INTERNATIONAL TELECOMMUNICATION UNION

ITU-T

G.7041/Y.1303

TELECOMMUNICATION
STANDARDIZATION SECTOR
OF ITU

(12/2001)

SERIES G: TRANSMISSION SYSTEMS AND MEDIA,

DIGITAL SYSTEMS AND NETWORKS
Digital terminal equipments General
SERIES Y: GLOBAL INFORMATION INFRASTRUCTURE
AND INTERNET PROTOCOL ASPECTS
Internet protocol aspects Transport

Generic framing procedure (GFP)

ITU-T Recommendation G.7041/Y.1303

ITU-T G-SERIES RECOMMENDATIONS

TRANSMISSION SYSTEMS AND MEDIA, DIGITAL SYSTEMS AND NETWORKS
INTERNATIONAL TELEPHONE CONNECTIONS AND CIRCUITS
GENERAL CHARACTERISTICS COMMON TO ALL ANALOGUE CARRIERTRANSMISSION SYSTEMS
GENERAL CHARACTERISTICS OF INTERNATIONAL CARRIER TELEPHONE
SYSTEMS ON RADIO-RELAY OR SATELLITE LINKS AND INTERCONNECTION WITH
METALLIC LINES
COORDINATION OF RADIOTELEPHONY AND LINE TELEPHONY
TESTING EQUIPMENTS
TRANSMISSION MEDIA CHARACTERISTICS
DIGITAL TERMINAL EQUIPMENTS
DIGITAL NETWORKS
DIGITAL SECTIONS AND DIGITAL LINE SYSTEM
QUALITY OF SERVICE AND PERFORMANCE
TRANSMISSION MEDIA CHARACTERISTICS
DIGITAL TERMINAL EQUIPMENTS
General
Coding of analogue signals by pulse code modulation
Coding of analogue signals by methods other than PCM
Principal characteristics of primary multiplex equipment
Principal characteristics of second order multiplex equipment
Principal characteristics of higher order multiplex equipment
Principal characteristics of transcoder and digital multiplication equipment
Operations, administration and maintenance features of transmission equipment
Principal characteristics of multiplexing equipment for the synchronous digital hierarchy
Other terminal equipment
DIGITAL NETWORKS
For further details, please refer to the list of ITU-T Recommendations.

G.100G.199
G.200G.299
G.400G.449
G.450G.499
G.500G.599
G.600G.699
G.700G.799
G.800G.899
G.900G.999
G.1000G.1999
G.6000G.6999
G.7000G.7999
G.7000G.7099
G.7100G.7199
G.7200G.7299
G.7300G.7399
G.7400G.7499
G.7500G.7599
G.7600G.7699
G.7700G.7799
G.7800G.7899
G.7900G.7999
G.8000G.8999

ITU-T Recommendation G.7041/Y.1303

Generic framing procedure (GFP)

Summary
This Recommendation defines a generic framing procedure (GFP) to delineate octet-aligned,
variable-length payloads from higher-level client signals for subsequent mapping into octetsynchronous paths such as those defined in ITU-T Recs G.707/Y.1322 and G.709/Y.1331. The
Recommendation definitions include the:

frame formats for protocol data units (PDUs) transferred between GFP initiation and
termination points;

mapping procedure for the client signals into GFP.

Source
ITU-T Recommendation G.7041/Y.1303 was prepared by ITU-T Study Group 15 (2001-2004) and
approved under the WTSA Resolution 1 procedure on 14 December 2001.

Keywords
Generic Framing Procedure, Optical Transport Network, Synchronous Digital Hierarchy.

ITU-T Rec. G.7041/Y.1303 (12/2001)

FOREWORD
The International Telecommunication Union (ITU) is the United Nations specialized agency in the field of
telecommunications. The ITU Telecommunication Standardization Sector (ITU-T) is a permanent organ of
ITU. ITU-T is responsible for studying technical, operating and tariff questions and issuing Recommendations
on them with a view to standardizing telecommunications on a worldwide basis.
The World Telecommunication Standardization Assembly (WTSA), which meets every four years,
establishes the topics for study by the ITU-T study groups which, in turn, produce Recommendations on these
topics.
The approval of ITU-T Recommendations is covered by the procedure laid down in WTSA Resolution 1.
In some areas of information technology which fall within ITU-T's purview, the necessary standards are
prepared on a collaborative basis with ISO and IEC.

NOTE
In this Recommendation, the expression "Administration" is used for conciseness to indicate both a
telecommunication administration and a recognized operating agency.

INTELLECTUAL PROPERTY RIGHTS

ITU draws attention to the possibility that the practice or implementation of this Recommendation may
involve the use of a claimed Intellectual Property Right. ITU takes no position concerning the evidence,
validity or applicability of claimed Intellectual Property Rights, whether asserted by ITU members or others
outside of the Recommendation development process.
As of the date of approval of this Recommendation, ITU had not received notice of intellectual property,
protected by patents, which may be required to implement this Recommendation. However, implementors are
cautioned that this may not represent the latest information and are therefore strongly urged to consult the
TSB patent database.

ITU 2002
All rights reserved. No part of this publication may be reproduced, by any means whatsoever, without the
prior written permission of ITU.

ii

ITU-T Rec. G.7041/Y.1303 (12/2001)

CONTENTS
Page
1

Scope ............................................................................................................................

References ....................................................................................................................

Terms and definitions ...................................................................................................

Abbreviations ...............................................................................................................

Conventions..................................................................................................................

Common aspects of GFP ..............................................................................................

6.1
Basic signal structure for GFP client frames..................................................
6.1.1
GFP Core Header ...........................................................................................
6.1.2
GFP Payload Area..........................................................................................
6.1.3
GFP client frames...........................................................................................
6.2
GFP Control frames .......................................................................................
6.2.1
GFP Idle frames .............................................................................................
6.2.2
Other control frames.......................................................................................
6.3
GFP frame-level functions .............................................................................
6.3.1
GFP frame delineation algorithm...................................................................
6.3.2
Frame multiplexing ........................................................................................
6.3.3
Client Signal Fail indication...........................................................................
6.3.4
Defect handling in GFP..................................................................................

4
5
6
7
13
15
15
15
15
16
17
18
18

Payload specific aspects for frame-mapped GFP.........................................................

7.1
Ethernet MAC payload...................................................................................
7.1.1
Ethernet MAC encapsulation .........................................................................
7.1.2
Ethernet Inter-Packet Gap (IPG) deletion and restoring ................................
7.2
IP/PPP payload...............................................................................................
7.2.1
PPP frame encapsulation................................................................................
7.2.2
GFP/HDLC delineation interworking ............................................................
7.2.3
PPP payload configuration options ................................................................
7.3
Error handling in frame-mapped GFP............................................................

19
19
19
19
20
20
20
21
21

Payload-specific aspects for transparent mapping of 8B/10B clients into GFP...........

8.1
Adapting 8B/10B client signals via 64B/65B block codes ............................
8.1.1
10B_ERR code...............................................................................................
8.1.2
Insertion of 65B_PAD code and GFP Client Management frames ...............
8.2
Adapting 64B/65B code blocks into GFP ......................................................
8.2.1
Error control with Transparent GFP...............................................................
8.3
Running disparity in 64B/65B codes .............................................................
8.3.1
Handling of running disparity on ingress.......................................................

21
22
24
25
25
26
26
27

ITU-T Rec. G.7041/Y.1303 (12/2001)

iii

Handling of running disparity on egress ........................................................

Client-specific running disparity aspects .......................................................
Rate adaptation in 64B/65B codes .................................................................
Egress rate adaptation procedures ..................................................................
Client-specific rate adaptation aspects ...........................................................
Client-specific Signal Fail aspects .................................................................
Fibre Channel payload ...................................................................................
ESCON Payload.............................................................................................
FICON Payload ..............................................................................................
Full-duplex Gigabit Ethernet payload ............................................................

Page
27
27
28
29
30
31
32
32
33
33

Appendix I Examples of functional models for GFP applications.......................................

33

Appendix II Sample GFP payload types ..............................................................................

36

Appendix III GFP frame example illustrating transmission order and CRC calculation.....

37

Appendix IV Number of superblocks used in Transparent GFP..........................................

40

8.3.2
8.3.3
8.4
8.4.1
8.4.2
8.5
8.5.1
8.5.2
8.5.3
8.5.4

iv

ITU-T Rec. G.7041/Y.1303 (12/2001)

Introduction
GFP provides a generic mechanism to adapt traffic from higher-layer client signals over a transport
network. Client signals may be PDU-oriented (such as IP/PPP or Ethernet MAC), block-code
oriented constant bit rate stream (such as Fibre Channel or ESCON/SBCON).
This Recommendation consists of both common and client-specific aspects. Common aspects of
GFP apply to all GFP-adapted traffic and they are specified in clause 6. Client-specific aspects of
GFP are specified in clauses 7 and 8. Currently, two modes of client signal adaptation are defined for
GFP.

A PDU-oriented adaptation mode, referred to as Frame-Mapped GFP (GFP-F), is specified

in clause 7.

