CAREER: Advanced interface circuits for communication systems through architecture, circuit, and software innovation

职业：通过架构、电路和软件创新实现通信系统的先进接口电路

• 批准号: 0238166
• 负责人: Michael Perrott
• 金额: $ 40万
• 依托单位: Massachusetts Institute of Technology
• 依托单位国家: 美国
• 项目类别: Continuing grant
• 财政年份: 2003
• 资助国家: 美国
• 起止时间: 2003-03-01 至 2008-02-29
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=0238166&HistoricalAwards=false
• 关键词: CAREER; Advanced; interface; circuits; communication

Efforts to create low cost,high performance communication systems in the future will be chal-lenged by the high cost of developing new interface circuits that are suited to their needs.Theseinterface circuits serve as the intermediary between real world signals,which are analog in nature,and the digital processing engines at the core of the system.Current interface circuits are primarilyanalog in their implementation,which makes their performance highly sensitive to manufacturingvariations in the process they are fabricated in,and to noise generated by digital circuits thatoperate on the same chip.The impact of this sensitivity is that interface circuits must be customdesigned for the process they are implemented in,cannot be easily integrated with their adjoiningdigital processor,and are often very restricted in their ability to serve a wide variety of applica-tions.The resulting lack of portability of these circuits leads to high development costs since theygenerally must be redesigned from scratch for new systems.The goal of the proposed research is to develop new interface circuit architectures that can beleveraged over a wide variety of fabrication processes and applications,and thereby dramaticallyreduce the cost of their implementation in future communication applications.This goal will beachieved by replacing a signi .cant portion of the analog circuitry in the interface with digitalcircuits that realize the required functionality through signal processing techniques that will bedeveloped as part of this e .ort.This implementation philosophy will lead to dramatically di .erentinterface architectures than in use today,and will provide new insights into the tradeo .s betweenusing analog versus digital circuitry for given function and performance requirements.New simulation models and software tools will be developed to facilitate the process of exploringnew system architectures and their accompanying circuits.The simulation modeling e .ort willbe focused on obtaining accurate behavioral level descriptions of analog components in digitalcircuit simulators,which will allow straightforward evaluation of the impact of di .erent signalprocessing algorithms on the overall system performance.The software development e .ort will bedirected toward better integrating system level and transistor level simulation tools,and providingestimation of the impact of extrinsic factors such as substrate noise.A new educational course will be developed at MIT that passes on the knowledge and tools ac-quired in this research program.The class will teach high speed circuit techniques for both analogand digital components.System level concepts will be explored through a combination of mathe-matical analysis and interactive numerical simulation using the custom models and software tools.It will examine the impact of circuit limitations on system performance,and explore ways of iter-ating between circuit and system designs to achieve the best overall system performance.ThroughMIT 's OpenCourseWare initiative,the course materials and software will be made available toother universities as well as industry.Dissemination of the course through the OpenCourseWareprogram will allow students and industry designers from a variety of ethnic,economic,and geo-graphic backgrounds to have access to state of the art techniques and tools for interface circuitdesign.

未来创造低成本、高性能通信系统的努力将受到开发适合其需求的新接口电路的高成本的挑战。这些接口电路作为真实世界信号（本质上是模拟信号）和系统核心的数字处理引擎之间的中介。当前的接口电路在其实现中主要是模拟的，这使得它们的性能对制造过程中的制造变化以及在同一芯片上运行的数字电路产生的噪声高度敏感。这种灵敏度的影响是接口电路必须根据其实现的过程进行定制设计，不能容易地与相邻的数字处理器集成，并且通常在服务于各种应用的能力方面受到限制。由于这些电路通常必须为新系统从头开始重新设计，因此缺乏可移植性导致开发成本高。提出的研究目标是开发新的接口电路架构，可以在各种制造工艺和应用中加以利用，从而大大降低其在未来通信应用中的实现成本。这一目标将通过用数字电路取代接口中的大部分模拟电路来实现，数字电路通过将作为本报告一部分开发的信号处理技术实现所需的功能。这种实现理念将导致显著的di。比目前使用的更先进的接口架构，并将为行业提供新的见解。根据给定的功能和性能要求，在使用模拟电路和数字电路之间进行选择。将开发新的仿真模型和软件工具，以促进探索新系统架构及其伴随电路的过程。仿真建模e。Ort将专注于获得数字电路模拟器中模拟元件的准确行为级别描述，这将允许直接评估di的影响。事件信号处理算法对系统整体性能的影响。软件开发e。Ort将致力于更好地集成系统级和晶体管级仿真工具，并提供对衬底噪声等外在因素影响的估计。麻省理工学院将开发一门新的教育课程，传授本研究项目中获得的知识和工具。本课程将教授模拟和数字元件的高速电路技术。系统级概念将通过使用自定义模型和软件工具的数学分析和交互式数值模拟相结合来探索。它将研究电路限制对系统性能的影响，并探索电路和系统设计之间的迭代方法，以实现最佳的整体系统性能。通过麻省理工学院的开放课程计划，课程材料和软件将提供给其他大学和行业。通过开放课程软件项目传播该课程将使来自不同种族、经济和地理背景的学生和工业设计师能够接触到界面电路设计的最新技术和工具。