillance system based on the TerraEchos Adelos S4 System with IBM InfoSphere Streams and an IBM System x3650 server.

Benefits:
Reduces time to process 275 Mbit of data from hours to just one-fourteenth of a second; captures and analyzes huge volumes of data in real time, providing unprecedented insight to detect, classify, locate, track and deter potential threats.

Case Study

A leading provider of covert intelligence and surveillance sensor systems, TerraEchos, Inc., provides organizations with advanced security solutions for critical infrastructures and extended borders. One TerraEchos client is a science-based, applied engineering national laboratory dedicated to supporting the mission of the U.S. Department of Energy (DOE) in nuclear and energy research, science and national defense.

Securing the scientific intelligence, technology and resources related to these initiatives is vital. To this end, this national lab recognized the need for a technology solution that would detect, classify, locate and track potential threats—above and below ground—to secure its perimeters and border areas. This solution would provide lab personnel and security staff with more situational awareness and enable a faster and more intelligent response to any threat.

Detecting and analyzing a wide range of sounds—even from miles away
The requirements of the ideal solution were considerable. The solution would have to continuously consume and analyze massive amounts of digital acoustic data from biological, mechanical and environmental objects-in-motion. In addition, because lab personnel lacked time to record the data and listen to it later, the solution had to gather and analyze information simultaneously.

The analysis could extract meaningful intelligence, as well as verify and validate the data, such as distinguishing between the sounds of a trespasser versus a grazing animal. To put the sophistication of the needed technology into perspective, the data consumption and analytical requirements would be akin to listening to 1,000 MP3 songs simultaneously and successfully discerning the word “zero” from every song—within a fraction of a second.

The solution would also serve as the lab’s central nervous system and would have to meet strict technical requirements, including:

  • Interoperability, enabling lab personnel to collect and analyze an array of data from video, acoustic, and other types of sensors to create a holistic view of a situation
  • Scalability to support new requirements as the lab’s fiber-optic arrays, surveillance areas, and security perimeters change
  • Extensibility, serving as a framework to fit into the lab’s existing IT architecture and integrating with signal processors and mobile and mapping applications

Advanced fiber-optics combine with real-time streaming data
To meet these requirements, the lab turned to IBM Business Partner, TerraEchos, to implement an advanced, covert security and surveillance system, based on the TerraEchos Adelos® S4 System, an IBM System x3650 server and IBM® InfoSphere® Streams software, part of the IBM big data platform. The TerraEchos Adelos S4 solution offers advanced fiber-optic acoustic sensor technology licensed from the United States Navy. InfoSphere Streams is the engine that processes digital acoustic data-in-motion continuously from fiber-optic sensor arrays.

Serving as the underlying analytics platform, the processing capacity of InfoSphere Streams enables the Adelos S4 solution to analyze and classify, streaming acoustic data in real time. InfoSphere Streams collects data from multiple sensor types and enables associated streams of structured and unstructured data to be integrated into an intelligence system for threat detection, classification, correlation, prediction and communication by means of a service oriented architecture (SOA). Based on this technology, TerraEchos provides one of the most robust surveillance classification systems in the industry, and is the first fiber-optic sensor company to incorporate InfoSphere Streams as the computational platform for sensor data analytics.

“InfoSphere Streams captures and feeds approximately 275 Mbit of acoustic data from 1,024 individual sensor channels into the Adelos S4 solution, where a computationally intensive Bayesian Data Reduction Algorithm performs calculations to classify the data in real time,” says Dr. Alex Philp, Founder and CTO of TerraEchos. “We are not only capturing and reducing a tremendous amount of digital acoustic data, which is exactly what InfoSphere Streams does very well, but we are also running intensive computational statistical analysis—and we are doing all of this in one-fourteenth of a second.”

“Before implementing our solution, many of our government customers required hours to process 275 Mbit of data, not in one-twelfth or one-fourteenth of a second,” Dr. Philp adds. “Our solution is deployed on a single IBM System x3650 server, running eight cores and capable of processing up to 42 terabytes every day, which would be costly to store. The cost-benefit ratio for using InfoSphere Streams to process and analyze streaming data in real time is phenomenal.”

Extending the security perimeter creates a strategic advantage
Because the solution captures and transmits real-time, streaming acoustical data from around the lab premises, security staff has unprecedented insight into any event. The system enables lab and security personnel to “hear” what is going on—even when the disturbance is miles away. In this way, it is possible to confidently identify and classify a potential security threat—and take appropriate action.

“We use the fiber-optic cable as a sensing array,” says Dr. Philp. “The listening devices are actually virtual segments of the cable, so think of a cable a mile long. We break it down digitally into individual microphones or individual listening areas of one meter, and these distances can change. The beauty of this extended perimeter security system is that it is completely passive. Using miles of fiber-optic cables and thousands of listening devices buried underground, the lab can extend its perimeter security and gain a strategic advantage.”

Correlating sensor data delivers a zero false-positive rate
The solution is part of a more comprehensive security system. With the ability to integrate and collect data from video and airborne surveillance systems, lab personnel gain a holistic view of potential threats and issues—or nonissues.

“In addition to detection, classification, localization and tracking, correlating the analysis from acoustic sensors and video cameras provides for verification and validation and a zero false-positive rate,” says Dr. Philp. “With these results, security staff can make confident decisions about responding to a threat—such as how many officers to deploy and which tactics to use—and can also thwart any plans intruders have to breach the property.”

Finally, in addition to meeting the lab’s requirements for extensibility, interoperability and scalability, the solution saves the lab costs associated with data storage because data does not have to be stored before being analyzed. “Capturing approximately 42 terabytes of data each day adds up fast and would be challenging and costly to store,” says Dr. Philp. “InfoSphere Streams offers advantages, especially when you have to capture data continuously in real time and analyze it as it passes by. Organizations can realize huge savings in storage.”

“Given the data processing and analytical challenges addressed using our Adelos Sensor Array, InfoSphere Streams is the right solution for us and our customers,” notes Dr. Philp. “We look forward to growing our strategic relationship with IBM across various sectors and markets to help revolutionize the concept of ‘Sensor as a Service.’”

Solution components
Software
  • IBM® InfoSphere® Streams
Servers
  • IBM System x3650

For more information
To learn more about IBM InfoSphere Streams, please contact your IBM sales representative or IBM Business Partner, or visit the following website: ibm.com/software/data/infosphere/streams

To learn more about IBM big data, visit: ibm.com/software/data/bigdata

To get involved in the conversation: www.smartercomputingblog.com/category/big-data

For information on TerraEchos, Inc. visit: http://www.terraechos.com

Products and services used

IBM products and services that were used in this case study.

Hardware:
System x: System x3650 M3

Software:
InfoSphere Streams

Legal Information

© Copyright IBM Corporation 2011 IBM Corporation Software Group Route 100 Somers, NY 10589 U.S.A. Produced in the United States of America December 2011 All Rights Reserved IBM, the IBM logo, ibm.com, InfoSphere, and System x are trademarks of International Business Machines Corporation in the United States, other countries or both. If these and other IBM trademarked terms are marked on their first occurrence in this information with a trademark symbol (® or ™), these symbols indicate U.S. registered or common law trademarks owned by IBM at the time this information was published. Such trademarks may also be registered or common law trademarks in other countries. A current list of IBM trademarks is available on the web at “Copyright and trademark information” at ibm.com/legal/copytrade.shtml Other company, product and service names may be trademarks or service marks of others. References in this publication to IBM products or services do not imply that IBM intends to make them available in all countries in which IBM operates.