A block-code oriented adaptation mode, referred to as Transparent GFP (GFP-T), is

specified in clause 8.
Figure 1 illustrates the relationship between the higher-layer client signals, GFP, and its transport
paths.

Ethernet

Other Client Signals

IP/PPP

GFP Client specific aspects

(Payload dependent)
GFP Common aspects
(Payload independent)

SDH VC-n Path

Other octetsynchronous
paths

OTN ODUk Path

T1545290-02

Figure 1/G.7041/Y.1303 GFP relationship to Client Signals and Transport Paths

Figure 2 illustrates the environment in which GFP operates.
In the Frame-Mapped adaptation mode, the Client/GFP adaptation function may operate at the data
link layer (or higher layer) of the client signal. Client PDU visibility is required. This visibility is
obtained when the client PDUs are received from either the data layer network (e.g. IP router fabric
or Ethernet switch fabric (C/C' in Figure 2)), or e.g. a bridge, switch or router function in a transport
network element (TNE). In the latter case, the client PDUs are received via e.g. an Ethernet interface
(A/A' in Figure 2).
For the Transparent adaptation mode, the Client/GFP adaptation function operates on the coded
character stream, rather than on the incoming client PDUs. Thus, processing of the incoming
codeword space for the client signal is required (B/B' in Figure 2).
Typically, interconnections can be set up between ports A and A', B and B', C and C', A and C'
and C and A'. Note that the physical port type of B and B' must be the same to support an
interconnection, while the physical port type of A and A' may be different.
Some high level functional models associated with the above GFP processing can be found in
Appendix I.

ITU-T Rec. G.7041/Y.1303 (12/2001)

Data layer network

Bridge or Switch or Router function in TNE

Bridge or Switch or Router function in TNE

C

GFP-T
PHY-T
A
10BASE
100BASE
1000BASE
10GBASE

GFP-F

C'
GFP-F

GFP-F

GFP-F

GFP-T

PHY-X

PHY-X

B'

ESCON
FICON
Fibre channel
1000BASE

ESCON
FICON
Fibre channel
1000BASE

Transport layer network

T1545300-02

Figure 2/G.7041/Y.1303 GFP Functional Model (Single Client)

vi

ITU-T Rec. G.7041/Y.1303 (12/2001)

PHY-T
A'
10BASE
100BASE
1000BASE
10GBASE

ITU-T Recommendation G.7041/Y.1303

Generic framing procedure (GFP)
1

Scope

This Recommendation defines a generic framing procedure (GFP) to encapsulate variable length
payload of various client signals for subsequent transport over SDH and OTN networks as defined
in ITU-T Recs G.707/Y.1322 and G.709/Y.1331. The Recommendation definitions include the:

frame formats for protocol data units (PDUs) transferred between GFP initiation and
termination points;

mapping procedure for the client signals into GFP.

The framing procedure described in this Recommendation can be applied to both the encapsulation
of entire client frames (frame mapped GFP), in which a single client frame is mapped into a single
GFP frame, and to character mapped transport (transparent GFP) in which a number of client data
characters are mapped into efficient block codes for transport within a GFP frame.
2

References

The following ITU-T Recommendations, and other references contain provisions which, through
referenced in this text, constitute provisions of this Recommendation. At the time of publication, the
editions indicated were valid. All Recommendations and other references are subject to revision;
users of this Recommendation are therefore encouraged to investigate the possibility of applying the
most recent edition of the Recommendations and other references listed below. A list of the
currently valid ITU-T Recommendations is regularly published.

ITU-T Recommendation G.707/Y.1322 (2000), Network node interface for the

synchronous digital hierarchy.

ITU-T Recommendation G.709/Y.1331 (2001), Interfaces for the optical transport network
(OTN).

ITU-T Recommendation G.783 (2000), Characteristics of synchronous digital hierarchy

(SDH) equipment functional blocks.

ITU-T Recommendation G.798 (2002), Characteristics of optical transport networks

hierarchy equipment functional blocks.

ITU-T Recommendation I.432 (1993), B-ISDN user-network interface Physical layer

specification.

ISO/IEC 3309:1993, Information Technology Telecommunications and information

exchange between systems High-level data link control (HDLC) procedures Frame
structure.

IEEE 802.3 (1998), Part 3: Carrier Sense Multiple Access with Collision Detection
(CSMA/CD) Access Method and Physical Layer Specifications.

ANSI X3.230 (1994), Information Technology Fibre Channel Physical and Signaling
Interface (FC-PH).

ANSI X3.296 (1997), Information Technology Single-Byte Command Code Sets

CONnection (SBCON) Architecture.

IETF RFC 1661 (1994), The Point-to-Point Protocol (PPP).

IETF RFC 1662 (1994), PPP in HDLC-like Framing.

ITU-T Rec. G.7041/Y.1303 (12/2001)

Terms and definitions

This Recommendation defines the following terms:

3.1
frame-mapped GFP: A type of GFP mapping in which a client signal frame is received
and mapped in its entirety into one GFP frame.
3.2
channel ID: The CID is an 8-bit binary number used to indicate one of 256
communications channels at a GFP initiation/termination point.
3.3
client data frame: A client data frame is a GFP frame that contains payload data from a
client signal.
3.4
client management frame: A client management frame is a GFP frame containing
information associated with the management of the GFP connection between the GFP source and
sink.
3.5
control frame: A control frame is a GFP frame used to control the GFP connection. The
only control defined at this time is the Idle frame.
3.6

Maximum Transmission Unit (MTU): Maximum size of the GFP Payload Area, in octets.

3.7
running disparity: A procedure used by block line codes, such as 8B/10B, to balance the
total of number of ones and zeros transmitted over time. The running disparity at the end of a line
code subblock is positive if more ones than zeros have been sent up to that point, and negative if
more zeros than ones have been sent. The encoder uses the running disparity value to choose which
of the two possible codes to transmit for the next character mapping in order to balance the number
of transmitted ones and zeros.
3.8

Source/Destination Port (SP/DP): A logical addressable entity on a physical interface.

3.9
superblock: A superblock refers to a Transparent GFP structure that combines multiple
64B/65B codes along with a CRC-16, for the purposes of providing payload octet alignment and
error control over the bits in the superblock. See Figure 8-3.
3.10
transparent GFP: A type of GFP mapping in which block-coded client characters are
decoded and then mapped into a fixed-length GFP frame and may be transmitted immediately
without waiting for the reception of an entire client data frame.
4

Abbreviations

This Recommendation uses the following abbreviations:

ANSI

American National Standards Institute

ATM

Asynchronous Transfer Mode

cHEC

Core HEC

CID

Channel ID

CoS

Class of Service

CRC

Cyclic Redundancy Check

CSF

Client Signal Fail

DE

Discard Eligibility

DP

Destination Port

DST

Destination

eHEC

Extension HEC

EOF

End Of Frame

ITU-T Rec. G.7041/Y.1303 (12/2001)

ESCON

Enterprise Systems Connection

EXI

Extension Header Identifier

FC

Fibre Channel

FCS

Frame-Check Sequence

FICON

Fibre Connection

GFP

Generic Framing Procedure

GFP-F

Frame-mapped GFP

GFP-T

Transparent GFP

HDLC

High-level Data Link Control

HEC

Header Error Check

IEEE

Institute of Electrical and Electronic Engineers

IFG

Inter-Frame Gap

IP

Internet Protocol

IPG

Inter-Packet Gap

ISDN

Integrated Services Digital Network

ISO

International Organization for Standardization

ITU-T

International Telecommunication Union Telecommunication Standardization Sector

LCC

Last Control Character

LOL

Loss Of Light

LOS

Loss Of Signal

LSB

Least Significant Bit

MAC

Media Access Control

MAPOS

Multiple Access Protocol Over SONET/SDH

MSB

Most Significant Bit

MTU

Maximum Transmission Unit

NE

Network Element

OA&M

Operations, Administration & Maintenance

ODU

Optical Data Unit

OTN

Optical Transport Network

PDU

Protocol Data Unit

PFI

Payload FCS Identifier

PLI

Payload Length Indicator

PPP

Point-to-Point Protocol

PTI

Payload Type Identifier

RD

Running Disparity

SBCON

Single-Byte Command Code Sets Connection

SDH

Synchronous Digital Hierarchy

ITU-T Rec. G.7041/Y.1303 (12/2001)

SOF

Start of Frame

SONET

Synchronous Optical Network

SP

Source Port

SPE

Synchronous Payload Envelope

Src

Source

SSF

Server Signal Failure

STS

Synchronous Transport Signal

tHEC

Type HEC

TSF

Trail Signal Fail

TTL

Time To Live

UPI

User Payload Identifier

Conventions

Transmission order: The order of transmission of information in all the diagrams in this
Recommendation is first from left to right and then from top to bottom. Within each byte the most
significant bit is transmitted first. The most significant bit is illustrated at the left of all the
diagrams.
Undefined field values: The default value for any undefined header fields is 0 unless otherwise
stated.
6

Common aspects of GFP

This clause discusses the common (protocol independent) aspects of GFP for octet-aligned
payloads. The mapping of the framed payloads into an SDH VC-n is specified in ITU-T
Rec. G.707/Y.1322. The mapping of the framed payloads into an OTN ODUk payload is specified
in ITU-T Rec. G.709/Y.1331.
GFP uses a variation of the HEC-based frame delineation mechanism defined for Asynchronous
Transfer Mode (ATM) (see ITU-T Rec. I.432.1). Two kinds of GFP frames are defined: GFP client
frames and GFP control frames. Frame formats for GFP client and control frames are defined in 6.1
and 6.2. GFP also supports a flexible (payload) header extension mechanism to facilitate the
adaptation of GFP for use with diverse transport mechanisms.Currently defined payload extension
header types are specified in 6.1.2.3.

