Quick Look

Innovative Software Architecture for Signal Analyzer, 16-9347

Printer Friendly Version

Principal Investigators
Rob Dollarhide
Ken Holladay

Inclusive Dates: 09/11/02 - 03/03/03

Background - Signal recognition and identification is a key technology area for the Signal Exploitation and Geolocation Division. Our current software product, the SR-200, is the result of several man-years of effort by SwRI® scientists and engineers. Although it has been successfully deployed in a number of sponsored projects, recent work has highlighted excessive software development and debug time as a major obstacle to rapidly satisfying our customers' requirements.

Approach - To reduce the cost and risk of SR-200 software development, this Quick Look program developed an innovative, object-oriented software architecture for signal recognition systems. Unlike existing SR-200 software that utilizes limited configuration parameters for modifying system behavior, this approach employs a structurally adaptive method for redirecting data flow (and thus signal-recognition algorithms) during run-time. The main objectives of the Quick Look research were to:

  • Determine feasibility of the architecture.
  • Provide software developers and engineers with an approach to quickly develop new recognition modules (i.e., major recognition functions).
  • Provide software developers and engineers with a means to easily modify recognition algorithms during run-time rather than compile-time.
  • Demonstrate compatibility with existing SR-200 code running on typical signal-recognition platforms.

Accomplishments - This project proved to be very successful. The initial UML Class Diagram led to the smooth implementation of test software and accomplished all of the primary objectives. The test software not only demonstrated the benefits of object-oriented programming during the implementation phase but also allows for easy algorithm adaptation during run-time unattainable by parameterized software. Furthermore, the architecture was incorporated into a SR-200 recognition module proving its compatibility with existing Signal Exploitation and Geolocation Division systems.

2003 Program Home