Elsevier

Integration

Volume 42, Issue 1, January 2009, Pages 10-23
Integration

ANTIGONE: Top-down creation of analog-to-digital converter architectures

https://doi.org/10.1016/j.vlsi.2008.07.001Get rights and content

Abstract

A new framework for high-level synthesis of analog and mixed-signal integrated systems is introduced. It focuses on the translation of a functional description into a behavioral model of a specific architecture with values for the parameters of its building blocks. An initial, simple, high-level solution is evolved into a more realistic low-level result by applying appropriate transformations of both architecture and parameters. This top-down heterogeneous optimization algorithm deals readily with multifarious performance characteristics and diverse types of objectives, and integrates various sources of design knowledge and types of transformations. Furthermore, it creates the architecture rather than selecting it. As illustration of the methodology, a tool, ANTIGONE, has been written that allows to generate different types of A/D converters depending on the specifications like speed and accuracy.

Introduction

Designing an analog or mixed-signal system involves two tasks: a choice of a topology and a selection of values for its parameters. In the last 25 years, many analog CAD tools have been developed dedicated to the second task [1], [2]. These tools assume the selection of the architecture has already been made by an experienced designer.

High-level synthesis tools, on the other hand, optimize both the architecture and its parameters. As a result, the design space is increased which can result in a more optimal solution compared to a manual selection of the architecture. Indeed, altering the topology may be a better idea to meet certain specifications than trying to set the parameters of an unsuited topology to some specific values. Section 2 briefly reviews methods that have been employed in analog CAD tools to deal with the selection or generation of the architecture.

A methodology suited for high-level analog and mixed-signal synthesis has to deal with complex systems, with several types of performance metrics and heterogeneous objectives, and with both parametric and architectural transformations of intermediate designs. All fundamental properties of an efficient high-level design strategy are summarized in Section 3. Based on these requirements, a top-down heterogeneous optimization algorithm has been developed in this work.

The new top-down evolutionary method aims to converting a functional specification of an analog system towards a low-level behavioral model of a specific architecture that implements the specified functionality. First, the functionality is translated into a behavioral model of a simple idealized architecture. This embryonic design acts as the ancestor of the population of designs. Each iteration of the optimization process starts with the calculation of the performance characteristics, and an evaluation of these properties. Then, transformations are chosen from a collection. They alter either the architecture, its parameters, or both. Finally, the transformations with the highest probability to improve the design characteristics are selected by preference. They are applied on the designs in the population, which results in a new generation. Section 4 elaborates on all aspects of the new optimization framework.

To illustrate the new methodology, a prototype EDA tool for some types of A/D converters has been written in C++. Section 5 presents this tool, ANTIGONE, and discusses the inputs expected from the user and the results of the tool of various optimizations run for different sets of specifications. Finally, conclusions are presented in Section 6.

Section snippets

Classification of design strategies

Design strategies suited for analog EDA differ from each other in several aspects, like the abstraction or description level they operate on [1], the flexibility to deal with multiple topologies or circuit types, the ability to explore different architectures during synthesis, or the optimization algorithm adopted to find the architecture or its parameters [2]. These properties can be used to distinguish different classes of design approaches.

The main contribution of the newly developed method,

Objectives for synthesis strategy

Different synthesis strategies are appropriate at different phases of the analog design process. For example, designing a basic op amp topology requires another approach than constructing an entire mixed-signal system with several basic building blocks. The main goal of the synthesis strategy presented in this work is to convert a functional description of an analog or mixed-signal system into a behavioral description at a lower abstraction level of a specific topology. More specifically, seven

Top-down heterogeneous optimization

The synthesis strategy tackles the complexity of designing analog and mixed-signal systems by employing a genetic evolutionary algorithm. Several types of components are defined within the optimization framework to fulfill specific tasks. Together, they perform the top-down heterogeneous optimization.

