DocumentCode
2664220
Title
A neural-like feed-forward ADC
Author
Chigusa, Yasutami ; Tanaka, Mamoru
Author_Institution
Dept. of Electr. & Electron. Eng., Sophia Univ., Tokyo, Japan
fYear
1990
fDate
1-3 May 1990
Firstpage
2959
Abstract
A fast, accurate, and compact neural-like feedforward analog-to-digital converter (ADC) with binary weighted synapses and sigmoidal neurons is presented. Since the feedforward ADC has no feedback lines which make the acquisition time of the system indefinite, the conversion time is finite like finite-impulse-response (FIR) filters. Therefore there is no implicit hysteresis in the feedforward ADC. The characteristics of the ADC are described. A prototype 8-bit feedforward ADC using analog parallel processing shows no errors and has a conversion rate of over 5 MHz
Keywords
analogue-digital conversion; neural nets; parallel processing; acquisition time; analog parallel processing; analog-to-digital converter; binary weighted synapses; conversion rate; conversion time; neural-like feed-forward ADC; sigmoidal neurons; Circuit testing; Design methodology; Feedforward systems; Finite impulse response filter; Hysteresis; Neural networks; Neurons; Prototypes; Robustness; Voltage;
fLanguage
English
Publisher
ieee
Conference_Titel
Circuits and Systems, 1990., IEEE International Symposium on
Conference_Location
New Orleans, LA
Type
conf
DOI
10.1109/ISCAS.1990.112631
Filename
112631
Link To Document