OIF 400ZR 01.0 - Reduced2

OIF-400ZR-01.
Implementation Agreement 400ZR
OIF-400ZR-01.0
March 10, 2020
Implementation Agreement created and approved

OIF
www.oiforum.com
www.oiforum.com 1
OIF-400ZR-01.0
The OIF is an international non-profit organization with over 100 member companies, including the world’s
leading carriers and vendors. Being an industry group uniting representatives of the data and optical worlds,
OIF’s purpose is to accelerate the deployment of interoperable, cost-effective and robust optical internetworks
and their associated technologies. Optical internetworks are data networks composed of routers and data
switches interconnected by optical networking elements.
With the goal of promoting worldwide compatibility of optical internetworking products, the OIF actively
supports and extends the work of national and international standards bodies. Working relationships or formal
liaisons have been established with CFP-MSA, COBO, EA, ETSI NFV, IEEE 802.3, IETF, INCITS T11, ITU SG-15, MEF,
ONF.
For additional information contact:

OIF
5177 Brandin Ct, Fremont, CA 94538
510-492-4040  [email protected]
www.oiforum.com
www.oiforum.com 2
OIF-400ZR-01.0
Working Group: Physical and Link Layer (PLL) Working Group
TITLE: Implementation Agreement 400ZR
SOURCE: TECHNICAL EDITOR WORKING GROUP CHAIR

OPTICAL CHAIR
Mike A. Sluyski David R. Stauffer, Ph D.

Acacia Communications Inc. Kandou Bus, SA
3 Mill and Main QI-I
Maynard, MA 01754, USA 1015 Lausanne, Switzerland
Phone: +1.978.938-4896 x2773 Phone: +1.802.316.0808
Email: [email protected] Email: [email protected]
WORKING GROUP CHAIR

OPTICAL VICE CHAIR
Karl Gass
Phone: +1.505.301.1511
Email: [email protected]
ABSTRACT: Implementation Agreement created and approved by the Optical Internetworking Forum for a 400ZR
Coherent Optical interface. The project start was approved at the Q3 Technical Meeting, October 2016 (San Jose
CA, USA). OIF2016.400.04 is the original project start document for this project
www.oiforum.com 3
OIF-400ZR-01.0
Notice: This Technical Document has been created by the Optical Internetworking Forum (OIF). This document is offered to the OIF
Membership solely as a basis for agreement and is not a binding proposal on the companies listed as resources above. The OIF reserves the
rights to at any time to add, amend, or withdraw statements contained herein. Nothing in this document is in any way binding on the OIF or any
of its members.
The user's attention is called to the possibility that implementation of the OIF implementation agreement contained herein may require the use
of inventions covered by the patent rights held by third parties. By publication of this OIF implementation agreement, the OIF makes no
representation or warranty whatsoever, whether expressed or implied, that implementation of the specification will not infringe any third party
rights, nor does the OIF make any representation or warranty whatsoever, whether expressed or implied, with respect to any claim that has
been or may be asserted by any third party, the validity of any patent rights related to any such claim, or the extent to which a license to use
any such rights may or may not be available or the terms hereof.
Copyright © 2020 Optical Internetworking Forum
This document and translations of it may be copied and furnished to others, and derivative works that comment on or otherwise explain it or
assist in its implementation may be prepared, copied, published and distributed, in whole or in part, without restriction other than the
following, (1) the above copyright notice and this paragraph must be included on all such copies and derivative works, and (2) this document
itself may not be modified in any way, such as by removing the copyright notice or references to the OIF, except as needed for the purpose of
developing OIF Implementation Agreements.
By downloading, copying, or using this document in any manner, the user consents to the terms and conditions of this notice. Unless the terms
and conditions of this notice are breached by the user, the limited permissions granted above are perpetual and will not be revoked by the OIF
or its successors or assigns.
This document and the information contained herein is provided on an “AS IS” basis and THE OIF DISCLAIMS ALL WARRANTIES, EXPRESS OR
IMPLIED, INCLUDING BUT NOT LIMITED TO ANY WARRANTY THAT THE USE OF THE INFORMATION HEREIN WILL NOT INFRINGE ANY RIGHTS OR
ANY IMPLIED WARRANTIES OF MERCHANTABILITY, TITLE OR FITNESS FOR A PARTICULAR PURPOSE.
www.oiforum.com 4
OIF-400ZR-01.0
1 Table of Contents
1 TABLE OF CONTENTS ........................................................................................... 5
2 LIST OF FIGURES ................................................................................................... 9
3 LIST OF TABLES................................................................................................... 11
4 DOCUMENT REVISION HISTORY ........................................................................ 13
5 INTRODUCTION .................................................................................................... 14
6 400ZR INTERFACES ............................................................................................. 15
6.1 400ZR Clocking Modes ................................................................................................................ 15
6.2 Media Interface - Black Link ........................................................................................................ 15
7 400ZR USE CASES ............................................................................................... 16
7.1 120 km or less, amplified, point-to-point, DWDM noise limited link ......................................... 16
7.2 Unamplified, single wavelength, loss limited link ....................................................................... 17
8 HOST TO 400ZR DATA PATH .............................................................................. 18
8.1 400G Host Side Interface ............................................................................................................ 19
8.2 PMA ............................................................................................................................................. 19
8.3 PCS (partial processes) ................................................................................................................ 19
8.3.1 PCS Rx direction (400ZR Tx datapath) ................................................................................. 19
8.3.2 PCS Tx direction (400ZR Rx datapath) ................................................................................. 20
8.4 400ZR frame structure ................................................................................................................ 20
8.4.1 400ZR Multi-Frame ............................................................................................................. 21
8.5 AM/PAD/OH insertion ................................................................................................................ 22
8.5.1 400ZR AM/PAD/OH Transmission order ............................................................................. 23
8.6 400ZR Alignment Markers (AM) ................................................................................................. 24
8.7 400ZR PAD................................................................................................................................... 25
8.8 400ZR OH .................................................................................................................................... 26
8.8.1 Multi-Frame Alignment Signal (MFAS)................................................................................ 26
8.8.2 Link error and Link degrade detection and marking ........................................................... 26
8.8.3 Link status monitoring and signaling (STAT) ....................................................................... 27
8.8.4 Link Degrade Indication (LDI) .............................................................................................. 28
8.8.5 Link Degrade Warning and Alarming. ................................................................................. 28
8.9 GMP mapping processes............................................................................................................. 30
8.9.1 Stuffing Locations................................................................................................................ 34
www.oiforum.com 5
OIF-400ZR-01.0
8.9.2 GMP overhead Encoding..................................................................................................... 36

8.9.3 GMP OH – CRC8 calculation ................................................................................................ 36
8.9.4 The JC6 OH CRC4 Calculation .............................................................................................. 36
9 400ZR FRAME TO SC ADAPTATION ................................................................... 37
9.1 Mapping 400ZR Frame Payload to Staircase FEC Blocks ............................................................ 38
9.2 400ZR CRC+MBAS Bit Insertion Block ......................................................................................... 38
9.2.1 400ZR Multi Block Alignment Signal (MBAS) ...................................................................... 39
10 400ZR FORWARD ERROR CORRECTION (FEC)............................................. 42
10.1 SC-FEC ......................................................................................................................................... 42
10.2 Sync Pad Insertion ....................................................................................................................... 42
10.3 Frame Synchronous Scrambling .................................................................................................. 44
10.4 Convolutional Interleave ............................................................................................................. 44
10.5 Inner Hamming Code .................................................................................................................. 46
11 DP-16QAM SYMBOL MAPPING AND POLARIZATION DISTRIBUTION ......... 48
11.1 Interleaving DP-16QAM Symbols ................................................................................................ 48
12 DSP FRAMING ................................................................................................... 50
12.1 First DSP sub-Frame .................................................................................................................... 50
12.1.1 FAW Sequence .................................................................................................................... 51
12.2 Subsequent DSP sub-frames. ...................................................................................................... 51
12.2.1 Training Sequence ............................................................................................................... 52
12.3 Pilot Sequence............................................................................................................................. 52
12.4 Channel Mappings ...................................................................................................................... 54
12.5 Frame Expansion Rate................................................................................................................. 55
13 OPTICAL SPECIFICATIONS.............................................................................. 57
13.1 400ZR, DWDM amplified - Application Code (0x01): .................................................................. 57
13.1.1 Optical channel specifications – Black Link ......................................................................... 57
13.1.2 Transmitter Optical Specifications ...................................................................................... 59
13.1.3 Receiver Optical Specifications ........................................................................................... 66
13.1.4 Module Requirements Tx - (Informative) ........................................................................... 68
13.1.5 Module Requirements Rx - (Informative) ........................................................................... 69
13.2 400ZR, Single wavelength, Unamplified - Application Code (0x02):........................................... 70
13.2.1 Optical channel specifications ............................................................................................ 70
www.oiforum.com 6
OIF-400ZR-01.0
13.2.2 Transmitter Optical Specifications ...................................................................................... 71

13.2.3 Receiver Optical Specifications ........................................................................................... 77
13.2.4 Module Requirements Tx - (Informative) ........................................................................... 78
13.2.5 Module Requirements Rx - (Informative) ........................................................................... 79
13.3 Optical Parameter Definitions .................................................................................................... 80
13.3.1 The Receiver Optical Signal-to-noise Ratio Tolerance ........................................................ 80
13.3.2 Spectral excursion ............................................................................................................... 80
13.3.3 Out-of-Band OSNR (OOB OSNR) ......................................................................................... 80
13.3.4 Ripple .................................................................................................................................. 80
13.3.5 Optical return loss at SS ....................................................................................................... 80
13.3.6 Discrete reflectance between SS and RS .............................................................................. 81
13.3.7 Differential Group Delay (DGD) .......................................................................................... 81
13.3.8 Polarization Dependent Loss (PDL) ..................................................................................... 81
13.3.9 Polarization rotation speed................................................................................................. 81
13.3.10 Inter-channel crosstalk.................................................................................................... 81
13.3.11 Interferometric crosstalk ................................................................................................ 81
14 INTEROPERABILITY TEST METHODOLOGY, DEFINITIONS ......................... 82
14.1 400ZR Test Features .................................................................................................................... 82
14.2 Loopback features, Test Generators and Checkers .................................................................... 82
14.2.1 Loopbacks ........................................................................................................................... 84
14.2.2 Test Generators/Checkers .................................................................................................. 85
14.3 Interoperability Test Vectors ...................................................................................................... 85
14.3.1 EVM PRBS ............................................................................................................................ 85
14.3.2 TV PRBS ............................................................................................................................... 86
14.3.3 GMP PCS test vectors .......................................................................................................... 86
14.3.1 PCS test vectors................................................................................................................... 87
14.3.1 Media loop testing .............................................................................................................. 88
15 OPERATING FREQUENCY CHANNEL DEFINITIONS...................................... 89
15.1 Normative 48 x 100 GHz DWDM Application Channels.............................................................. 89
15.2 Optional 64 x 75 GHz DWDM Application Channels ................................................................... 89
15.3 Optional Flexible DWDM Grid..................................................................................................... 90
15.3.1 Example 100GHz Flexible Grid Offset ................................................................................. 91
www.oiforum.com 7
OIF-400ZR-01.0
15.3.2 Example 75GHz Flexible Grid Offset ................................................................................... 91

15.3.3 Flexible Grid Provisioning.................................................................................................... 91
16 SUMMARY .......................................................................................................... 92
17 REFERENCES .................................................................................................... 93
17.1 Normative references ................................................................................................................. 93
17.2 Informative references ............................................................................................................... 93
18 APPENDIX A: GLOSSARY ................................................................................ 94
19 APPENDIX B: FUTURE WORK ITEMS.............................................................. 95
20 APPENDIX C: ERROR VECTOR MAGNITUDE ................................................. 96
20.1 Maximum error vector magnitude ............................................................................................. 96
20.2 Maximum I-Q DC offset .............................................................................................................. 96
20.3 Reference receiver for EVM and I-Q DC offset ........................................................................... 96
20.3.1 Hardware characteristics: ................................................................................................... 96
20.3.2 Processing steps: ................................................................................................................. 96
20.4 EVM evaluation ........................................................................................................................... 97
20.5 Reference Algorithms for EVM Test of 400ZR transmitters. ...................................................... 98
21 APPENDIX D: LIST OF COMPANIES BELONGING TO OIF WHEN DOCUMENT
IS APPROVED .............................................................................................................. 99
www.oiforum.com 8
OIF-400ZR-01.0
2 List of Figures
Figure 1: 400ZR reference diagram............................................................................................................. 14
Figure 2: Transceiver line card with 400ZR amplified point-to-point interface.......................................... 16
Figure 3: Router switch line card with 400ZR DWDM Interfaces ............................................................... 16
Figure 4: Transceiver line card with 400ZR DWDM interfaces ................................................................... 16
Figure 5: Router/Switch line card with 400ZR unamplified point-to-point Interface................................. 17
Figure 6: Data path detail ........................................................................................................................... 18
Figure 7: 400ZR frame structure without parity bits .................................................................................. 20
Figure 8: 400ZR multi-frame structure with parity bits .............................................................................. 21
Figure 9: 400ZR Frame overhead ................................................................................................................ 22
Figure 10: 400ZR overhead ......................................................................................................................... 23
Figure 11: Alignment Marker transmission order – 10b interleaved ......................................................... 24
Figure 12: Alignment Marker format .......................................................................................................... 24
Figure 13: PAD transmission order – 10b interleaved ................................................................................ 25
Figure 14: Over Head transmission order – 10b interleaved ...................................................................... 26
Figure 15: STAT Over Head byte definitions ............................................................................................... 27
Figure 16: Local/Remote Degrade interworking between Switch/Router and 400ZR transceiver ............ 27
Figure 17: Error marking ............................................................................................................................. 30
Figure 18: GMP mapping/de-mapping process .......................................................................................... 31
Figure 19: GMP mapping over four 400ZR frames with Cm=10216 ............................................................ 35
Figure 20: 400ZR frame to SC-FEC relationship .......................................................................................... 37
Figure 21: CRC32 + MBAS ........................................................................................................................... 39
Figure 22: Multi-block alignment signal overhead ..................................................................................... 39
Figure 23: CRC32 + MBAS transmission order ............................................................................................ 40
Figure 24: 400ZR frame adaptation SC FEC block ....................................................................................... 41
Figure 25: 400ZR HD-FEC processes............................................................................................................ 42
Figure 26: Pad insertion/removal ............................................................................................................... 42
Figure 27: 6 × 119 Pad Insertion ................................................................................................................. 43
Figure 28: Frame synchronous scrambler ................................................................................................... 44
Figure 29: Convolution interleave............................................................................................................... 44
Figure 30: Convolution interleave............................................................................................................... 45
Figure 31: Hamming FEC frame format ...................................................................................................... 46
www.oiforum.com 9
OIF-400ZR-01.0
Figure 32: Hamming code ........................................................................................................................... 46

Figure 33: Hamming code 8-way interleave ............................................................................................... 49
Figure 34: First DSP sub-frame of super-frame .......................................................................................... 50
Figure 35: DSP sub-frames 2-49 of the DSP super-frame ........................................................................... 51
Figure 36: QPSK mapped Pilot Sequence .................................................................................................... 52
Figure 37: Pilot seed and sequence ............................................................................................................ 53
Figure 38: 400ZR expansion rates ............................................................................................................... 55
Figure 39: 400ZR test features .................................................................................................................... 83
Figure 40: Test vector PRBS31 generator ................................................................................................... 86
Figure 41: Test vector detail ....................................................................................................................... 88
Figure 42: Flexible grid provisioning example ............................................................................................ 90
www.oiforum.com 10
OIF-400ZR-01.0
3 List of Tables
Table 1: 400ZR IA document revision history ............................................................................................. 13
Table 2: 400ZR host interface ..................................................................................................................... 15
Table 3: 400ZR Alignment Marker encodings ............................................................................................. 25
Table 4: Replacement signal ...................................................................................................................... 26
Table 5: Host interface and its GMP parameter values .............................................................................. 31
Table 6: GMP parameter values ................................................................................................................. 33
Table 7: GMP stuff locations of 400ZR ....................................................................................................... 34
Table 8: In-phase (I) and quadrature phase (Q) symbol amplitude ............................................................ 48
Table 9: FAW sequence............................................................................................................................... 51
Table 10: Training symbol sequence........................................................................................................... 52
Table 11: Pilot polynomial and seed ........................................................................................................... 53
Table 12: Pilot Sequence ............................................................................................................................. 54
Table 13: Channel mappings ....................................................................................................................... 55
Table 14: 400ZR expansion rate table......................................................................................................... 56
Table 15: 400ZR application codes ............................................................................................................. 57
Table 16: Optical channel specifications..................................................................................................... 58
Table 17: Tx optical specifications .............................................................................................................. 65
Table 18: Rx optical specifications .............................................................................................................. 67
Table 19: 400ZR module – Tx specifications ............................................................................................... 68
Table 20: 400ZR module – Rx specifications ............................................................................................... 69
Table 21: Optical channel specifications..................................................................................................... 70
Table 22: Tx Optical specifications .............................................................................................................. 76
Table 23: Rx Optical specifications.............................................................................................................. 77
Table 24: 400ZR module – Tx specifications ............................................................................................... 78
Table 25: 400ZR module – Rx specifications ............................................................................................... 79
Table 26: Loopbacks.................................................................................................................................... 84
Table 27: Test generator/checker descriptions .......................................................................................... 85
Table 28: Test vector PRBS files .................................................................................................................. 86
Table 29: GMP PCS test vector files ............................................................................................................ 87
Table 30: 100GHz channel spacing ............................................................................................................. 89
Table 31: 75GHz channel spacing ............................................................................................................... 89
www.oiforum.com 11
OIF-400ZR-01.0
Table 32: Example 100GHz flexible grid...................................................................................................... 91

Table 33: Example 75GHz flexible grid........................................................................................................ 91
Table 34: Acronyms..................................................................................................................................... 94
Table 35: EVM algorithms ........................................................................................................................... 98
www.oiforum.com 12
OIF-400ZR-01.0
4 Document Revision History
Table 1 provides the 400ZR Implementation Agreement revision history.
Document Date Revisions/Comments

OIF-400ZR-01.0 March 10, 2020 Initial release
Table 1: 400ZR IA document revision history
www.oiforum.com 13
OIF-400ZR-01.0
5 Introduction
This Implementation Agreement (IA) specifies a Digital Coherent 400ZR interface for two applications:
• 120 km or less, amplified, point-to-point, DWDM noise limited links.
• Unamplified, single wavelength, loss limited links.
The IA aims to enable interoperable, cost-effective, 400Gb/s implementations based on single-carrier
coherent DP-16QAM modulation, low power DSP supporting absolute (Non-Differential) phase
encoding/decoding, and a Concatenated FEC (C-FEC) with a post-FEC error floor <1.0E-15. 400ZR operates
as a 400GBASE-R PHY.
Figure 1 shows the scope of this IA.
Timing and Code Word Transparent Transport of Ethernet signal, including FlexE and SyncE
Single Channel line Single Channel line

Ethernet Interface Ethernet Interface
interface with +/-20ppm interface with +/-20ppm
with +/-100ppm with +/-100ppm
400ZR Module 400ZR Module

Ss Rs
400ZR Black Link 400ZR
HOST Optics Optics HOST
DSP 80 km DSP
Rs Ss
1x400GAUI-8
400ZR link 1x400GAUI-8
Figure 1: 400ZR reference diagram

No restriction on the physical form factor is implied by this IA (QSFP-DD, OSFP, COBO, CFP2, CFP8), but
the specifications target a pluggable DCO architecture with port densities equivalent to grey client optics.
400ZR builds upon the work of other standards bodies including IEEE 802.3TM-2018 and ITU-T SG-15.
www.oiforum.com 14
OIF-400ZR-01.0
6 400ZR interfaces
The 400ZR IA supports the following host interface functions.
Host protocol support Sublayer Capabilities
FEC coding RS(544,514), lane distribution, AM lock

PCS
and deskew, per clause 119.1, Extender sublayer.
IEEE Std 802.3™-2018

Mux’ing, clock and data recovery, clock generation,
400GBASE-R PMA
modulation.
Optionally physically instantiated as 400GAUI-8 C2M;

AUI
8 x CEI-56G-VSR PAM-4.
Table 2: 400ZR host interface

This IA does NOT define support of other host interfaces, nor the aggregation of multiple host interfaces.
This IA, however, should not limit the ability to extend the host interfaces in the future.
6.1 400ZR Clocking Modes

