A Neuro Fuzzy Solution in the Design of Analog Circuits

Pedro MIRANDA-ROMAGNOLI  Norberto HERNANDEZ-ROMERO  Juan C. SECK-TUOH-MORA  

Publication
IEICE TRANSACTIONS on Fundamentals of Electronics, Communications and Computer Sciences   Vol.E94-A   No.1   pp.434-439
Publication Date: 2011/01/01
Online ISSN: 1745-1337
DOI: 10.1587/transfun.E94.A.434
Print ISSN: 0916-8508
Type of Manuscript: LETTER
Category: VLSI Design Technology and CAD
Keyword: 
neuro fuzzy design,  analog circuits,  operational amplifier,  optimization,  

Full Text: PDF>>
Buy this Article




Summary: 
A neuro fuzzy method to design analog circuits is explained, where the universe of discourse of the fuzzy system is adjusted by means of a self-organized artificial neural network. As an example of this approach, an op-amp is optimized in order to hold a predetermined aim; where the unity gain bandwidth is an objective of design, and the restrictions of open-loop gain and margin phase are treated as objectives too. Firstly, the experience of the behavior of the circuit is obtained, hence an inference system is constructed and a neural network is applied to achieve a faster convergence into a desired solution. This approach is characterized by having a simple implementation, a very natural understanding and a better performance than static methods of fuzzy optimization.