ITU-T Rec. G.7041/Y.1303 (12/2001)

6.1

Basic signal structure for GFP client frames

The format for GFP frames is shown in Figure 6-1. GFP frames are octet-aligned and consist of a
GFP Core Header and, except for GFP Idle frames, a GFP Payload Area.
1
2
3
4
5

Core header

Payload area

4-65535

.
Octet
transmission
order

.
.

n
1

Octet

Bit
Bit transmission order
a) Frame size and transmission order

16-bit Payload
length indicator

Core
header

cHEC
(CRC-16)

Payload
headers
(4-64 bytes)
Payload
area
Client
payload
information
field

Optional
payload FCS
(CRC-32)
T1545310-02

b) Fields constituting a GFP client frame

Figure 6-1/G.7041/Y.1303 Frame format for GFP client frames

ITU-T Rec. G.7041/Y.1303 (12/2001)

6.1.1

GFP Core Header

The GFP Core Header format is shown in Figure 6-2. The four octets of the GFP Core Header
consist of a 16-bit PDU Length Indicator field and a 16-bit Core Header Error Check (cHEC) field.
This header allows GFP frame delineation independent of the content of the higher layer PDUs.

Octet
transmission
order

PLI

<15:08>

PLI

<07:00>

cHEC

<15:08>

cHEC

<07:00>

Octet

1
Bit

T1545320-02

Bit transmission order

Figure 6-2/G.7041/Y.1303 GFP Core Header format

6.1.1.1

PDU Length Indicator (PLI) field

The two-octet PLI field contains a binary number representing the number of octets in the GFP
Payload Area. The absolute minimum value of the PLI field in a GFP client frame is 4 octets. PLI
values 0-3 are reserved for GFP control frame usage (see 6.2).
6.1.1.2

Core HEC (cHEC) field

The two-octet Core Header Error Control field contains a CRC-16 error control code that protects
the integrity of the contents of the Core Header by enabling both single-bit error correction and
multi-bit error detection. The cHEC sequence is calculated over the octets of the Core Header as
defined in 6.1.1.2.1.
6.1.1.2.1 HEC processing
16

12

The HEC generating polynomial is G(x) = x + x + x + 1, with an initialization value of zero,

16
0
where x corresponds to the MSB and x corresponds to the LSB.
The cHEC field is generated by the source adaptation process using the following steps (see
Appendix I/V.41):
1)
The first two octets of the GFP frame are taken in network octet order, most significant bit
first, to form a 16-bit pattern representing the coefficients of a polynomial M(x) of
degree 15.
16

and divided (modulo 2) by G(x), producing a remainder R(x) of

2)

M(x) is multiplied by x
degree 15 or less.

3)

The coefficients of R(x) are considered to be a 16-bit sequence, where x is the most
significant bit.
This 16-bit sequence is the CRC-16 where the first bit of the CRC-16 to be transmitted is
15
0
the coefficient of x and the last bit transmitted is the coefficient of x .

4)

15

The sink adaptation process performs steps 1-3 in the same manner as the source adaptation
process. In the absence of bit errors, the remainder shall be 0000 0000 0000 0000.
This single error correction shall be performed on the Core Header. The GFP sink adaptation
process shall discard any of those GFP frames where multi-bit errors are detected. The sink
adaptation process also updates any relevant system records for performance monitoring purposes.

ITU-T Rec. G.7041/Y.1303 (12/2001)

6.1.1.3

Core header scrambling

The Core Header is scrambled for DC balanced by an exclusive-OR operation (modulo 2 addition)
with the hexadecimal number B6AB31E0. This number is the maximum transition, minimum sidelobe, Barker-like sequence of length 32. The scrambling of the GFP Core Header improves the
robustness of the GFP frame delineation procedure and provides a sufficient number of 0-1 and 1-0
transitions during idle transmission periods.
6.1.2

GFP Payload Area

The GFP Payload Area, which consists of all octets in the GFP frame after the GFP Core Header, is
used to convey higher layer specific protocol information. This variable length area may include
from 4 to 65 535 octets. As shown in Figure 6-3, the GFP Payload Area consists of two common
components: a Payload Header and a Payload Information field. An optional Payload FCS (pFCS)
field is also supported.

Octet
transmission
order

5
6
7
8
9
6
.
.
.
n

Octet

Payload
information
field

0 to 65535-X

Payload FCS
(Optional)

1
Bit

X = 4 to 64

Payload header

T1545330-02

Bit transmission order

Figure 6-3/G.7041/Y.1303 GFP Payload Area format

Practical GFP MTU sizes for the GFP Payload Area are application specific. An implementation
should support transmission and reception of GFP frames with GFP Payload Areas of at least
1600 octets. By prior arrangement, consenting GFP implementations may use other MTU values.
6.1.2.1

Payload Header

The Payload Header is a variable-length area, 4 to 64 octets long, intended to support data link
management procedures specific to the higher-layer client signal. The structure of the GFP Payload
Header is illustrated in Figure 6-4. The area contains two mandatory fields, the Type and the tHEC
fields, and a variable number of additional payload header fields. This group of additional payload
header fields are referred to as the Extension Header. The presence of the Extension Header, and its
format, and the presence of the optional Payload FCS are specified by the Type field. The tHEC
protects the integrity of the Type field.

ITU-T Rec. G.7041/Y.1303 (12/2001)

Type

tHEC

6
7
8
Octet
transmission
order

Extension
header
field

.
.

0 to 60

.
2

eHEC

Octet

Bit

Bit transmission order

T1545340-02

Figure 6-4/G.7041/Y.1303 GFP Payload Header format

An implementation shall support reception of a GFP frame with a Payload Header of any length in
the range 4 to 64 octets.
6.1.2.1.1 GFP Type field
The GFP Type field is a mandatory two-octet field of the Payload Header that indicates the content
and format of the GFP Payload Information field (see 6.1.2.2). The Type field distinguishes
between GFP frame types and between different services in a multi-service environment. As shown
in Figure 6-5, the Type field consists of a Payload Type Identifier (PTI), a Payload FCS Indicator
(PFI), a Extension Header Identifier (EXI) and a User Payload Identifier (UPI).
15 14 13 12 11 10
Octet
5
transmission
order
6

PTI

Bit

EXI

PFI
UPI

2 3 4 5 6 7
Bit transmission order

Bit

T1545350-02

Figure 6-5/G.7041/Y.1303 GFP Type Field format

The interpretation of the UPI field for PTI values different from 000 or 100 is for further study.
Sample Type field values are illustrated in Appendix II.
6.1.2.1.1.1

Payload Type Identifier

A 3-bit subfield of the Type field identifying the type of GFP client frame. Two kinds of client
frames are currently defined, User Data frames (PTI = 000) and Client Management frames
(PTI = 100). PTI codepoints are given in Table 6-1.

ITU-T Rec. G.7041/Y.1303 (12/2001)

Table 6-1/G.7041/Y.1303 GFP Payload Type Identifiers

Payload Type Identifiers
Type Bits <15:13>

6.1.2.1.1.2

Usage

000

Client Data

100

Client Management

Others

Reserved

Payload FCS Indicator (PFI)

A one bit subfield of the Type field indicating the presence (PFI = 1) or absence (PFI = 0) of the
Payload FCS field.
6.1.2.1.1.3

Extension Header Identifier (EXI)

A 4-bit subfield of the Type field identifying the type of Extension Header GFP. Three kinds of
Extension Headers are currently defined, a Null Extension Header, a Linear Extension Header, and
a Ring Extension Header. EXI codepoints are given in Table 6-2.
Table 6-2/G.7041/Y.1303 GFP Extension Header Identifiers
Extension Header Identifiers
Type Bits <11:8>

6.1.2.1.1.4

Usage

0000

Null Extension Header

0001

Linear Frame

0010

Ring Frame

Others

Reserved

User Payload Identifier (UPI)

An 8-bit field identifying the type of payload conveyed in the GFP Payload Information field.
Interpretation of the UPI field is relative to the type of GFP client frame as indicated by the PTI
subfield. UPI values for client data frames are specified in 6.1.3.1 and UPI values for Client
Management frames are specified in 6.1.3.2.
6.1.2.1.2 Type HEC (tHEC) field
The two-octet Type Header Error Control field contains a CRC-16 error control code that protects
the integrity of the contents of the Type Field by enabling both single-bit error correction and multibit error detection.
The contents of the tHEC field is generated using the same steps as the cHEC (see 6.1.1.2.1) with
the following exception:

For the tHEC step 1) is modified such that M(x) is formed from all the octets in the Type
field, but excluding the tHEC field itself.

