Automatic Design of Operational Amplifier Utilizing both Equation-Based Method and Genetic Algorithm

Kento SUZUKI  Nobukazu TAKAI  Yoshiki SUGAWARA  Masato KATO  

IEICE TRANSACTIONS on Fundamentals of Electronics, Communications and Computer Sciences   Vol.E100-A   No.12   pp.2750-2757
Publication Date: 2017/12/01
Online ISSN: 1745-1337
DOI: 10.1587/transfun.E100.A.2750
Type of Manuscript: Special Section PAPER (Special Section on VLSI Design and CAD Algorithms)
automatic design,  circuit block,  equation-based,  Genetic Algorithm,  operational amplifier,  

Full Text: PDF(1.7MB)
>>Buy this Article

Automatic design of analog circuits using a programmed algorithm is in great demand because optimal analog circuit design in a short time is required due to the limited development time. Although an automatic design using equation-based method can design simple circuits fast and accurately, it cannot solve complex circuits. On the other hand, an automatic design using optimization algorithm such as Ant Colony Optimization, Genetic Algorithm, and so on, can design complex circuits. However, because these algorithms are based on the stochastic optimization technique and determine the circuit parameters at random, a lot of circuits which do not operate in principle are generated and simulated to find the circuit which meets specifications. In this paper, to reduce the search space and the redundant simulations, automatic design using both equation-based method and a genetic algorithm is proposed. The proposed method optimizes the bias circuit blocks using the equation-based method and signal processing blocks using Genetic Algorithm. Simulation results indicate that the evaluation value which considers the trade-off of the circuit specification is larger than the conventional method and the proposed method can design 1.4 times more circuits which satisfy the minimum requirements than the conventional method.