A Distributed Approach to Array Optimization, 16-9275

Printer Friendly Version

Principal Investigator
Christopher Smith

Inclusive Dates: 10/01/01 - 10/01/03

Background - The deployment of antenna arrays has classically been an ad hoc, manpower intensive process. Experience has shown that some arrays are very well behaved while others are very ill behaved. While theory exists to explain the behavior of particular antenna arrays, no methods have been developed that can automatically design an antenna array based on a given environment.

Over the past half-decade, the Signal Exploitation and Geolocation Division has made significant advances in the development of numerical methods for generating antenna patterns and antenna-array manifolds. Active development is continuing to produce significant advances in this technology. The application of these modeling techniques to intelligently select the best location for antennas within a given environment would be a significant step forward in the state of the art in direction finding (DF) antenna-array development.

Approach - The goal of this program is to develop the capability to optimize the design of an antenna array for a given environment. Numerical modeling methods along with an AI-based optimization technique are used to generate array designs for a given environment. This program will result in an enhanced ability to design and deploy DF antenna systems in complex environments.

Accomplishments - This project has successfully designed a number of new antenna arrays. The optimization methods developed have shown to be very effective in this problem domain. In addition a significant parallel-processing cluster has been developed which can continue to be applied to new problems for internal or external project work.

2003 Program Home