Automatic Synthesis of Analog Integrated Circuits Using Genetic Algorithms and Electrical Simulations (original) (raw)
Related papers
Automatic Synthesis of Analog Integrated Circuits Including Efficient Yield Optimization
Computational Intelligence in Analog and Mixed-Signal (AMS) and Radio-Frequency (RF) Circuit Design, 2015
The goal of this paper is to present a tool for automatic synthesis of analog basic integrated blocks using the genetic algorithm heuristic and external electrical simulator. The methodology is based on the minimization of a cost function and a set of constraints in order to size circuit individual transistors. The synthesis methodology was implemented in Matlab and used GAOT (Genetic Algorithm Optimization Toolbox) as heuristic. For transistors simulation we used the Smash simulator and ACM MOSFET model, which includes a reduced set of parameters and is continuous in all regions of operation, providing an efficient search in the design space. As circuit design example, this paper shows the application of this methodology for the design of an active load differential amplifier in AMS 0.35um technology.
AN OPTIMIZATION-BASED TOOL FOR CIRCUIT LEVEL SYNTHESIS ANALOG INTEGRATED CIRCUITS
The goal of this paper is to present a tool for automatic sizing of analog basic integrated blocks using heuristics of non-linear optimization and an external electrical simulator. The methodology is based on the minimization of a cost function and constraint parameters related to circuit electrical characteristics. The methodology was implemented in Matlab using two heuristics of non-linear optimization, Simulated Annealing (SA) and Genetic Algorithms (GA). As electrical simulations we used the Smash simulator and the ACM MOSFET model. As design example, this paper shows the application of this methodology for the design of an active load differential amplifier using AMS 0.35µm technology.
This paper proposes a methodology for automatic synthesis of analog basic building blocks. The methodology is based on the minimization of a cost function composed by the variables of an analog block (transistor lengths and widths). The implementation was made in Matlab, with a generic non-linear optimization function and an external electrical simulator for the evaluation of circuit specifications. The MOSFET model used was ACM, which has a reduced set of technology parameters and is continuous in all regions of operation. As example, we present the design of a differential amplifier.
ANALOG INTEGRATED CIRCUITS DESIGN BY MEANS OF GENETIC ALGORITHMS
This paper applies the Artificial Intelligence technique called Genetic Algorithms (GA) to perform automatic analog integrated circuits design, synthesis and optimization in order to reduce the development time of this kind of integrated circuits. We have used in this work the stage-single, single-ended SOI CMOS Operational Transconductance Amplifier (OTA), where the transistors dimensions are determined focusing on achieving the smallest OTA die area possible. As future works, we intend to use SPICE simulations to validate the promising results obtained by the GA method proposed.
An evolutionary approach to automatic synthesis of high-performance analog integrated circuits
IEEE Transactions on Evolutionary Computation, 2003
This paper presents an analog integrated circuit synthesis system based on an evolutionary approach. The system contains several novel features. One of these is the high-performance optimization algorithm, which is a combination of evolutionary strategies and simulated annealing. Modeling of dc parameters is done via a fast dc simulator developed for this purpose whereas modeling of ac parameters can be done either with user-defined equations or with neural-fuzzy performance models trained from SPICE simulations. Another novel feature of the system is the incorporation of matching properties of devices. This way, the optimized circuit becomes tolerant to process variations. The synthesis system has been tested on several independent examples and synthesized circuits have been verified functionally with SPICE simulations. Finally, a prototype chip composed of the three examples has been manufactured. The measurement results have demonstrated the validity of the synthesis system on silicon.
2008 IEEE International Symposium on Circuits and Systems, 2008
We present a methodology for efficient design of analog circuits using an automated simulation based synthesis tool. In this methodology, the designer chooses a suitable circuit topology and defines the performance criteria of the circuit. The synthesis tool provides optimized device dimensions which guarantees that the circuit meets the specified performance criteria across process corners. This methodology is independent of the circuit type (as long as the performance criteria can be quantified and measured from simulation data), the fabrication process being used, and also the circuit simulator of choice. We show that this design methodology reduces the design cycle time by a significant amount and helps analyzing the trade off between different performance criteria. It also helps in analyzing the suitability of several alternative topologies for a given purpose in a short time. In this paper we substantiate this claim with the help of an operational amplifier design.
Design Optimization of Analog Active Circuits Using the Genetic Algorithm
2012
This paper presents an optimization tool for the design of analog circuits implemented with commercially-available discrete devices. It takes into consideration the standard values available for the passive components, as well as the non- idealities of commercially-available active devices. The multi-criteria optimization is based on a genetic algorithm developed in Matlab specifically for this task, that works in conjunction with components of the OrCAD package, namely the OrCAD CIS Capture and PSpice. The effectiveness of the proposed procedure is demonstrated by a real-life example, the optimized design of a fourth-order low-pass filter implemented with Operational Amplifiers and discrete resistors and capacitors.
Genetic algorithm application to analog integrated circuit design
International Journal of Reasoning-based Intelligent Systems, 2012
Genetic Algorithms (GAs) are basically search algorithms that are based on the mechanics of natural selection and natural genetics. The search mechanism is based on the dividing it into generations and each generation consists of individuals, these individuals describe the problem parameters. The advantages of this method are: zeroth-order search algorithm, extremely robust and it is based on population that can search many different areas on the search space. This paper presents an application of GAs in analog integrated circuit design. Several important design parameters and performances can be significantly improved. The GAs is developed and applied to CMOS OP-AMP structure. The GAs results a design parameters that minimise the required chip area and reduce the power consumption, while preserving the specified performances. A comparison with the conventional analytical method shows the advantage of using GAs.
2012
Analog integrated circuits (ICs) design is a complex task due to the large number of input variables that must be determined simultaneously in order to achieve different multiple design goals of an analog integrated circuit design, such as voltage gain (A V ), unit voltage gain frequency (f T ), slew-rate (SR), harmonic distortion (THD), etc. By using an evolutionary system based on Genetic Algorithm (AG) integrated to the SPICE simulator, named AGSPICE, this work aims to study, understand and calculate the different correlations between the MOSFETs inversion regimes and the design goals of an operational transconductance amplifier (OTA) operating in different design features. We believe that the AGSPICE can provide new design recommendations for the designers and reduce the design cycle time as well. Our experimental results with the AGSPICE are also compared to the results obtained manually and present compatible solutions to other works available in the related literature.