Accession Number : AD0903027

Title :   Applications of Learning Techniques to AMTI Radar Problems.

Descriptive Note : Final rept. 30 Jun 71-30 Jun 72,

Corporate Author : HUGHES AIRCRAFT CO CULVER CITY CA ANTENNA DEPT

Personal Author(s) : Howard, James E. ; Birgenheier, Raymond A. ; Young, Gregory O.

Report Date : AUG 1972

Pagination or Media Count : 123

Abstract : A study of the application of learning and adaptive techniques to AMTI radars was conducted. A dynamic doppler filter was conceived in block diagram form to allow a radar to learn from and adapt to its environment. A multiple-target doppler finder was developed to learn the spectral characteristics of the environment. Recursive algorithms and efficient search procedures were developed to make the doppler finder practicable for rapid processing during flight. Means of updating the doppler finder to make use of trends that are recognized in the clutter spectrum were conceived. The performance of the doppler finder was analyzed by computer simulation; It was found that the doppler frequencies for up to four reflectors could be located accurately on the basis of just eight repeated pulses of data, even when the dopplers fell within the same resolution cell. A technique for rapid filter synthesis was developed to adapt the response of a doppler filter to the learned clutter spectrum. The responses of filters that were synthesized using this technique were recorded. The results show that eight pulses are sufficient to provide up to four nulls at selected frequencies without appreciably distorting the in-band response of the filters. The work needed complete integration of the doppler finder and doppler filter into the dynamic doppler filter for an AMTI radar is outlined.

Descriptors :   (*MOVING TARGET INDICATORS, AIRBORNE), (*LEARNING MACHINES, MOVING TARGET INDICATORS), (*BIONICS, RADAR EQUIPMENT), ARTIFICIAL INTELLIGENCE, SEARCH THEORY, ADAPTIVE SYSTEMS, DOPPLER SYSTEMS, ELECTRIC FILTERS, RADAR CLUTTER, RADAR PULSES, ALGORITHMS, DATA PROCESSING, SIMULATION, COMPUTER PROGRAMS, MATHEMATICAL MODELS, SPECTRUM ANALYZERS, TARGET ACQUISITION.

Subject Categories : Active & Passive Radar Detection & Equipment
      Bionics

Distribution Statement : APPROVED FOR PUBLIC RELEASE