The 400ZR data path is mapped asynchronously using a local clock reference. Simplified GMP mapping
per ITU-T G.709.1 Annex D is used to rate-adapt the payload to the local reference, supporting data and
timing transparency. The local clock tolerance is +/- 20ppm.
For timing transparent applications digital phase-interpolation is used to recover the timing information
from the GMP mapped Cm bytes.
6.2 Media Interface - Black Link
400ZR provides timing and codeword transparent transmission of a 400GBASE-R interface. 400ZR uses a
"black link" approach, to define the optical interface parameters for a (single-channel) optical tributary as
shown in Figure 1.
The black link may contain neighboring channels and optical amplifiers in the optical path. Black link
specifications are provided in Sections 13.1.1 and 13.2.1. The black link methodology enables longitudinal
mode compatibility at the single-channel points (Ss, Rs), however, it does not enable longitudinal mode
compatibility at multichannel points.
www.oiforum.com 15
OIF-400ZR-01.0
7 400ZR use cases

400ZR is intended for the use cases summarized here. The different 400ZR use cases can be addressed
with different 400ZR DCO module implementations.
7.1 120 km or less, amplified, point-to-point, DWDM noise limited link
There are 3 use cases of amplified point-to-point links (no OADM) identified for 400ZR in Figure 2 through
Figure 4. For amplified links the reach is dependent on the OSNR at the receiver (noise limited). The 400ZR
targeted reach for these applications is 80km or more. These use cases are covered by Application Code
0x01 in this IA.
Figure 2: Transceiver line card with 400ZR amplified point-to-point interface
Figure 3: Router switch line card with 400ZR DWDM Interfaces
Figure 4: Transceiver line card with 400ZR DWDM interfaces
www.oiforum.com 16
OIF-400ZR-01.0
7.2 Unamplified, single wavelength, loss limited link

For an unamplified link as shown in Figure 5, the reach is dependent on the transmit output power, input
receive sensitivity, and the channel’s loss characteristics. This use case is covered by Application Code
0x02 in this IA.
Figure 5: Router/Switch line card with 400ZR unamplified point-to-point Interface
www.oiforum.com 17
OIF-400ZR-01.0
8 Host to 400ZR data path

Figure 6 shows the functional blocks in the Tx and Rx data path.
400GBASE-R
8 8
PMA sublayer(*) (*)

per IEEE Std.
16 16 -
Alignment lock and lane
deskew
Lane reorder and de-

Distribution & Interleave
interleave
FEC Decode FEC Encode

RS-FEC
detected Local PHY 400GXS
Degrade
Post-FEC Interleave Pre-FEC Distribution
Rx Client rx_am_sf tx_am_sf Tx Client
FEC <2:0> <2:0> FEC
Alignment Removal Alignment Insertion
Degrade Degrade
signaling signaling
Descramble Scramble
256b/257b blocks 256b/257b blocks
256b/257b Transcode 400GAUI-8

Reverse Transcode
Decode & No rate No rate PMA
matching matching Encode
Error Marking PHY 400GXS
400GMII*
No rate No rate Decode & PCS

Encode
matching matching Error Marking PMA PHY 400ZR
256b/257b transcode Reverse transcode PMD
256b/257b blocks System-side CLOCK DOMAIN 256b/257b blocks MDI
(+/- 100 ppm)
GMP Mapping GMP de-mapping MEDIUM
(4*257b stuffing) Line-Side CLOCK DOMAIN (4*257b de-stuffing)
(+/- 20 ppm)
rx_oh_sf
OH/AM insertion AM/OH detect & removal <2:0> 400ZR
tx_oh_sf (20*257b) (20*257b)
<2:0> *Logical Equivalent Implementaion. The rate matching functions shall be
disabled to/from the 400GMII interface in both directions. No Idle deletion/
CRC-32 Calculation CRC-32 checking SC-FEC insertion shall be performed.
detected
Local
SC-FEC Adapt, ED & SC-FEC Adapt, ED & Degrade
encoding PCS (ZR) Decoding
Pad Insertion Pad removal

(6*119b) (6*119b)
Scramble Descramble
Convolution de-
Convolution interleaving
interleaving
128b/119b SD-FEC 128b/119b SD-FEC

Hamming encode Hamming decode
Gray mapping, symbol Pol combining & Symbol

Interleaving, Pol distribution de-interleaving
2 2
FAW & TS Insertion Align C-FEC &
FAW/TS removal
2
2
Pilot Insertion Pilot removal
2 2
TX DSP PMA (ZR) RX DSP
DAC ADC
PMD (ZR)
Photonic Interface Photonic Interface
Figure 6: Data path detail

www.oiforum.com 18
OIF-400ZR-01.0
8.1 400G Host Side Interface

The 400GbE data enters the transceiver using the 400GBASE-R PMA sublayer, where the electrical
interface, for example, may be 400GAUI-8 C2M. The characteristic information of the adapted and
mapped 400GBASE-R host interface signal consists of a scrambled sequence of 256b/257b encoded blocks
with a nominal bit-rate of 425 000 000 kbit/s, 100 ppm.
NOTE – 425 000 000 kbit/s is the nominal bit-rate of the aggregate 400GBASE-R PCS signal consisting of
16 PCS lanes with 256b/257b encoding and FEC at the PMA service interface.
8.2 PMA
The PMA provides a medium-independent means for the PCS to support the use of a range of physical
media. The 400GBASE-R PMA performs the mapping of transmit and receive data streams between the
PCS and PMA via the PMA service interface, and the mapping and multiplexing of transmit and receive
data streams between the PMA and PMD via the PMD service interface. In addition, the PMA performs
retiming of the received data stream when appropriate. The 400GBASE-R PMA service interface is defined
in IEEE Std -2018 Clause 120.3 as an instance of the inter-sublayer service interface definition in
clause 116.3.
8.3 PCS (partial processes)
The 400ZR application implements only a portion of the full PCS processes defined in IEEE Std -
2018 clause 119.
8.3.1 PCS Rx direction (400ZR Tx datapath)
The 400GBASE-R PCS Rx direction (400ZR Tx data path) is defined for 400ZR to include the following
services:
• Alignment lock and lane de-skew (reference IEEE Std -2018 119.2.5.1).
• Lane reorder and de-interleave (reference IEEE Std -2018 119.2.5.2).
• Reed-Solomon FEC decoding the 257-bit blocks and signaling of RS-FEC (544,514) detected local
degrade (Reference IEEE Std -2018 119.2.5.3).
• Post FEC interleave (Reference IEEE Std -2018 119.2.5.4).
• Alignment Marker removal and signaling of Alignment Marker Signal Fail (rx_am_sf<2:0>).
Reference IEEE Std -2018 119.2.5.5.
• Descramble (Reference IEEE Std -2018 119.2.5.5).
• Error Marking – Signaling Tx link degrade (Reference ITU-T G.709/Y.1331 Amendment 2
(06/2018) Annex K and Section 8.8.4).
The RS-FEC decoder may provide the option to perform error detection without error correction to reduce
the latency contributed by the RS-FEC sublayer.
www.oiforum.com 19
OIF-400ZR-01.0
8.3.2 PCS Tx direction (400ZR Rx datapath)

The 400GBASE-R PCSs Tx direction (400ZR Rx data path) defined for 400ZR include the following services:
• Scramble (Reference IEEE Std -2018 Clause 119.2.4.3).
• Alignment Marker Insertion and signaling of Alignment Marker Signal Fail (tx_am_sf<2:0>).
(Reference IEEE Std -2018 Clause 119.2.4.4).
• Pre-FEC distribution (Reference IEEE Std -2018 Clause 119.2.4.5).
• Reed-Solomon FEC encoding the 257-bit blocks (reference IEEE Std -2018 Clause
119.2.4.6).
• Distribution and interleave (Reference IEEE Std -2018 Clause 119.2.4.6).
• Error Marking – Signaling Rx link degrade (Reference ITU-T G.709/Y.1331 Amendment 2
(06/2018) Annex K and Section 8.8.4).
8.4 400ZR frame structure
400GBASE-R, FlexO-4-DSH, and 400ZR frames have similar structures. They are all block formats, 10280
columns×4096 rows (1×4096 or 16×256). The 400ZR frame OH area is the same as the FlexO-4-DSH,
however, fewer OH fields are defined as required for 400ZR than for FlexO-4-DSH. Bonding across multiple
PHY’s is not supported by the 4 ZR frame structure Figure 7 shows the 400ZR frame structure.
10280
5140
5141
5120
5121
1920
1921
3840
3841
1
rows Columns
1 AM PAD OH Payload (5140 bits)
2 1920b 1280b 20b Pad
3 Payload (10280 bits)
. Frame
.
. Payload area
256 Payload (10280 bits)
Figure 7: 400ZR frame structure without parity bits
www.oiforum.com 20
OIF-400ZR-01.0
8.4.1 400ZR Multi-Frame

The 400ZR multi-frame structure with FEC parity field (columns 10281 to 10970) is shown in Figure 8 and
contains a frame Alignment Marker (AM) sequence every 256 rows. Columns are defined as 1-bit wide
and a frame consists of 10970 columns. This results in a bit-oriented structure. The 400ZR multi-frame can
be viewed as a binary matrix with n×256 rows of 10970 bits.
10970
10280
Columns
20×257b
1 AM Pad OH
Row
1
Frame
256
1 AM Pad OH
Frame
FEC
parity
256
1 AM Pad OH
Frame
256
Figure 8: 400ZR multi-frame structure with parity bits
www.oiforum.com 21
OIF-400ZR-01.0
8.5 AM/PAD/OH insertion

5120-bits of AM/PAD/OH, plus 20-bits of additional pad for 257b alignment, are inserted in columns 1 to
5140 of the first row of each 400ZR frame. This leaves 10220 × 257b of additional payload area in the
frame. See Figure 9 below.
4 × 5b (20b) pad
Columns
10281
10970
10280
5141
1921
5120
3841
1920
3840
1
rows
1 16 × 20-bit AM 1920-bit PAD 4 × 320-bit OH 5140 payload bits = (20 × 257-bits) FEC
10220 (257b payload)

10280 payload bits = (40 × 257 bits) FEC
2
400ZR
Frame
255 10280 payload bits = (40 × 257 bits) FEC
256 10280 payload bits = (40 × 257 bits) FEC
Figure 9: 400ZR Frame overhead

The AM field is a set of 16×120-bit blocks that are 10-bit interleaved. PAD is a 1920-bit all-zeros field. The
400ZR OH consists of 4×320b blocks (1280-bits) that are 10b interleaved and transmitted immediately
after the 1920-bits of PAD. Figure 10 details the first 320-bit OH block. The remaining 3×320-bit 400ZR OH
blocks are reserved for future standardization (transmitted as all-zeros and ignored on receipt).
The required 400ZR OH fields are highlighted in black text; the optional fields, in gray text. The 400ZR OH
area includes GMP mapping control bytes (JCx Bytes). See section 8.9 for GMP processing details. The
undefined 400ZR frame OH can provide additional OAM fields, or be set to zero and ignored at the 400ZR
receiver.
www.oiforum.com 22
OIF-400ZR-01.0
8.5.1 400ZR AM/PAD/OH Transmission order
The 400ZR frame structure carries 514 blocks of 10-bit interleaved (5140 bits) of AM/PAD/OH + 20-bits of
additional PAD. The transmission order for each of these fields is defined in Section 8.6 through Section
8.8 and is the same as IEEE Std -2018 and 400G FlexO.
BI TS 1 120
{CM0,CM1,CM2} UP0 {CM3,CM4,CM5} UP1 {UM0 ,UM1,UM2} UP2 {UM3 ,UM4,UM5}
CM: Common alignment marker x16

UP: Unique pad
UM: Unique alignment marker
10280
10970
Columns
1
20 x 257b
rows
1 AM PAD OH
1920b 1920b 1280b 20b Pad
2
3
. Frame 1
.
.
256
FEC Parity
1 AM PAD OH
2 4x320 b
3
. Frame 2
.
.
256
MFAS: Multi-Frame Alignment RES: Reserved/Unused
STAT: RPF MAP: PHY Member Map (RES)
JC1-JC6: Justification Control CRC: Cyclic Redundancy Check (RES)
GID: Group Identification(RES) FCC: Frame Communications Channel (RES)
PID: PHY member Identification(RES)
OSMC: Synchronization Message (RES)
Byte 1 2 29 40
MFAS STAT GID GID GI D RE S PID MAP CRC FCC OSMC
MFAS STAT JC4 JC1 MAP CRC FCC OSMC
RES
MFAS STAT RES MAP CRC FCC OSMC
GMP justification overhead – Carried

only in 1st OH instance. Reserved in
x4
Instances 2 through 4
123 6 7 8 GMP Cm enco din g
1 2 3 4 5 6 7 8
RES
RES
RPF
RD
LD
JC1 C1 C2 C3 C4 C5 C6 C7 C8 14-bit C1028

Link Status overhead – Carried
only in the 1st OH instance. JC2 C9 C10 C11 C12 C13 C14 II DI GMP O verhead &
Parameters (4x257b
Reserved in instances 2 blocks stuffing granularity
JC3 CRC8
through 4.
1 2 3 4 5 6 7 8
JC4 RES RES RES RES D1 D2 D3 D4 7 -bit S CnD
JC5 RES RES RES RES D5 D6 D7 RES Pr ovides 7-bit

granularity
JC6 RES RES RES RES CRC4
Figure 10: 400ZR overhead
www.oiforum.com 23
OIF-400ZR-01.0
8.6 400ZR Alignment Markers (AM)

The role of AM is to find the FEC block boundary. Alignment markers are inserted before FEC encoding
and removed after FEC decoding. The 400ZR modem operates across two domains. IEEE Std -2018
clause 119 defines the AM requirements at the 400GBASE-R interface. This IA defines the 400ZR frame
AM requirements for the coherent single-carrier media interface.
400ZR frame defines a 16×120 = 1920 bits AM field which contains the 10-bit interleaving result of 16
lane alignment markers of 120-bits each. Frame alignment, however, can be done across a subset of
these fields. The alignment marker field is carried at the beginning of each frame (1st row). 400ZR AM is
protected by the SC-FEC and its value is scrambled. AM alignment is processed post FEC decode (after
descrambling) to locate the row number corresponding to the start of the 400ZR frame (SC-FEC being
already 10970b row aligned).
Figure 11 illustrates the AM transmission order. The 192×10b (1920 bits total) blocks are transmitted left
to right starting with the 1st 10-bits of am0, followed by the 1st 10-bits of am1, etc.., until the 12th 10-bits
of am14 (1920 bits total).
1911
1920
11
21
1
AM
12 10 bits of am13
12th 10 bits of am12
1st 10 bits of am10
1st 10 bits of am11
1st 10 bits of am12
1st 10 bits of am13
1st 10 bits of am14
1st 10 bits of am15
2nd 10 bits of am1
2nd 10 bits of am0
2nd 10 bits of am3
2nd 10 bits of am2
2nd 10 bits of am5
2nd 10 bits of am4
2nd 10 bits of am7
2nd 10 bits of am6
1st 10 bits of am0
1st 10 bits of am1
1st 10 bits of am2
1st 10 bits of am3
1st 10 bits of am4
1st 10 bits of am5
1st 10 bits of am6
1st 10 bits of am7
1st 10 bits of am8
1st 10 bits of am9
th
Figure 11: Alignment Marker transmission order – 10b interleaved

The 400ZR AM field consists of 16 logical lane alignment marker indicators (am, where = 0,1...15).
Each lane carries a 120-bit lane alignment marker. Figure 12, and rows of Table 3 give the values of am
transmitted over lane .
BITS 1 120
{ CM0, CM1, CM2} UP0 { CM3, CM4, CM5} UP1 { UM0, UM1, UM2} UP2 { UM3, UM4, UM5}
CM: Common alignment marker

x16
UP: Unique pad
UM: Unique alignment marker
Figure 12: Alignment Marker format
www.oiforum.com 24
OIF-400ZR-01.0
The alignment marker encoding is shown in Table 3.
Logical Encoding
Lane am {CM0, CM1, CM2, UP0, CM3, CM4, CM5, UP1, UM0, UM1, UM2, UP2, UM3, UM4, UM5}
0 0x59,0x52,0x64,0x6D,0xA6,0xAD,0x9B,0x9B,0x80,0x8E,0xCF,0x64,0x7F,0x71,0x30
1 0x59,0x52,0x64,0x20,0xA6,0xAD,0x9B,0xE6,0x5A,0x7B,0x7E,0x19,0xA5,0x84,0x81
2 0x59,0x52,0x64,0x62,0xA6,0xAD,0x9B,0x7F,0x7C,0xCF,0x6A,0x80,0x83,0x30,0x95
3 0x59,0x52,0x64,0x5A,0xA6,0xAD,0x9B,0x21,0x61,0x01,0x0B,0xDE,0x9E,0xFE,0xF4
4 0x59,0x52,0x64,0x87,0xA6,0xAD,0x9B,0x98,0x54,0x8A,0x4F,0x67,0xAB,0x75,0xB0
5 0x59,0x52,0x64,0x4F,0xA6,0xAD,0x9B,0x72,0x48,0xF2,0x8B,0x8D,0xB7,0x0D,0x74
6 0x59,0x52,0x64,0xBC,0xA6,0xAD,0x9B,0x77,0x42,0x39,0x85,0x88,0xDB,0xC6,0x7A
7 0x59,0x52,0x64,0x44,0xA6,0xAD,0x9B,0x4C,0x6B,0x6E,0xDA,0xB3,0x94,0x91,0x25
8 0x59,0x52,0x64,0x06,0xA6,0xAD,0x9B,0xF9,0x87,0xCE,0xAE,0x06,0x78,0x31,0x51
9 0x59,0x52,0x64,0xD6,0xA6,0xAD,0x9B,0x45,0x8E,0x23,0x3C,0xBA,0x71,0xDC,0xC3
10 0x59,0x52,0x64,0x5F,0xA6,0xAD,0x9B,0x20,0xA9,0xD7,0x1B,0xDF,0x56,0x28,0xE4
11 0x59,0x52,0x64,0x36,0xA6,0xAD,0x9B,0x8E,0x44,0x66,0x1C,0x71,0xBB,0x99,0xE3
12 0x59,0x52,0x64,0x81,0xA6,0xAD,0x9B,0xDA,0x45,0x6F,0xA9,0x25,0xBA,0x90,0x56
13 0x59,0x52,0x64,0x28,0xA6,0xAD,0x9B,0x33,0x8C,0xE9,0xC3,0xCC,0x73,0x16,0x3C
14 0x59,0x52,0x64,0x0B,0xA6,0xAD,0x9B,0x8D,0x53,0xDF,0x65,0x72,0xAC,0x20,0x9A
15 0x59,0x52,0x64,0x2D,0xA6,0xAD,0x9B,0x6A,0x65,0x5D,0x9E,0x95,0x9A,0xA2,0x61
NOTE – The value in each byte of this table is in MSB-first transmission order. Note that this per-byte
bit ordering is the reverse of AM values found in [IEEE Std -2018], which uses an LSB-first bit
transmission format.
Table 3: 400ZR Alignment Marker encodings
8.7 400ZR PAD

Immediately following the 1920-bit AM is a 1920-bit field of PAD, transmitted as all-zeros and ignored on
receipt.
1920
1
PAD
Figure 13: PAD transmission order – 10b interleaved
www.oiforum.com 25
OIF-400ZR-01.0
8.8 400ZR OH
Four 320-bit blocks of overhead are transmitted immediately after the 1920 bit of PAD. These are 10-bit
interleaved.
1271
1280
21
11
1
OH
32nd 10 bits of OH0

32nd 10 bits of OH1
32nd 10 bits of OH2
32nd 10 bits of OH3
2nd 10 bits of OH0
2nd 10 bits of OH1
2nd 10 bits of OH2
2nd 10 bits of OH3
3rd 10 bits of OH0
3rd 10 bits of OH1
3rd 10 bits of OH2
3rd 10 bits of OH3
4th 10 bits of OH0
4th 10 bits of OH1
4th 10 bits of OH3

4th 10 bits of OH2
1st 10 bits of OH1
1st 10 bits of OH2
1st 10 bits of OH0
1st 10 bits of OH3
Figure 14: Over Head transmission order – 10b interleaved

