Title :
A kind of neural network similar to ART network with application to radar signal sorting
Author :
Tang, Jinsong ; Zhu, Zhaoda
Author_Institution :
Nanjing Univ. of Aeronaut. & Astronaut., China
Abstract :
In order to solve the problem of radar signal sorting in real time, this paper presents a kind of neural network (NN) similar to ART network. Its characteristic is to finish heuristic clustering algorithm of pattern recognition by using adaptive resonance theory (ART) on the basis of making good use of data statistic property. This NN has not the self-stabilized top-down structure of the general ART network. The self-stabilized learning is finished by bottom-up structure. Although this NN has less adaptability than ART NN, it can solve a wide variety of problems as well. The vigilance and weight updating have more strict mathematical formulas. The result of digital simulation shows this network can work well in sorting radar signals. It can also be used to solve the similar problem of clustering
Keywords :
heuristic programming; neural nets; radar theory; signal processing; ART network; adaptive resonance theory; data statistic property; heuristic clustering algorithm; neural network; pattern recognition; radar signal sorting; self-stabilized top-down structure; vigilance; weight updating; Clustering algorithms; Digital simulation; Heuristic algorithms; Neural networks; Pattern recognition; Radar; Resonance; Sorting; Statistics; Subspace constraints;
Conference_Titel :
Aerospace and Electronics Conference, 1994. NAECON 1994., Proceedings of the IEEE 1994 National
Conference_Location :
Dayton, OH
Print_ISBN :
0-7803-1893-5
DOI :
10.1109/NAECON.1994.332871
