서브메뉴
검색
Machine Learning Meets Analog Circuit Design: Intelligent Automation of IC Design.
Machine Learning Meets Analog Circuit Design: Intelligent Automation of IC Design.
상세정보
- 자료유형
- 학위논문
- Control Number
- 0017164482
- International Standard Book Number
- 9798384044420
- Dewey Decimal Classification Number
- 621.3
- Main Entry-Personal Name
- Fayazi, Morteza.
- Publication, Distribution, etc. (Imprint
- [S.l.] : University of Michigan., 2024
- Publication, Distribution, etc. (Imprint
- Ann Arbor : ProQuest Dissertations & Theses, 2024
- Physical Description
- 180 p.
- General Note
- Source: Dissertations Abstracts International, Volume: 86-03, Section: B.
- General Note
- Advisor: Afshari, Ehsan;Dreslinski, Ronald G.
- Dissertation Note
- Thesis (Ph.D.)--University of Michigan, 2024.
- Summary, Etc.
- 요약Custom Analog and Mixed-Signal (AMS) circuits have broad applications in wireless communication, biosensors, etc. Traditionally, the design parameters of AMS circuits e.g. transistor size and biasing are calculated manually by designers. However, the complexity of physical models and severe process variations with downscaling of technology node for large AMS circuits leads to inefficiency and hardship of robust manual design especially in corner extraction. Moreover, generating post-layout simulation is very time-consuming. Therefore because of the growing demand of high performance, low power and time-to-market Integrated Circuits (ICs), there is a crucial need of autonomous AMS circuit design. This would catch up with the digital circuit designs, which have been automated for a long time.Automating AMS circuit design procedure has always been challenging as it tightly ties to human expertise and intuition to make a relationship between various parameters and performances. Given the main circuit topology, in order to satisfy the desired specifications, the circuit parameters should be chosen optimally. So, this circuit optimization problem, i.e. determining the circuit parameter values to meet the required specifications, can be solved using mathematical optimization methods. As a result, a Computer-Aided Design (CAD) tool would be able to automate this circuit sizing optimization procedure leveraging techniques such as gradient-based, convex optimization, and evolutionary algorithms.Recently, Machine Learning (ML) has shown promising results in different areas from image categorization to speech recognition. ML algorithms are suitable for yield estimation and making high-order models which can be used instead of complex physical models with a less computational cost for AMS circuit design. Furthermore, ML can be leveraged to replace lengthy and costly measurements by a set of simpler measurements to ease the AMS production test. The main ML challenge to have an accurate model is providing robust and enough simulation data for training sets due to essential high-dimensional variation space to model the process variation and costly simulation in a growing AMS system size. Also, providing Intellectual Properties (IPs) from different IC companies is another barrier of dataset collection. Automatically optimized AMS circuit sizing is another application of ML. In such approaches, Electronic Design Automation (EDA) tools try combinations of design variables to find a sizing that meets the desired specifications instead of the traditional circuit sizing approach i.e. going from the target specification to the corresponding device sizes. After automation of each AMS block sizing, automation of whole System on Chip (SoC) is doable with integrating these pieces.This thesis proposes a series of works that uncover unique methods for automating the design of AMS circuits. First, we propose a fully automated Single-Board Computer (SBC) generator tool, FASCINET. FASCINET uses a Neural Network (NN) model to design customized peripheral circuits for SBCs. The tool creates a large Commercial Off-the-Shelf Database (COTS DB) of existing components, efficiently searches through them, and selects optimal components for both main and peripheral components based on the user's requirements. Creating such a broad COTS DB requires processing abundant datasheets. In order to automate this process, we describe a novel NN-based approach for automatically categorizing datasheets and propose an extraction technique for parsing relevant functional information. Our evaluations show that FASCINET is able to design SBCs that are identical to the manually-designed ones except for minor differences. Second, we go over our table extraction tool, Tablext, which can be used on datasheets or any other documents.Third, we present AnGeL, a fully-automated analog circuit generator framework. AnGeL performs all the schematic circuit design steps from deciding the circuit topology to determining the circuit parameters i.e. sizing. Furthermore, we present a method to reduce the size of AnGeL's database. For this purpose, we use NNs to determine the behavior of complicated circuit topologies by combining the more simple ones. By generating such unlabeled data, the time for providing the training set is significantly reduced compared to the conventional approaches. Our results show that for multiple circuit topologies, in comparison to the state-of-the-art works while maintaining the same accuracy, the required labeled data is reduced by 4.7x - 1090x. Also, the runtime of AnGeL is 2.9x - 75x faster. Fourth, we propose FuNToM, a functional modeling method for Radio Frequency (RF) circuits. FuNToM leverages the two-port analysis method along with NNs for modeling multiple topologies using a single main dataset and multiple small datasets. This significantly reduces the required number of training data. Our results show that for multiple RF circuits, in comparison to the state-of-the-art works while maintaining the same accuracy, the required training data is reduced by 2.8x - 10.9x. In addition, FuNToM needs 176.8x - 188.6x less time for collecting the training set in post-layout modeling.Finally, we conclude our work and give meaningful insights about the current challenges and open issues.