ITU-T Rec. G.7041/Y.1303 (12/2001)

The GFP sink adaptation process may perform single-bit error correction on all of the fields
protected by a tHEC field. This single error correction shall be performed for the Type Header. The
GFP sink adaptation process shall discard any of those GFP frames where multi-bit errors are
detected, or where any error occurs in a header field that does not make use of single error
correction. The sink adaptation process also updates any relevant system records for performance
monitoring purposes.
6.1.2.1.3 GFP Extension Headers
The payload Extension Header is a 0-to-60 octet extended field (including the eHEC) that supports
technology specific data link headers such as virtual link identifiers, source/destination addresses,
port numbers, Class of Service, extension header error control, etc. The type of the extension header
is indicated by the content of the EXI bits in the Type Field of the payload header.
Three Extension Header variants are currently defined to support client specific data over a logical
Ring or logical Point-to-Point (Linear) configurations.
This clause describes the various fields in each Extension Header. The default value for any
undefined fields is 0 unless otherwise stated.
6.1.2.1.3.1

Null Extension Header

The Payload header for a frame with a Null Extension Header is shown in Figure 6-6. This
Extension Header applies to a logical point-to-point configuration. It is intended for scenarios where
the transport path is dedicated to one client signal.

Octet
transmission
order

Type

<15:08>

Type

<07:00>

tHEC

<15:08>

tHEC

<07:00>

Octet

1
Bit

T1545360-02

Bit transmission order

Figure 6-6/G.7041/Y.1303 Payload Header for a GFP frame

with a Null Extension Header
6.1.2.1.3.2

Extension Header for a linear frame

The Payload Header for a Linear (Point-to-Point) frame with an Extension Header, shown in
Figure 6-7, is intended for scenarios where there are several independent links requiring aggregation
onto a single transport path.

10

ITU-T Rec. G.7041/Y.1303 (12/2001)

Octet
transmission
order

Type

<15:08>

Type

<07:00>

tHEC <15:08>

tHEC <07:00>

CID

10

Spare <07:00>

11

eHEC <15:08>

12

eHEC <07:00>

Octet

1
Bit

<07:00>

T1545370-02

Bit transmission order

Figure 6-7/G.7041/Y.1303 Payload Header for a Linear (Point-to-Point)

Frame including the Extension Header
6.1.2.1.3.2.1

Channel ID (CID) field

The CID is an 8-bit binary number used to indicate one of 256 communications channels at a GFP
termination point.
6.1.2.1.3.2.2

Spare field

The 8-bit spare field is reserved for future use.

6.1.2.1.3.2.3

Extension HEC (eHEC) field

See 6.1.2.1.4.
6.1.2.1.3.3

Extension Header for a ring frame

For further study.

6.1.2.1.4 Extension HEC (eHEC) field
The two-octet Extension Header Error Control field contains a CRC-16 error control code that
protects the integrity of the contents of the extension headers by enabling both single-bit error
correction (optional) and multi-bit error detection.
The contents of the eHEC field is generated using the same steps as the cHEC (see 6.1.1.2.1) with
the following exception:

For the eHEC step 1) is modified such that M(x) is formed from all the octets in the
Extension Header, but excluding the eHEC field itself.

The GFP sink adaptation process may perform single-bit error correction on all of the fields
protected by a tHEC field. Single error correction is optional for the Extension Header. The GFP
sink adaptation process shall discard any of those GFP frames where multi-bit errors are detected,
or where any error occurs in a header field that does not make use of single error correction. The
sink adaptation process also updates any relevant system records for performance monitoring
purposes.

ITU-T Rec. G.7041/Y.1303 (12/2001)

11

6.1.2.2

Payload Information Field

The Payload Information field contains the framed PDU for frame-mapped GFP or, in the case of
transparent GFP, a group of client signal characters. This variable length field may include from 0
to 65,535-X octets, where X is the size of the Payload Header. This field may include an optional
Payload FCS field. The client PDU/signal is always transferred into the GFP Payload Information
field as an octet-aligned packet stream.
6.1.2.2.1 Payload Frame Check Sequence (pFCS) field
The GFP Payload FCS, as shown in Figure 6-8, is an optional, four-octet long, frame check
sequence. It contains a CRC-32 sequence that protects the contents of the GFP Payload Information
field. The FCS generation process is defined in 6.1.2.2.1.1. A value of 1 in the PFI bit within the
Type field identifies the presence of the payload FCS field.

Octet
transmission
order

Octet

Bit

pFCS

<31:24>

pFCS

<23:16>

pFCS

<15:08>

pFCS

<07:00>

T1545380-02

Bit transmission order

Figure 6-8/G.7041/Y.1303 GFP Payload Frame Check Sequence format

6.1.2.2.1.1

Payload FCS generation

The Payload FCS is generated using the CRC-32 generating polynomial (ISO/IEC 3309)
32
26
23
22
16
12
11
10
8
7
5
4
2
1
32
G(x) = x + x + x + x + x + x + x + x + x + x + x + x + x + x + 1 where x
0
corresponds to the MSB and x corresponds to the LSB.
The Payload FCS field is generated using the following steps:
1)
The N octets from the GFP Payload Information field, excluding the FCS are taken in
network octet order, most significant bit first, to form a 8N-bit pattern representing the
coefficients of a polynomial M'(x) of degree 8N 1.
2)

32

31

M'(x) is multiplied by x , added to the all-ones polynomial U(x) = 1 + x + x + + x ,

and divided (modulo 2) by G(x), producing a remainder R(x) of degree 31 or less.
8N

31

NOTE The addition of x [1 + x + x + + x ] is equivalent to presetting the shift register to

all 1s for typical shift register implementations using presets.

3)
4)

31

The coefficients of R(x) are considered to be a 32-bit sequence, where x

significant bit.
The complement of this 32-bit sequence is the CRC-32.

is the most

The sink adaptation process performs steps 1)-3) in the same manner as the source adaptation
process.
In
the
absence
of
errors,
the
remainder
shall
be
31
0
11000111_00000100_11011101_01111011, in the order x to x .

12

ITU-T Rec. G.7041/Y.1303 (12/2001)

6.1.2.3

Payload area scrambling

Scrambling of the GFP Payload Area is required to provide security against payload information
replicating scrambling word (or its inverse) from a frame synchronous scrambler such as those used
in the SDH RS layer or in an OTN OPUk channel). Figure 6-9 illustrates the scrambler and
descrambler processes.
"x 43 + 1" Scrambler

"x 43 + 1" Descrambler

X(t)

Y(t)

Y(t)

X(t)

Y(t 43)
D 43

Y(t 43)
D2

D1

D1

D2

D 43
T1545390-02

43

Figure 6-9/G.7041/Y.1303 x +1 Scrambler and descrambler processes for GFP

43

All octets in the GFP Payload Area are scrambled using a 1 + x

Scrambling is done in network bit order.

self-synchronous scrambler.

At the source adaptation process, scrambling is enabled starting at the first transmitted octet after
the cHEC field, and is disabled after the last transmitted octet of the GFP frame. When the
scrambler or descrambler is disabled, its state is retained. Hence, the scrambler or descrambler state
at the beginning of a GFP frame Payload Area will thus be the last 43 Payload Area bits of the GFP
frame transmitted in that channel immediately prior to the current GFP frame.
The activation of the sink adaptation process descrambler also depends on the present state of the
cHEC check algorithm:
a)
In the HUNT and PRESYNC states, the descrambler is disabled.
b)
In the SYNC state, the descrambler is enabled only for the octets between the cHEC field
and the end of the candidate GFP frame.
NOTE the GFP sink adaptation process can reliably forward GFP frames to the higher layer entity only
when the sink adaptation process is in the SYNC state.