8.8.1 Multi-Frame Alignment Signal (MFAS)
The Multi-frame alignment signal (MFAS) is in the first of the four 320-bit OH instances. It is present and
incremented in every 400ZR frame. It counts from 0x00 to 0xFF and provides a 256-frame multi-frame
sequence following [ITU-T G.709.1] Clause 9.2.1 definition.
8.8.2 Link error and Link degrade detection and marking
Table 4 specifies the replacement signal to the host interface in the event of DSP framing or ZR
frame/multi-frame loss.
Host interface signal Replacement signal Bit-rate tolerance (ppm)

Continuous 400GBASE-R local fault sequence +/- 100
400GBASE-R
ordered sets as 256b/257b encoded blocks
Table 4: Replacement signal

The FEC decoder can also detect a degrading link and signal Link Degrade (LD). A degraded link condition
data may be passing data without error, however, the BER may be high and approaching FEC exhaust.
Causes of link degrade may be component wear-out due to aging or stress. A user may want to take pre-
emptive actions based on programmable BER thresholds. Consequent actions can include re-routing
traffic away from the impaired link.
www.oiforum.com 26
OIF-400ZR-01.0
8.8.3 Link status monitoring and signaling (STAT)

The status (STAT) overhead byte is present in every 400ZR frame, but only carried in the first of the four
320-bit OH instances. It includes the 1-bit RPF and 3-bit LDI fields:
• The Remote PHY Fault (RPF) bit indicates signal fail status detected at the remote 400ZR sink
function in the upstream direction and follows the definition in [ITU-T G.709.1] Clause 9.2.5.1. RPF
is set to ''1'' to indicate a remote 400ZR PHY defect indication; otherwise, it is set to '' ’ The RPF
field is in bit 1 of the STAT field as per Figure 15.
• The 3-bit host Link Degrade Indication (LDI) field is defined to indicate to the downstream device
the quality of the host interface signal or the media interface signal.
STAT OH Byte
Bits# 1 2 3 4 5 6 7 8
RES RD LD
RPF RES =Am_ =Am_ =Am_
sf<0> sf<2> sf<1>
LDI
Figure 15: STAT Over Head byte definitions
The 400ZR link shall provide detection and signaling of Link Degrade (LD) for use by switch/routers with
soft reroute capabilities. Figure 16 illustrates the bidirectional signaling between a 400ZR transceiver and
two Routers (A and B). Pre-FEC BER monitors are used to detect and insert link degrade at both the 400ZR
optical link and the 400GBASE-R interface.
400ZR Transponder Application
Router A 400ZR Transponder X 400ZR Transponder Y Router B

Reserved Reserved Reserved Reserved
0 am_sf<0> STAT[6] am_sf<0> am_sf<0>
RD RD RD RD
am_sf<2> STAT[7] am_sf<2> am_sf<2>
LD LD
0 am_sf<1>
LD LD
STAT[8] am_sf<1> am_sf<1>
ZR
PCS
400GBASE-R
(De) Mapper
400GBASE-R
PHY XS
FEC
DTE XS
FEC
FEC
FEC FEC
FEC FEC
FEC FEC
FEC FEC
FEC FEC
MII
(De)Mapper
Single optical
PCS
PCS
AUI
PCS
span and FEC

FEC
FEC
FEC
ZR
LD LD
am_sf<1>
LD LD
STAT[8] am_sf<1> am_sf<1>
0
RD RD RD RD
Reserved Reserved Reserved Reserved
0
+`- FEC Ethernet FEC FEC 400ZR FEC
Figure 16: Local/Remote Degrade interworking between Switch/Router and 400ZR

transceiver
www.oiforum.com 27
OIF-400ZR-01.0
8.8.4 Link Degrade Indication (LDI)

[IEEE 802.3] has specified three bits in the AM field (am_sf<2:0>) to carry Link Degrade Indication (LDI).
Bit am sf<2> is defined as a Remote Degrade (RD) signal, bit am_sf<1> is defined as a Local Degrade (LD)
signal and bit am_sf<0> is reserved.
The 400ZR transceiver X and Y shall forward the information in the Reserved (am_sf<0>) and RD
(am_sf<2>) bits between transceivers as illustrated in Figure 16. The information in am_sf<0> shall be
carried in 400ZR STAT overhead bit 6. The status information in am_sf<2> shall be carried in 400ZR STAT
overhead bit 7. The status information in the LD (am_sf<1>) bit shall be carried after some additional
processing in the 400ZR STAT overhead, bit 8 to the downstream device.
In the host-to-media datapath, the additional processing consists of ORing the ingress LD status in the
am_sf<1> bit of the 400GBASE-R signal with the local host interface RS(544,514) FEC degrade status and
signaling LD in STAT[8] to the media interface. In the media-to-host datapath, the STAT<8> bit from the
media interface is ORed with the 400ZR FEC degrade status and signaled on the am_sf<1> bit to the local
host.
8.8.5 Link Degrade Warning and Alarming.
FEC Detected Degrade (FDD) and FEC Excessive Degrade (FED) is an optional [user configurable] link
monitoring feature, indicating a link degrade condition to the local host and remote transmitter. It can
be used, for example, to pre-emptively move traffic away from a degraded link (e.g. traffic re-route). This
feature requires capturing the pre-FEC BER from the FEC decoder block over a Performance Monitor (PM)
interval. Statistics are gathered by HW and reported by SW. FED and FDD are determined by comparing
the HW BER reported statistics against [user configurable] thresholds.
Link Degrade (LD) signaling shall be based on the FEC decoder statistics (number of corrected errored bits,
and uncorrectable blocks). Fault detection calculation and threshold settings may be implementation
dependent (e.g. based on FEC decoder pre-FEC BER detection capabilities).
The following Performance Monitoring (PM) parameters are defined for determining a Link Degrade (LD)
condition over a PM interval. The PM interval and the collection of the statistics to determine LD is defined
by the Management Interface Spec specific to the module which this IA is implemented.
FEC decoder block, bit counters:
• pFECblkcount = FEC blocks counted over PM interval
• pFECbitcount = total number of bits counted over PM interval = (pFECblkcount × bits per FEC block), 64-
bit value
• pFECcorrbitblk = FEC corrected bits per block (min., avg., max.) over PM interval
• pFECcorrbit = total number of FEC corrected bits over PM interval = ∑pFECcorrbitblk over PM interval.
(64-bit value).
Pre-FEC BER block, bit counters:
• pFECblkBER = FEC block BER (min., avg., max.) over PM interval = (pFECcorrbitblk/pFECblkcount)
• pFECBER = FEC BER over PM interval = (pFECcorrbit / pFECbitcount)
www.oiforum.com 28
OIF-400ZR-01.0
Pre-FEC threshold settings:

• FEC_excessive_BER_activate_threshold (programmable)
• FEC_excessive_BER_deactivate_threshold (programmable)
• FEC_degraded_BER_activate_threshold (programmable)
• FEC_degraded_BER_deactivate threshold (programmable)
FEC degrade settings:
• FECdetectdegraded = FEC degraded status condition over PM interval.
• FECexcessdegraded = FEC excessively degraded status condition over PM interval.
Each of the above registers shall have a corresponding enable, status, and latch bit settings.
FECdetectdegraded and FECexcessdegraded shall also be a maskable interrupt.
PM interval:
• PM_Interval = (programmable); default = 1 second.
The FEC decoder counts and reports the number of bits detected in error over the PM interval per FEC
block (min., max., avg.).
• When the (avg) number of bit errors exceeds the threshold set in
FEC_degraded_BER_activate_threshold, FECdetectdegraded is set and latched.
• When the (avg) number of bit errors falls below the threshold
FEC_degraded_BER_deactivate_threshold, FECdetectdegraded is cleared.
• When the (avg) number of bit errors exceeds the threshold set in
FEC_excessive_BER_activate_threshold, FECexcessdegraded is set and latched.
• When the (avg) number of bit errors falls below the threshold
FEC_excessive_BER_deactivate_threshold, FECexcessdegraded is cleared.
When errors are detected after C-FEC error correction in the Rx data path (e.g. uncorrected block status
from the C-FEC decoder or CRC32 checking), the entire base block of 30592×8 bits is considered corrupted
and all 952×257-bits of information must be marked as being in error using transcoded error control
blocks.
If the link input BER is much lower than C-FEC limit under normal operational conditions, error marking
using post-FEC statistics (i.e. CRC32 checking) could be turned-off to lower Rx latency. In this case, a
programmable pre-FEC excessive error threshold status could be used for error marking at the FEC
decoder output.
Per 802.3TM-2018 Clause 119.2.5.3, if bypass error indication is not supported or not enabled, when the
Reed-Solomon decoder determines that a codeword contains errors that were not corrected, it shall cause
the PCS receive function to set every 66-bit block within the two associated codewords to an error block
(EBLOCK_R) as in Figure 17.
www.oiforum.com 29
OIF-400ZR-01.0
The encoding of a 64b/66b error control block is: [sync="10", control block type=0x1e, and eight 7-bit
/E/control characters.
257-bit error marking block encoding

0 0000 0xE /E/E/E/E/E/E/E/E/ 0x1E /E/E/E/E/E/E/E/E/ 0x1E /E/E/E/E/E/E/E/E/ 0x1E /E/E/E/E/E/E/E/E/
Error Control Block Error Control Block Error Control Block Error Control Block
/E/ control code value (7-bit) = 0x1E →
0 01110000 0111100 01111000 01111000
8×7-bit 8×7-bit 8×7-bit 8×7-bit
257-bit error marking block encoding (binary value)

Transmission order
Figure 17: Error marking
8.9 GMP mapping processes
The 400GBASE-R is asynchronously mapped into a 400ZR container using GMP. The timing is de-correlated
from the 400GBASE-R host clock to simplify ASIC design. Even though the mapping is asynchronous, the
400GBASE-R stream is treated as CBR data (including preamble, and IPG). Data and timing transparency
shall be supported using information fields which are inserted by the GMP process for use upon de-
mapping.
The GMP Justification Control bytes (JC1-6) are carried in the first of the four 320-bit OH instances, and
nd rd th
present in the 2 , 3 , and 4 frames of a 400ZR 4-frame multi-frame to signal the GMP parameters Cm
and ∑CnD from mapper to de-mapper. The 4 frames are identified by MFAS bits 7 and 8 being 00, 01, 10
and 11.
Reference ITU-T G.709 annex D for the general principles of the Generic Mapping Procedure (GMP).
For the purpose of 400ZR the GMP parameters shall be defined as:
• m = GMP data/stuff granularity = 4 × 257 = 1028 bit;
• n = 1028/128 = 8.03125-bit unit and represents the timing granularity of the GMP mapping
present in Cn and ∑CnD parameters;
• Pm,server = maximum number of m-bit data entities in 4-frame multi-frame server payload = 10220.
• Cm = number of client m-bit data entities in 4-frame multi-frame server payload. It is encoded with
14 bits and carried in JC1 and JC2 control OH bytes.
• Cn = number of equivalent client n-bit data entities in 4-frame multi-frame server payload. This
value provides additional ‘n’-bit timing information.
• ∑CnD = accumulated value of the remainder of Cn and Cm. It is encoded with 7-bits and carried in
JC4 and JC5 control OH bytes.
• Cn and Cm being integer values, then:
Cn(t) = 128 × Cm(t) + (∑CnD(t) - ∑CnD(t-1))
The support for n-bit timing information (∑CnD) in the JC4/JC5/JC6 OH is required.
www.oiforum.com 30
OIF-400ZR-01.0
The mapper shall first recover the 400GBASE-R stream. The 400GBASE-R is a sequence of 256b/257b
encoded blocks as per IEEE Std -2018 after the partial PCS processing defined in Figure 6 and
Section 8.3. The 400ZR frame payload area is a direct multiple of 257 bits (10220×257b).
The 400GBASE-R signal is mapped to the 400ZR frame as a 257b block stream, with 20 blocks of
AM/PAD/OH every 10240 blocks. The payload area for this mapping consists of the payload of a 4-frame
400ZR multi-frame (40880 257b blocks) for host interface data. Groups of 1028 successive bits (4×257b),
of the client signal are mapped into a group of 4 successive 257b blocks of the 4-frame 400ZR multi-frame
payload area under control of the GMP data/stuff control mechanism. Each group of 4×257b in the 4-
frame 400ZR multi-frame payload area may either carry 1028 host interface bits or carry 1028 stuff bits.
The stuff bits shall be set to zero.
No rate No rate Decode &

Encode matching
matching Error Marking
256b/257b transcode Reverse transcode
System-side CLOCK DOMAIN 256b/257b blocks
(+/- 100 ppm)
GMP Mapping GMP de-mapping
(+/- 20 ppm)
OH/AM insertion AM/OH detect & removal
(20*257b) (20*257b)
Figure 18: GMP mapping/de-mapping process
Table 5 specifies the host interface and its GMP m, n and CnD parameter values.
Nominal
information bit Bit-rate
Host nominal bit
rate (kbits/s) after tolerance m n CnD
rate (kbits/s)
FEC and AM (ppm)
removal
425 000 000 401 542 892 +/- 100 1028 8.03125 Yes
Table 5: Host interface and its GMP parameter values
The server input nominal bit rate of 401 542 892 kbit/s equals the 400GBASE-R interface signal after
RS(544/514) FEC decode and AM removal.
www.oiforum.com 31
OIF-400ZR-01.0
The de-mapping process decodes Cm(t) and CnD(t) from JC1/JC2/JC3 and JC4/JC5/JC6 and interprets Cm(t)
and CnD(t) according to ITU-T G.709 Annex D. CRC8 shall be used to protect against an error in JC1/JC2/JC3
and CRC4 protect against an error in the JC4/JC5/JC6 signals.
Ref GMP Parameter Formula Value Units

425.00 Gbit/s × 514/544
fclient nominal client information bit rate
× 20479/20480
401,542,892,456.055 bit/s
fclient client bit rate tolerance 100 ppm
fserver server nominal bit rate f 402,489,753,309.729 bit/s
fserver server bit rate tolerance 20 ppm

period of the server multi- 425.00 Gbit/s × 514/544
Tserver 26.154 µs
frame, × 20479/20480
number of bits per server
Bserver 10,526,720 bits
multi-frame
number of overhead bits per
Oserver 20,560 bits
server multi-frame
maximum number of bits in
Pserver 10,506,160 bits
the server payload area
nominal server payload bit
fp,server Bserver / fserver 401,703,640,510.296 bits/s
rate
m GMP data/stuff granularity (4×257 =) 1,028 bits
M m and n ratio 128
maximum number of (m bits)
Pm,server data entities in the server Bserver - Oserver 10220
payload area
fserver × Pserver / Bserver =
number of client m-bit data 478.75 × 28/29 ×
Cm 119/128 × 5140/5488 ×
entities per server multi-frame
511/512
cm value at nominal client and 1028b
Cm,nom m bit data entity 10,215.910
server bit rates blocks
cm value at minimum client
Cm,min m/n 10,214.684
and maximum server bit rates
cm value at maximum client
Cm,max Pserver / m 10,217.136
and minimum server bit rates
Cm,min integer value of cm,min 10,214
Cm,max rounded up value of cm,max (fclient / fp,server) × Pm,server 10,218
GMP justification accuracy, n Cm,nom × (1 - fclient) / (1
n 8.03125 bits
bit data entity + fserver)
maximum number of (n bits)
Cm,nom × (1 + fclient) / (1 8.03125b
Pn,server data entities in the server 1,308,160.000
- fserver) blocks
payload area
www.oiforum.com 32
OIF-400ZR-01.0
Ref GMP Parameter Formula Value Units

number of client n-bit data
Cn C,min 
entities per server multi-frame
Cn value at nominal client and
Cn,nom  Cm,max  1,307,636.519
server bit rates
Cn value at minimum client
Cn,min 1,307,479.603
and maximum server bit rates
Cn value at maximum client
Cn,max Pserver / n 1,307,793.436
and minimum server bit rates
CnD remainder of Cn and Cm 0.910305637
CnD integer value of cnD (fclient / fp,server) × Pn,server
Cn,nom × (1 - fclient) / (1
ƩCnD accumulated value of CnD 127
+ fserver)
Table 6: GMP parameter values

Where,
• Client information rate is 400GBASE-R after RS(544,514) FEC and AM removal with fclient nominal
bit rate and ∆fclient bit rate tolerance.
• Server is 400ZR 4-frame multi-frame (both payload and overhead) with fserver nominal bit rate
∆fserver bit rate tolerance and Bserver number of bits per server 4-frame multi-frame.
• Server payload is 400ZR 4-frame multi-frame payload (before AM/PAD/OH insert) with fp,server
nominal bit rate, ∆fserver bit rate tolerance and Pserver number of bits per server 4-frame multi-frame
payload area.
• The maximum number_of_m [=1028] bit GMP data entities per 4-frame multi-frame payload is
Pm,server [=10220].
• For 400ZR, we use n = [m / 128] = [4×257-bit]/128 = 8.03125 UI that is used as a phase unit “n-bit
equivalent” for Cn parameter. Cn indicates the number of “n-bit equivalent” of the 400GBASE-R
client per 400ZR 4-frame multi-frame server payload. It can be used as a finer phase indicator to
encode the client clock at the GMP mapper.
• So, Cn, nom = 128 × Cm, nom; Cn, min = 128 × Cm, min; Cn, max = 128 × Cm, max
• C = Pm,server × [client_bit_rate / Server_Payload_bit_rate].
m
• C is an integer value indicating to every 400ZR frame the number of m-bit client blocks carried
m
[m = 4×257b = 1028b] in this 400ZR 4-frame server multi-frame payload =
int(Pm,server×[client_bit_rate/Server_Payload_bit_rate]).
• Cm ≤ Pm,server and is a value varying between Cm,min and Cm,max for the given client and
payload type, due to client and payload bit rate tolerance range (+/- 100 ppm and +/-20
ppm).
www.oiforum.com 33
OIF-400ZR-01.0
8.9.1 Stuffing Locations

Stuff location determination for GMP uses a delta-sigma algorithm based on the Cm value over the total
number of payload location. GMP is a positional mapping with non-fixed stuff. So, the stuff location will
vary on a GMP payload-by-payload basis, based on the Cm(t) value. In the case of 400ZR the GMP payload
covers four 400ZR frames.
Table 7 shows the location of the "stuff" GMP blocks for a few specific Cm values.
Cm GMP Blocks Number of Stuff Locations
10220 N/A
10219 1
10218 1, 5111
10217 1, 3407, 6814
10216 1, 2556, 5111, 7666
10215 1, 2045, 4089, 6133, 8177
10214 1, 1704, 3407, 5111, 6814, 8517
Table 7: GMP stuff locations of 400ZR
www.oiforum.com 34
OIF-400ZR-01.0
Figure 19 shows an example of GMP stuff opportunities over four 400ZR frames.
10280
Columns
5141
3841
5140
Row
1
1 ZR OH PAD ZR OH 1 1 1 1 2 2 2 2 5 5 5 5
2 6 6 6 6 7 7 7 7 15 15 15 15
St uff Dat a
4×257b 4×257b
3 Dat a
4×257b
4
400ZR Frame 1
Dat a Dat a
4×257b 4×257b
256 2554 2554 2554 2554 2555 2555 2555 2555
257 ZR OH PAD ZR OH 2556 2556 2556 2556 2557 2557 2557 2557 2560 2560 2560 2560
258 2561 2561 2561 2561 2562 2562 2562 2562 2600 2600 2600 2600
St uff
4×257b Dat a
259 4×257b
Dat a Dat a Dat a
4×257b 4×257b 4×257b
260
400ZR Frame 2
Dat a Dat a
4×257b 4×257b
512 5109 5109 5109 5109 5110 5110 5110 5110
769 ZR OH PAD ZR OH 7666 7666 7666 7666 7667 7667 7667 7667 7686 7686 7686 7686
770 7687 7687 7687 7687 7688 7688 7688 7688 7720 7720 7720 7720
Stuff Dat a
4×257b 4×257b
771 Dat a Dat a
Dat a
4×257b 4×257b
4×257b
772
400ZR Frame 4
Dat a Dat a
4×257b 4×257b
1024 10119 10119 10119 10119 10220 10220 10220 10220
Figure 19: GMP mapping over four 400ZR frames with Cm=10216
www.oiforum.com 35
OIF-400ZR-01.0
8.9.2 GMP overhead Encoding

