NOTE: This item is not available outside the Texas A&M University network. Texas A&M affiliated users who are off campus can access the item through NetID and password authentication or by using TAMU VPN. Non-affiliated individuals should request a copy through their local library's interlibrary loan service.
New methodologies for top-down statistical modeling and optimization of integrated circuits
Abstract
New methodologies intended to facilitate system level design and optimization are developed. Specifically, these methodologies allow: (1) development of system level models for the purpose of gathering statistical data on the performance of large systems and performance dependencies on sub-block parameters and (2) the development of system level top-down optimization strategies for hierarchical propagation of optimal sub-block performance specifications. Together, these new methodologies expand the current scope of Design for Quality, allowing statistical design and optimization techniques to be applied early in the product design cycle. In both of these areas, new methodologies, techniques and concepts are presented. In the area of system level modeling, criteria for judging model appropriateness are developed and explained. A new method of system level modeling intended for use in top-down investigations is introduced, exploiting the flexibility and efficiency of behavioral modeling techniques. A complicated IC system (a PLL filter tuning circuit) is thoroughly analyzed and models suited to a variety of analysis methods are developed and tested. Model examples using two behavioral modeling languages, Spectre and MAST, are developed to further demonstrate the validity of the methods introduced. In the area of top-down design, a new methodology suited to system level optimization tasks is introduced. The methodology relies on cost definitions and tolerance assignment originally proposed by M.A. Styblinski and the relevant algorithms developed for the optimal assignment of sub-block performance by A. Achab as a specifications. In addition, the nominal values of sub-block performances are optimized in order to improve the performance of the overall system while reducing sensitivity to sub-block performance variations. Two examples are investigated using the new methodology. The first example is a MOSFET-C bandpass filter, which utilizes macromodels and SPICE analysis to gain insights into the performance de pendencies of the system. The second example uses the models created for the PLL to complete a system level optimization and optimal specification assignment process using models developed and analyzed in the Saber environment. Following the development and verification of the new methodologies, conclusions are drawn, and possibilities for future work in the area are suggested.
Description
Due to the character of the original source materials and the nature of batch digitization, quality control issues may be present in this document. Please report any quality issues you encounter to digital@library.tamu.edu, referencing the URI of the item.Includes bibliographical references.
Issued also on microfiche from Lange Micrographics.
Collections
Citation
Alexander, Daniel D. (1996). New methodologies for top-down statistical modeling and optimization of integrated circuits. Master's thesis, Texas A&M University. Available electronically from https : / /hdl .handle .net /1969 .1 /ETD -TAMU -1996 -THESIS -A446.
Request Open Access
This item and its contents are restricted. If this is your thesis or dissertation, you can make it open-access. This will allow all visitors to view the contents of the thesis.