6.1.3

GFP client frames

Two types of GFP client frames are currently defined, Client Data and Client Management. GFP
client data frames are used to transport data from the client signal. GFP Client Management Frames
are used to transport information associated with the management of the client signal or GFP
connection.
6.1.3.1

Client data frames

Client data is transported over GFP using client data frames. Client data frames are GFP client
frames consisting of a Core Header and a Payload Area. The Type field of the client data frames
uses the following Type subfield values:

PTI = 000

PFI = Payload specific

EXI = Payload specific

UPI = Payload specific

The Payload FCS Indicator (PFI) shall be set as required depending whether FCS is enabled or not.
The Extension Header Identifier (EXI) shall be set consistently with the frame multiplexing and
topology requirements for the GFP connection. The User Payload Identifier shall be set according
to the transported client signal type. Defined UPI values for client data frames are given Table 6-3.
ITU-T Rec. G.7041/Y.1303 (12/2001)

13

Table 6-3/G.7041/Y.1303 User Payload Identifiers for GFP Client frames

PTI = 000
User Payload Identifier (binary)
TYPE Bits <7:0>

GFP Frame
Payload Area

0000 0000

Reserved and not available

1111 1111
0000 0001

Frame-mapped Ethernet

0000 0010

Frame-mapped PPP

0000 0011

Transparent fiber channel

0000 0100

Transparent FICON

0000 0101

Transparent ESCON

0000 0110

Transparent Gb Ethernet

0000 0111

Reserved for future

0000 1000

Frame-mapped multiple access protocol over

SDH (MAPOS)

0000 1001
through

Reserved for future standardization

1110 1111
1111 0000
through

Reserved for proprietary use

1111 1110

6.1.3.2

GFP Client Management frames

Client Management frames provide a generic mechanism for the GFP client specific source
adaptation process to optionally send Client Management frames to the GFP client specific sink
adaptation process. The frame consists of:
Client Management frames are GFP client frames consisting of a Core Header and a Payload Area.
The Type field of the client data frames uses the following Type subfield values:

PTI = 100

PFI = Payload specific

EXI = Payload specific

UPI = Payload specific

For use as a GFP Client Management frame the Payload FCS Indicator (PFI) shall be set as required
depending whether FCS is enabled or not. (Note that the use of FCS in GFP Client management
frames reduces the amount of 'spare' bandwidth that can be used for such frames.) The Extension
Header Indicator (EXI) shall be set as required depending on whether the extension header is
employed or not. (Note that the use of Extension Header in GFP Client Management frame will
significantly reduce the amount of 'spare' bandwidth that can be used for such frames.)
The UPI defines the use of the GFP Client Management frame payload. In this way the GFP Client
management frame may be used for multiple purposes. Table 6-4 defines the GFP Client
Management frame payload uses.

14

ITU-T Rec. G.7041/Y.1303 (12/2001)

Table 6-4/G.7041/Y.1303 GFP Client Management frame User Payload Identifier

PTI = 100

6.2

UPI value

Usage

0000 0000 and 1111 1111

Reserved

0000 0001

Client signal fail (Loss of client signal)

0000 0010

Client signal fail (Loss of character synchronization)

0000 0011 thru 1111 1110

Reserved for future use

GFP Control frames

GFP control frames are used in the management of the GFP connection. The only control frame
specified at this time is the GFP Idle frame.
6.2.1

GFP Idle frames

The GFP Idle frame is a special four-octet GFP control frame consisting of only a GFP Core Header
with the PLI and cHEC fields (see 6.1.1) set to 0, and no Payload Area. The Idle frame is intended
for use as a filler frame for the GFP source adaptation process to facilitate the adaptation of the GFP
octet stream to any given transport medium where the transport medium channel has a higher
capacity than required by the client signal. The GFP Idle frame format is shown in Figure 6-10,
with the parenthetical values indicating the values after the Barker-like scrambling has been
performed.

Octet
transmission
order

00 (B6) hex

00 (AB) hex

00 (31) hex

00 (E0) hex

Octet

1
Bit

T1545400-02

Bit transmission order

Figure 6-10/G.7041/Y.1303 GFP Idle frame (Barker-like scrambled frame)

6.2.2

Other control frames

Control frames with PLI = 1, 2, or 3 are for further study.

6.3

GFP frame-level functions

This clause discusses frame-level processes common to all payloads that are framed via GFP.
Processes specific to particular payloads are discussed in clauses 7 and 8. The relationships among
these processes are illustrated in Figure 6-11.

ITU-T Rec. G.7041/Y.1303 (12/2001)

15

Core Header
scrambler

Octet streams from

client-specific
source adaptation
processes

Payload
scrambler

Client management
Idle insert
Octet streams
to/from
payload
mapping

Frame multiplex

Octet streams to
client-specific
sink adaptation
processes

Payload
descrambler

Client management
Core Header
check

Idle termination
Frame demultiplex

T1545410-02

Figure 6-11/G.7041/Y.1303 GFP common (Protocol Independent) procedures

6.3.1

GFP frame delineation algorithm

GFP uses a modified version of the HEC check algorithm specified in 4.5.1.1/I.432, to provide GFP
frame delineation. The frame delineation algorithm used in GFP differs from that in ITU-T
Rec. I.432 in two basic ways:
a)
The algorithm uses the PDU Length Indicator field of the GFP Core Header to find the end
of the GFP frame; and
b)
HEC field calculation uses a 16-bit polynomial and, consequently, generates a two-octet
cHEC field.
GFP frame delineation is performed based on the correlation between the first two octets of the GFP
frame and the embedded two-octet cHEC field. Figure 6-12 shows the state diagram for the GFP
frame delineation method.
The state diagram works as follows:
1)
In the HUNT state, the GFP process performs frame delineation by searching, octet by
octet, for a correctly formatted Core Header over the last received sequence of four octets.
The Core Header single error correction is disabled while in this state. Once a correct cHEC
match is detected in the candidate PLI and cHEC fields, a candidate GFP frame is identified
and the receive process enters the PRESYNC state.
2)
In the PRESYNC state, the GFP process performs frame delineation by checking, frame by
frame, for a correct cHEC match in the presumed Core Header of the next candidate GFP
frame. The PLI field in the Core Header of the preceding GFP frame is used to find the
beginning of the next candidate GFP frame. Core Header single error correction remains
disabled while in this state. The process repeats until DELTA consecutive correct cHECs
are confirmed, at which point the process enters the SYNC state. If an incorrect cHEC is
detected, the process returns to the HUNT state. The total number of consecutive correct
cHECs required to move from the HUNT state to the SYNC state is therefore DELTA + 1.
16

ITU-T Rec. G.7041/Y.1303 (12/2001)

3)

4)

In the SYNC state, the GFP process performs frame delineation by checking for a correct
cHEC match on the next candidate GFP frame. The PLI field in the Core Header of the
preceding GFP frame is used to find the beginning of the next candidate GFP frame.
Single-bit Core Header error correction is enabled while in this state. Frame delineation is
lost whenever multiple bit errors are detected in the Core Header by the cHEC. In this case,
a GFP Loss of Frame Delineation event is declared, the framing process returns to the
HUNT state, and a client Server Signal Failure (SSF) is indicated to the client adaptation
process.
Idle GFP frames participate in the delineation process and are then discarded.
Frame-by-frame
(Error correction disabled)

Virtual
framers
(up to M)

PRESYNC
(cHEC1D)

PRESYNC
(cHECMD)

Incorrect
cHEC
PRESYNC

PRESYNC

(cHEC11)

(cHECM1)

Correct
cHEC

DELTA
consecutive
correct
cHECs

Correct
cHEC
HUNT

SYNC
Incorrect HEC
(Multi-bit errors)

Octet-by-octet
(Error correction disabled)

T1545420-02

Frame-by-frame
(Error correction disabled)

Figure 6-12/G.7041/Y.1303 GFP Frame Delineation state diagram

Robustness against false delineation in the resynchronization process depends on the value of
DELTA. A value of DELTA = 1 is suggested.
Frame delineation acquisition speed can be improved by the implementation of multiple "virtual
framers", whereby the GFP process remains in the HUNT state and a separate PRESYNC substate
is spawned for each candidate GFP frame detected in the incoming octet stream, as depicted in
Figure 6-12.
6.3.2

Frame multiplexing

GFP frames from multiple ports and multiple client types are multiplexed on a frame-by-frame
basis. The choice of scheduling algorithms is outside the scope of this Recommendation.
When there are no other GFP frames available for transmission, GFP Idle frames shall be inserted,
thus providing a continuous stream of frames for mapping into an octet aligned physical layer.

ITU-T Rec. G.7041/Y.1303 (12/2001)

17

6.3.3

Client Signal Fail indication

GFP provides a generic mechanism for a GFP client-specific source adaptation process to propagate
a Client Signal Fail (CSF) indication to the far-end GFP client-specific sink-adaptation process on
detection of failure defect in the ingress client signal.
Detection rules for client signal fail events is, by definition, client-specific (see clauses 7 and 8).
Upon detection, a GFP source adaptation process should generate a Client Management frame
(PTI = 100). The PFI subfield is set to 0 (no Payload Information field FCS), and the EXI subfield
is set to the appropriate Extension Header type as applicable. The two types of CSF, use the
following UPI field values:

Loss of Client Signal (UPI = 0000 0001).

Loss of Client Character Synchronization (UPI = 0000 0010).