GMP overhead (JC Bytes OH) is carried once per GMP payload envelope (combining four consecutive
400ZR frame payloads), so once per 4-frame multi-frame. GMP overhead carries the encoded 14-bit Cm(t)
(i.e. 4×257b block count value) in C1-14 bits of JC1 & JC2 (C1 = MSB, ..,C14= LSB) and the encoded 7-bit
ƩCnD(t) (cumulative value of CnD(t)) in D1-D7 bits of JC4 & JC5 (D1=MSB,.. , D7 = LSB) GMP parameters.
Cm(t) is protected by a CRC8 (carried in JC3 OH byte) and ƩCnD(t) is protected by a CRC4 (carried in the four
LSBs of JC6 OH byte).
The JC3 OH CRC8 calculation is described in ITU-T G.709 Annex D.3, and an example of a parallel
implementation can be found in ITU-T G.709 Appendix VI.
8.9.3 GMP OH – CRC8 calculation
The CRC8 located in JC3 is calculated over the JC1 and JC2 bits. The CRC8 uses the generator polynomial:
g(x) = x8 + x3 + x2 + 1
• The JC1 and JC2 octets are taken in order, most significant bit first, to form a 16-bit pattern
representing the coefficients of a polynomial M(x) of degree 15.
• M(x) is multiplied by x8 and divided (modulo 2) by G(x), producing a remainder R(x) of degree 7 or
less.
• The coefficients of R(x) are considered to be an 8-bit sequence, where x7 is the most significant
bit.
• This 8-bit sequence is the CRC8 where the MSB of the CRC8 is the coefficient of x7 and the LSB is
the coefficient of x0.
The de-mapper process performs steps 1-3 in the same manner as the mapper process, except that here,
the M(x) polynomial of step 1 includes the CRC bits of JC3, resulting in M(x) having degree 23. In the
absence of bit errors, the remainder shall be 0000 0000.
8.9.4 The JC6 OH CRC4 Calculation
The CRC4 located in JC6 uses the generator polynomial:
g(x) = x4 + x + 1.
• The four least significant bits of the JC4 and JC5 octets (JC4 D1-D4 and JC5 D5-D7 + RES) are taken
in order, most significant bit first, to form an 8-bit pattern representing the coefficients of a
polynomial M(x) of degree 7.
• M(x) is multiplied by x4 and divided (modulo 2) by G(x), producing a remainder R(x) of degree 4 or
less.
• The coefficients of R(x) are considered to be a 4-bit sequence, where x3 is the most significant bit.
• This 4-bit sequence is the CRC4 where the MSB of the CRC4 is the coefficient of x3 and the LSB is
the coefficient of x0.
The de-mapper process performs steps 1-3 in the same manner as the mapper process, except that here,
the M(x) polynomial of step 1 includes the CRC bits of JC6, resulting in M(x) having degree 11. In the
absence of bit errors, the remainder shall be 0000.
www.oiforum.com 36
OIF-400ZR-01.0
9 400ZR frame to SC adaptation

Figure 20 and Figure 24 show the relationship of the 400ZR frame mapping to the SC-FEC block.
• 119 rows×[2×5140-bit] of information (1223320 bits) + 119×[2×345 bit] of FEC parity (81920 bits)
and pad (190 bits) is mapped to 5×SC-FEC Blocks.
• One SC-FEC frame [510b×512b] carries 952×257-bit blocks of information + CRC32 + 6-bit MBAS
+ 34-bit zero stuff (261120 bits).
columns
rows 1
5140b =20× 257b
10280 10281 10970
| | |
1 AM PAD OH 10
4× 960b
2 11
FEC
Payload area (mapped 400GE) Parity
MBAS
Five consecutive Staircase FEC Blocks
CRC/
110 119
510
111 1 478
Payload Information
952x257- bit block of

payload (framed/
MBAS
CRC/
134 24
mapped 400GE) +
5 x SC-FEC
512
FEC Parity
135 25 119 Rows
256 rows (400ZR Frame)
CRC32 + 6b MBAS +
Blocks
Payload Area 34b all zero (not
FEC
(400GE 256b/257b host blocksX+x Parity
transmitted)
MBAS
CRC/
257b stuff blocks)
Bi
Zero
MBAS
CRC/
158 48
MBAS
CRC/
159 49
Bi+1
Zero
s
MBAS
CRC/
Bi+2
.
Zero
s
MBAS
Format 119 rows of
CRC/
10970- bit into 5 Bi+3
Zero
contiguous SC- FEC
s
blocks (Bi to Bi+4) of 512
-
MBAS
CRC/
bit x 510-bit
Bi+4 72
MBAS
CRC/
229 119
Zero
s
230 1
MBAS
CRC/
254 24
255 FEC 25
256 Parity 26
1 AM PAD OH 27
2× 5140bits 2× 345bits
2× 5485bits
Figure 20: 400ZR frame to SC-FEC relationship
www.oiforum.com 37
OIF-400ZR-01.0
9.1 Mapping 400ZR Frame Payload to Staircase FEC Blocks

The payload of a 400ZR frame is mapped into units of 244,664 bits, where 244,664 = 512×478-72
consecutive bits. For every 244,664 consecutive bits a 32-bit CRC (ref. section 9.2) is calculated, plus a 6-
bit Multi-Block Alignment Signal (MBAS) is added forming the CRC(32B)+MBAS(6b) block. The CRC+MBAS
(38b) block is inserted at the end of each parity block (ref. Figure 21).
Each SC-FEC frame contains 261120 bits (244664b of payload + 16384b of FEC parity bits + 32 bits of CRC
+ 6b of MBAS + 34b of pad). The 34b of additional pad is not transmitted. In the Figure 23, the 38-bit
CRC+MBAS is shown located at the end of each parity block.
Information and parity bits in 119 400ZR frame rows (119×10970 bits) or (1305430 bits), can be
represented in 5 SC blocks organized as 5×32640x8 bits – 34 bits of pad that is not transmitted). See Figure
24 left and right side.
• 400ZR Information block boundaries are thus located at the 23.8th, 47.6th, 71.4th, 95.2th and 119th
rows and at columns 8224, 6184, 4112, 2056 and 10280.
• Parity block boundaries are thus at parity columns 138, 276, 414, 552 and 690 (or columns 10418,
10556, 10694, 10832 and 10970).
9.2 400ZR CRC+MBAS Bit Insertion Block
A 32-bit CRC is calculated over the 244,664 input bits with the generator polynomial IEEE 802.3
(Hammond, et.al. [1]).
𝐺(𝑥) = 𝑥 32 + 𝑥 26 + 𝑥 23 + 𝑥 22 + 𝑥 16 + 𝑥 12 + 𝑥 11 + 𝑥 10 + 𝑥 8 + 𝑥 7 + 𝑥 5 + 𝑥 4 + 𝑥 2 + 𝑥 + 1,
appended to the end of the sequence.
Mathematically, the CRC value corresponding to the 244,664 input bits is defined by the following
procedures:
• The first 32 bits of the frame are complemented.
• The 244,664 bits of the protected fields are the coefficients of a polynomial 𝑀(𝑥) of degree
244,663. (The first bit of the 244,664 input bits corresponds to the 𝑥 244,663 term and the last bit
of the 244,664 input bits corresponds to the 𝑥 0 term).
• 𝑀(𝑥) is multiplied by 𝑥 32 and divided (modulo 2) by 𝐺(𝑥), producing a remainder R(x) of degree
< 31.
• The coefficients of R(x) are a 32-bit sequence.
• The bit sequence is complemented, and the result is the CRC.
www.oiforum.com 38
OIF-400ZR-01.0
The 32 bits of the CRC value are placed with the 𝑥 31 term as the left-most bit of the CRC32 field and the
𝑥 0 term as the right-most bit of the CRC32 field. (The bits of the CRC are thus transmitted in the order:
𝑥 31 , 𝑥 30 , … , 𝑥 1 , 𝑥 0 ). The 6-bit MBAS is appended after the 32-bit CRC
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38
CRC32 MBAS
Figure 21: CRC32 + MBAS
9.2.1 400ZR Multi Block Alignment Signal (MBAS)
To synchronize the state of the Error De-correlator (ED) controllers between the receiver and the
transmitter, the Staircase FEC scheme uses a 7-bit SC FEC Multi Block Alignment Signal (MBAS) which
provides a 128-block sequence.
The six most significant bits of the 7-bit MBAS are transferred between source and sink in the 6-bit MBAS
overhead, which is located in bits 33 to 38 in Figure 21.
The numerical value represented in the six MBAS overhead bits will be incremented every two SC FEC
blocks and provides as such a 128-block multi-block as illustrated in Figure 22.
Figure 22: Multi-block alignment signal overhead
www.oiforum.com 39
OIF-400ZR-01.0
Figure 23 shows the location and transmission order of the CRC32 and MBAS.
2056
6168
8224
4112
10280
10418
10832
10556
10694
10970
columns
rows
1
i+1
16384
Parity bits CRC32
244664 Information bits of Bj of Bj-1
i+24
CRC32 of Bj
16384
244664 Information bits of Bj+1 Parity bits CRC32
of Bj
i+48
CRC32 of Bj+1
16384
of Bj+1
i+72
CRC32 of Bj+2
16384
of Bj+2
i+96
CRC32 of Bj+3
16384
of Bj+3
i+119
CRC32 of Bj+4
AM/PAD/OH and payload SC FEC and
38 PAD
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38
CRC32 MBAS
Figure 23: CRC32 + MBAS transmission order
www.oiforum.com 40
OIF-400ZR-01.0
Figure 24 details the 400ZR frame to SC-FEC adaptation.
2560 bit information 2432 bit information 2048 bit parity

columns blocks blocks blocks
10280
10418
10556
10694
10832
10970
2056
4112
6184
8224
rows
1
32640
1.0 1.1 1.2 1.3
i+1 1.4 1.5 1.6 1.7 1.12 1.0 1.10 1.11 1.12
1.8 1.9 1.10 1.11 2.0 2.0 2.10 2.11 2.12
2.0 3.0 3.10 3.11 3.12
4.0 244664 information 4.10 4.11 4.12
16384 5.0 5.10 5.11 5.12
8
bits of Bj-1
244664 Information bits of Bj Parity bits 6.0 6.10 6.11 6.12
7.0 7.10 7.11 7.12
of Bj-1 8.0 8.10 8.11 8.12
16384 parity
8.12 bits of Bj-1
i+24 8.10 8.11* 1.0
1.1 1.2 1.3 1.4 1.12
1.5 1.6 1.7 1.8
1.9 1.10 1.11 2.0
1.0 1.10 1.11 1.12
16384 2.0 2.10 2.11 2.12
3.0 3.10 3.11 3.12
244664 Information bits of Bj+1 Parity bits 4.0 244664 information 4.10 4.11 4.12
of Bj 5.0 bits of Bj 5.10 5.11 5.12
6.0 6.10 6.11 6.12
7.0 7.10 7.11 7.12
8.0 8.10 8.11 8.12
8.12
i+48 16384 parity
8.10 8.11* 1.0 1.1
119 rows  5 × FEC Blocks
1.2 1.3 1.4 1.5 bits of Bj

1.6 1.7 1.8 1.9 1.12
1.10 1.11 2.0
1.0 1.10 1.11 1.12
16384 2.0 2.10 2.11 2.12
244664 Information bits of Bj+2 Parity bits 3.0 3.10 3.11 3.12
4.0 244664 information 4.10 4.11 4.12
of Bj+1 5.0 bits of B j+1 5.10 5.11 5.12
Information and 6.0 6.10 6.11 6.12
7.0 7.10 7.11 7.12
8.12 parity bits in 119 8.0 8.10 8.11 8.12
i+72 8.11* 8.11 1.0 1.1 1.2 400ZR
1.
frame rows 16384 parity
1.3 1.4 1.5 1.6 1.12 can be represented bits of Bj+1
1.7 1.8 1.9 1.10
1.11 2.0 in 5 blocks of
32640×8 bit with 72
16384 XXX bits at the end
244664 Information bits of Bj+3 Parity bits of subblock 8.12 1.0 1.10 1.11 1.12
of Bj+2 2.0 2.10 2.11 2.12
3.0 3.10 3.11 3.12
4.0 244664 information 4.10 4.11 4.12
5.0 bits of B j+2 5.10 5.11 5.12
8.10 8.12 6.0 6.10 6.11 6.12
i+96 8.11* 1.0 1.1 1.2 7.0 7.10 7.11 7.12
1.3 1.4 1.5 1.6 8.0 8.10 8.11 8.12
1.7 1.8 1.9 1.0 1.11 1.12 16384 parity
2.0 bits of Bj+2
16384
244664 Information bits of Bj+4 Parity bits
of Bj+3
1.0 1.10 1.11 1.12
8.12 2.0 2.10 2.11 2.12
i+119 3.0 3.10 3.11 3.12
8.10 8.11 4.0 244664 information 4.10 4.11 4.12
1.0 1.1 1.2 1.3 5.0 5.10 5.11 5.12
1.4 1.5 1.6 1.7 1.12 bits of B j+3
1.8 1.9 1.10 1.11 2.0 6.0 6.10 6.11 6.12
2.1 7.0 7.10 7.11 7.12
8.0 8.10 8.11 8.12
CRC32
16384
400ZR: 6-bit MBAS
244664 Information bits of Bj+5 Parity bits
CRC32 34 PAD
of Bj+4 6 PAD
16384 parity
bits of Bj+3
8.12
8.10 8.11* 1.0 1.10 1.11 1.12
2.0 2.10 2.11 2.12
3.0 3.10 3.11 3.12
AM/PAD/OH and payload 4.0 244664 information 4.10 4.11 4.12
5.0 bits of B j+4 5.10 5.11 5.12
8.11* block shown in grey contains only 6.0 6.10 6.11 6.12
payload bits. The remaining 32-bit CRC and 7.0 7.10 7.11 7.12
8.0 8.10 8.11 8.12
6-bit pad are shown in yellow in the FEC 16384 parity
parity area bits of Bj+4
Figure 24: 400ZR frame adaptation SC FEC block
www.oiforum.com 41
OIF-400ZR-01.0
10 400ZR Forward Error Correction (FEC)

The 400ZR Forward Error Correction (FEC) algorithm is a Concatenated FEC (C-FEC) that combines a HD-
FEC (255,239) outer code and an inner double-extended SD-FEC (128,119) Hamming code resulting in
~10.8dB of NCG with ~14.8% overhead (e.g. BER_in = 1.25E-2 results in BER_out = 1.0E-15).
The HD-FEC is a (512-bit × 510-bit) generalized staircase code that works in conjunction with an error de-
correlator. The error de-correlator function randomizes the position of the symbols to reduce the impact
of correlated errors on the FEC performance.
10.1 SC-FEC
For the purposes of this IA and to minimize the possibility of misinterpretation that might deviate from a
common implementation the SC-FEC, Adapt, ED and Encoding/Decoding processes shown in Figure 25 are
defined by reference [5] Annex A.
CRC-32 Calculation CRC-32 checking
SC-FEC Adapt, ED & SC-FEC Adapt, ED &


(6*119b) (6*119b)
Figure 25: 400ZR HD-FEC processes

10.2 Sync Pad Insertion
For the purpose of alignment and synchronization 6×119 bits are appended/removed from the tail end of
the 5xSC-FEC block.


(6*119b) (6*119b)
Scramble Descramble
Figure 26: Pad insertion/removal
www.oiforum.com 42
OIF-400ZR-01.0
Figure 27 shows the location of 6 ×119b pad relative to the 400ZR Frame.
Scrambler reset to 0xFFFF
columns
rows 1
5140b = 20×257b |
10280 10281 10970
| |
1 AM PAD OH 10
4×960b
2 11
FEC
Payload area (mapped 400GE) Parity
110 6x119b 119
CRC
111 1
134 24
CRC
5 x SC-FEC
135 25
119 Rows
256 rows (400ZR Frame)
Blocks
Payload Area
FEC
(400GE 256b/257b host blocks + X x Parity
257b stuff blocks)
158 48
CRC
159 49
229 6x119b 119

CRC
230 1
254 24
CRC
255 FEC 25
256 Parity 26
1 AM PAD OH 27
2× 5140 bits 2× 345 bits
2× 5485 bits
Figure 27: 6 × 119 Pad Insertion
www.oiforum.com 43
OIF-400ZR-01.0
10.3 Frame Synchronous Scrambling

The scrambler/descrambler is located after/before the SC-FEC encoding and 6 ×119b pad insertion.

(6*119b) (6*119b)
Scramble Descramble
Convolution de-
interleaving
Figure 28: Frame synchronous scrambler

The operation of the scrambler shall be functionally equivalent to that of a frame-synchronous scrambler
of sequence 65535 and the generating polynomial shall be:
16 12 3
x + x + x + x + 1.
The scrambler/descrambler resets to 0xFFFF on row 1, column 1 of the five SC-FEC block structure and
subsequent 714-bit (6 x 119b) pad insertion and the scrambler state advances during each bit of the 5xSC-
FEC blocks. In the source function, all payload bits (included SC-FEC parity) are scrambled. At the sink the
scrambler is synchronized (initialized) at the start of each payload.
10.4 Convolutional Interleave
The staircase encoded frame + 6 Sync/Pad, which consists of 10976×119 bits, is first interleaved (in units
of 119 bits) by a convolutional interleave (CI). The CI serves to spread out the transmission order of
consecutive units of 119 bits from the staircase encoded frame, which increases the resilience of the
system to bursts of errors.
Figure 29: Convolution interleave
www.oiforum.com 44
OIF-400ZR-01.0
The CI is of depth 16, that is, it consists of 16 parallel delay lines, as illustrated in Figure 30.
Figure 30: Convolution interleave

Each delay operator “D” represents a storage element of 9b From one row to the next lower row, two
delays operators are deleted.
At time 𝑖, the input and output switches are aligned at row 𝑏𝑖 :
• A block of 119b is read from row 𝑏𝑖
• The contents of row 𝑏𝑖 are shifted to the right by 119b
• A block of 119b is written to row 𝑏𝑖
• The switch position is updated to 𝑏𝑖+1 = 𝑏𝑖 + 1 (𝑚𝑜𝑑 16)
www.oiforum.com 45
OIF-400ZR-01.0
Initialization of the convolutional interleave switches (to their topmost positions) is defined to occur at
the start of every DSP super frame, which contains 5 SC-FEC blocks (i.e. immediately prior to processing
the first row in Figure 27). Since 10976 is evenly divisible by the depth of the CI (i.e. 16), the switches will
wrap around to this position at the start of every ZR frame. The start of the DSP super frame emitted from
the CI will align with the first block of data emitted following a re-initialization of the interleaving switches.
119 119 9
Convolutional Hamming
119b Interleaver 119b
Encoder
Hamming Parity
10970
10976
ZR Frame bits assigned
Hamming Payload
row-by-row, read out
Area
row-by-row to
convolutional
interleaver
6x119 Pad
6
Figure 31: Hamming FEC frame format