- Subject Added Entry-Topical Term
- Computer engineering.
- Subject Added Entry-Topical Term
- Computer science.
- Subject Added Entry-Topical Term
- Electrical engineering.
- Index Term-Uncontrolled
- Electronic Design Automation
- Index Term-Uncontrolled
- Analog circuit
- Index Term-Uncontrolled
- Physical models
- Index Term-Uncontrolled
- Machine Learning
- Added Entry-Corporate Name
- University of Michigan Electrical and Computer Engineering
- Host Item Entry
- Dissertations Abstracts International. 86-03B.
- Electronic Location and Access
- 로그인을 한후 보실 수 있는 자료입니다.
- Control Number
- joongbu:656948
MARC
008250224s2024 us ||||||||||||||c||eng d■001000017164482
■00520250211153008
■006m o d
■007cr#unu||||||||
■020 ▼a9798384044420
■035 ▼a(MiAaPQ)AAI31631416
■035 ▼a(MiAaPQ)umichrackham005653
■040 ▼aMiAaPQ▼cMiAaPQ
■0820 ▼a621.3
■1001 ▼aFayazi, Morteza.
■24510▼aMachine Learning Meets Analog Circuit Design: Intelligent Automation of IC Design.
■260 ▼a[S.l.]▼bUniversity of Michigan. ▼c2024
■260 1▼aAnn Arbor▼bProQuest Dissertations & Theses▼c2024
■300 ▼a180 p.
■500 ▼aSource: Dissertations Abstracts International, Volume: 86-03, Section: B.
■500 ▼aAdvisor: Afshari, Ehsan;Dreslinski, Ronald G.
■5021 ▼aThesis (Ph.D.)--University of Michigan, 2024.