Upon detection of the CSF condition, the GFP client-specific source adaptation process should send
CSF indications to the far end GFP client-specific sink adaptation process once every
100 ms T 1000 ms, beginning at the next GFP frame. Interim frames shall be GFP Idle frames.
Upon reception of the CSF indication, the GFP client sink adaptation process declares a sink client
signal failure. Defect handling is discussed in 6.3.4.
The GFP client-specific sink adaptation process should clear the defect condition either:
1)
after failing to receive N CSF indications in N 1000 ms, (a value of 3 is suggested for N);
or
2)
upon receiving a valid GFP client data frame.
Handling of incomplete GFP frames at the onset of a CSF event should be consistent with the error
handling procedures specified in 7.3 for Frame-mapped GFP and 8.5 for Transparent-mapped GFP.
6.3.4

Defect handling in GFP

Figure 6-13 depicts the causal relationship between various defects detected or indicated by the
GFP process. Trail Signal Fail (TSF) events refer to failure events detected in the SDH or OTN
transport network as defined in ITU-T Recs G.783 and G.798. GFP Server Signal Fail events refer
to GFP Loss of Frame Delineation events as defined in the GFP state machine (see 6.3.1) or
propagation of TSF events to the GFP clients. CSF events refer to failure events detected in the
client signal on ingress (communicated to far-end by a CSF client management frame) or egress
(client-specific mapping defects such as payload errors, see clauses 7 and 8).
Ingress client process

Egress client process

CSF
GFP client-specific
source adaptation process

GFP client-specific
sink adaptation process

GFP common process

(Client source adaptation)

GFP common process

(Client sink adaptation)

SSF

TSF
Transport network

Transport network
T1545430-02

Figure 6-13/G.7041/Y.1303 Defect Signal Propagation in GFP

18

ITU-T Rec. G.7041/Y.1303 (12/2001)

Upon detection of a TSF event or a GFP Loss of Frame Delineation event, the GFP sink adaptation
process generates a GFP SSF indication to its client-specific sink adaptation processes. These
failure events are cleared as soon as the GFP process regains link synchronization.
Upon detection of CSF events other than a far-end CSF indication, the GFP client-specific sink
adaptation processes should take client-specific (as well as server-specific) actions to deal with
those failure events.
7

Payload specific aspects for frame-mapped GFP

This clause describes those aspects of the generic encapsulation specific to the adaptation of client
signals using a frame-by-frame mapping of the client payload into GFP.
7.1

Ethernet MAC payload

The format of Ethernet MAC frames is defined in IEEE 802.3, section 3.1. There is a one-to-one
mapping between a higher-layer PDU and a GFP PDU. Specifically, the boundaries of the GFP
PDU are aligned with boundaries of the framed higher layer PDUs. This relationship between
Ethernet MAC frames and GFP frames is illustrated in Figure 7-1.
Ethernet MAC Frame

7
1
6
6
2

Octets
Bits

GFP Frame
Octets
2
2
2
2
0-60

Preamble
Start of frame delimiter

PLI
cHEC
Type
tHEC
GFP Extension Hdr

Destination Address (DA)

Source Address (SA)
Length/Type
MAC client data
Pad
Frame Check Sequence (FCS)

GFP
Payload
T1545440-02

1 2 3

7 8

Figure 7-1/G.7041/Y.1303 Ethernet and GFP frame relationships

7.1.1

Ethernet MAC encapsulation

The Ethernet MAC octets from Destination Address through Frame Check Sequence, inclusive, are
placed in the GFP Payload Information field. Octet-alignment is maintained and bit identification
within octets is maintained. Specifically, on an octet-by-octet basis, bits 0 and 7 in clause 3 of
IEEE 802.3 correspond to bits 8 and 1, respectively, in this GFP Recommendation.
7.1.2

Ethernet Inter-Packet Gap (IPG) deletion and restoring

The following rules apply to the deletion and restoration of Ethernet IPGs when the client is not a
native frame-mapped GFP client:
1)
IPGs are deleted before the Ethernet MAC frame is processed by the GFP source adaptation
process and restored after the GFP frame is processed by the GFP sink adaptation process.
2)
IPGs are deleted as the Ethernet MAC frame is extracted from the client bit-stream. The
extracted (decoded) Ethernet MAC frame is then forwarded to the GFP source adaptation
process for subsequent encapsulation into a GFP frame.

ITU-T Rec. G.7041/Y.1303 (12/2001)

19

3)

IPGs are restored after the Ethernet MAC frame is extracted from the GFP frame by the
GFP termination element. The extracted (uncoded) Ethernet MAC frame is then forwarded
to the client layer for subsequent processing. IPGs are restored by ensuring that sufficient
octets containing an idle pattern of 00 hex are present between consecutive received
Ethernet MAC frames to meet the minimum receiver IFG requirements. Minimum receiver
IFG requirements are stated in IEEE 802.3, section 4.4.

7.2

IP/PPP payload

IP/PPP payloads are first encapsulated in an HDLC-like frame. The format of a PPP frame is
defined in IETF RFC 1661, section 2. The format of the HDLC-like frame is defined in IETF RFC
1662, section 3. Unlike IETF RFC 1662, no octet stuffing procedure is performed on flag or control
escape characters. There is a one-to-one mapping between a higher-layer PPP/HDLC PDU and a
GFP PDU. Specifically, the boundaries of the GFP PDU are aligned with boundaries of the framed
higher layer PPP/HDLC PDUs. This relationship between PPP/HDLC frame and the GFP frame is
illustrated in Figure 7-2.
Similar clients, such as MAPOS, are mapped in the same manner as PPP frames.
PPP/HDLC Frame

1
1
1
2

Octets
Bits

GFP Frame
Octets
2
2
2
2
0-60

Flag
Address
Control
PPP Type
Length/Type
PPP Information
(Pad)
Frame Check Sequence (FCS)

PLI
cHEC
Type
tHEC
GFP Extension Hdr

GFP
Payload
T1545450-02

1 2 3

7 8

Figure 7-2/G.7041/Y.1303 PPP/HDLC and GFP frame relationships

7.2.1

PPP frame encapsulation

All octets from the PPP/HDLC frame, including any optional PPP Information field padding, are
placed in the Payload Information field of a GFP frame. Octet alignment is maintained and bit
identification within octets is also maintained.
7.2.2

GFP/HDLC delineation interworking

GFP does not rely on flag characters, and associated control escape octet, for frame delineation
purposes. The following rules apply to the processing of Octet-Synchronous HDLC frames by a
GFP/HDLC interworking function:
1)
Flags and associated control escape octets are removed (as specified in IETF RFC 1662,
section 4.2) as the PPP/HDLC frame is extracted from the incoming client octet stream.
The extracted (decoded) PPP/HDLC frame is then forwarded to the GFP source adaptation
process for subsequent encapsulation into a GFP frame.

20

ITU-T Rec. G.7041/Y.1303 (12/2001)

2)

The GFP extracts the PPP/HDLC frame from the GFP frame. The extracted (uncoded)
PPP/HDLC frame is then forwarded to the client layer for subsequent processing. Flags and
control escape characters are then restored by inserting flag characters (e.g. hexadecimal
0x7e) and escape control characters (e.g. hexadecimal 0x7d) as specified in IETF RFC
1662, section 4.

7.2.3

PPP payload configuration options

Modifications to the PPP/HDLC-like frame format may be negotiated using the Link Configuration
Protocol (LCP) Configuration Options procedures as defined in IETF RFC 1661, section 6. For
example, the format of the GFP frame after a successful negotiation of the Address-and-ControlField-Compression (ACFC) Configuration Option is illustrated in Figure 7-3. Such configuration
procedures are client-specific and transparent to GFP process.
PPP/HDLC-like Frame

GFP Frame
Octets
2
2
2
2
0-60

1-2
1
2

PPP Information

PPP Padding (Optional)

PPP/HDLC FCS

Octets
Bits

PLI
cHEC
Type
tHEC
GFP Extension Hdr

PPP Protocol

GFP
Payload

T1545460-02

1 2 3

7 8

Figure 7-3/G.7041/Y.1303 PPP/HDLC/HDLC and GFP frame relationships

(with PPP's ACFC configuration option)
7.3

Error handling in frame-mapped GFP

On ingress, PDUs detected in error before transmission by the client source adaptation process
should be discarded. PDUs detected in error while in transmission by the client source adaptation
process should be padded up with an all ones bit sequence, and transmitted with a Payload FCS
which has all 32-bits complemented, when present. These actions ensure that the termination GFP
process, or the client end, will drop the errored PDU.
8

Payload-specific aspects for transparent mapping of 8B/10B clients into GFP

Transparent mapping of 8B/10B payloads into GFP is intended to facilitate the transport of 8B/10B
block-coded client signals for scenarios that require very low transmission latency. Examples of
such client signals include Fibre Channel, ESCON, FICON, and Gigabit Ethernet. Rather than
buffering an entire frame of the client data into its own GFP frame, the individual characters of the
client signal are demapped from the client block codes and then mapped into periodic, fixed-length
GFP frames. The mapping occurs regardless of whether the client character is a data or a control
character, which thus preserves the client 8B/10B control codes. Frame multiplexing is not
precluded with transparent GFP.
The transparent GFP frame uses the same frame structure as the frame-mapped GFP, including the
required Payload Header. The Payload FCS is optional. The transparent GFP frame format is
depicted in Figure 8-1.