The 119b outputs of the convolutional interleave are encoded by a systematic (128,119) double-extended
Hamming code.
10.5 Inner Hamming Code
The inner FEC of C-FEC is a double-extended Hamming Code SD-FEC (128,119), increasing NCG from 9.4
dB to ~10.8 dB with ~7.56% added overhead.
Figure 32: Hamming code
www.oiforum.com 46
OIF-400ZR-01.0
The systematic double-extended Hamming code is most naturally defined in terms of its parity-check
matrix. Consider the function 𝑔 which maps an integer 𝑖, 0 ≤ 𝑖 ≤ 127, to the column vector:
𝑠0,𝑖
𝑠1,𝑖
𝑔(𝑖) = 𝑠 ⋮ ,
6,𝑖
𝑠7,𝑖
[1]
where,
𝑖 = 64𝑠6,𝑖 + 32𝑠5,𝑖 + ⋯ + 2𝑠1,𝑖 + 𝑠0,𝑖 ,
and,
𝑠7,𝑖 = (𝑠0,𝑖 ∧ 𝑠2,𝑖 ) ∨ (𝑠̅̅̅̅
0,𝑖 ∧ ̅̅̅̅
𝑠1,𝑖 ∧ ̅̅̅̅)
𝑠2,𝑖 ∨ (𝑠0,𝑖 ∧ 𝑠1,𝑖 ∧ ̅̅̅̅).
𝑠2,𝑖
The parity-check matrix is then a 9×128 binary matrix:
𝐻 = [𝑔(0): 𝑔(62), 𝑔(64): 𝑔(94), 𝑔(96): 𝑔(110), 𝑔(112): 𝑔(118), 𝑔(120), 𝑔(122), 𝑔(124),
𝑔(63), 𝑔(95), 𝑔(111), 𝑔(119), 𝑔(121), 𝑔(123), 𝑔(125): 𝑔(127)]
where g(a):g(b) represents:
[g(a),g(a+1),g(a+2),…,g(b)]
To obtain the encoder matrix 𝐺, we calculate
𝑃 = 𝐵[𝑔(0): 𝑔(62), 𝑔(64): 𝑔(94), 𝑔(96): 𝑔(110), 𝑔(112): 𝑔(118), 𝑔(120), 𝑔(122), 𝑔(124)]
where,
B=[ 𝑔(63), 𝑔(95), 𝑔(111), 𝑔(119), 𝑔(121), 𝑔(123), 𝑔(125): 𝑔(127)]−1
Finally, the generator matrix of the Hamming code is,
𝐺 = [𝐼; 𝑃𝑇 ],
and a 119-bit message,
𝑏 = [𝑏0 , 𝑏1 , … , 𝑏118 ]
is encoded to the 128-bit code word.
𝑐 = [𝑐0 , 𝑐1 , … , 𝑐127 ] = 𝑏𝐺
www.oiforum.com 47
OIF-400ZR-01.0
11 DP-16QAM Symbol mapping and polarization distribution

Each 128-bit code word is mapped to 16 DP-16QAM symbols (S),
𝑆 = [𝑠0 , 𝑠1 , … , 𝑠15 ],
where,
• (𝑐8𝑖 , 𝑐8𝑖+1 ) maps to the in-phase (I) component of the X-pol of 𝑠𝑖
• (𝑐8𝑖+2 , 𝑐8𝑖+3 ) maps to the quadrature-phase (Q) component of the X-pol of 𝑠𝑖
• (𝑐8𝑖+4 , 𝑐8𝑖+5 ) maps to the I component of the Y-pol of 𝑠𝑖
• (𝑐8𝑖+6 , 𝑐8𝑖+7 ) maps to the Q component of the Y-pol of 𝑠𝑖
In each signaling dimension, we define the following mapping from binary label to symbol amplitude:
( , )→− ,( , )→− ,( , )→+ ,( , )→+
This mapping is further detailed in Table 8 below:
(𝒄𝟖𝒊 , 𝒄𝟖𝒊+𝟏, 𝒄𝟖𝒊+𝟐, 𝒄𝟖𝒊+𝟑 ) 𝑜𝑟 (𝒄𝟖𝒊+𝟒 , 𝒄𝟖𝒊+𝟓, 𝒄𝟖𝒊+𝟔, 𝒄𝟖𝒊+𝟕 ) I Q
(0,0,0,0) -3 -3
(0,0,0,1) -3 -1
(0,0,1,0) -3 3
(0,0,1,1) -3 1
(0,1,0,0) -1 -3
(0,1,0,1) -1 -1
(0,1,1,0) -1 3
(0,1,1,1) -1 1
(1,0,0,0) 3 -3
(1,0,0,1) 3 -1
(1,0,1,0) 3 3
(1,0,1,1) 3 1
(1,1,0,0) 1 -3
(1,1,0,1) 1 -1
(1,1,1,0) 1 3
(1,1,1,1) 1 1
Table 8: In-phase (I) and quadrature phase (Q) symbol amplitude
11.1 Interleaving DP-16QAM Symbols
The DP-16QAM symbols are time-interleaved, to de-correlate the noise between consecutively received
symbols, as well as to uniformly distribute the symbols (mapped from a single Hamming code word)
between pilot symbols.
www.oiforum.com 48
OIF-400ZR-01.0
Prior to Fame Alignment Word (FAW) and pilot insertion, each frame consists of 10976×16 DP-16QAM
symbols. The symbol interleave performs an 8-way interleaving of symbols from Hamming code words.
S0,0 S0,1 .. . S0,14 S0,15

S1,0 S1,1 .. . S1,14 S1,15
S2,0 S2,1 .. . S2,14 S2,15
.. .
8-way S7,0 S7,1 .. . S7,14 S7,15
interleaved
Hamming
Codewords
S8,0 S8,1 . .. S8,14 S8,15
S9,0 S9,1 . .. S9,14 S9,15

S10,0 S10,1 . .. S10,14 S10,15
. ..
S15,0 S15,1 . .. S15,14 S15,15

...
Figure 33: Hamming code 8-way interleave
www.oiforum.com 49
OIF-400ZR-01.0
12 DSP framing
A DSP super-frame is defined as a set of 181888 symbols in each of the X/Y polarization. A DSP sub-frame
consists of 3712 symbols. A DSP super-frame thus consists of 49 DSP sub-frames.
Pilot symbols are inserted every 32 symbols, starting with the first symbol of each DSP super-frame. The
first 11 symbols of the DSP sub-frame can also be used for training (e.g. frame acquisition). The first symbol
of the Training Sequence (TS) is a Pilot Symbol (PS).
• Every DSP subframe has the same structure based on a fixed TS with the first symbol processed
as a pilot.
• The TS includes 11 QPSK symbols for each polarization. The TS is different between X and Y
polarizations
• The PS sequence includes (1+115) QPSK symbols based on PRBS. The first TS symbol is also the
first symbol of the PS sequence.
12.1 First DSP sub-Frame
The first DSP sub-frame of the super-frame includes a 22 symbol Frame Alignment Word (FAW) used to
align to the 5 SC-FEC Frames. 76 additional symbols are reserved for future use/innovation.
The First DSP sub-frame includes:
• 22 symbols used as the Super Frame Alignment Word (FAW). The FAW is different between X and
Y polarizations.
• 76 symbols are reserved to be used for future proofing and for innovation. These symbols should
be randomized to avoid strong tones. These symbols should be selected from 16QAM modulation.
Since 1st symbol is 11 symbols as training sequen ce Pilot Symbols

known QPSK symbol it
can be processed as a 22 symbols as sup er-frame alignment word Information, FEC parity Symbols
Pilot
76 symbols as reserved for future use
Seeds for pilot PRBS
selected so that this is a
sequence
FAW + reserved +3 pilot = 98 symbols
11 21 11 30 1 31 1 15 16 1 31 1 31
32 Symbols 32 Symbols 32 Symbols 32 Symbols 32 Symbols 32 Symbols

3712 symbols = 116 × 32 symbols
FAW
(76 reserved + 2 pilot) = 78
(22 + 1 Pilot)
Figure 34: First DSP sub-frame of super-frame
www.oiforum.com 50
OIF-400ZR-01.0
12.1.1 FAW Sequence
Index FAW X FAW Y Index FAW X FAW Y
1 3 - 3j 3 + 3j 12 3 - 3j -3 + 3j
2 3 + 3j -3 + 3j 13 -3 - 3j -3 + 3j
3 3 + 3j -3 - 3j 14 -3 - 3j 3 + 3j
4 3 + 3j -3 + 3j 15 -3 + 3j -3 - 3j
5 3 - 3j 3 - 3j 16 3 + 3j 3 + 3j
6 3 - 3j 3 + 3j 17 -3 - 3j -3 - 3j
7 -3 - 3j 3 - 3j 18 3 - 3j -3 + 3j
8 3 + 3j 3 - 3j 19 -3 + 3j 3 - 3j
9 -3 - 3j -3 - 3j 20 3 + 3j -3 - 3j
10 -3 + 3j 3 - 3j 21 -3 - 3j 3 - 3j
11 -3 + 3j 3 + 3j 22 -3 + 3j -3 + 3j
Table 9: FAW sequence
12.2 Subsequent DSP sub-frames.
Each subsequent DSP sub-frame after the first includes an 11 symbol TS, the first symbol of which is a PS.
Since 1 st symbol is
11 symbols as training sequen ce
known QPSK symbol it
can be processed as a
Pilot Pilot Symbols
Seeds for pilot PRBS Information, FEC parity Symbols

selected so that this is a
sequence
11 21 1 31 1 31
32 Symbols 32 Symbols 32 Symbols
Figure 35: DSP sub-frames 2-49 of the DSP super-frame
www.oiforum.com 51
OIF-400ZR-01.0
12.2.1 Training Sequence

The TS is defined by the following table:
Index Training X Training Y

1* -3 + 3j -3 - 3j
2 3 + 3j -3 - 3j
3 -3 + 3j 3 - 3j
4 3 + 3j -3 + 3j
5 -3 - 3j -3 + 3j
6 3 + 3j 3 + 3j
7 -3 - 3j -3 - 3j
8 -3 - 3j -3 + 3j
9 3 + 3j 3 -3j
10 3 - 3j 3 + 3j
11 3 - 3j 3 - 3j
Table 10: Training symbol sequence
*The first symbol of the TS is processed as a pilot
12.3 Pilot Sequence
Training symbols and pilot symbols shall be set at the outer 4 points of the 16QAM constellation. See
Figure 36.
The PS is a fixed PRBS10 sequence mapped to QPSK with different seed values for X/Y.
• Seeds are selected so that the pilot and training sequence combined are DC balanced
• Seeds are selected so that the first symbol in the training sequence is also the first symbol in
the pilot sequence
• The seed is reset at the start of every DSP sub-frame
01 11
PRBS10 2 × 116 = 232 bits
with seed = 0x19E are mapped to QPSK
PRBS10 2 × 116 = 232 bits

with seed = 0x0D0 are mapped to QPSK
00 10
Figure 36: QPSK mapped Pilot Sequence
www.oiforum.com 52
OIF-400ZR-01.0
Table 11 shows the pilot generator polynomial and seed values.
Generator polynomial Seed X Seed Y
𝒙𝟏𝟎 + 𝒙𝟖 + 𝒙𝟒 + 𝒙𝟑 + 𝟏 0x19E 0x0D0

Table 11: Pilot polynomial and seed
Figure 37 shows the sequencing.
Out X Out Y
0 0
Output -3+3i -3-3i
1 0
1 0
Seed X = 0x19E 0 1 1 0 0 1 1 1 1 0 +3+3i -3 -3i
1 0
Seed Y = 0x0D0 0 0 1 1 0 1 0 0 0 0 1 1
0 +3 -3i +3-3i
0
0 1
-3+3i +3 +3i
1 1
1 0
+3-3i -3-3i
0 0
1 1
+3-3i +3+3i
0 1
. .
. .
. .
Figure 37: Pilot seed and sequence
www.oiforum.com 53
OIF-400ZR-01.0
The complete table is shown below:
Index Pilot X Pilot Y Index Pilot X Pilot Y Index Pilot X Pilot Y Index Pilot X Pilot Y
1 -3 + 3j -3-3j 30 3 - 3j 3-3j 59 3 - 3j 3-3j 88 3 - 3j -3+3j
2 3 + 3j -3-3j 31 -3 - 3j -3+3j 60 3 + 3j -3+3j 89 -3 - 3j -3+3j
3 3 - 3j 3-3j 32 3 + 3j -3-3j 61 3 - 3j 3+3j 90 3 - 3j 3-3j
4 -3 + 3j 3+3j 33 -3 + 3j 3-3j 62 -3 - 3j -3-3j 91 3 - 3j 3+3j
5 3 - 3j -3-3j 34 -3 + 3j -3-3j 63 3 - 3j 3+3j 92 -3 + 3j 3-3j
6 3 - 3j 3+3j 35 -3 + 3j -3-3j 64 -3 + 3j -3+3j 93 -3 - 3j 3-3j
7 -3 - 3j -3+3j 36 3 - 3j 3-3j 65 3 - 3j 3-3j 94 3 + 3j -3+3j
8 3 + 3j -3+3j 37 3 - 3j 3-3j 66 3 + 3j 3+3j 95 -3 - 3j 3-3j
9 -3 + 3j -3-3j 38 -3 - 3j -3-3j 67 3 - 3j -3-3j 96 -3 - 3j 3-3j
10 3 + 3j 3+3j 39 -3 - 3j 3+3j 68 -3 + 3j 3-3j 97 3 + 3j -3+3j
11 3 + 3j 3+3j 40 3 - 3j -3-3j 69 3 - 3j -3+3j 98 -3 + 3j 3-3j
12 -3 - 3j -3-3j 41 -3 - 3j 3-3j 70 -3 + 3j -3+3j 99 3 - 3j -3-3j
13 3 + 3j 3+3j 42 3 - 3j 3-3j 71 3 + 3j -3+3j 100 -3 - 3j 3+3j
14 3 - 3j 3+3j 43 -3 + 3j -3-3j 72 -3 - 3j -3-3j 101 3 + 3j -3-3j
15 3 + 3j 3-3j 44 -3 + 3j -3-3j 73 -3 - 3j -3+3j 102 -3 + 3j -3+3j
16 3 - 3j 3+3j 45 -3 - 3j 3+3j 74 3 - 3j 3+3j 103 -3 - 3j -3+3j
17 3 + 3j 3+3j 46 -3 + 3j -3+3j 75 -3 + 3j -3-3j 104 -3 - 3j 3+3j
18 3 - 3j -3+3j 47 -3 - 3j 3+3j 76 3 - 3j -3-3j 105 3 + 3j -3+3j
19 -3 + 3j -3-3j 48 3 + 3j -3+3j 77 -3 + 3j -3-3j 106 3 - 3j 3-3j
20 -3 - 3j 3-3j 49 3 + 3j 3-3j 78 -3 - 3j 3+3j 107 3 + 3j 3+3j
21 3 + 3j 3-3j 50 -3 + 3j -3+3j 79 3 + 3j -3-3j 108 -3 + 3j -3+3j
22 -3 + 3j 3+3j 51 3 - 3j 3+3j 80 3 + 3j -3-3j 109 -3 - 3j 3+3j
23 -3 + 3j -3+3j 52 3 - 3j -3+3j 81 3 + 3j 3-3j 110 -3 + 3j -3-3j
24 3 - 3j 3-3j 53 3 - 3j -3+3j 82 -3 - 3j -3-3j 111 -3 - 3j -3+3j
25 -3 + 3j 3-3j 54 -3 - 3j 3+3j 83 -3 - 3j 3+3j 112 -3 + 3j 3-3j
26 -3 + 3j 3+3j 55 3 - 3j -3+3j 84 3 + 3j -3-3j 113 -3 + 3j -3+3j
27 -3 + 3j -3+3j 56 3 + 3j -3+3j 85 3 - 3j -3-3j 114 3 + 3j 3+3j
28 -3 + 3j 3+3j 57 -3 + 3j -3-3j 86 -3 + 3j -3-3j 115 3 + 3j 3-3j
29 -3 - 3j 3+3j 58 -3 - 3j 3-3j 87 3 + 3j 3-3j 116 -3 - 3j 3-3j
Table 12: Pilot Sequence
12.4 Channel Mappings
X and Y indicate a pair of mutually orthogonal polarizations of any orientation and I and Q are mutually
orthogonal phase channels in each polarization. The four data path channels are therefore labeled XI, XQ,
YI, and YQ.
All coherent channel mappings provided in Table 13 are allowed for the Tx signal. The Rx should work in
all cases because the Rx can unambiguously identify the signals polarization and phase, based on the FAW.
The Tx mapping is specified in Table 13 by two designations: [X:Y ; I,Q], where a “:” is used to separate X
& Y, a “,” is used to separate I & Q.
www.oiforum.com 54
OIF-400ZR-01.0
Table 13 does not allow interleaving of the channels by polarization since this would add a non-essential
level of complexity to the Rx digital processing.
Mapping X:Y I,Q Notes
[0,x] X:Y Pol. cannot be interleaved

[1,x] Y:X
[x,0] I,Q:I,Q Same across Pol.
[x,1] Q,I:Q,I
[x,2] I,Q:Q,I Flip across Pol.
[x,3] Q,I:I,Q
Table 13: Channel mappings

12.5 Frame Expansion Rate
The 400ZR optical signal is DP-16QAM with a symbol rate of 59.843750000 Gbaud (478.750 Gbps) per
polarization. Figure 38 and Table 14 provide details on expansion for each functional block.
400ZR
100GE Domain 400ZR Domain
~478 750 Gb/Sec
(+/- 100ppm) (+/- 20ppm)
+/- 20ppm
~402 489 Gb/Sec ~429 505 Gb/Sec ~429 740 Gb/Sec ~462 241 Gb/Sec
~401.562 Gb/Sec ~401.703 Gb/Sec +/- 20ppm +/- 20ppm +/- 20ppm +/- 20ppm
+/- 100ppm +/- 20ppm
20479/20480 512/511 5485/5140 5488/5485 128/119 29/28
514/544
255/239
244 702 261 086
400ZR Container SC-FEC
RS FEC AM/PAD/OH Insert Remove 5 x 261 086
425 000 000 AM GMP 5 x 244,702 Encoder
Decode +Payload 34b 34b bits
Gb/Sec Remove map bits &
(544/514) +CRC32 (Zero s) (Zero s)
Interleaver
+MBAS
5x 244 736 261 120
401 562 500 401 542 892 456 401 703 640 510 402 489 753 310
bit/s bit/s bit/s bit/s
1 305 430
Bits
(10970 x 119)
5 x 261 086 Insert PAD Convolution Hamming

Scrambler
bits 6 x 119b Interleave (128/119)
1 306 144 1 306 144 1 306 144

Bits Bits Bits
(10976 x 119) (10976 x 119) (10976 x 119)
32/31
Insert 588
175 616 176 204
Symbols
FAW/TS/RES
Symbols
Insert Pilots 178176 X pol
Gray Map Symbols
Polarization
and
Distribution
Interleave Insert 588
175 616 176 204
Symbols FAW/TS/RES Symbols
Insert Pilots 178176 Y pol
Symbols
1 404 928 351 232 351 232 176 204 181 888
Bits Symbols Symbols Symbols Symbols
(10976 x 128) (3b/Sym) (3b/Sym) (3b/Sym) (3b/Sym)
400GE – ZR bit rate; 156.25Mbit/s x 383 x 8 +/-20ppm
Figure 38: 400ZR expansion rates
www.oiforum.com 55
OIF-400ZR-01.0
Table 14 details the bit level expansion.
GMP
FlexO/400ZR frame
400GBASE-R After Client FEC After 400GE AM Before 400ZR Before SC-FEC + Before [6 x Before Before Before Pilot
400ZR Bit rate 400ZR Baud rate
Client -rate Termination Removal AM/PAD/OH MBAS + CRC32 119b ] pad Hamming FAW/TS/RES Symbol insertion
[bps] [bps]
[bps] [bps] [bps] insert [bps] [bps] insert [bps] [bps] [bps] [bps]
+100ppm 425 042 500 000 401 602 656 250 401 583 046 745 401 711 674 583 402 497 803 105 429 513 706 231 429 748 627 128 462 250 624 138 463 798 338 281 478 759 575 000 059 844 946 875 +20ppm
Nominal 425 000 000 000 401 562 500 000 401 542 892 456 401 703 640 510 402 489 753 310 429 505 116 129 429 740 032 328 462 241 379 310 463 789 062 500 478 750 000 000 059 843 750 000 Nominal
-100ppm 424 957 500 000 401 522 343 750 401 502 738 167 401 695 606 437 402 481 703 515 429 496 526 027 429 731 437 527 462 232 134 483 463 779 786 719 478 740 425 000 059 842 553 125 -20ppm
Table 14: 400ZR expansion rate table
www.oiforum.com 56
OIF-400ZR-01.0
13 Optical Specifications
The 400ZR optical parameters are organized by Application Code (defined in Table 15) for Tx, Rx, and the
Optical Channel (black link).
Ref. Application Description Minimum Reach Application Code - Name
120 km or less, amplified, point- 80 km

13.0.100 to-point, DWDM noise limited 0x01 – 400ZR, DWDM amplified
links.
Unamplified, single wavelength, 11dB loss budget minus 0x02 – 400ZR, Single wavelength,
13.0.110
loss limited links. link impairments Unamplified
Table 15: 400ZR application codes

Note: All specifications are defined after calibration and compensation, at EOL over temperature
and wavelength. All specifications are based on default grid spacing (defined in 13.1.110).
Bold italicized items found in tables indicate a reference to a Coherent Management Interface Spec[1]
(CMIS) defined function, state, or status condition.
13.1 400ZR, DWDM amplified - Application Code (0x01):

