Subthreshold analog circuit for computing the maximal principal component of 3-D data

We present the analysis, design, and experimental results of a circuit that computes the maximal principal component of three dimensional data. To reduce power consumption, the implemented model uses circuit elements and differential pairs which operate in the subthreshold regime of MOS operation.<<ETX>>

[1]  Fathi M. A. Salam,et al.  Data-driven subthreshold analog circuit that computes principal components , 1993, Proceedings of 36th Midwest Symposium on Circuits and Systems.

[2]  Carver Mead,et al.  Analog VLSI and neural systems , 1989 .

[3]  Terence D. Sanger,et al.  Optimal unsupervised learning in a single-layer linear feedforward neural network , 1989, Neural Networks.

[4]  Lex A. Akers,et al.  A neural processing node with on-chip learning , 1993, 1993 IEEE International Symposium on Circuits and Systems.

[5]  Hong Chen,et al.  An on-line unsupervised learning machine for adaptive feature extraction , 1993, 1993 IEEE International Symposium on Circuits and Systems.

[6]  P. Wintz,et al.  Information Extraction, SNR Improvement, and Data Compression in Multispectral Imagery , 1973, IEEE Trans. Commun..

[7]  Paul Hasler,et al.  Circuit implementation of a trainable neural network using the generalized Hebbian algorithm with supervised techniques , 1992, [Proceedings 1992] IJCNN International Joint Conference on Neural Networks.