ITU-T Rec. G.7041/Y.1303 (12/2001)

21

16-bit Payload
length indicator
cHEC
(CRC-16)

Core
header

16-bit Payload
type field

Payload
headers
(4-64 bytes)

tHEC
(CRC-16)

Payload
area
Payload
information
field
[N (8 65B + 16)]

Optional
payload FCS
(CRC-32)
T1545470-02

Figure 8-1/G.7041/Y.1303 Transparent GFP frame format

8.1

Adapting 8B/10B client signals via 64B/65B block codes

As depicted in the Functional Model in Figure 2, the first step in the client adaptation process is
decoding the physical layer of the client signal. For 8B/10B line codes, the received 10-bit character
is decoded into its original 8-bit value, if it is an 8B/10B data codeword, or into a control character
if it is an 8B/10B control codeword. The 8B/10B control codewords are mapped into one of the 16
possible 4-bit Control Code Indicators for the 8-bit control characters available in transparent GFP.
(See Table 8-1.)
Table 8-1/G.7041/Y.1303 Mapping between 8B/10B Control Characters
and the 64B/65B Control Code Indicators

22

Name

Octet Value

10B Codeword (RD

)
abcdei fghj

10B Codeword (RD+)

abcdei fghj

64B/65B
4-bit Mapping

/K28.0/

1C

001111 0100

110000 1011

0000

/K28.1/

3C

001111 1001

110000 0110

0001

/K28.2/

5C

001111 0101

110000 1010

0010

/K28.3/

7C

001111 0011

110000 1100

0011

/K28.4/

9C

001111 0010

110000 1101

0100

/K28.5/

BC

001111 1010

110000 0101

0101

/K28.6/

DC

001111 0110

110000 1001

0110

/K28.7/

FC

001111 1000

110000 0111

0111

/K23.7/

F7

111010 1000

000101 0111

1000

/K27.7/

FB

110110 1000

001001 0111

1001

ITU-T Rec. G.7041/Y.1303 (12/2001)

Table 8-1/G.7041/Y.1303 Mapping between 8B/10B Control Characters

and the 64B/65B Control Code Indicators
Name

Octet Value

10B Codeword (RD

)
abcdei fghj

10B Codeword (RD+)

abcdei fghj

64B/65B
4-bit Mapping

/K29.7/

FD

101110 1000

010001 0111

1010

/K30.7/

FE

011110 1000

100001 0111

1011

10B_ERR

N/A

Unrecognized RD

Unrecognized RD+

1100

65B_PAD

N/A

N/A

N/A

1101

Spare

N/A

N/A

N/A

1110

Spare

N/A

N/A

N/A

1111

NOTE 1 While all 256 data characters must be supported, only 12 special 8B/10B control codewords
are recognized and used for 64B/65B control characters in Gigabit Ethernet, Fibre Channel, FICON,
and ESCON. Hence, the compression of special 8B/10B control codewords into 4-bit values is possible
without restricting client signals, or providing protocol-specific handling of 8B/10B control codewords.
NOTE 2 The recoding process is entirely unaware of the meaning of control words or ordered sets. It
simply generically recodes data and control words into 65B blocks. No knowledge of start-of-frame,
end-of-frame, errors, idles, control codes, ordered sets, etc. is required.

The decoded 8B/10B characters are then mapped into a 64-bit/65-bit (64B/65B) block code. The
structure of the 64B/65B block code is shown in Figure 8-2. The leading bit of the 65-bit block, the
Flag bit, indicates whether that block contains only 64B/65B 8-bit data characters or whether client
control characters are also present in that block. (Flag bit = 0 indicates data octets only and Flag
bit = 1 indicates at least one control octet in the block). Client control characters, which are mapped
into 8-bit 64B/65B control characters, are located at the beginning of the 64-bit block payload if
they are present in that block. The first bit of the 64B/65B control character contains a Last Control
Character (LCC) flag bit which indicates whether this control character is the last one in this block
(LCC = 0), or whether there is another control character in the next octet (LCC = 1). The next three
bits contain the Control Code Locator, which indicates the original location of the 8B/10B control
code character within sequence of the eight client characters contained in the block. The last 4 bits,
the Control Code Indicator, give the 4-bit representation of the 8B/10B control code character. The
explicit mapping of 8B/10B control code characters into the 4-bit Control Codes is defined in
Table 8-1. The control codes are mapped into the payload bytes of the 64B/65B code in the order in
which they were received. Note that, as a result, the control code addresses aaa-hhh in Figure 8-2
will be in ascending order.

ITU-T Rec. G.7041/Y.1303 (12/2001)

23

Input Client
Characters

Flag
Bit

All data

D1

D2

D3

D4

D5

D6

D7

D8

7 data,
1 control

0 aaa C1

D1

D2

D3

D4

D5

D6

D7

6 data,
2 control

1 aaa C1

0 bbb C2

D1

D2

D3

D4

D5

D6

5 data,
3 control

1 aaa C1

1 bbb C2

0 ccc C3

D1

D2

D3

D4

D5

4 data,
4 control

1 aaa C1

1 bbb C2

1 ccc C3

0 ddd C4

D1

D2

D3

D4

3 data,
5 control

1 aaa C1

1 bbb C2

1 ccc C3

1 ddd C4

0 eee C5

D1

D2

D3

2 data,
6 control

1 aaa C1

1 bbb C2

1 ccc C3

1 ddd C4

1 eee C5

0 fff C6

D1

D2

1 data,
7 control

1 aaa C1

1 bbb C2

1 ccc C3

1 ddd C4

1 eee C5

1 fff C6

0 ggg C7

D1

8 control

1 aaa C1

1 bbb C2

1 ccc C3

1 ddd C4

1 eee C5

1 fff C6

1 ggg C7

0 hhh C8

64-Bit (8-Octet) Field

Leading bit in a control octet (LCC) = 1 if there are more control octets and = 0 if this payload octet contains the last
control octet in that block.
aaa = 3-bit representation of the 1st control code's original position (1st Control Code Locator).
bbb = 3-bit representation of the 2nd control code's original position (2nd Control Code Locator).

hhh = 3-bit representation of the 8th control code's original position (8th Control Code Locator).
Ci = 4-bit representation of the ith control code (Control Code Indicator).
Di = 8-bit representation of the ith data value in order of transmission.

Figure 8-2/G.7041/Y.1303 Transparent GFP 64B/65B code components

(see Figure 8.3 for actual superblock structure)
For example, if there is a single 64B/65B control character in a block and it was originally located
between 8B/10B data codewords D2 and D3, the first octet of the 64B/65B block will contain
0.010.C1. The LCC value of 0 indicates that this 64B/65B control character is the last one in that
block and the value of aaa = 010 indicates C1's location between D2 and D3. At the demapper, the
64B/65B data characters are remapped as 8-bit data octets and then encoded back into the
8B/10B data codewords. For 64B/65B control characters, the four-bit Control Code Indicators are
remapped into the appropriate 8B/10B control codewords with their positions within the original
character stream restored based on the three-bit Control Code Locator.
8.1.1

10B_ERR code

Certain client signal defects may produce 8B/10B codewords on ingress to the GFP source
adaptation process that cannot be recognized by the 64B/65B adaptation process (e.g. a client signal
failure, an illegal 8/10B codeword or a legal codeword with a running disparity error, see 8.3). A
special 64B/65B control character, the 10B_ERR code, is provided to convey such "unrecognized
8B/10B codeword" client signal defects.
When reconstructing the client signal on egress from the transport network, received 10B_ERR
code are typically recoded by the demapper into either 001111 0001 (RD) or 110000 1110 (RD+)
(fixed illegal 8B/10B codewords with neutral disparity), depending on running disparity (see 8.3 for
other client-specific running disparity considerations). Although the actual value of the
unrecognized 8B/10B codeword is not retained, the occurrence and location of the client signal
defect are preserved.
24

ITU-T Rec. G.7041/Y.1303 (12/2001)

8.1.2

Insertion of 65B_PAD code and GFP Client Management frames

Since the Transparent GFP application requires that the available path (channel) capacity is at least
that of the client signal base (pre-encoding) data rate, the input receive (ingress) buffer at the
mapper will regularly approach underflow. For rate adaptation purposes, if a transparent GFP frame
is currently being transmitted, and if there are no client characters ready for transmission by the
transparent GFP mapper, the mapper shall insert a 65B_PAD padding character. The pad character
is mapped into the GFP frame in the same manner as a control character and is recognized and
removed by the GFP demapper. Client-specific considerations for 65B_PAD code handling are
given in 8.4.
Client data frames are transmitted with priority over Client Management frames. If a GFP Client
Management frame is available to transmit, and the ingress buffer is nearly empty (e.g. if a
65B_PAD character has been sent during the current client data frame), then the Client
Management frame may be sent after the current client data frame. In order to maintain low latency,
it is recommended that, for a right-sized channel only, a single Client Management frame be sent
between client data frames. It is also recommended that Client Management frames used with
Transparent GFP be limited to a Payload Information Field of eight bytes or less. Note that low
latency may also be maintained by increasing the channel size to allow the exchange of additional
Client Management frames.
8.2

Adapting 64B/65B code blocks into GFP

To preserve the octet alignment of the Transparent GFP signal with the transport SDH/ODUk
frame, the first step in the adaptation process is to group eight 64B/65B codes into a superblock as
shown in Figure 8-3. The leading (Flag) bits of each of the eight 64B/65B codes are grouped
together into a first trailing octet. The sixteen bits of the last two trailing octets are used for a
CRC-16 error check over the bits of this superblock.
Octet 1, 1
Octet 1, 2
Octet 1, 3
.
.
.
Octet 8, 7
Octet 8, 8
L1

L2

L3

L4

L5

L6

L7

L8

CRC1

CRC2

CRC3

CRC4

CRC5

CRC6

CRC7

CRC8

CRC9

CRC10

CRC11

CRC12

CRC13

CRC14

CRC15

CRC16

where:

th

th

Octet j, k is the k octet of the j 64B/65B code in the superblock

th

Lj is the leading (Flag) bit j 64B/65B code in the superblock

th

CRCi is the i error control bit where CRC1 is the MSB of the CRC

Figure 8-3/G.7041/Y.1303 Superblock structure for mapping 64B/65B

code components into the GFP frame
NOTE To minimize latency, the transparent GFP mapper can begin transmitting data as soon as the first
64B/65B code in the group has been formed rather than waiting for the entire superblock to be formed.