This section defines the optical parameters for the DWDM amplified application code (0x01).
13.1.1 Optical channel specifications – Black Link
Ref. Parameter Default Min Max Unit Conditions/Comments

Channel
13.1.100 193.7 191.3 196.1 THz
frequency
13.1.110 100 GHz See Section 15.1
13.1.111
Channel 75 GHz See Section 15.2
(optional)
spacing†
13.1.112
75 GHz See Section 15.3
(optional)
13.1.120 Post FEC BER 10-15 Pre-FEC BER 1.25E-2 or lower.
Single mode fiber. Specified for link
13.1.130 Fiber type G.652
budgeting purposes only.
13.1.140 Target reach 80 - km Amplified Link – Noise limited
www.oiforum.com 57
OIF-400ZR-01.0
Nominal central Nominal central

frequency minus Nominal central frequency frequency plus
Maximum Spectral Maximum Spectral
Excursion Excursion
Maximum
Ripple
-10Log(Loss) dB
13.1.150 Ripple 2.0 dB
0
Offset from nominal
Central Frequency
(GHz)
(See definition 13.3.4)

Chromatic Frequency dependent change in
13.1.160 0 2400 ps/nm
Dispersion phase velocity due to fiber.
Optical Return
13.1.161 24 dB (See definition 13.3.5)
Loss at Ss
Discrete
Reflectance
13.1.162 -27 dB (See definition 13.3.6)
between Ss
and Rs
Maximum
Instantaneous
13.1.170 Differential 28 ps (See definition 13.3.7)
Group Delay
(DGD)
Polarization
13.1.171 Dependent 2 dB (See definition 13.3.8)
Loss (PDL)
Polarization
13.1.172 50 krad/s (See definition 13.3.9)
Rotation Speed
(See definition 13.3.10)
Inter-channel
13.1.180 -8 dB Ref. Clause 9.6 and Fig 9-17 in ITU-T
Crosstalk at Rs
G.sup39.
Interferometric
13.1.181 -35 dB (See definition 13.3.11)
Crosstalk
Table 16: Optical channel specifications

†
For channel spacing of 100 GHz on a fiber, the allowed channel frequencies (in THz) are defined by 193.1 + n × 0.1
where n is a positive or negative integer including 0. For 400ZR modules, n = 30 to -17 in steps of 1. The specified 48
× 100GHz DWDM application channels are as defined in Section 15.1.
†
For channel spacing of 75 GHz or more on a fiber, the allowed channel frequencies (in THz) are defined by 193.1 +
3n × 0.025 where n is a positive or negative integer including 0. For 400ZR modules, 3n = 120 to -69. The reference
64 × 75 GHz DWDM application channels are as defined in Section 15.2.
www.oiforum.com 58
OIF-400ZR-01.0
†
For the flexible DWDM grid, the allowed frequency slots have a nominal central frequency (in THz) defined by 193.1
+ n × 0.00625 where n is a positive or negative integer including 0 and 0.00625 is the nominal central frequency
granularity in THz. Slot width is defined by 12.5 × m where m is a positive integer and 12.5 is the slot width granularity
in GHz. Any combination of frequency slots is allowed if no two frequency slots overlap. Example 100 GHz and 75
GHz DWDM applications with offset grid channels are defined in Section 15.3.
13.1.2 Transmitter Optical Specifications

Note: All Tx optical specifications are based on default grid spacing of 100GHz (see 13.1.110). For this
grid spacing no Tx shaping is required. If optical shaping is applied on Tx (e.g. for operation at other grid
settings) a matched Rx equalizer setting must also be applied on Rx or additional OSNR penalty may be
incurred.
Ref. Parameter Min Max Unit Conditions/Comments

Laser Offset from channel frequency set point.
13.1.200 frequency -1.8 1.8 GHz The receiver LO has the same frequency
accuracy accuracy.
Nominal central Nominal central
frequency minus Nominal central frequency frequency plus
Maximum Spectral Maximum Spectral
Excursion Excursion
0
-2 32 GHz 32 GHz
-4
Power Relative to Peak (dB)
-6
-8
-10
-12
-14
-16
-18
Tx Spectral Offset from nominal
13.1.201 32 GHz Central Frequency
Excursion (See definition 13.3.2)
Measured between the nominal central

frequency of the channel and the -3.0dB
points of the transmitter spectrum
furthest from the nominal central
frequency measured at point Ss.
Includes Laser frequency accuracy

(13.1.200) error value from nominal
center frequency.
www.oiforum.com 59
OIF-400ZR-01.0
Frequency 1- sided Noise power

[Hz] spectral density
Laser [Hz2/Hz]
See
13.1.210 frequency 1.0e+02 1.0e+11
Mask
noise 1.0e+04 1.0e+09
1.0e+06 1.0e+06
1.0e+07 6.0e+05
1.0e+08 1.6e+05
1.0e+09 1.6e+05
Mask does not apply to spurs.
Measurement Resolution BW shall be
between 10-1 and 10-6 of the frequency
of interest.
High frequency component of the phase

noise (100MHz and above) is consistent
with a 500 kHz laser line width. The
receiver LO has the same linewidth.
-145 0.2GHz < f < 10GHz Avg
13.1.212 Laser RIN dB/Hz
-140 0.2GHz < f < 10GHz Peak
www.oiforum.com 60
OIF-400ZR-01.0
PN [dBc/Hz] Frequency [Hz]

Tx clock phase
-100 1.00E+04
noise (PN):
See -120 1.00E+05
13.2.213a Maximum PN dBc/Hz
mask -130 1.00E+06
mask for low
frequency PN -140 1.00E+07
Phase noise, ℒ(𝑓),
fbaud
fc = = ~467.53 MHz
128
Mask does not apply to spurs,

broadband phase noise only. Spurs are
considered separately as per 13.1.213b
and 13.1.213c
www.oiforum.com 61
OIF-400ZR-01.0

rms random jitter:
ℒ(𝑓)
1 𝑓
𝜎𝑟𝑗 = 2𝜋𝑓 √2 ∙ ∫𝑓 2 10 10 𝑑𝑓
𝑐 1
rms periodic jitter (spurs):
1 𝑠𝑖
𝜎𝑝𝑗,𝑖 = ⋅ 1020
√2𝜋𝑓𝑐
Tx clock phase where,

noise (PN);
Maximum f1 = 10kHz,
total f2 = 10MHz,
13.1.213b
integrated 600 fs fbaud
RMS phase fc = = ~467.53MHz
128
jitter between ℒ(f) = phase noise (PN)
10kHz and 𝑠𝑖 = individual spur in [dBc]
10MHz
rms total jitter:
𝜎𝑡𝑗 = √𝜎𝑟𝑗 2 + ∑ 𝜎𝑝𝑗,𝑖 2

𝑖=1
where,
N = total number of spurs.
www.oiforum.com 62
OIF-400ZR-01.0

rms random jitter:
ℒ(𝑓)
1 𝑓
𝜎𝑟𝑗 = 2𝜋𝑓 √2 ∙ ∫𝑓 2 10 10 𝑑𝑓
𝑐 1
1 𝑠𝑖
𝜎𝑝𝑗,𝑖 = ⋅ 1020
√2𝜋𝑓𝑐
Tx clock phase
noise (PN): where,
Maximum
total f1 = 1MHz,
13.1.213c f2 = 200MHz,
integrated 250 fs
RMS phase 𝑓𝑏𝑎𝑢𝑑
fc = = 467.53𝑀𝐻𝑧,
jitter between 128
ℒ(f) = phase noise (PN),
1MHz and
𝑠𝑖 = individual spur in [dBc]
200MHz) Tx
rms total jitter:
𝑁
𝜎𝑡𝑗 = √𝜎𝑟𝑗 + ∑ 𝜎𝑝𝑗,𝑖 2

2
𝑖=1
where,
www.oiforum.com 63
OIF-400ZR-01.0

The baseband Tx spectral shape in this
excess bandwidth shall meet or exceed
the following conditions:
The magnitude of the spectrum in the
frequency range:
1 9
≤𝑓≤
2𝑇 16𝑇
shall meet
1 8 7
|𝐻(𝑓)| ≥ 𝐻(0)√ {1 + cos [8𝜋𝑇 (( )− )]} ,
2 15𝑇 16𝑇
1 8
≤ |𝑓| ≤
2𝑇 15𝑇
1 7
|𝐻(𝑓)| ≥ 𝐻(0)√ {1 + cos [8𝜋𝑇 (|𝑓| − )]} ,
2 16𝑇
Minimum 8 9
≤ |𝑓| ≤
Excess 15𝑇 16𝑇
13.1.215 12.5 % where T denotes the symbol period of the
Bandwidth1
(See Mask) signal.
Allowable
13.1.220 output signal -10 -6 dBm Measured at optical connector.
power window
Total output
13.1.221 power with Tx -20 dBm Tx Disable == false
disabled
Total output
power during Applicable to modules with tunable
13.1.222 -20 dBm
wavelength optics.
switching
www.oiforum.com 64
OIF-400ZR-01.0

Inband OSNR is defined within the
bandwidth of the Tx spectral excursion
Inband (IB) given in (13.1.201)
13.1.230 34 dB/0.1nm
OSNR The IB OSNR is referenced to an optical
bandwidth of 0.1nm @ 193.7 THz or 12.5
GHz.
Channel total power over peak noise
power in the whole frequency range
measured with 0.1 nm resolution
Out-of-band
13.1.231 23 dB/0.1nm bandwidth.
(OOB) OSNR
The OOB OSNR is referenced to an
optical bandwidth of 0.1nm @ 193.7 THz
or 12.5 GHz.
Transmitter
13.1.240 -20 dB Looking into the Tx
reflectance
Light reflected relative to Tx output
Transmitter
power back to transmitter while still
13.1.241 back reflection -24 dB
meeting Tx optical performance
tolerance
requirements.
Transmitter
polarization Power difference between X and Y
13.1.250 1.5 dB
dependent polarization.
power
13.1.260 X-Y Skew 5 ps
𝟐
𝐈𝐦𝐞𝐚𝐧 + 𝐐𝟐𝐦𝐞𝐚𝐧
DC I-Q offset 𝑷𝒆𝒙𝒄𝒆𝒔𝒔 =
𝐏𝐒𝐢𝐠𝐧𝐚𝐥
13.1.270a (mean per -26 dB
polarization)
𝑰𝑸𝒐𝒇𝒇𝒔𝒆𝒕 = 𝟏𝟎 𝐥𝐨𝐠 𝟏𝟎 (𝐏𝐞𝐱𝐜𝐞𝐬𝐬 )
Same formula definition as 13.1.270a,

however, any averaging period shall be <
I-Q
1us to be consistent with the timescales
13.1.270b instantaneous -20 dB
of Rx DSP operations. Specification
offset
applies at any point in time. Allows for
modulator bias controls/errors.
Mean I-Q
13.1.271 amplitude 1 dB
imbalance
I-Q phase
13.1.272 -5 +5 degrees
imbalance
13.1.273 I-Q Skew 0.75 ps
Table 17: Tx optical specifications
www.oiforum.com 65
OIF-400ZR-01.0
1
The minimum excess bandwidth is specified to guarantee multi-vendor clock recovery
interoperability. It is required because the Tx spectrum mask is not defined by this IA. For operation on
a 75 GHz grid this specification will be modified or removed and replaced by a Tx spectrum mask.
13.1.3 Receiver Optical Specifications
The receiver optical tolerance specifications include margin for Tx and line impairments.
Note: All Rx optical specifications are based on default grid spacing of 100GHz (see 13.1.110). When
operating at other grid settings additional compensation may be required or additional penalties may be
incurred.

Frequency
offset
13.1.300 between -3.6 +3.6 GHz Acquisition Range.
received
carrier and LO
Signal power of the channel at the
Input power
13.1.310 -12 0 dBm OSNR performance defined in
range
(13.1.330).
Input power needed to achieve post
Input
13.1.320 -12 dBm FEC BER per (13.1.120) when OSNR
sensitivity
Tolerance > (13.1.330).
At C-FEC threshold (ref. Section 10).
See Definition in Section 13.3.2
OSNR
13.1.330 26 dB/0.1nm The OSNR tolerance is referenced to
Tolerance
an optical bandwidth of 0.1nm @
193.7 THz or 12.5 GHz.
Optical return Optical reflectance at Rx connector
13.1.340 20 dB
loss input.
13.1.341 CD Tolerance 2400 ps/nm Tolerance to Chromatic Dispersion.
OSNR penalty tolerance over
Optical path
(13.1.330) due to interferometric
13.1.342 OSNR penalty 0.5 dB
crosstalk (13.1.181) and chromatic
tolerance
dispersion (13.1.160).
www.oiforum.com 66
OIF-400ZR-01.0

Tolerance to PMD with < 0.5 dB
penalty to OSNR sensitivity (13.1.330).
when change in SP is < 1 rad/ms.
10 ps of average PMD (DGD, SOPMD)
corresponds to:
PMD • 33 ps of DGDmax when
tolerance SOPMD = 0 ps2.
13.1.350 10 - ps
(DGD, • 272 ps2 of SOPMD when DGD
SOPMD) = 23.3 ps.
Due to the statistical nature of PMD
the DGDmax to DGDmean Ratio is
calculated at 3.3 (4.1 x 10-6 probability
that DGDmean being greater than
DGDmax).
Tolerance to peak PDL with < 1.3 dB
penalty to OSNR sensitivity (13.1.330)
Peak PDL
13.1.351 3.5 - dB when change in SOP is < =1 rad/ms.
tolerance
Test configuration - PDL emulator
applied before noise loading.
Tolerance to change in SOP with <0.5
Tolerance to dB additional OSNR penalty over all
13.1.352 50 - krad/s
change in SOP PMD and PDL values defined in
(13.1.350) and (13.1.351).
Tolerance to change in input power
with < 0.5 dB penalty to OSNR
Optical input sensitivity (13.1.330).
power When transient received power is
13.1.353 +/-2 - dB
transient within range defined by input power
tolerance range (13.1.310).
Rise/fall times of power change
defined by 20-80% of 50 us or slower.
Table 18: Rx optical specifications
1
Italicized items indicate a reference to a Coherent Management Interface Spec [1] (CMIS) defined
function, state, or status condition.
www.oiforum.com 67
OIF-400ZR-01.0
13.1.4 Module Requirements Tx - (Informative)

The following specifications provide guidance for modules based on the 400ZR IA.

The maximum transmitter
turn-off time from any
condition that results in
Tx_Disable==true to reach
Transmitter laser disable
13.1.400 100 ms the Tx output power given by
time
(13.1.221).
Rx shall remain locked and
thus LO must remain
enabled.
The maximum time from
Transmitter turn-up time
13.1.410 180 s ModuleLowPwr to
from warm start
DataPathActivated state.
The maximum time from de-
Transmitter turn-up time assertion of ResetS == false to
13.1.411 200 s
from cold start DataPathActivated state
while LoPwrS == false.
The maximum time to change
wavelengths including turn-
Transmitter wavelength
13.1.420 180 s up time.
switching time
Applicable to modules with
tunable optics.
Total output power
measurement including all
Output power monitor - ASE contribution.
13.1.430 -2.0 2.0 dB
Accuracy Measurement accuracy does
not contribute to allowable
output power signal window.
Table 19: 400ZR module – Tx specifications
www.oiforum.com 68
OIF-400ZR-01.0
13.1.5 Module Requirements Rx - (Informative)

Upon Rx_LOS de-assert,
Receiver turn-up time
13.1.510 10 s Receiver has been turned
from warm start
up previously.
From module reset, with
Receiver turn-up time
13.1.511 200 s valid optical input signal
from cold start
present.
Over the superset of
input power range
(13.1.310), receiver
sensitivity 13.1.320) and
the optical Rx_LOS Assert
Input total power
13.1.530 -4.0 4.0 dB threshold range (13.1.532
monitor - Accuracy
assuming Min accuracy –
i.e. Real input total power
range of 0dBm to -14dBm
at the default Optical
Rx_LOS Assert Threshold.
The module reports the
Input Channel power
13.1.531 -4.0 4.0 dB channel power as
monitor - Accuracy
received by the module.
Optical Rx_LOS Assert
13.1.532 -18 -20 -16 dBm Channel Power.
Threshold†
Optical Rx_LOS
13.1.533 1.0 2.5 dBm Rx_LOS cleared.
Hysteresis
Table 20: 400ZR module – Rx specifications
†
If a module supports both amplified and unamplified use cases, Optical Rx_LOS thresholds must be
programmable to support different ranges for each application.
www.oiforum.com 69
OIF-400ZR-01.0
13.2 400ZR, Single wavelength, Unamplified - Application Code (0x02):

This section defines the optical parameters for application code 0x02.
13.2.1 Optical channel specifications
13.2.100 Channel frequency 193.7 THz
Pre-FEC BER 1.25E-2 or
13.2.120 Post FEC BER 10-15
lower.
13.2.130 Fiber type G.652 Single mode fiber.
Frequency dependent
Chromatic
13.2.160 0 1200 ps/nm change in phase
Dispersion
velocity due to fiber
Optical Return
13.2.161 24 dB (See definition 13.3.5)
Loss at Ss
Discrete
13.2.162 Reflectance -27 dB (See definition 13.3.6)
between Ss and Rs
Maximum
Instantaneous
13.2.170 16 ps (See definition 13.3.7)
Differential Group
Delay (DGD)
Polarization
13.2.172 50 krad/s (See definition 13.3.9)
Rotation Speed
Table 21: Optical channel specifications
www.oiforum.com 70
OIF-400ZR-01.0
13.2.2 Transmitter Optical Specifications

Laser Offset from channel frequency set point.
13.2.200 frequency -1.8 1.8 GHz The receiver LO has the same frequency
accuracy accuracy
Frequency Frequency
[Hz] Noise [Hz2/Hz]
Laser 1.0e+02 1.0e+11
See
13.2.210 frequency 1.0e+04 1.0e+09
Mask
noise 1.0e+06 1.0e+06
1.0e+07 6.0e+05
1.0e+08 3.2e+05
1.0e+09 3.2e+05
Mask does not apply to spurs.
Measurement Resolution BW shall be
between 10-1 and 10-6 of the frequency of
interest.
High frequency component of the phase

noise (100MHz and above) is consistent
with a 1 MHz laser line width. The receiver
LO has the same linewidth.
-145 0.2GHz < f < 10GHz - Avg
13.2.212 Laser RIN dBc/Hz
-140 0.2GHz < f < 10GHz - Peak
www.oiforum.com 71
OIF-400ZR-01.0
Tx clock
phase noise PN [dBc/Hz] Frequency [Hz]
(PN): -100 1.00E+04
Maximum See -120 1.00E+05
13.2.213a dBc/Hz
PN mask for mask -130 1.00E+06
low -140 1.00E+07
frequency
PN Phase noise, ℒ(𝑓),
fbaud
fc = = ~467.53 MHz
128
Mask does not apply to spurs, broadband

phase noise only. Spurs are considered
separately as per 13.2.213b and 13.2.213c
www.oiforum.com 72
OIF-400ZR-01.0

rms random jitter:
ℒ(𝑓)
1 𝑓
𝜎𝑟𝑗 = 2𝜋𝑓 √2 ∙ ∫𝑓 2 10 10 𝑑𝑓
𝑐 1
1 𝑠𝑖
𝜎𝑝𝑗,𝑖 = ⋅ 1020
√2𝜋𝑓𝑐
Tx clock
phase noise where,
(PN);
Maximum f1 = 10kHz,
total f2 = 10MHz,
13.2.213b
integrated 600 fs fbaud
RMS phase fc = = ~467.53MHz
128
jitter ℒ(f) = phase noise (PN)
between 𝑠𝑖 = individual spur in [dBc]
10kHz and
10MHz rms total jitter:
𝜎𝑡𝑗 = √𝜎𝑟𝑗 2 + ∑ 𝜎𝑝𝑗,𝑖 2

𝑖=1
where,
www.oiforum.com 73
OIF-400ZR-01.0

rms random jitter:
ℒ(𝑓)
1 𝑓
𝜎𝑟𝑗 = 2𝜋𝑓 √2 ∙ ∫𝑓 2 10 10 𝑑𝑓
𝑐 1
1 𝑠𝑖
𝜎𝑝𝑗,𝑖 = ⋅ 1020
Tx clock √2𝜋𝑓𝑐
phase noise
(PN): where,
Maximum
total f1 = 1MHz,
13.2.213c f2 = 200MHz,
integrated 250 fs
RMS phase 𝑓𝑏𝑎𝑢𝑑
fc = = 467.53𝑀𝐻𝑧,
jitter 128
ℒ(f) = phase noise (PN),
between
𝑠𝑖 = individual spur in [dBc]
1MHz and
200MHz) Tx rms total jitter:
𝑁
𝜎𝑡𝑗 = √𝜎𝑟𝑗 + ∑ 𝜎𝑝𝑗,𝑖 2

