AN 835: PAM4 Signaling Fundamentals

Download PDF
ID 683852
Date 3/12/2019
Public
Document Table of Contents
Document Table of Contents x
1. Introduction 2. CEI-56G-MR Transmitter 3. CEI-56G-MR-PAM4 Interface Details 4. CEI-56G-MR-PAM4 Receiver 5. PAM4 Link Case Study 6. Medium Reach (MR) PAM4 System Design Study 7. Glossary and Acronyms 8. References 9. Document Revision History for AN 835: PAM4 Signaling Fundamentals
1. Introduction x
1.1. NRZ Fundamentals 1.2. Standards Using PAM4 Coding Scheme 1.3. CEI-56G Interconnect Reaches and Application Distances
1.3. CEI-56G Interconnect Reaches and Application Distances x
1.3.1. VSR (Very Short Reach) Chip-to-Module 1.3.2. MR (Mid-Range Reach) - Chip to Chip Within a PCBA 1.3.3. LR (Long Reach) - Chip to Chip Across a Backplane/Midplane or a Cable
2. CEI-56G-MR Transmitter x
2.1. Naming Conventions 2.2. Eye Height (EH6) and Eye Width (EW6) 2.3. Eye Linearity 2.4. Electrical Characteristics 2.5. PAM4 Jitter Methodology 2.6. TX Pre-Emphasis Method
2.4. Electrical Characteristics x
2.4.1. Transmitter Characteristics 2.4.2. Transmitter Return Loss 2.4.3. Transmitter Linearity (RLM) 2.4.4. Signal-to-Noise-and-Distortion Ratio (SNDR)
3. CEI-56G-MR-PAM4 Interface Details x
3.1. COM Introduction 3.2. Normative Channel Specification COM 3.3. Informative Channel Insertion Loss 3.4. Informative Channel Return Loss
3.1. COM Introduction x
3.1.1. Backplane Measurements
4. CEI-56G-MR-PAM4 Receiver x
4.1. Challenges in Analyzing PAM4 Signals 4.2. PAM4 Receiver Architecture 4.3. Equalization Technique 4.4. CEI-56G-MR-PAM4 Receiver Details
4.2. PAM4 Receiver Architecture x
4.2.1. Analog vs. Digital Receiver 4.2.2. Slicer 4.2.3. Clock and Data Recovery (CDR)
4.4. CEI-56G-MR-PAM4 Receiver Details x
4.4.1. Electrical Characteristics 4.4.2. Receiver Input Return Loss 4.4.3. Receiver Interference Tolerance 4.4.4. Receiver Jitter Tolerance
5. PAM4 Link Case Study x
5.1. OIF_Stressed 5.2. OIF_Compliant
5.1. OIF_Stressed x
5.1.1. Channel Characteristics 5.1.2. OIF_Stressed PAM4 Link Simulation with the Advanced Link Analyzer
5.2. OIF_Compliant x
5.2.1. Channel Characteristics 5.2.2. OIF_Compliant PAM4 Link Simulation with the Advanced Link Analyzer
6. Medium Reach (MR) PAM4 System Design Study x
6.1. System Power 6.2. System Cost 6.3. Board Space 6.4. Conclusion
1. Introduction
2. CEI-56G-MR Transmitter
3. CEI-56G-MR-PAM4 Interface Details
4. CEI-56G-MR-PAM4 Receiver
5. PAM4 Link Case Study
6. Medium Reach (MR) PAM4 System Design Study
7. Glossary and Acronyms
8. References
9. Document Revision History for AN 835: PAM4 Signaling Fundamentals

2.4.3. Transmitter Linearity (RLM)

The Level Separation Mismatch Ratio, commonly referred to as RLM, is a measurement that is not required in normative or informative VSR tests, but is required by most other variants.

RLM is conceptually similar to eye linearity but measured differently. An ideal PAM4 eye has RLM equal to 1, but it does not scale in the same way as eye linearity. The four voltage levels of PAM4 are V0, V1, V2 and V3 respectively. The mid-range level Vmid is defined in Equation 6. The mean signal levels are normalized and offset adjusted so that Vmin corresponds to 0, V0 to -1, V1 to -ES1, V2 to ES2 and V3 to 1.

The mean signal levels described above are measured from a waveform captured while the transmitter is transmitting the QPRBS13-CEI test pattern. The waveform consists of M samples per unit interval and is aligned such that the first M samples of the waveform correspond to the first PAM4 symbol of the test pattern, the second M samples to the second PAM4 symbol, and so on. This allows each sample of the waveform to be associated with a specific PAM4 symbol in the test pattern.

Equation 6. Vmin Calculation
Equation 7. ES1 Calculation
Equation 8. ES2 Calculation
Equation 9. RLM Calculation

For an ideal eye, three eye heights are the same. The distance of V1 to Vmin is one third of the distance of V0 to Vmin. In the same manner, the distance of V2 to Vmin is one third of the distance of V3 to Vmin. Using Equation 9, RLM equals to 1. The second figure shows an eye that does not have very good eye linearity. V1 moved lower to a point that the distance of V1 to Vmin is two thirds of V0 to Vmin. V2 moved lower to a point that the distance of V2 to Vmin is one quarter of the distance of V3 to Vmin.

When using Equation 9, the calculated RLM is equal to 0. For the MR standard, the minimal acceptable linearity is 0.95, which roughly translates to a 5% error on the linearity of the eye. The closer RLM is to 1, the better the eye linearity is.

Figure 15. RLM
Level Two Title