Application: analog-to-digital conversion

As application of the optimization framework, a custom tool has been developed to explore a part of the design space of analog-to-digital converters. The implementation consists of two parts (see Fig. 9): a library with basic elements of the framework and the actual program for A/D converters. The following subsections elaborate on this application showing how the developed framework is used for the design of mixed-signal systems.

Conclusions

The high-level synthesis of analog and mixed-signal systems is a complex process characterized by several heterogeneous aspects. The top-down heterogeneous optimization strategy presented in this paper offers the flexibility to work with different design representations, performance functions, and optimization objectives. An evolutionary process is defined which evolves a high-level design into a low-level description of an architecture that fulfills the specifications and implements the

Ewout S. J. Martens was born in Genk, Belgium, in 1978. He received his M.S. and Ph.D. degrees in electrical engineering from the Katholieke Universiteit Leuven (KUL), Belgium in 2001 and 2007, respectively. He participated in summer internships at the analog design company AnSem and at the research center IMEC.

From 2001 to 2007, he was a research assistant with the ESAT-MICAS Laboratories, KUL, as member of the CAD group of Prof. Georges Gielen. His research focused on the development of

References (30)

  • E. Martens et al.

    Classification of analog synthesis tools based on their architecture selection mechanisms

    Integration VLSI J.

    (2008)
  • L.R. Carley, et al., Synthesis tools for mixed-signal ICs: progress on frontend and backend strategies, in: IEEE Design...
  • G.G.E. Gielen et al.

    Computer-aided design of analog and mixed-signal integrated circuits

    Proc. IEEE

    (2000)
  • H.Y. Koh

    OPASYN: a compiler for CMOS operational amplifiers

    IEEE Trans. Comput. Aided Des.

    (1990)
  • A. Torralba

    FASY: a fuzzy-logic based tool for analog synthesis

    IEEE Trans. Comput. Aided Des. Integrated Circuits Syst.

    (1996)
  • G. Van der Plas

    AMGIE—a synthesis environment for CMOS analog integrated circuits

    IEEE Trans. Comput. Aided Des. Integrated Circuits Syst.

    (2001)
  • G.G.E. Gielen

    Analog circuit design optimization based on symbolic simulation and simulated annealing

    IEEE J. Solid-State Circuits

    (1990)
  • G. Alpaydın

    An evolutionary approach to automatic synthesis of high-performance analog integrated circuits

    IEEE Trans. Evol. Comp.

    (2003)
  • W. Nye

    DELIGHT.SPICE: an optimization-based system for the design of integrated circuits

    IEEE Trans. Comput. Aided Des.

    (1988)
  • F. El-Turky et al.

    BLADES: an artificial intelligence approach to analog circuit design

    IEEE Trans. Comput. Aided Des.

    (1989)
  • R. Harjani

    OASYS: a framework for analog circuit synthesis

    IEEE Trans. Comput. Aided Des.

    (1989)
  • C. Toumazou et al.

    Analog IC design automation: part I—automated circuit generation: new concepts and methods

    IEEE Trans. Comput. Aided Des. Integrated Circuits Syst.

    (1995)
  • T. Eeckelaert, et al., Hierarchical bottom–up analog optimization methodology validated by a delta–sigma A/D converter...
  • P.C. Maulik

    Integer programming based topology selection of cell-level analog circuits

    IEEE Trans. Comput. Aided Des. Integrated Circuits Syst.

    (1995)
  • H. Tang et al.

    High-level synthesis of ΔΣ modulator topologies optimized for complexity, sensitivity, and power consumption

    IEEE Trans. Comput. Aided Des. Integrated Circuits Syst.

    (2006)
  • Cited by (5)

    • Automated topology synthesis of analog and RF integrated circuits: A survey

      2017, Integration, the VLSI Journal
      Citation Excerpt :

      With the ever-growing complexity of the analog/RF circuits, it is now inevitable to embed some kind of knowledge in the EDA procedure of analog/RF circuits. The EA-based methods, which are employed as a remedy in some research publications, aim to contribute more of the automation sense and make the synthesis procedure more independent of the human effort [25,35]. GA or GP techniques conduct the synthesis procedure while evolving the circuit topology by leveraging genetic operators such as mutation, crossover, and selection.