2
𝑖=1
where,
www.oiforum.com 74
OIF-400ZR-01.0

The baseband Tx spectral shape in this
excess bandwidth shall meet or exceed the
following conditions:
The magnitude of the spectrum in the
frequency range:
1 9
≤𝑓≤
2𝑇 16𝑇
shall meet
1 8 7
|𝐻(𝑓)| ≥ 𝐻(0)√ {1 + cos [8𝜋𝑇 (( )− )]} ,
2 15𝑇 16𝑇
1 8
≤ |𝑓| ≤
2𝑇 15𝑇
1 7
|𝐻(𝑓)| ≥ 𝐻(0)√ {1 + cos [8𝜋𝑇 (|𝑓| − )]} ,
2 16𝑇
Minimum 8 9
≤ |𝑓| ≤
15𝑇 16𝑇
Excess
13.2.215 12.5 %
Bandwidth1 where T denotes the symbol period of the
(See Mask) signal.
Allowable
output signal
13.2.220 -9 0 dBm Measured at optical connector.
power
window
Total output
13.2.221 power with -20 dBm Tx Disable == false
Tx disabled
The 0.1nm bandwidth for the IB OSNR
Inband (IB) dB/0.1n
13.2.230 34 refers to 193.7 THz or 12.5 GHz optical
OSNR m
bandwidth.
Transmitter
13.2.240 -20 dB Looking into the Tx
reflectance
www.oiforum.com 75
OIF-400ZR-01.0

Transmitter
Light reflected relative to Tx output power
back
13.2.241 -24 dB to transmitter while still meeting Tx optical
reflection
performance requirements.
tolerance
Transmitter
polarization Power difference between X and Y
13.2.250 1.5 dB
dependent polarization.
power
13.2.260 X-Y Skew 5 ps
𝟐
𝐈𝐦𝐞𝐚𝐧 + 𝐐𝟐𝐦𝐞𝐚𝐧
DC I-Q offset 𝑷𝒆𝒙𝒄𝒆𝒔𝒔 =
𝐏𝐒𝐢𝐠𝐧𝐚𝐥
13.2.270a (mean per -26 dB
polarization)
𝑰𝑸𝒐𝒇𝒇𝒔𝒆𝒕 = 𝟏𝟎 𝐥𝐨𝐠 𝟏𝟎 (𝐏𝐞𝐱𝐜𝐞𝐬𝐬 )
Same formula definition as 13.2.270a,

however, any averaging period shall be <
I-Q
1us to be consistent with the timescales of
13.2.270b instantaneou -20 dB
Rx DSP operations. Specification applies at
s offset
any point in time. Allows for modulator bias
controls/errors.
I-Q
13.2.271 amplitude 1 dB
imbalance
I-Q phase
13.2.272 -5 +5 degrees
imbalance
13.2.273 I-Q Skew 0.75 ps
Table 22: Tx Optical specifications
1
The minimum excess bandwidth is specified to guarantee multi-vendor clock recovery
interoperability. It is required because the Tx spectrum mask is not defined by this IA.
www.oiforum.com 76
OIF-400ZR-01.0
13.2.3 Receiver Optical Specifications

Frequency offset
13.2.300 between received -3.6 +3.6 GHz Acquisition Range.
carrier and LO
13.2.310 Input power range -20 0 dBm Signal power of the channel.
Input power needed to achieve post
13.2.320 Input sensitivity -20 dBm FEC BER per (13.2.120) when Inband
(IB) OSNR > (13.2.230).
13.2.340 Optical return loss 20 dB Optical reflectance at connector input.
13.2.341 CD Tolerance 1200 ps/nm Tolerance to Chromatic Dispersion.
Rx Sensitivity penalty over (13.2.320)
Optical path
13.2.342 0.5 dB due to reflections and the combined
power penalty
effects of dispersion (13.2.341).
Tolerance to PMD with < 0.5dB Rx
sensitivity penalty (13.2.320) when
change in SOP is <= 1 rad/ms.
7 ps of average PMD (DGD, SOPMD)
corresponds to:
• 23 ps of DGDmax when
Average PMD
SOPMD = 0 ps2.
13.2.350 (DGD, SOPMD) 7 - ps
• 132 ps2 of SOPMD when DGD
tolerance
= 16.3 ps.
Due to the statistical nature of PMD
the DGDmax to DGDmean Ratio is
calculated at 3.3 (4.1 x 10-6 probability
that DGDmean being greater than
DGDmax).
Tolerance to change in peak PDL with <
Peak PDL
13.2.351 1.5 - dB 0.4dB Rx sensitivity penalty (13.2.320)
tolerance
when change in SOP is <= 1 rad/ms.
Tolerance to change in SOP with <0.3
Tolerance to dB additional power penalty over all
13.2.352 50 - krad/s
change in SOP PMD and PDL values defined in
(13.2.350) and (13.2.351).
Table 23: Rx Optical specifications
www.oiforum.com 77
OIF-400ZR-01.0
13.2.4 Module Requirements Tx - (Informative)


The maximum transmitter
turn-off time from any
condition that results in
Tx_Disable==true to reach
Transmitter laser disable the Tx output power given by
13.2.400 100 ms
time (13.2.221).
Rx shall remain locked and

thus LO must remain
enabled.
The maximum time from
Transmitter turn-up time
13.2.410 180 s ModuleLowPwr to
from warm start
DataPathActivated state.
The maximum time from de-
Transmitter turn-up time assertion of ResetS == false to
13.2.411 200 s
from cold start DataPathActivated state
while LoPwrS == false.
Total output power
measurement including all
Output power monitor - ASE contribution.
13.2.430 -2.0 2.0 dB
Accuracy Measurement accuracy does
not contribute to allowable
output power signal window.
Table 24: 400ZR module – Tx specifications
www.oiforum.com 78
OIF-400ZR-01.0
13.2.5 Module Requirements Rx - (Informative)


Receiver turn-up Upon Rx_LOS de-assert,
13.2.510 time from warm 10 s Receiver has been
start turned up previously.
Receiver turn-up From module reset,
13.2.511 time from cold 200 s with valid optical input
start signal present.
Over the superset of
input power range
(13.2.310), receiver
sensitivity (13.2.320),
and the optical Rx_LOS
Input Channel or
Assert threshold range
Total power
13.2.531 -4.0 4.0 dB (13.2.532) assuming
monitor -
Min accuracy (i.e. real
Accuracy
input total power range
of 0dbm to -22dBm at
the default Optical
Rx_LOS Assert
Threshold).
Optical Rx_LOS Channel or Total Input
13.2.532 -26 -28 -24 dBm
Assert Threshold† Power.
Optical Rx_LOS
13.2.533 1.0 2.5 dBm Rx_LOS cleared.
Hysteresis
Table 25: 400ZR module – Rx specifications
†
If a module supports both amplified and unamplified use cases, Optical Rx_LOS thresholds must be
programmable to support different ranges for each application.
www.oiforum.com 79
OIF-400ZR-01.0
13.3 Optical Parameter Definitions

13.3.1 The Receiver Optical Signal-to-noise Ratio Tolerance
The receiver OSNR tolerance is defined as the minimum value of OSNR (referred to 0.1 nm @193.7 THz or
12.5 GHz) that can be tolerated while maintaining the maximum BER of the application. This must be met
for all powers between the maximum and minimum mean input power with a transmitter with worst-
case values of:
• Transmitter optical return loss,
• Receiver connector degradations
• Measurement tolerances
The receiver OSNR tolerance does not have to be met in the presence of chromatic dispersion, non-linear
effects, reflections from the optical path, PMD, and PDL or optical crosstalk. These effects are specified
separately but contribute to total optical path OSNR penalty.
System integrators need to account for these path penalties when evaluating network performance.
13.3.2 Spectral excursion
Spectral excursion is defined as the difference between the nominal central frequency of the channel
and the -3.0 dB points of the transmitter spectrum furthest from the nominal central frequency
measured at point Ss.
13.3.3 Out-of-Band OSNR (OOB OSNR)
Out-of-Band OSNR (OOB OSNR) is the ratio of the peak transmitter power to the integrated power
outside the transmitter spectral excursion. The spectral resolution of the measurement shall be better
than the maximum spectral width of the peak.
13.3.4 Ripple
Ripple is defined as the peak-to-peak difference in insertion loss between the input and output ports
of the black link over that channel in the frequency (or wavelength) range of the channel +/- the
maximum spectral excursion.
13.3.5 Optical return loss at SS
Reflections are caused by refractive index discontinuities along the optical path. If not controlled, they
can degrade system performance through their disturbing effect on the operation of the optical source,
or through multiple reflections which lead to interferometric noise at the receiver. Reflections from the
optical path are controlled by specifying the:
• minimum optical return loss of the cable plant at the source reference point (Ss), including
any connectors; and
• maximum discrete reflectance between source reference point (Ss) and receive reference
point (Rs)
Reflectance denotes the reflection from any single discrete reflection point, whereas the optical return
loss is the ratio of the incident optical power to the total returned optical power from the entire fiber
including both discrete reflections and distributed backscattering such as Rayleigh scattering.
www.oiforum.com 80
OIF-400ZR-01.0
13.3.6 Discrete reflectance between SS and RS

Optical reflectance is defined to be the ratio of the reflected optical power present at a point, to the
optical power incident to that point. The maximum number of connectors or other discrete reflection
points which may be included in the optical path must be such as to allow the specified overall optical
return loss to be achieved.
13.3.7 Differential Group Delay (DGD)
Differential group delay (DGD) is the time difference between the fractions of an optical signal transmitted
in the two principal states of polarization. For distances greater than several kilometers, and assuming
random (strong) polarization mode coupling, DGD in a fiber can be statistically modelled as having a
Maxwellian distribution.
Due to the statistical nature of polarization mode dispersion (PMD), the relationship between maximum
instantaneous DGD and mean DGD can only be defined probabilistically. The probability of the
instantaneous DGD exceeding any given value can be inferred from its Maxwellian statistics.
For purposes of this IA the ratio of maximum instantaneous DGD to mean DGD is defined as 3.3,
corresponding to the probability of exceeding the maximum DGD 4.1 × 10-6.
13.3.8 Polarization Dependent Loss (PDL)
The polarization dependent loss (PDL) is the difference (in dB) between the maximum and minimum
values of the channel insertion loss (or gain) of the black link from point Ss to Rs due to a variation of the
State Of Polarization (SOP) over all state of polarizations.
13.3.9 Polarization rotation speed
The polarization rotation speed is the rate of rotation in Stokes space of the two polarizations of the
optical signal at point RS measured in krad/s.
13.3.10 Inter-channel crosstalk
Inter-channel crosstalk is defined as the ratio of total power in all the disturbing channels to that in the
wanted channel, where the wanted and disturbing channels are at different wavelengths.
Specifically, the isolation of the link shall be greater than the amount required to ensure that when any
channel is operating at the minimum mean output power at point Ss and all of the others are at the
maximum mean output power, then the inter-channel crosstalk at the corresponding point Rs is less than
the maximum inter-channel crosstalk value.
13.3.11 Interferometric crosstalk
Interferometric crosstalk is defined as the ratio of the disturbing power to the wanted power within a
single channel, where the disturbing power is the power (not including ASE) within the optical channel
that would remain if the wanted signal were removed from the link while leaving all of the other link
conditions the same.
Specifically, the isolation of the link shall be greater than the amount required to ensure that when any
channel is operating at the minimum mean output power at point Ss and all of the others are at the
maximum mean output power, then the interferometric crosstalk at the corresponding point Rs is less
than the maximum interferometric crosstalk value.
www.oiforum.com 81
OIF-400ZR-01.0
14 Interoperability Test Methodology, Definitions

Interoperability is achievable by complying with all required aspects of this IA. Digital datapath verification
is measured through a combination of interoperability Test Vectors and the use of common sets of test
generators and checkers. The generators and checkers can be configure using looped back pairs for self-
testing or in a cross-linked configuration.
Optical interworking is achieved through strict adherence to the discrete Tx/Rx optical specifications over
a compliant channel (ref Section 13). Error Vector Magnitude Testing (Section 20, Appendix C) is intended
for future integration to the normative sections of this IA.
14.1 400ZR Test Features
To verify the design for interoperability, a full set of test vectors is made available to OIF member
companies. Lower level diagnostic capabilities in the form of loopbacks and insertion points for test
generators/checkers is also described in Section 14.2.
14.2 Loopback features, Test Generators and Checkers
Figure 39 shows the various diagnostic and test capabilities overlaid on the data path. Generators and
checkers are provided and can be used in conjunction with the loopbacks for self-diagnostic, or they can
be used in conjunction with external test equipment to verify the data path.
www.oiforum.com 82
OIF-400ZR-01.0
8 400GBASE-R (400GAUI-8) 8
PMA sublayer(*)
16
Per IEEE 802.3TM-2018
Host Side Rx
Alignment lock and lane Loopback
16
deskew Host Side Tx
Loopback
Lane reorder and de- TX Datapath

Distribution & Interleave
interleave
RX Datapath
FEC Decode FEC Encode

RS-FEC
detected Local PHY 400GXS
Degrade
Post-FEC Interleave Pre-FEC Distribution
Rx Client rx_am_sf tx_am_sf Tx Client

FEC <2:0> <2:0> FEC
Alignment Removal Alignment Insertion
Degrade Degrade
signaling signaling
Descramble Scramble
256b/257b Transcode 400GAUI-8

Reverse Transcode
PCS Decode & No rate No rate PCS PMA
Checker matching matching Encode Generator
Error Marking PHY 400GXS
400GMII*
GMP PCS No rate No rate Decode & GMP PCS PCS

Generator Encode matching Checker
matching Error Marking PMA PHY 400ZR
256b/257b transcode Rx Modem Loopback Reverse transcode PMD
256b/257b blocks System-side CLOCK DOMAIN 256b/257b blocks MDI
(+/- 100 ppm)
GMP Mapping GMP de-mapping MEDIUM
(+/- 20 ppm)
tx_oh_sf rx_oh_sf
<2:0> OH/AM insertion AM/OH detect & removal <2:0> 400ZR
(20*257b) Tx Modem Loopback (20*257b)
*Logical Equivalent Implementaion. The rate matching functions shall
TV PRBS be disabled to/from the 400GMII interface in both directions. No Idle
TV PRBS CRC-32 Calculation CRC-32 checking Checker deletion/insertion shall be performed.
Generator

SC-FEC
detected Local
Pad Insertion Pad removal Degrade
(6*119b) (6*119b)
Scramble Descramble
Convolution de-
interleaving
128b/119b SD-FEC 128b/119b SD-FEC

Hamming encode Hamming decode
Gray mapping, symbol Pol combining & Symbol

Interleaving, Pol distribution de-interleaving
2 2
FAW & TS Insertion Align C-FEC &
FAW/TS removal
2
2
Pilot Insertion Pilot removal

2
2
EVM PMA (ZR)
PRBS TX DSP Media Side Tx RX DSP
Generator Loopback
Media Side Rx
2 2
Loopback
DAC ADC
Photonic Interface PMD (ZR) Photonic Interface
Figure 39: 400ZR test features

www.oiforum.com 83
OIF-400ZR-01.0
14.2.1 Loopbacks
A 400ZR module must be capable to minimally support one of the following loopback sets. The sets are
defined such that when two 400ZR modules are cross-connected over a black link a near-end and a far-
end loopback path exists across the black link. The CMIS supported loopback modes are shown in Italic.
Each set has 1 Rx path and 1 Tx path.
• Modem Tx loopback + Modem Rx Loopback

• Modem Tx loopback + Host Side Rx Loopback
• Media Side Tx loopback + Modem Rx Loopback
• Host Side Tx loopback + Host Side Rx Loopback
The specific loopback mode enabled must be coordinated at each end of the link by each host.
The following loopback modes are defined:
Loopback Name Description

Host Side Tx Loopback Loopback after Alignment lock and lane De-skew → PMA sublayer.
Host loop timed.
Modem Tx Loopback Loopback after GMP mapping → GMP De-mapping. Data re-
transmitted relative to local clock
Media Side Tx Loopback Loopback after Tx DSP processing blocks and before Rx DSP
processing blocks
Media Side Rx Loopback Loopback in DSP. After polarity split and symbol de-interleave →
Grey mapper, symbol Interleave. Media loop timed.
Modem Rx Loopback Loopback after GMP De-mapping → GMP mapping. Data
retransmitted relative to local clock.
Host Side Rx Loopback Loopback after distribution/interleaving block on host ingress path,
and before lane reorder and interleave
Table 26: Loopbacks
www.oiforum.com 84
OIF-400ZR-01.0
14.2.2 Test Generators/Checkers

The test generators and checker requirements are described below: Required modes are highlighted with
Bold text.
Generator/Checker Type Description

EVM PRBS Tx Generator only - Used for EVM
• PRBS-7 - Optional
TV PRBS Tx Generator, Rx Checker, ZR Frame replacement
to/from SC-FEC, Used for Test Vectors, and FEC
characterization.
• PRBS-31 - Required
GMP PCS Tx Generator, Rx Checker. PCS Test pattern. Data
retransmitted relative to local clock.
PCS Rx Generator, Tx Checker. PCS Test pattern: IEEE
Std 802.3™-2018 Clause 119.2.4.9 idle control blocks
(block type 0x1E)
Table 27: Test generator/checker descriptions

14.3 Interoperability Test Vectors
The Interoperability generators/checkers are primarily used during design development (e.g. simulation).
Test vectors are used to guarantee the design integrity and the datapath interoperability.
14.3.1 EVM PRBS
The EVM PRBS is intended for EVM measurements. The EVM PRBS will overwrite all transmit symbols. No
specific algorithm, synchronization or seed is required.
www.oiforum.com 85
OIF-400ZR-01.0
14.3.2 TV PRBS
The TV PRBS is used for validating C-FEC/DSP framing, symbol mapping, and FAW/TS/PS insertion. The
required PRBS31 is per IEEE 802.3 with initial state being all ’s.
• Generation/checking is to/from the media interface (see Figure 39).
• The PRBS test vector generator is inserted in the Tx data path after the GMP mapper. Test vector
generation data is a PRBS31 sequence replacing the entire 400ZR frame.
• The TV PRBS test vector checker is inserted in the Rx data path before the GMP de-mapper. The
TV PRBS checker shall recover and verify the PRBS31 sequence.
• The TV PRBS test vector generator can be looped back to the TV PRBS test vector checker as a
self-test.
Figure 40: Test vector PRBS31 generator
The TV PRBS test vector files are attached in Table 28.
Order Polynomial Seed value Test Vector File
31 Z31+Z28+1 No seed value required.

testVector.txt
Table 28: Test vector PRBS files

14.3.3 GMP PCS test vectors
The GMP PCS Test Vectors are used for validating the PCS (ZR) datapath. This includes the GMP mapping
process, C-FEC generation and DSP framing. The GMP PCS generator inserts a continuous stream of Idle
control characters /I/ per IEEE Std -2018 Clause 119.2.3.5 prior to GMP mapping on the Tx
datapath. The checker is after the GMP de-mapper on the Rx datapath. See Figure 39.
GMP PCS Test vectors should be longer than 26 super frames. The test vector attached in this document
have a length of 52 super frames. However, 52 super frames may not be enough length to find GMP
stuffing event. The vector of 256 400ZR frames (as input of C-FEC) is also included.
Cm OH value may be mismatched due to Cm fluctuation (between 10215 min. and 10216 max.) depending
on the ppm offset and the initialization process of the Cm calculation.
www.oiforum.com 86
OIF-400ZR-01.0
Reserved symbols in the super frame are set to (0,0) for the test vectors. Although these symbols are
permitted for the proprietary usage, these symbols must be mapped with the following considerations:
• Randomized,
• DC Balanced,
• Low cross correlation on the symbol stream of TS, FAW and RES
The GMP PCS test vector files are attached in Table 29.
Description Test Vector File
Readme
Readme.txt
Idle test pattern into 400ZR frame

