A VLSI architecture for fast clustering with fuzzy ART neural networks

The hardware implementation of the fuzzy ART neural network applied to a demanding real time radar signal clustering problem is investigated. To obtain efficient solutions for implementing this neural network with dedicated hardware, the network's algorithm is reformulated, and then a novel fuzzy ART system architecture is proposed. This system architecture is composed of a global comparator and several identical elementary modules (EMs), each one emulating a number of neurons. The general architecture of each EM consists of a local comparator, dividers, neural processors, and a block of memory.

[1]  Simon R. Jones,et al.  Learning in linear systolic neural network engines: analysis and implementation , 1994, IEEE Trans. Neural Networks.

[2]  Karl Sammut,et al.  Toroidal Neural Network: Architecture and Processor Granularity Issues , 1991 .

[3]  Yvon Savaria,et al.  Fast convergence with low precision weights in ART1 networks , 1994, Proceedings of IEEE International Symposium on Circuits and Systems - ISCAS '94.

[4]  Jonathan S. Kane,et al.  POPART: partial optical implementation of adaptive resonance theory 2 , 1993, IEEE Trans. Neural Networks.

[5]  Anil K. Jain,et al.  Algorithms for Clustering Data , 1988 .

[6]  Michael R. Anderberg,et al.  Cluster Analysis for Applications , 1973 .

[7]  U. Ramacher.  Guide Lines to VLSI Design of Neural Nets , 1991 .

[8]  Stephen Grossberg,et al.  Fuzzy ART: Fast stable learning and categorization of analog patterns by an adaptive resonance system , 1991, Neural Networks.

[9]  Gerald Sommer,et al.  Pattern Recognition by Self-Organizing Neural Networks , 1994 .

[10]  D. Curtis Schleher,et al.  Introduction to Electronic Warfare , 1986 .

[11]  N. El-Leithy,et al.  CMOS realization of a class of Hartline neural pools , 1990, IEEE International Symposium on Circuits and Systems.

[12]  Joseph Cavanagh,et al.  Digital Computer Arithmetic , 1983 .

[13]  Robert W. Newcomb,et al.  VLSI implementation of ART1 memories , 1991, IEEE Trans. Neural Networks.

[14]  R. O. Grondin,et al.  VLSI Implementation of Neural Classifiers , 1990, Neural Computation.

[15]  James A. Anderson,et al.  Radar signal categorization using a neural network , 1990, Proc. IEEE.

[16]  Behrooz Kamgar-Parsi,et al.  Automatic Data Sorting Using Neural Network Techniques. , 1996 .

[17]  Stephen Grossberg,et al.  A massively parallel architecture for a self-organizing neural pattern recognition machine , 1988, Comput. Vis. Graph. Image Process..

[18]  Ulrich Rückert,et al.  VLSI Design of Neural Networks , 1990 .

[19]  Tony R. Martinez,et al.  Digital Neural Networks , 1988, Proceedings of the 1988 IEEE International Conference on Systems, Man, and Cybernetics.

[20]  T P Caudell,et al.  Hybrid optoelectronic adaptive resonance theory neural processor, ART1. , 1992, Applied optics.

[21]  Lex A. Akers,et al.  Integrated circuit emulation of ART1 networks , 1989 .

[22]  Christos G. Christodoulou,et al.  Mixed analog/digital VLSI design and simulation of an adaptive resonance theory (ART) neural network architecture , 1995, SPIE Defense + Commercial Sensing.

[23]  B. Moore,et al.  ART1 and pattern clustering , 1989 .

[24]  Joseph J. F. Cavanagh.  Digital Computer Arithmetic: Design And Implementation , 1984 .

[25]  José Luis Huertas,et al.  A real time clustering CMOS neural engine , 1995, 38th Midwest Symposium on Circuits and Systems. Proceedings.