DocumentCode :
1905869
Title :
Implementation of O(n) complexity max/min circuits for fuzzy and connectionist computing
Author :
Hassoun, Mohamad H. ; Nabha, Ali M.
Author_Institution :
Dept. of Electr. & Comput. Eng., Wayne State Univ., Detroit, MI, USA
fYear :
1993
fDate :
1993
Firstpage :
998
Abstract :
O(n) complexity electronic circuits are proposed for maximum (minimum) selection/propagation of n analog input voltages. The proposed min/max circuits employ electronic amplifiers and diodes and are easily realizable with CMOS VLSI technology. Theoretical analysis and experimental results with off-the-shelf components are presented. These circuits are of low complexity in the sense that the number of transistors and/or connections required scales linearly in the number of inputs, n. An analytical expression for the accuracy of the proposed circuits in terms of the number of inputs and amplifier gain is derived. It shows that high-resolution low-complexity analog max/min modules are feasible. The authors´ results hinge on the ability to map a sequential algorithm for the computation of Max (Min) functions onto a simple collective computing architecture
Keywords :
CMOS integrated circuits; VLSI; analogue computer circuits; analogue processing circuits; fuzzy logic; neural nets; CMOS VLSI technology; O(n) complexity max/min circuits; amplifier gain; connectionist computing; electronic amplifiers; fuzzy computing; sequential algorithm; Analog computers; CMOS technology; Circuits; Computer architecture; Computer networks; Fuzzy systems; Hardware; Neural networks; Neurons; Very large scale integration;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Neural Networks, 1993., IEEE International Conference on
Conference_Location :
San Francisco, CA
Print_ISBN :
0-7803-0999-5
Type :
conf
DOI :
10.1109/ICNN.1993.298694
Filename :
298694
Link To Document :
بازگشت