400ZR_PCS_test_FlexO_out.txt
Test Pattern into DSP

400ZR_PCS_test_symbol_out.txt
Table 29: GMP PCS test vector files

14.3.1 PCS test vectors
Generation/checking is to/from host interface (see Figure 39).
The PCS test vector generator is inserted in the Rx data path after the GMP de-mapper. The test pattern
is based on IEEE Std -2018 Clause 119.2.4.9 (Idle Insert). Downstream logic in the 400ZR data path
shall support transcoding, scrambling, PCS alignment marker insertion and RS(544,514) FEC
encapsulation. The host loop Rx data path vector check monitor point is at the 400GBASE-R PMA sublayer.
The PCS test vector checker is inserted in the Tx data path before the GMP mapper. The test pattern used
within the 400GBASE-R PCS sublayer is IEEE Std -2018 Clause 119.2.4.9 idle control blocks (block
type 0x1E). Downstream logic in the 400ZR data path shall support RS(544,514) FEC termination, PCS de-
skew, descrambling. The host loop Tx data path vector check monitor is pre GMP mapping. The expected
string is per IEEE Std -2018 Clause 119.2.3.5 (Idle Insert).
Once the host loop Rx and Tx data path are confirmed the PCS test vector generator can be looped back
to the PCS test vector checker as a self-test.
www.oiforum.com 87
OIF-400ZR-01.0
14.3.1 Media loop testing

Test vector generation/checking shall be run on the complete data path bypassing the on-board test
vector generators/checkers to verify end-to-end interoperability. Test vector generation/checking in this
case is done at both the media and host interfaces.
Host Loop
Descramble Scramble
Reverse Transcode TX_PHY_FACILITY 256b/257b Transcode

LOOPBACK
Decode & No rate No rate
matching matching Encode
Error Marking
PCS Test Vector

PCS Test Vector Checker
Generator
No rate No rate Decode &

Encode matching
matching Error Marking
256b/257b transcode Reverse transcode
256b/257b blocks System-side CLOCK DOMAIN 256b/257b blocks

(+/- 100 ppm)
GMP Mapping GMP de-mapping
(4*257b stuffing) (4*257b de-stuffing)
Network-side CLOCK DOMAIN
(+/- 20 ppm)
Line side PRBS
Line side PRBS Checker
generator
Media Loop
Figure 41: Test vector detail
www.oiforum.com 88
OIF-400ZR-01.0
15 Operating frequency channel definitions

15.1 Normative 48 x 100 GHz DWDM Application Channels.
Application Code (0x1) requires 400ZR modules at the 48 frequencies provided in Table 30. The Channel
spacing in Table 30 is based on ITU-T G 694 Section 6 “Fixed grid nominal central frequencies for dense
WDM systems”
index n (from ITU-T G.694.1) freq. [THz]

1 30 196.100
2 29 196.000
3 28 195.900
…
46 -15 191.600
47 -16 191.500
48 -17 191.400
Table 30: 100GHz channel spacing
15.2 Optional 64 x 75 GHz DWDM Application Channels
The preferred optional Application Code (0x1) 75 GHz Channel spacing in Table 31 is based on ITU-T
G.694 Section 6 “Fixed grid nominal central frequencies for Dense WDM systems ”
index n (from ITU-T G.694.1) freq. [THz]

1 120 196.100
2 117 196.025
3 114 195.950
…
62 -63 191.525
63 -66 191.450
64 -69 191.375
Table 31: 75GHz channel spacing
www.oiforum.com 89
OIF-400ZR-01.0
15.3 Optional Flexible DWDM Grid

Flexible DWDM grids are defined in ITU-T G 694 Section 7 “Flexible DWDM grid definition”, where center
frequencies are determined by,
(193.1 + n × 0.00625) THz
and each channel occupies a slot width,
12.5 GHz × m
Such that adjacent channel’s frequency slots differ in n by,
∆n = 2 x m
There are two example grids in Section 15.3.1 and 15.3.2 that allow the maximum channel fill with 100
GHz and 75 GHz spaced channels on a previously defined spectrum of 96 50 GHz spaced channels. This
allows the re-use of previously designed and deployed DWDM hardware such as WSS and amplifier
components without wasting spectrum.
The offset in the grids in Table 32 and Table 33 relative to Table 30 and Table 31 allow for channel plans
(e.g. 48 channels spaced at 100 GHz or 64 Channels spaced at 75 GHz), which are edge-aligned to the band
of 96 channels spaced at 50 GHz where the center frequencies are not offset.
Figure 42: Flexible grid provisioning example
www.oiforum.com 90
OIF-400ZR-01.0
15.3.1 Example 100GHz Flexible Grid Offset

This 48-channel grid has 100 GHz frequency slots corresponding to m = 8. The grid is offset by 25 GHz
from the normative fixed 100 GHz grid defined in Section 15.1. The following table shows the values of n
in steps of ∆n = 16, and an offset at 193.1 THz of n = -4 which corresponds to -25 GHz.
n (from ITU-T G.694.1 Sect. 7) center freq. [THz]

476 196.075
460 195.975
444 195.875
…
12 193.175
-4 193.075
-20 193.975
Table 32: Example 100GHz flexible grid
15.3.2 Example 75GHz Flexible Grid Offset
This 64-channel grid has 75 GHz frequency slots corresponding to m = 6. The grid is offset by 12.5 GHz
from fixed 75 GHz grid defined in Section 15.2. The following table shows the values of n in steps of ∆n =
12, and an offset at 193.1 THz of n = -2 which corresponds to -12.5 GHz.
n (from ITU-T G.694.1 Sect. 7) center freq. [THz]

478 196.0875
466 196.0125
454 195.9375
…
10 193.1625
-2 193.0875
-14 193.0125
…
-254 191.5125
-266 191.4375
-278 191.3625
Table 33: Example 75GHz flexible grid
15.3.3 Flexible Grid Provisioning
The 400ZR coherent MIS IA [1] defines an alternate frequency provisioning model to allow a provisioning
method for flexible DWDM grid systems. This alternative frequency provisioning model allows a direct
setting of the frequency (i.e. no grid limitations) once enabled by a control register. The grid provisioning
model remains the default model.
www.oiforum.com 91
OIF-400ZR-01.0
16 Summary
This 400ZR IA specifies the requirements of a 400GBASE-R PHY. The 400ZR PHY provides timing and code-
word transparent transmission of a 400GBASE-R host signal over a single carrier optical interface (Black
Link) with less than 1.0E-15 bit-errors. This coherent interface uses DP-16QAM, non-differential phase
encoding/decoding, and a Concatenated FEC (C-FEC). The two application codes defined for this IA are:
• 120 km or less, amplified, point-to-point, DWDM noise limited links.
• Unamplified, single wavelength, loss limited links.
No restrictions are placed on the physical form factor by this IA. This 400ZR IA builds upon the work of
other standards bodies including IEEE 802.3TM-2018 and ITU-T SG-15.
www.oiforum.com 92
OIF-400ZR-01.0
17 References
17.1 Normative references
[1] Implementation Agreement for Coherent CMIS, IA # oif2019.015.06
[2] Standard for Ethernet: IEEE Std 802.3TM-2018
[3] ITU-T G.709/Y.1331 (2019), Amendment 3, Interfaces for the optical transport network.
[4] ITU-T G.709.1/Y.1331.1 (2018), Flexible OTN short-reach interfaces.
[5] ITU-T G.709.2/Y.1331.2 (2018), OTU4 long-reach interfaces.
[6] ITU-T G.709.3/Y.1331.3 (2018), Flexible OTN long-reach interfaces.
[7] ITU-T G.sup39 (02/2016), Optical system design and engineering considerations.
17.2 Informative references
[8] ITU-T G.694.1 (2012): Spectral grids for WDM applications: DWDM Frequency grid.
[9] EIC/TR 61282-10, Ed. 1.0, 201: Fibre optic communication system design guides- Part 10:
Characterization of the quality of optical vector-modulated signals with the error vector
magnitude ”
www.oiforum.com 93
OIF-400ZR-01.0
18 Appendix A: Glossary
Acronym Definition Acronym Definition

AM Alignment Marker NA Not Applicable
BER Bit Error Ratio NCG Net Coding Gain
CD Chromatic Dispersion OADM Optical Add/Drop
Multiplexer
C-FEC Concatenated FEC OSNR Optical Signal-to-Noise
(Staircase FEC + Ratio
Hamming)
DGD Differential Group PDL Polarization Dependent
Delay Loss
DP-mQAM Dual Polarization – m PMD Polarization Mode
state Quadrature Dispersion
Amplitude Modulation
DSP Digital Signal QAM Quadrature Amplitude
Processing Modulation
DWDM Dense Wavelength- Rs Single-Channel
Division Multiplexing Reference point at the
DWDM network
element tributary
output
EOL End of Life SC-FEC StairCase FEC
EVM Error Vector SD-FEC Soft-Decision FEC
Magnitude
FEC Forward Error Ss Single-Channel
Correction Reference point at the
DWDM network
element tributary input
FFS For Further Study SNR Signal-to-Noise Ratio
GMP Generic Mapping SOP State of Polarization
Procedure
HD-FEC Hard-Decision FEC SOPMD Second Order
Polarization Mode
Dispersion
IA Implementation TBD To Be Decided
Agreement
LD Local Degrade WDM Wavelength-Division
Multiplexing
LO Local Oscillator WSS Wavelength Selective
Switching
LOS Loss of Signal
Table 34: Acronyms
www.oiforum.com 94
OIF-400ZR-01.0
19 Appendix B: Future work items

These items below will be considered as part of a maintenance update. Additional contributions are
required to define these items.
1. EVM specifications
2. 75 GHz grid operation
a. Once the Transmit Spectrum is defined it can replace Minimum Excess Bandwidth.
3. ZR+ definitions and specifications
www.oiforum.com 95
OIF-400ZR-01.0
20 Appendix C: Error Vector Magnitude

20.1 Maximum error vector magnitude
The Error vector magnitude is measured using a reference receiver as defined in Section 20.3. EVMrms uses
the peak ref. vector (not average) for normalization.
20.2 Maximum I-Q DC offset
The I-Q DC offset of a modulated signal relates to the average signal amplitudes in the I and Q phases of
that signal. The relative excess (unmodulated) power, Pexcess, is a measure of this impairment and is
obtained from the parameters Imean and Qmean and Psignal, which are intermediate results during the
evaluation of the Error Vector Magnitude:
2
Imean + Q2mean
𝑃𝑒𝑥𝑐𝑒𝑠𝑠 =
PSignal
𝐼𝑄𝑜𝑓𝑓𝑠𝑒𝑡 = 10 log10 (Pexcess )
20.3 Reference receiver for EVM and I-Q DC offset
The reference receiver includes the following hardware characteristics and processing steps:
20.3.1 Hardware characteristics:
• Dual-polarization coherent receiver. Ideally, the receiver should be calibrated over wavelength
for:
o Frequency response
o Channel imbalances
o IQ phase angle error
o Timing skew
• Real-time Nyquist sampler with sampling rate equal to or larger than the 400ZR symbol rate.
20.3.2 Processing steps1:
• Polarization demultiplex.
• Retime and resample to one sample per symbol using a Gaussian-shaped low pass filter anti-
aliasing filter with a 3-dB bandwidth of 0.5 times the symbol rate.
• Clock phase recovery.
• Frequency offset estimation and removal assuming a constant frequency offset over the given
block size N.
• Carrier phase recovery.
• IQ-offset evaluation and compensation.
• Noise loading for EQ training and EVM evaluation.
The amplitude ARMS of the noise for each quadrature is calculated from the following equation:
1 The processing is done block wise with block size N = 1000. It is possible to group multiple blocks for some of the processing steps. The
processing steps should perform only the tasks mentioned in the description. Processing steps can be consolidated and changed in order but
not perform any additional signal processing with the purpose of compensating for signal distortions resulting for example from CD, PMD,
skews, crosstalk, etc.
www.oiforum.com 96
OIF-400ZR-01.0
0.814 ∙ 𝑅𝑠𝑦𝑚𝑏𝑜𝑙
𝐴𝑅𝑀𝑆 = √ 𝑂𝑆𝑁𝑅
10 10 ∙ ∆𝑓𝑟𝑒𝑓
where OSNR is 26 dB and,
𝑐
∆𝑓𝑟𝑒𝑓 = ∙ 𝑅𝐵
𝜆2
where c is the velocity of light in vacuum, λ is the optical wavelength and RB is the resolution
bandwidth that is 0.1 nm.
• Apply a 7-tap T-spaced FIR filter with the tap coefficients optimized for BER
The sum of all filter tap coefficients is equal to one, and the largest coefficient can be for any of
the 7 taps. The individual filter taps are found by minimizing the EVMRMS value.
20.4 EVM evaluation
Find the peak vector normalization scaling factor2:
2
max (𝐼ref (𝑘)2 + 𝑄ref (𝑘)2 )
0≤𝑘<𝐾
𝛼= √
1 𝐾−1
∑ (𝐼 (𝑘)2 + 𝑄ref (𝑘)2 )
𝐾 𝑘=0 ref
o Normalize the sample pairs Iẟ and Qẟ in each of the polarizations using the average power
multiplied by the peak vector constellation scaling factor3:
𝑁−1
1
𝛼peak = 𝛼√ ∑(𝐼𝛿 (𝑛)2 + 𝑄𝛿 (𝑛)2 )
𝑁
𝑛=0
o Find the nearest constellation pair Iref(n) and Qref(n) for each normalized sample pair Iẟ and Qẟ
in each of the polarizations.
o Calculate the error vector magnitude for each normalized sample pair Iẟ and Qẟ in each of the
polarizations:
EVM(𝑛) = √(𝐼𝛿 (𝑛) − 𝐼ref (𝑛))2 + (𝑄𝛿 (𝑛) − 𝑄ref (𝑛))2

where n is the symbol number within the block starting at 0
o Using all the N samples from the x-polarization calculate EVMRMS,x:
𝑁−1
1
EVMRMS,𝑥 = √ ∑ EVM(𝑛)2
𝑁
𝑛=0
o Using all the N samples from the y-polarization and calculate EVMRMS,y:
2
k runs over all points in the constellation
3
This assumes that all constellation points have equal probability in the sample pairs
www.oiforum.com 97
OIF-400ZR-01.0
𝑁−1
1
EVMRMS,𝑦 = √ ∑ EVM(𝑛)2
𝑁
𝑛=0
o Then calculate EVMRMS in percent from:
2 2
(EVMRMS,𝑥 + EVMRMS,𝑦 )
EVMRMS = √ × 100%
2
20.5 Reference Algorithms for EVM Test of 400ZR transmitters.
The EVM algorithms are attached in Table 35.
Description Test Vector File

400ZR EVM Test
Error Vector Magnitude Algorithms1
Vectors
Table 35: EVM algorithms
1
Only available to OIF members at this time.
www.oiforum.com 98
OIF-400ZR-01.0
21 Appendix D: List of companies belonging to OIF when document is approved
Acacia Communications Google

ADVA Optical Networking Hewlett Packard Enterprise (HPE)
Alibaba IBM Corporation
Alphawave IP Inc. Idea Sistemas Electronicos S.A.
Amphenol Corp. II-VI Incorporated
AnalogX Inc. Infinera
Applied Optoelectronics, Inc. InnoLight Technology Limited
Arista Networks Innovium
BizLink Technology Inc. Inphi
Broadcom Inc. Integrated Device Technology
Cadence Design Systems Intel
China information and communication IPG Photonics Corporation
technology Group Corporation
China Telecom Global Limited Juniper Networks
Ciena Corporation Kandou Bus
Cisco Systems KDDI Research, Inc.
Corning Keysight Technologies, Inc.
Credo Semiconductor (HK) LTD Lumentum
Dell, Inc. MACOM Technology Solutions
EFFECT Photonics B.V. Marvell Semiconductor, Inc.
Elenion Technologies, LLC Maxim Integrated Inc.
Epson Electronics America, Inc. MaxLinear Inc.
eSilicon Corporation MediaTek
Facebook Mellanox Technologies
Foxconn Interconnect Technology, Ltd. Microsemi Inc.
Fujikura Microsoft Corporation
Fujitsu Mitsubishi Electric Corporation
Furukawa Electric Japan Molex
Global Foundries Multilane SAL Offshore
www.oiforum.com 99
OIF-400ZR-01.0
NEC Corporation SiFotonics Technologies Co., Ltd.

NeoPhotonics Socionext Inc.
Nokia Spirent Communications
NTT Corporation Sumitomo Electric Industries, Ltd.
O-Net Communications (HK) Limited Sumitomo Osaka Cement
Open Silicon Inc. Synopsys, Inc.
Optomind Inc. TE Connectivity
Orange Tektronix
PETRA Telefonica SA
Precise-ITC, Inc. TELUS Communications, Inc.
Rambus Inc. UNH InterOperability Laboratory (UNH-IOL)
Ranovus Verizon
Rianta Solutions, Inc. Viavi Solutions Deutschland GmbH
Rosenberger Hochfrequenztechnik GmbH & Co. Xelic
KG
Samsung Electronics Co. Ltd. Xilinx
Samtec Inc. Yamaichi Electronics Ltd.
Semtech Canada Corporation ZTE Corporation
[EOD] End of Document
www.oiforum.com 100

OIF 400ZR 01.0 - Reduced2

Uploaded by

Copyright:

Available Formats

OIF 400ZR 01.0 - Reduced2

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

OIF 400ZR 01.0 - Reduced2

Uploaded by

Copyright:

Available Formats

OIF-400ZR-01.

Implementation Agreement 400ZR

March 10, 2020

Implementation Agreement created and approved

For additional information contact:

Working Group: Physical and Link Layer (PLL) Working Group

TITLE: Implementation Agreement 400ZR

SOURCE: TECHNICAL EDITOR WORKING GROUP CHAIR

Mike A. Sluyski David R. Stauffer, Ph D.

WORKING GROUP CHAIR

8.9.2 GMP overhead Encoding..................................................................................................... 36

13.2.2 Transmitter Optical Specifications ...................................................................................... 71

15.3.2 Example 75GHz Flexible Grid Offset ................................................................................... 91

Figure 32: Hamming code ........................................................................................................................... 46

Table 32: Example 100GHz flexible grid...................................................................................................... 91

4 Document Revision History

Table 1 provides the 400ZR Implementation Agreement revision history.

Document Date Revisions/Comments

Table 1: 400ZR IA document revision history

Single Channel line Single Channel line

400ZR Module 400ZR Module

Figure 1: 400ZR reference diagram

Host protocol support Sublayer Capabilities

FEC coding RS(544,514), lane distribution, AM lock

IEEE Std 802.3™-2018

Optionally physically instantiated as 400GAUI-8 C2M;

Table 2: 400ZR host interface

6.1 400ZR Clocking Modes

7 400ZR use cases

Figure 2: Transceiver line card with 400ZR amplified point-to-point interface

Figure 3: Router switch line card with 400ZR DWDM Interfaces

Figure 4: Transceiver line card with 400ZR DWDM interfaces

7.2 Unamplified, single wavelength, loss limited link

Figure 5: Router/Switch line card with 400ZR unamplified point-to-point Interface

8 Host to 400ZR data path

PMA sublayer(*) (*)

Lane reorder and de-

FEC Decode FEC Encode

256b/257b Transcode 400GAUI-8

No rate No rate Decode & PCS

Pad Insertion Pad removal

128b/119b SD-FEC 128b/119b SD-FEC

Gray mapping, symbol Pol combining & Symbol

TX DSP PMA (ZR) RX DSP

Figure 6: Data path detail

8.1 400G Host Side Interface

8.3.2 PCS Tx direction (400ZR Rx datapath)

3 Payload (10280 bits)

256 Payload (10280 bits)

Figure 7: 400ZR frame structure without parity bits

8.4.1 400ZR Multi-Frame

Figure 8: 400ZR multi-frame structure with parity bits

8.5 AM/PAD/OH insertion

10220 (257b payload)

255 10280 payload bits = (40 × 257 bits) FEC

256 10280 payload bits = (40 × 257 bits) FEC

Figure 9: 400ZR Frame overhead

8.5.1 400ZR AM/PAD/OH Transmission order

{CM0,CM1,CM2} UP0 {CM3,CM4,CM5} UP1 {UM0 ,UM1,UM2} UP2 {UM3 ,UM4,UM5}

PMA sublayer() ()