基於圖形方法之嵌入式系統設計組件化解決方案

簡易檢索 / 詳目顯示

回結果列表

研究生：	楊文彣 Yang, Wen-Wen
論文名稱：	基於圖形方法之嵌入式系統設計組件化解決方案 Enhancing Embedded Systems Design: A Graph-Centric Approach to Component-Based Solutions
指導教授：	周百祥 Chou, Pai-Hsiang
口試委員:	韓永楷 Hon, Wing-Kai 謝孫源 Hsieh, Sun-Yuan 李皇辰 Lee, Huang-Chen
學位類別：	碩士 Master
系所名稱：	電機資訊學院 - 資訊系統與應用研究所 Institute of Information Systems and Applications
論文出版年：	2024
畢業學年度：	112
語文別：	英文
論文頁數：	76
中文關鍵詞：	嵌入式系統、電子設計自動化、知識圖譜、特徵模型、元件篩選、演算法
外文關鍵詞：	Embedded System, Electronic Design Automation, Knowledge Graph, Feature Model, Component Selection, Algorithm
相關次數：	點閱：84 下載：0
分享至:	分享至facebook 分享至twitter

查詢本校圖書館目錄查詢臺灣博碩士論文知識加值系統勘誤回報

本論文提出了一種創新的商用現成 (COTS) 元件建模方法，旨在解決現有嵌入式系統設計工具在功能理解和介面描述上的不足。傳統設計工具主要依賴原圖繪製，缺乏對元件深度資訊的掌握，導致設計過程繁瑣且易錯。過往將元件建模為模組化單元的方式容易受限，限制了設計空間的探索。

為此，我們提出基於特徵的知識圖譜建模方法，將 COTS 元件表示為特徵及其關係，全面捕捉功能、結構、介面等多方面資訊。這種模型能夠在不受分類限制的情況下，實現基於功能需求的組件選擇。此外，模型整合了假設-保證 (assume-guarantee) 資訊，支援其他系統模組的自動合約式驗證，有助於及早發現並解釋設計錯誤。

此研究經由驗證多個案例，證實我們的建模方式在設計空間探索方面具有顯著優勢，能夠發現其他建模方法無法企及的優化設計方案，因此我們相信此研究提出的元件庫設計方法，能夠為嵌入式系統設計提供一個革新的設計實踐，提升設計效率與可靠性，同時為開發更複雜與高效的嵌入式系統提供堅實的基礎。

This thesis proposes a way to model commercial off-the-shelf (COTS) components for a new generation of design tools for embedded systems. Tools today are mainly schematic drawing tools with little or no knowledge about the components' functionality or even interfaces, thereby forcing designers to manually handle many intricate, error-prone details. Previous attempts at modeling components as modular units in class hierarchy may allow exploration of alternative components but only if their functionality is similarly packaged. Unfortunately, such models prevent exploration of the same set of functionality packaged in very different ways, such as the case of systems-on-chip (SoC). To overcome these challenges, we propose modeling COTS components as features and their relationships within a knowledge graph, serving as a unified internal representation for various component aspects. Features are units of information that capture not only the functions offered by a component and flow relationships between them but also their relations to structural features such as ports on the interface and inner structures, plus parameter features as attributes associated with these features.

Our proposed feature-based model offers several significant advantages over previous attempts at component modeling. First, the use of functional features and the associated inner structures enables coverage of requirements by selecting COTS components without being limited to how they are packaged, a problem we call the modularity trap. Second, our component model captures assumptions and guarantees for contract-based validations of interfaces and specifications. This is a crucial capability, as the knowledge captured in our model can not only be used to find errors in the design but also to explain why, whether the design is created manually by a person or automatically by generative AI. We have conducted a number of case studies to show the effectiveness of our component models to enable design space exploration to discover superior design points that would otherwise be impossible for other models to find automatically.

Introduction 1
1 Motivation ............................................. 1
1.1 COTS-Based Design of Embedded Systems ................ 1
1.2 Lack of Smart Design Tools for COTS-Based Design ..... 2
1.3 Need for Design Space Exploration .................... 3
2 Component Library in Sysmaker .......................... 4
3 Contributions .......................................... 6
4 Outline ................................................ 6

Related Work 7
1 Modular Component Approach ............................. 7
1.1 IP-XACT and Kactus2 .................................. 8
1.2 Polymorphic Blocks ................................... 9
2 Feature Approach ....................................... 10
2.1 Feature Technology and Ontology ...................... 10
2.2 Feature Algebra ...................................... 11

System Overview 13
1 Sysmaker Design Framework .............................. 13
1.1 Purpose and Objective ................................ 13
1.2 Microservice Framework ............................... 14
1.3 Five-Level Abstraction ............................... 16
1.4 Feature Representation and Projection ................ 17
2 Component Library ...................................... 18
2.1 Component Model Representation ....................... 18
2.2 APIs for COTS Data Retrieval ......................... 19
2.3 Integration with Design Flow ......................... 19

Technical Approach 21
1 Feature-Based Modeling ................................. 21
1.1 Feature Hierarchy .................................... 22
1.2 Types of Features .................................... 23
1.3 Contract-Based Characteristics ....................... 24
1.4 Container Concept .................................... 25
1.5 Feature Coverage ..................................... 26
2 Graph-Centric Representation ........................... 29
2.1 COTS Component Graph ................................. 29
2.2 Graph Construction Algorithm ......................... 30
2.3 Graph Traversal and Query Operations ................. 33
2.4 Similarity Measurement ............................... 37

