数据传输完整性_基于IBIS模型的FPGA信号完整性仿真验证方法

人工智能与深度学习等领域的快速发展,使得FPGA等器件应用范围愈加广泛,同时也要求器件的开关速率加快、引脚数量增多。但陡峭的时钟边沿和增加的引脚数使得杂散、耦合、寄生电容电感会对器件产生诸多信号完整性(Signal Integrity,SI)问题。这不仅会降低器件的应用可靠性,对通信网络引入噪声,严重时会致使系统功能失效[1]

一个高速数字系统的信号完整性与以下三种设计层次上的因素有关:(1)芯片级:I/O buffer和信号回流的路径设计不当等[2];(2)封装级:封装的高电感系数,阻抗不匹配[3],布线不当和信号回流路径布局不合理等;(3)PCB板级:链路串扰,端口反射,信号衰减,电磁兼容问题等[4]

目前,对于包括FPGA在内的高速数字器件的信号完整性研究较多,但是均局限于在设计过程中如何改善器件信号完整性。如顾炯炯等人分析了高速集成电路的封装对信号完整性的影响[5];尚玉玲等人通过建立TSV三维物理模型来分析信号完整性影响因素[6];YE Y等人使用模块化建模获得等效电流从而进行信号完整性分析[7]。可以看出,这些工作缺少针对设计师选用器件的角度的考虑,而器件自身引入的信号完整性关系到设计系统的鲁棒性,因而,开展器件信号完整性的验证是很有意义的。

本文首先通过对信号完整性问题产生机理的分析,提出了器件本身信号完整性仿真验证方法,然后使用HyperLynx软件针对SRAM型FPGA器件进行了基于IBIS模型的器件级的信号完整性仿真,再通过对类似的FPGA器件的仿真结果进行对比分析,给出了模型参数的差异对器件信号完整性的影响。

1 仿真原理

1.1 仿真模型

信号完整性仿真工作是基于模型的计算来预测实际信号的传输情况。在四种常用的模型中的行为模型里[8],元器件可以被看成黑盒子,使用中只测量或者模拟其端口的电气特性,而不涉及器件的详细描述,另外它与电路模型相比,在保持了精确性的同时,仿真时间大大缩减。

目前行为模型中有一种IBIS(Input/Output Buffer Information Specification)模型,它通过输入和输出引脚的电压电流关系和电压时间关系来描述器件的行为[9],源文件可以进行修改且易于获取。基于这些原因,本文将选用这种行为模型。

1.2 仿真工具

EDA厂商提供了多种多样的信号完整性仿真工具,其中,HyperLynx软件与IBIS模型的接口较好[10],不需要格式转换即可直接使用[11],并且集成有IBIS模型编辑工具IBIS Editor 3.2。仿真结果的分析可通过测试工具直接测出信号的峰峰值、过冲/下冲的最大幅值以及信号的上升/下降时间等参数,也可实现标准模式下的仿真,或信号眼图的仿真[12]等,因此本文将选用这种软件。

