Title :
CMOS implementation and fabrication of the pseudo analog neuron
Author :
Taheri, Babak A.
Author_Institution :
SRI Int., Menlo Park, CA, USA
Abstract :
The pseudo-analog neuron (PAN) uses a transresistance amplifier and multivalued logic design techniques to implement the basic building blocks of artificial neural networks. The performance characteristics of PAN building blocks that were implemented in a standard 2-μm CMOS process are described. A two-layer network of PANs is used to implement various Boolean functions. The two-layer circuit computes ten weighted sum of inputs plus four thresholds in the first layer, and a four weighted sum plus one threshold in the second layer. All thresholding and weighted sums are propagated in 10 ns through both layers. This circuit dissipates 250 μW of average power, and occupies 288 μm×136 μm of silicon area
Keywords :
CMOS integrated circuits; logic circuits; many-valued logics; neural nets; PAN; artificial neural networks; multivalued logic design; pseudo analog neuron; thresholding; transresistance amplifier; two-layer network; weighted sum; Artificial neural networks; Biological neural networks; Biology computing; CMOS process; Circuits; Fabrication; Neural network hardware; Neurons; Silicon; Threshold voltage;
Conference_Titel :
Multiple-Valued Logic, 1993., Proceedings of The Twenty-Third International Symposium on
Conference_Location :
Sacramento, CA
Print_ISBN :
0-8186-3350-6
DOI :
10.1109/ISMVL.1993.289549