Title :
Neural networks for target detection
Author :
Wang, Chia-Jiu ; Wu, Chwan-Hwa
Author_Institution :
Dept. of Electr. & Comput. Eng., Colorado Univ., Colorado Springs, CO, USA
Abstract :
A back-propagation neural network and an adaptive filter are used to detect moving targets for pulse-Doppler radar applications. The back-propagation network has one hidden layer. The adaptive filter consists of an instar layer and a shunting competitive layer. Significant performance improvements have been achieved compared to the conventional fast Fourier transform method. An example with 16 input-signal samples per moving target is used for demonstration purposes
Keywords :
adaptive filters; fast Fourier transforms; neural nets; radar theory; adaptive filter; back-propagation neural network; fast Fourier transform method; hidden layer; input-signal samples; instar layer; moving targets; pulse-Doppler radar applications; shunting competitive layer; target detection; Adaptive filters; Clutter; Doppler radar; Filter bank; Frequency; Neural networks; Object detection; Radar detection; Radar signal processing; Springs;
Conference_Titel :
Circuits and Systems, 1990., IEEE International Symposium on
Conference_Location :
New Orleans, LA
DOI :
10.1109/ISCAS.1990.112030