■520 ▼aCustom Analog and Mixed-Signal (AMS) circuits have broad applications in wireless communication, biosensors, etc. Traditionally, the design parameters of AMS circuits e.g. transistor size and biasing are calculated manually by designers. However, the complexity of physical models and severe process variations with downscaling of technology node for large AMS circuits leads to inefficiency and hardship of robust manual design especially in corner extraction. Moreover, generating post-layout simulation is very time-consuming. Therefore because of the growing demand of high performance, low power and time-to-market Integrated Circuits (ICs), there is a crucial need of autonomous AMS circuit design. This would catch up with the digital circuit designs, which have been automated for a long time.Automating AMS circuit design procedure has always been challenging as it tightly ties to human expertise and intuition to make a relationship between various parameters and performances. Given the main circuit topology, in order to satisfy the desired specifications, the circuit parameters should be chosen optimally. So, this circuit optimization problem, i.e. determining the circuit parameter values to meet the required specifications, can be solved using mathematical optimization methods. As a result, a Computer-Aided Design (CAD) tool would be able to automate this circuit sizing optimization procedure leveraging techniques such as gradient-based, convex optimization, and evolutionary algorithms.Recently, Machine Learning (ML) has shown promising results in different areas from image categorization to speech recognition. ML algorithms are suitable for yield estimation and making high-order models which can be used instead of complex physical models with a less computational cost for AMS circuit design. Furthermore, ML can be leveraged to replace lengthy and costly measurements by a set of simpler measurements to ease the AMS production test. The main ML challenge to have an accurate model is providing robust and enough simulation data for training sets due to essential high-dimensional variation space to model the process variation and costly simulation in a growing AMS system size. Also, providing Intellectual Properties (IPs) from different IC companies is another barrier of dataset collection. Automatically optimized AMS circuit sizing is another application of ML. In such approaches, Electronic Design Automation (EDA) tools try combinations of design variables to find a sizing that meets the desired specifications instead of the traditional circuit sizing approach i.e. going from the target specification to the corresponding device sizes. After automation of each AMS block sizing, automation of whole System on Chip (SoC) is doable with integrating these pieces.This thesis proposes a series of works that uncover unique methods for automating the design of AMS circuits. First, we propose a fully automated Single-Board Computer (SBC) generator tool, FASCINET. FASCINET uses a Neural Network (NN) model to design customized peripheral circuits for SBCs. The tool creates a large Commercial Off-the-Shelf Database (COTS DB) of existing components, efficiently searches through them, and selects optimal components for both main and peripheral components based on the user's requirements. Creating such a broad COTS DB requires processing abundant datasheets. In order to automate this process, we describe a novel NN-based approach for automatically categorizing datasheets and propose an extraction technique for parsing relevant functional information. Our evaluations show that FASCINET is able to design SBCs that are identical to the manually-designed ones except for minor differences. Second, we go over our table extraction tool, Tablext, which can be used on datasheets or any other documents.Third, we present AnGeL, a fully-automated analog circuit generator framework. AnGeL performs all the schematic circuit design steps from deciding the circuit topology to determining the circuit parameters i.e. sizing. Furthermore, we present a method to reduce the size of AnGeL's database. For this purpose, we use NNs to determine the behavior of complicated circuit topologies by combining the more simple ones. By generating such unlabeled data, the time for providing the training set is significantly reduced compared to the conventional approaches. Our results show that for multiple circuit topologies, in comparison to the state-of-the-art works while maintaining the same accuracy, the required labeled data is reduced by 4.7x - 1090x. Also, the runtime of AnGeL is 2.9x - 75x faster. Fourth, we propose FuNToM, a functional modeling method for Radio Frequency (RF) circuits. FuNToM leverages the two-port analysis method along with NNs for modeling multiple topologies using a single main dataset and multiple small datasets. This significantly reduces the required number of training data. Our results show that for multiple RF circuits, in comparison to the state-of-the-art works while maintaining the same accuracy, the required training data is reduced by 2.8x - 10.9x. In addition, FuNToM needs 176.8x - 188.6x less time for collecting the training set in post-layout modeling.Finally, we conclude our work and give meaningful insights about the current challenges and open issues.
■590 ▼aSchool code: 0127.
■650 4▼aComputer engineering.
■650 4▼aComputer science.
■650 4▼aElectrical engineering.
■653 ▼aElectronic Design Automation
■653 ▼aAnalog circuit
■653 ▼aPhysical models
■653 ▼aMachine Learning
■690 ▼a0544
■690 ▼a0984
■690 ▼a0464
■690 ▼a0800
■71020▼aUniversity of Michigan▼bElectrical and Computer Engineering.
■7730 ▼tDissertations Abstracts International▼g86-03B.
■790 ▼a0127
■791 ▼aPh.D.
■792 ▼a2024
■793 ▼aEnglish
■85640▼uhttp://www.riss.kr/pdu/ddodLink.do?id=T17164482▼nKERIS▼z이 자료의 원문은 한국교육학술정보원에서 제공합니다.
미리보기
내보내기
chatGPT토론
Ai 추천 관련 도서
Buch Status
- Reservierung
- 캠퍼스간 도서대출
- 서가에 없는 책 신고
- Meine Mappe