Implementations 41
1 Software Framework and Tools Used ...................... 41
2 Graph Data Structure for COTS Modeling ................. 42
2.1 Feature Hierarchy .................................... 43
2.2 Query Examples ....................................... 47
3 Component Information Retrieval ........................ 48
3.1 Get Component Information ............................ 49
3.2 Get Feature Information .............................. 49
3.3 Get Component Similarity ............................. 51
4 Component Library Indexing.............................. 53
4.1 Leveraging Neo4j’s Indexing Capabilities ............. 53
4.2 Indexing Strategy .................................... 54

Case Study 55
1 The EcoMini System ..................................... 55
2 COTS Component Pre-Screening ........................... 56
3 Component Alternative Searching ........................ 58
3.1 Voltage Regulator Selection .......................... 59
3.2 Motion Sensor Selection .............................. 59
3.3 Comparative Analysis of Selected Components .......... 61
3.4 System-Level Optimization and Evaluation ............. 61
4 Discussion ............................................. 62

Conclusions and Future Work 66
1 Conclusions ............................................ 66
2 FutureWork ............................................. 67
                                

[HC15] Hen-Pai Hsu and Pai H. Chou. EcoMini: Low-power, miniature bluetooth low energy motion sensor node. Master's thesis, National Tsing Hua University, 2015.

[HC21] Chih-En Hsu and Pai H. Chou. Top-down design of embedded systems using feature algebra. Master's thesis, National Tsing Hua University, 2021.

[IP-23] IP-XACT Working Group. IEEE standard for IP-XACT, standard structure for packaging, integrating, and reusing IP within tool flows. IEEE Std 1685-20
22 (Revision of IEEE Std 1685-2014), pages 1-750, 2023.

[KMM+11] Antti Kamppi, Lauri Matilainen, Joni-Matti Maatta, Erno Salminen, Timo D. Hamalainen, and Marko Hannikainen. Kactus2: Environment for embedded product development using IP-XACT and MCAPI. In 2011 14th Euromicro Conference on Digital System
Design, pages 262-265, 2011.

[LRC+20] Richard Lin, Rohit Ramesh, Connie Chi, Nikhil Jain, Ryan Nuqui, Prabal Dutta, and Björn Hartmann. Polymorphic Blocks: Unifying high-level specification and low-level control for circuit board design
. In Proceedings of the 33rd Annual ACM Symposium on User Interface Software and Technolo
gy, UIST '20, page 529-540, New York, NY, USA, 2020. Association for Computing Machinery.

[Neo07] Neo4j. Neo4j: The leader in graph databases. https://neo4j.com/, 2007. Accessed: 2024-07-17.

[RD19] Avi Rana and K. Deeba. Online book recommendation system using collaborative filtering (with Jaccard Similarity). Journal of Physics: Conference Series, 13
62:012130, 11 2019.

[Rie03] M. Riebisch. Towards a more precise definition of feature models. In M. Riebisch, J. O. Coplien, and D. Streitferdt, editors, Modeling Variability for Object-Oriented Product Lines, pages 64-76, Norderstedt, 2003. Book On Demand Publ. Co.

[Ron10] Armin Ronacher. Flask: A microframework for python. https://flask.palletsprojects.com/, 2010. Accessed: 2024-07-17.

[RSP04] Matthias Riebisch, Detlef Streitferdt, and Ilian Pashov. Modeling variability for object-oriented product lines. In Frank Buschmann, Alejandro P. Buchmann, and Mariano A. Cilia, editors, Object-Oriented Technology. ECOOP 2003 Workshop Reader, pages 1
65-178, Berlin, Heidelberg, 2004. Springer Berlin Heidelberg.

[TDK16] TDK InvenSense. MPU-9250 Nine-Axis (Gyro + Accelerometer + Compass) MEMS MotionTracking ™™ Device, June 2016. Available at https://product.tdk.com/system/files/dam/doc/product/sensor/mortion-inertial/imu/data_sheet/ps-mpu-9250a-01-v1.1.pdf, Revision 1.1.

[TDK20] TDK InvenSense. MPU-6500 Six-Axis (Gyro + Accelerometer) MEMS Motion Tracking ™™ Device, May 2020. Available at https://product.tdk.com/system/files/dam/doc/product/sensor/mortion-inertial/imu/data_sheet/mpu-6500-datasheet2.pdf, Revision 1.3.

[Tex13] Texas Instruments. CC2541 SimpleLink™ Bluetooth® low energy and proprietary wireless MCU datasheet, June 2013. Available at https://www.ti.com/lit/ds/symlink/cc2541.pdf, Rev. D.

[Tex14a] Texas Instruments. TPS6274x 360nA IQ Step Down Converter For Low Power Applications datasheet, July 2014. Available at https://www.ti.com/lit/ds/symlink/tps62740.pdf, Rev. B.

[Tex14b] Texas Instruments. TPS783xx 500-nA IQ, 150-mA, Ultralow Quiescent Current Low-Dropout Linear Regulator, November 2014. Available at https://www.ti.com/lit/ds/symlink/tps783.pdf, Rev. A.

[VST+22] Sandeep Varma, Shivam Shivam, Aakash Thumu, Apuroop Bhushanam, and Debjit Sarkar. Jaccard based similarity index in graphs: A multi-hop approach. In 2022 IEEE Delhi Section Conference (DELCON), pages 1-4. IEEE, 02 2022.

[ZFS05] Xuan F. Zha, Steven J. Fenves, and Ram D. Sriram. A Feature-Based Approach to Embedded System Hardware and Software Co-Design. In ASME 2005 International Design Engineering Technical Conferences and Computers and Information in Engineeri
ng and Information in Engineering Conference, pages 609-620, 09 2005.

[ZS06] Xuan F. Zha and Information in Engineering Conference, pages 609-620, 09 2006.

簡易檢索 / 詳目顯示

相關論文