2 仿真实例

2.1 仿真对象

当FPGA的传输速率达到Gb/s时,其数字信号的有效频谱已经扩展至毫米波频段,会在通信

  • 0
    点赞
  • 9
    收藏
    觉得还不错? 一键收藏
  • 0
    评论
Publisher: Prentice Hall PTR Pub Date: September 15, 2003 ISBN: 0-13-066946-6 Pages: 608 Section 2.7. The Spectrum of an Ideal Square Wave Section 2.8. From the Frequency Domain to the Time Domain Section 2.9. Effect of Bandwidth on Rise Time Section 2.10. Bandwidth and Rise Time Section 2.11. What Does "Significant" Mean? Section 2.12. Bandwidth of Real Signals Section 2.13. Bandwidth and Clock Frequency Section 2.14. Bandwidth of a Measurement Section 2.15. Bandwidth of a Model Section 2.16. Bandwidth of an Interconnect Section 2.17. Bottom Line Chapter 3. Impedance and Electrical Models Section 3.1. Describing Signal-Integrity Solutions in Terms of Impedance Section 3.2. What Is Impedance? Section 3.3. Real vs. Ideal Circuit Elements Section 3.4. Impedance of an Ideal Resistor in the Time Domain Section 3.5. Impedance of an Ideal Capacitor in the Time Domain Section 3.6. Impedance of an Ideal Inductor in the Time Domain Section 3.7. Impedance in the Frequency Domain Section 3.8. Equivalent Electrical Circuit Models Section 3.9. Circuit Theory and SPICE Section 3.10. Introduction to Modeling Section 3.11. The Bottom Line Chapter 4. The Physical Basis of Resistance Section 4.1. Translating Physical Design into Electrical Performance Section 4.2. The Only Good Approximation for the Resistance of Interconnects Section 4.3. Bulk Resistivity Section 4.4. Resistance per Length Section 4.5. Sheet Resistance Section 4.6. The Bottom Line Chapter 5. The Physical Basis of Capacitance Section 5.1. Current Flow in Capacitors Section 5.2. The Capacitance of a Sphere Section 5.3. Parallel Plate Approximation Section 5.4. Dielectric Constant Section 5.5. Power and Ground Planes and Decoupling Capacitance Section 5.6. Capacitance per Length Section 5.7. 2D Field Solvers Section 5.8. Effective Dielectric Constant Section 5.9. The Bottom Line Chapter 6. The Physical Basis of Inductance Section 6.1. What Is Inductance? Section 6.2. Inductance Principle #1: There Are Circular Magnetic-Field Line Loops Around All Currents Section 6.3. Inductance Principle #2: Inductance Is the Number of Webers of Field Line Loops Around a Conductor per Amp of Current Through It Section 6.4. Self-Inductance and Mutual Inductance Section 6.5. Inductance Principle #3: When the Number of Field Line Loops Around a Conductor Changes, There Will Be a Voltage Induced Across the Ends of the Conductor Section 6.6. Partial Inductance Section 6.7. Effective, Total, or Net Inductance and Ground Bounce Section 6.8. Loop Self- and Mutual Inductance Section 6.9. The Power-Distribution System (PDS) and Loop Inductance Section 6.10. Loop Inductance per Square of Planes Section 6.11. Loop Inductance of Planes and Via Contacts Section 6.12. Loop Inductance of Planes with a Field of Clearance Holes Section 6.13. Loop Mutual Inductance Section 6.14. Equivalent Inductance Section 6.15. Summary of Inductance Section 6.16. Current Distributions and Skin Depth Section 6.17. High-Permeability Materials Section 6.18. Eddy Currents Section 6.19. The Bottom Line Chapter 7. The Physical Basis of Transmission Lines Section 7.1. Forget the Word Ground Section 7.2. The Signal Section 7.3. Uniform Transmission Lines Section 7.4. The Speed of Electrons in Copper Section 7.5. The Speed of a Signal in a Transmission Line Section 7.6. Spatial Extent of the Leading Edge Section 7.7. "Be the Signal" Section 7.8. The Instantaneous Impedance of a Transmission Line Section 7.9. Characteristic Impedance and Controlled Impedance Section 7.10. Famous Characteristic Impedances Section 7.11. The Impedance of a Transmission Line Section 7.12. Driving a Transmission Line Section 7.13. Return Paths Section 7.14. When Return Paths Switch Reference Planes Section 7.15. A First-Order Model of a Transmission Line Section 7.16. Calculating Characteristic Impedance with Approximations Section 7.17. Calculating the Characteristic Impedance with a 2D Field Solver Section 7.18. An n-Section Lumped Circuit Model Section 7.19. Frequency Variation of the Characteristic Impedance Section 7.20. The Bottom Line Chapter 8. Transmission Lines and Reflections Section 8.1. Reflections at Impedance Changes Section 8.2. Why Are There Reflections? Section 8.3. Reflections from Resistive Loads Section 8.4. Source Impedance Section 8.5. Bounce Diagrams Section 8.6. Simulating Reflected Waveforms Section 8.7. Measuring Reflections with a TDR Section 