Comparison of NRZ, PR-2, and PR-4 signaling Presented by: Rob Brink Contributors: Pervez Aziz Qasim Chaudry Adam Healey Greg Sheets
Scope and Purpose Operation over electrical backplanes at 10.3125Gb/s is investigated using NRZ, PR2, and PR4 signaling. A common equalizer architecture is used in all cases. Estimated BER, as well as voltage and timing margin at 1E-12, is reported. 2 March 15, 2005 (r1.0)
Agenda Simulator Overview Link Model Transmitter Model Receiver Model Equalization Strategy Test Cases Sample Results Results Summary 3 March 15, 2005 (r1.0)
Simulator Overview
Link Model TP1 TP4 50 50 2 1 1 2 1 2 50 TX Package Backplane RX Package 50 4 3 3 4 3 4 TX Package Model Mellitz Capacitor-Like Model Backplane Model As described in Test Cases section RX Package Model Mellitz Capacitor-Like Model 5 March 15, 2005 (r1.0)
Crosstalk (1/2) For each crosstalk aggressor The response to a PRBS-15 pattern (with an additional trailing 0 ) is computed. This response is sampled at baud-spaced intervals at 16 offsets from 0 to (15/16)T in T/16 steps. At each offset, the amplitude distribution of sampled response is computed. The aggressor amplitude distribution is defined as the average of the amplitude distributions computed at each sample offset. 6 March 15, 2005 (r1.0)
Crosstalk (2/2) The overall crosstalk distribution is defined as the convolution of the individual aggressor distributions. The effect of crosstalk on the eye is modeled as the convolution of the overall crosstalk distribution and amplitude distribution of the thru path at each sample phase. This methodology is has been previously described by Moore: http://ieee802.org/3/ap/public/channel_adhoc/moore_c1_0305.pdf Computed RMS and peak-peak crosstalk will be reported. 7 March 15, 2005 (r1.0)
Transmitter Model Transmitter differential output voltage fixed at 800mV p-p. Transmit filter is Gaussian. Rise Time (20-80%): 24 ps Transmitter output jitter Duty Cycle Distortion: Deterministic Jitter: Random Jitter: 0.05 UI p-p (even-odd) 0.10 UI p-p (sinusoidal) 0.15 UI p-p (at 1E-15), 9.4mUI rms Parameters defined at package model input and do not include package parasitics. The impact of the package model is investigated in the next slide. 8 March 15, 2005 (r1.0)
Transmitter Output at TP1 800 mv p-p 800 mv p-p 50 50 0.70 UI p-p at 1E-15 2 TX Package 1 TP1 50 50 4 3 9 March 15, 2005 (r1.0)
Receiver Model Receiver modeled as a single pole at 75% fbaud. Noise Bandwidth (B n ): 11.4 GHz Noise Figure: 18 db sqrt( 4kTRB n ): 1.08 mv rms Receiver jitter: = 2 B n H f ) 0 ( df Random Jitter: 0.15 UI p-p (at 1E-15), 9.4 mui rms No gain stages have been included in the receive path. 10 March 15, 2005 (r1.0)
Equalization Strategy Transmitter Finite Impulse Response filter. 3 taps, T-spaced with infinite tap weight resolution. Receiver Decision Feedback Equalizer 5 taps, unconstrained with infinite tap weight resolution. Sequential adaptation Transmit FIR is adapted first, then the DFE. Sample phase chosen to minimize mean-squared error. T/32 resolution 11 March 15, 2005 (r1.0)
Test Cases
