Compadre: A Device Independent Voice-to-Voice Language Translator Software Solution

Award Information
Department of Defense
Award Year:
Phase I
Agency Tracking Number:
Solicitation Year:
Solicitation Topic Code:
Solicitation Number:
Small Business Information
12701 Fair Lakes Circle, Fairfax, VA, 22033
Hubzone Owned:
Minority Owned:
Woman Owned:
Principal Investigator:
Robert Kellogg
Senior Scientist
(703) 322-0881
Business Contact:
Victor Sellier
Vice President
(703) 322-0881
Research Institution:
There are many approaches to improving IW systems' ability to provide Indications and Warning (I&W) for the Information Warfare (IW) Picture. Traditional approaches have relied on rule based systems to infer the RF environment. But in the modern IWbattlespace the rules may change too quickly to effectively write IW heuristics.This SBIR takes an innovative approach to deriving the IW Picture by investigating self-organizing algorithms to quickly classify and parse the RF environment in a high-order parametric space only available to the sensor itself.A number of algorithms have been tentatively identified as prospective candidates for investigation, including hard and soft Hebbian learning and neural-gas algorithms. It is expected that these algorithms have the potential to separate signals not only bytheir individual parameters (frequency, bandwidth, etc.), but also by important (and a priori unknown) joint parameters such as transmission timing, secondary signal follow on, and diffuse RF environmental changes.During Phase I, the self-organizing algorithms will be selected and put into a mathematical foundation suitable for inferencing using IW system high-order data sets. The algorithms will be coded (JAVA is the current language of choice for machineindependence) and subjected to well controlled data sets with parameter vectors that match IW systems used in both the National and Tactical RF environment.The performance of the self-organizing algorithms will be evaluated and compared. It is anticipated that one or more algorithms will be recommended for further testing in a prototype system. Initially this software system will be demonstrated on the newgeneration of IW systems, such as those built on LIGHTHOUSE Technology. However, by using JAVA as machine independent code, the software can migrate to other generations of IW equipment. It is expected during Phase II that a successful IW InferencingEngine will become a GCCS-M/CUB software segment.We envision multiple opportunities for commercializing the software products that are developed under this SBIR. The immediate opportunities for commercialization involve other government agencies that have similar problems with rapidly interpretingchanges in the RF environment, whether to identify hostile threat or to recognize illicit communications. There are extensive applications in the telecommunication industry for monitoring and predicting traffic loads and managing predictable interference.Further, the same concepts of self-organization may be applied to other disciplines including weather prediction and stock-market forecasting.

* information listed above is at the time of submission.

Agency Micro-sites

SBA logo

Department of Agriculture logo

Department of Commerce logo

Department of Defense logo

Department of Education logo

Department of Energy logo

Department of Health and Human Services logo

Department of Homeland Security logo

Department of Transportation logo

Enviromental Protection Agency logo

National Aeronautics and Space Administration logo

National Science Foundation logo
US Flag An Official Website of the United States Government