8.8. Transmission Lines and Unintentional Discontinuities Section 8.9. When to Terminate Section 8.10. The Most Common Termination Strategy for Point-to-Point Topology Section 8.11. Reflections from Short Series Transmission Lines Section 8.12. Reflections from Short-Stub Transmission Lines Section 8.13. Reflections from Capacitive End Terminations Section 8.14. Reflections from Capacitive Loads in the Middle of a Trace Section 8.15. Capacitive Delay Adders Section 8.16. Effects of Corners and Vias Section 8.17. Loaded Lines Section 8.18. Reflections from Inductive Discontinuities Section 8.19. Compensation Section 8.20. The Bottom Line Chapter 9. Lossy Lines, Rise-Time Degradation, and Material Properties Section 9.1. Why Worry About Lossy Lines Section 9.2. Losses in Transmission Lines Section 9.3. Sources of Loss: Conductor Resistance and Skin Depth Section 9.4. Sources of Loss: The Dielectric Section 9.5. Dissipation Factor Section 9.6. The Real Meaning of Dissipation Factor Section 9.7. Modeling Lossy Transmission Lines Section 9.8. Characteristic Impedance of a Lossy Transmission Line Section 9.9. Signal Velocity in a Lossy Transmission Line Section 9.10. Attenuation and the dB Section 9.11. Attenuation in Lossy Lines Section 9.12. Measured Properties of a Lossy Line in the Frequency Domain Section 9.13. The Bandwidth of an Interconnect Section 9.14. Time-Domain Behavior of Lossy Lines Section 9.15. Improving the Eye Diagram of a Transmission Line Section 9.16. Pre-emphasis and Equalization Section 9.17. The Bottom Line Chapter 10. Cross Talk in Transmission Lines Section 10.1. Superposition Section 10.2. Origin of Coupling: Capacitance and Inductance Section 10.3. Cross Talk in Transmission Lines: NEXT and FEXT Section 10.4. Describing Cross Talk Section 10.5. The SPICE Capacitance Matrix Section 10.6. The Maxwell Capacitance Matrix and 2D Field Solvers Section 10.7. The Inductance Matrix Section 10.8. Cross Talk in Uniform Transmission Lines and Saturation Length Section 10.9. Capacitively Coupled Currents Section 10.10. Inductively Coupled Currents Section 10.11. Near-End Cross Talk Section 10.12. Far-End Cross Talk Section 10.13. Decreasing Far-End Cross Talk Section 10.14. Simulating Cross Talk Section 10.15. Guard Traces Section 10.16. Cross Talk and Dielectric Constant Section 10.17. Cross Talk and Timing Section 10.18. Switching Noise Section 10.19. Summary of Reducing Cross Talk Section 10.20. The Bottom Line Chapter 11. Differential Pairs and Differential Impedance Section 11.1. Differential Signaling Section 11.2. A Differential Pair Section 11.3. Differential Impedance with No Coupling Section 11.4. The Impact from Coupling Section 11.5. Calculating Differential Impedance Section 11.6. The Return-Current Distribution in a Differential Pair Section 11.7. Odd and Even Modes Section 11.8. Differential Impedance and Odd-Mode Impedance Section 11.9. Common Impedance and Even-Mode Impedance Section 11.10. Differential and Common Signals and Odd- and Even-Mode Voltage Components Section 11.11. Velocity of Each Mode and Far-End Cross Talk Section 11.12. Ideal Coupled Transmission-Line Model or an Ideal Differential Pair Section 11.13. Measuring Even- and Odd-Mode Impedance Section 11.14. Terminating Differential and Common Signals Section 11.15. Conversion of Differential to Common Signals Section 11.16. EMI and Common Signals Section 11.17. Cross Talk in Differential Pairs Section 11.18. Crossing a Gap in the Return Path Section 11.19. To Tightly Couple or Not to Tightly Couple Section 11.20. Calculating Odd and Even Modes from Capacitance- and Inductance-Matrix Elements Section 11.21. The Characteristic Impedance Matrix Section 11.22. The Bottom Line Appendix A. 100 General Design Guidelines to Minimize Signal-Integrity Problems Section A.1. Minimize Signal-Quality Problems on One Net Section A.2. Minimize Cross Talk Section A.3. Minimize Rail Collapse Section A.4. Minimize EMI Appendix B. 100 Collected Rules of Thumb to Help Estimate Signal-Integrity Effects Section B.1. Chapter 2 Section B.2. Chapter 3 Section B.3. Chapter 4 Section B.4. Chapter 5 Section B.5. Chapter 6 Section B.6. Chapter 7 Section B.7. Chapter 8 Section B.8. Chapter 9 Section B.9. Chapter 10 Section B.10. Chapter 11 Appendix C. Selected References About the Author
评论
添加红包

请填写红包祝福语或标题

红包个数最小为10个

红包金额最低5元

当前余额3.43前往充值 >
需支付:10.00
成就一亿技术人!
领取后你会自动成为博主和红包主的粉丝 规则
hope_wisdom
发出的红包
实付
使用余额支付
点击重新获取
扫码支付
钱包余额 0

抵扣说明:

1.余额是钱包充值的虚拟货币,按照1:1的比例进行支付金额的抵扣。
2.余额无法直接购买下载,可以购买VIP、付费专栏及课程。

余额充值