    • An automated high-level topology generation procedure for continuous-time ΣΔ modulator

      2010, Integration, the VLSI Journal
      Citation Excerpt :

      A completely new topology with suitable topology parameters may yield better results. Therefore, in recent days, the researchers are giving more attention in developing an integrated framework of topology generation and parameter optimization [2,4,5]. The important advantages of the present procedure in comparison to similar other procedures are:

    • Automatic filter synthesis based on tree generation and evolutionary optimization

      2011, Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
    • Evolutionary algorithms in the optimal sizing of analog circuits

      2011, Studies in Computational Intelligence
    • Electronic design automation of multi-scroll chaos generators

      2010, Electronic Design Automation of Multi-scroll Chaos Generators

    Ewout S. J. Martens was born in Genk, Belgium, in 1978. He received his M.S. and Ph.D. degrees in electrical engineering from the Katholieke Universiteit Leuven (KUL), Belgium in 2001 and 2007, respectively. He participated in summer internships at the analog design company AnSem and at the research center IMEC.

    From 2001 to 2007, he was a research assistant with the ESAT-MICAS Laboratories, KUL, as member of the CAD group of Prof. Georges Gielen. His research focused on the development of architectural-level analysis and modeling methods, and of automated synthesis approaches for data converters, in particular Delta-Sigma A-to-D modulators, and RF transceiver architectures. For this work, he has been awarded a research fellowship from the Fund for Scientific Research—Flanders (Belgium) (F.W.O.-Vlaanderen).

    Since 2007, he has been affiliated as chief scientist with Mephisto Design Automation, a spin-off of KUL that provides solutions for analysis and optimization of electrical circuits.

    Georges G.E. Gielen received his M.Sc. and Ph.D. degrees in Electrical Engineering from the Katholieke Universiteit Leuven, Belgium, in 1986 and 1990, respectively. In 1990, he was appointed as a postdoctoral research assistant and visiting lecturer at the department of Electrical Engineering and Computer Science of the University of California, Berkeley. From 1991 to 1993, he was a postdoctoral research assistant of the Belgian National Fund of Scientific Research at the ESAT laboratory of the Katholieke Universiteit Leuven. In 1993, he was appointed assistant professor at the Katholieke Universiteit Leuven, where he was promoted to full professor in 2000.

    His research interests are in the design of analog and mixed-signal integrated circuits, and especially in analog and mixed-signal CAD tools and design automation (modeling, simulation and symbolic analysis, analog synthesis, analog layout generation, analog and mixed-signal testing). He is coordinator or partner of several (industrial) research projects in this area. He has authored or coauthored two books and more than 300 papers in edited books, international journals and conference proceedings. He regularly is a member of the Program Committees of international conferences (DAC, ICCAD, ISCAS, DATE, CICC, etc.), and served as General Chair of the DATE conference in 2006 and of the ICCAD conference in 2007. He serves regularly as member of editorial boards of international journals (IEEE Transactions on Circuits and Systems, Springer international journal on Analog Integrated Circuits and Signal Processing, Elsevier Integration). He received the 1995 Best Paper Award in the John Wiley international journal on Circuit Theory and Applications, and was the 1997 Laureate of the Belgian Royal Academy on Sciences, Literature and Arts in the discipline of Engineering. He received the 2000 Alcatel Award from the Belgian National Fund of Scientific Research for his innovative research in telecommunications, and won the DATE 2004 Best Paper Award. He is a Fellow of the IEEE, served as elected member of the Board of Governors of the IEEE Circuits And Systems (CAS) society and as chairman of the IEEE Benelux CAS chapter. He served as the President of the IEEE Circuits And Systems (CAS) Society in 2005.

    View full text