Logic gate formed neuron type processing element
- 6 January 2003
- proceedings article
- Published by Institute of Electrical and Electronics Engineers (IEEE)
Abstract
The authors present the design of neuron-type circuit elements based on logic circuits, mainly NAND gates. Three types of subcircuits were introduced, one called the cell body with its dendritic inputs, another representing the axon base, and the final one mimicking the axon. The resulting design was an asynchronous sequential circuit. This neuron-type processing element (NTPE) computes the sum of binary weighted input pulses incorporating temporal as well as spatial summation and thresholding. This process has been accomplished by introducing a time constraint variable in the operation of the NTPE. With the availability of VLSI techniques to implement these NTPEs, full neural-type systems, including computers, can be conceived and constructed using presently available IC technology. Alternatively, they can be assembled from off-the-shelf logic circuits.Keywords
This publication has 3 references indexed in Scilit:
- Neurons with graded response have collective computational properties like those of two-state neurons.Proceedings of the National Academy of Sciences, 1984
- Neural networks and physical systems with emergent collective computational abilities.Proceedings of the National Academy of Sciences, 1982
- Analysis and simulation of networks of mutually inhibiting neuronsBiological Cybernetics, 1972