论文信息 - Architectures for analog VLSI implementation of neural networks for solving linear equations with inequality constraints

Architectures for analog VLSI implementation of neural networks for solving linear equations with inequality constraints

A novel algorithm for solving generalized systems of linear equations subject to linear inequalities is proposed. The algorithm is robust with respect to wild (spiky) noise and outliers. The algorithm is expressed completely in differential equations. An OTA (operational transconductance amplifier)-based implementation of the algorithm as an analog VLSI circuit in CMOS technology is discussed. One of the many possible applications of the proposed techniques is the implementation of cellular neural networks with real-time on-chip learning capabilities.<<ETX>>

[1] Edgar Sanchez-Sinencio,et al. Operational transconductance amplifier-based nonlinear function syntheses , 1989 .

[2] Joos Vandewalle,et al. Application of relaxation methods to the adaptive training of neural networks , 1989 .

[3] F. Zou,et al. Design method for cellular neural network with linear relaxation , 1991, 1991., IEEE International Sympoisum on Circuits and Systems.

[4] B. Linares-Barranco,et al. CMOS circuit implementations for neuron models , 1990, IEEE International Symposium on Circuits and Systems.

[5] Lin-Bao Yang,et al. Cellular neural networks: theory , 1988 .

[6] J J Hopfield,et al. Neural networks and physical systems with emergent collective computational abilities. , 1982, Proceedings of the National Academy of Sciences of the United States of America.

[7] A. Cichocki,et al. Neural networks for solving systems of linear equations and related problems , 1992 .

[8] Andrzej Cichocki,et al. MOS Switched-Capacitor and Continuous-Time Integrated Circuits and Systems: Analysis and Design , 1989 .

[9] Jaime Ramirez-Angulo,et al. Extended transconductance adjustment/linearisation technique , 1991 .