Analog Integrated Circuits and Signal Processing

, Volume 49, Issue 3, pp 343–358

Platform-based mixed signal design: Optimizing a high-performance pipelined ADC

  • Pierluigi Nuzzo
  • Fernando De Bernardinis
  • Alberto Sangiovanni Vincentelli
Regular Paper

DOI: 10.1007/s10470-006-9067-8

Cite this article as:
Nuzzo, P., De Bernardinis, F. & Vincentelli, A.S. Analog Integr Circ Sig Process (2006) 49: 343. doi:10.1007/s10470-006-9067-8

Abstract

We apply Platform-Based Design (PBD) to the power optimization of a 14 bit, 80 MS/s pipelined Analog-to-Digital Converter (ADC) in a mixed signal formulation. A platform is a library of components and interconnects, each characterized with a set of behavioral, performance and composition models, that is used to raise the level of abstraction to enable system-level design. PBD is a meet-in-the-middle methodology that consists of two phases. The bottom-up phase generates a set of platform libraries that are exploited in the system hierarchy. The top-down phase allows exploring feasible solutions within the platform libraries and selecting the optimum implementation. To evaluate the cost of each implementation, the behavioral models available through platform abstraction are used both for digital and analog components. We provide an example of the use of the methodology and its features for analog circuits by modeling two amplifiers with different topologies as analog components, showing details of the analog characterization process. Then, we create a mixed signal platform library as a combination of an analog and a digital platform (bottom-up phase). The top-down phase performs optimization across the analog/digital boundary to minimize power consumption constrained to given noise and linearity requirements. Simulation results show that interesting power saving can be achieved, as much as 64% compared with an original hand-optimized ADC.

Keywords

Platform-based designPipelined analog-to-digital converterDigital calibrationSystem level optimization

Copyright information

© Springer Science + Business Media, LLC 2006

Authors and Affiliations

  • Pierluigi Nuzzo
    • 2
  • Fernando De Bernardinis
    • 1
    • 2
  • Alberto Sangiovanni Vincentelli
    • 1
  1. 1.Department of Electrical Engineering and Computer ScienceUniversity of CaliforniaBerkeleyUSA
  2. 2.Dipartimento di Ingegneria, dell'InformazioneUniversità, di PisaItaly