Test Patterns Equalizer trained with PN-11 pattern with a trailing 0. Equalizer settings are then frozen. Voltage and timing margin is estimated based on PN-15 pattern with a trailing 0. Thru, NEXT, and FEXT channels share the same output amplitude (800 mv ppd ) and transmit FIR settings. Decision threshold set at the mid-point between nominal signal levels, as determined by the 1E-4 contour. Reported margins are twice the minimum distance from the sample phase (or threshold) to the BER contour of interest. 13 March 15, 2005 (r1.0)
Channels Studied All Tyco Electronics channels Test Cases 1 through 7 Note that the FEXT channels were included twice due to connector symmetry. Recommended subset of Intel channels Test Cases 8, 11, and 12 map to T1, T12, T20 Test Cases 14, 17, and 18 map to B1, B12, B20 Test Cases 20 and 24 map to M1, and M20 Other channels not simulated due to a lack of time 14 March 15, 2005 (r1.0)
Sample Results
Disclaimer The following results are based on PRBS-11 pattern and are included for illustrative purposes. 16 March 15, 2005 (r1.0)
NRZ Sample Results: Eye at Slicer Input Test Case #6 Test Case #1 17 March 15, 2005 (r1.0)
PR2 Sample Results: Eye at Slicer Input Test Case #6 Test Case #1 18 March 15, 2005 (r1.0)
PR4 Sample Results: Eye at Slicer Input Test Case #6 Test Case #1 19 March 15, 2005 (r1.0)
Results Summary
BER Estimates BER 1.0E+00 1.0E-02 1.0E-04 1.0E-06 1.0E-08 1.0E-10 1.0E-12 1.0E-14 1.0E-16 1.0E-18 1.0E-20 1.0E-22 1.0E-24 1.0E-26 1.0E-28 1.0E-30 1.0E-32 1.0E-34 1.0E-36 0 5 10 15 20 25 Test Case NRZ PR2 PR4 21 March 15, 2005 (r1.0)
Voltage and Timing Margin at 1E-12 Vertical Margin at 1E-12 Horizontal Margin at 1E-12 0.050 0.300 0.045 0.040 0.250 Margin (V p-p ) 0.035 0.030 0.025 0.020 0.015 Margin (UI p-p ) 0.200 NRZ 0.150 PR2 0.100 NRZ PR2 0.010 0.005 0.050 0.000 0 5 10 15 20 25 0.000 0 5 10 15 20 25 Test Case Test Case 22 March 15, 2005 (r1.0)
Crosstalk Environment RMS and Peak-Peak (at 1E-12) Crosstalk 0.1400 0.0120 0.1200 0.0100 Peak-Peak Crosstalk (V) 0.1000 0.0800 0.0600 0.0400 0.0080 0.0060 0.0040 RMS Crosstalk (V) NRZ (p-p) PR2 (p-p) PR4 (p-p) NRZ (rms) PR2 (rms) PR4 (rms) 0.0200 0.0020 0.0000 0 5 10 15 20 25 0.0000 Test Case 23 March 15, 2005 (r1.0)
Required FFE Boost FFE Gain at (Symbol Rate)/2 30.0 25.0 20.0 Gain (db) 15.0 10.0 NRZ PR2 PR4 5.0 0.0 0 5 10 15 20 25 Test Case 24 March 15, 2005 (r1.0)
Conclusions (1/2) The target response for PR4 is a poor fit to the channel and therefore higher equalizer complexity is required to achieve acceptable performance. NRZ and PR2 both support 1E-12 operation over the Tyco channels. In general, PR2 requires considerably less boost to achieve this objective. In the majority of test cases studied, NRZ offered superior voltage and timing margin. 25 March 15, 2005 (r1.0)
Conclusions (2/2) The Intel T channels were not supported by any of the signaling schemes studied with the chosen equalizer architecture. NRZ signaling may be feasible for select Intel B and M channels. Crosstalk is a significant impairment on these channels. 26 March 15, 2005 (r1.0)
Backup
Relationship Between Crosstalk and Boost RMS Crosstalk vs. FFE Boost 0.0120 0.0100 RMS Crosstalk (V) 0.0080 0.0060 0.0040 NRZ (rms) PR2 (rms) PR4 (rms) 0.0020 0.0000 0.0 5.0 10.0 15.0 20.0 25.0 FFE Gain at (Sym bol Rate)/2 28 March 15, 2005 (r1.0)