ITU-T Rec. G.7041/Y.1303 (12/2001)

25

Assuming no Payload FCS and a Null Extension header, the resulting GFP frame is
[(N ((65 8) + 16) + (8 8)] bits long, where N is the number of superblocks in the GFP frame.
The value of N depends on the base, uncoded, rate of the client signal and on the transport channel
capacity. Suggested SDH virtually concatenated channel capacities and the associated minimum
values for N are shown in Appendix IV. Suggested channel capacities for other transport paths are
for further study. The minimum value of N depends on the data rate of the client signal, the number
of GFP frame overhead octets (e.g. 8 with no optional Payload FCS and a Null Extension Header),
and the size of the payload envelope, as shown in Appendix IV. Specifically, Nmin must be chosen
such that for the fastest tolerance client clock rate and slowest tolerance SDH/OTN clock rate, the
time required to transmit the GFP frame containing the N 8 8 client characters is less than the
time in which the client can deliver these N 8 8 characters to the GFP mapper.
Note that N may be optionally configurable according to spare bandwidth requirements for the
transport of Client Management frames. See Appendix IV.
8.2.1

Error control with Transparent GFP

The 16 error control bits in a superblock (see Figure 8-3) contain a CRC-16 error check code over
the 536 bits in that superblock. If the demapper detects an error, it should output either 10B error
characters or unrecognized 10B characters in place of all of the client characters contained in that
superblock. The 10B error and unrecognized characters are described for disparity errors in the
client-specific aspects (see 8.4). This replacement guarantees that the client receiver will be able to
detect the presence of the error.
16

15

12

10

The generator polynomial for the CRC-16 is G(x) = x + x + x + x + x + x + x + x + 1 with

16
0
an initialization value of zero, where x corresponds to the MSB and x to the LSB. The
superblock CRC is generated by the source adaptation process using the following steps:
1)
The first 65 octets of the superblock are taken in network octet order (see Figure 8-3), most
significant bit first, to form a 520-bit pattern representing the coefficients of a polynomial
M(x) of degree 519.
16

2)

M(x) is multiplied by x
degree 15 or less.

3)

The coefficients of R(x) are considered to be a 16-bit sequence, where x

significant bit.
This 16-bit sequence is the CRC-16.

4)

and divided (modulo 2) by G(x), producing a remainder R(x) of

15

is the most

NOTE Single error correction is also possible with this CRC-16. However, since the sink adaptation
process performs the CRC-16 check after the payload descrambling is performed, the error correction circuit
should account for single bit errors as well as double errors spaced 43 bits apart coming out of the
descrambler.

The sink adaptation process performs steps 1)-3) in the same manner as the source adaptation
process. In the absence of bit errors, the remainder shall be 0000 0000 0000 0000.
8.3

Running disparity in 64B/65B codes

8B/10B codewords are designed to facilitate error-free transmission by maintaining DC balance,

providing significant transitions for clock recovery, and limiting run-length of consecutive 1s or 0s.
DC balance is measured on a codeword by codeword basis by keeping track of "running disparity."
Running disparity is either positive (indicating more 1s than 0s have been sent), or negative (more
0s than 1s sent).

26

ITU-T Rec. G.7041/Y.1303 (12/2001)

In order to maintain DC balance in 8B/10B codewords, each 8-bit data character and each of the 12
recognized "special control characters" have two 10-bit encodings. Depending on current running
disparity, the 8B/10B encoder will select which of the two encodings to transmit for next data or
control character in order to either flip the running disparity, or to maintain the current running
disparity. Specifically, the new codeword flips the running disparity from negative to positive if
there have been more 0s than 1s transmitted, from positive to negative if there have been more 1s
than 0s transmitted, or maintains the running disparity if there has been an equal number of 1s
and 0s.
Transmission bit errors may cause a received 8B/10B codeword to have the wrong disparity for the
current beginning running disparity state. In these cases, a running disparity error is detected.
Independent of the received character's validity, the received transmission character shall be used to
calculate a new value of running disparity. The new value shall be used as the receiver's current
running disparity for the next received transmission character.
NOTE Transmission bit errors may also result in the errored codeword being received with correct
disparity and a corrupted but legal 8B/10B codeword that results in some later non-errored codeword being
detected with a running disparity error. In some cases, protocol-specific running disparity rules have been
created to ensure each data packet begins or ends with defined disparity so that errors will not be propagated
across data packets.

8.3.1

Handling of running disparity on ingress

On ingress, the initial running disparity, upon power-on, reset, or transition from a loss of signal or
loss of codeword synchronization phase, may be assumed either positive or negative.
A match to the received 10B character is searched for in the appropriate RD+ or RD column of the
8B/10B valid codeword lookup table, depending on the current beginning running disparity. If no
match is found, either an illegal codeword or a legal codeword with a running disparity error has
been detected. Both are treated as 8B/10B code violations, and are replaced with the 10B_ERR
code in the 64B/65B mapping process.
8.3.2

Handling of running disparity on egress

On egress, the initial running disparity upon power-on, reset, or transition from a loss of signal or
loss of codeword synchronization phase, shall be assumed to be negative.
Transparent transport implementations must generate correct running disparity using any applicable
protocol-specific rules. References are provided in 8.3.3 to the standard(s) that define each currently
applicable protocol's disparity rules.
10B_ERR codes are recoded into client signals either as an unrecognized codeword with valid
running disparity, or as a protocol-specific error, depending on the protocol, as described in 8.3.3.
8.3.3

Client-specific running disparity aspects

This clause describes the client-specific running disparity rules for each of the identified, supported
8B/10B client protocols.
8.3.3.1

Fibre Channel payload

Running disparity rules for Fibre Channel are found in ANSI X3.230, Fibre Channel Physical and
Signaling Interface (FC-PH), Rev 4.3, section 11. In addition to the "generic" running disparity
rules specified in section 11.2, Fibre Channel specific rules in section 11.4 provide two versions of
each EOF ordered set, and dictate their use to ensure that negative running disparity will result after
processing of the final character of the EOF ordered set. Ordered sets defined for the primitive
signals and primitive sequences preserve this negative disparity, ensuring that the ordered sets
associated with SOF delimiters, primitive signals, and primitive signals will also always be
transmitted with negative beginning running disparity. This restriction allows Fibre Channel Idle
ITU-T Rec. G.7041/Y.1303 (12/2001)

27

words to be removed and added from an encoded bit stream one word at a time without altering the
beginning running disparity.
To prevent subsequent valid Fibre Channel frames from being declared invalid, the K28.5 character
associated with all ordered sets except EOF should be generated assuming beginning negative
running disparity. In the event that a previous transmission error results in an incorrect EOF for the
current running disparity, the next ordered set will be generated with RD K28.5, forcing ending
running disparity to be negative. As a result, transmission errors will not cause a running disparity
error to be propagated across frames.
For "transparent transport" of Fibre Channel payloads, 10B_ERR shall be recoded into one of the
following unrecognized 10B neutral disparity codewords, depending on beginning running
disparity: 001111 0001 (RD) or 110000 1110 (RD+).
8.3.3.2

ESCON payload

Running disparity rules for ESCON are found in ANSI X3.296, Information
Technology Single-Byte Command Code Sets Connection (SBCON) Architecture, section 6.2.2.
Since ESCON does not define an error code to substitute for code violations, on egress, 10B_ERR
shall be recoded into one of the following unrecognized 10B neutral disparity codewords,
depending on beginning running disparity: 001111 0001 (RD) or 110000 1110 (RD+).
8.3.3.3

FICON payload

For purposes of mapping into transparent GFP, the running disparity rules for FICON are identical
to those specified for Fibre Channel in ANSI X3.230, Rev 4.3.
8.3.3.4

Gigabit Ethernet payload

Running disparity rules for Gigabit Ethernet are found in IEEE 802.3, section 36.2.4. Two Idle
encodings are provided, indicated as /I1/ and /I2/. The first /I/ following a packet or Configuration
ordered set restores the current running disparity to a negative value. All subsequent /I/s are /I2/ to
ensure negative ending running disparity. This restriction allows single /I2/s to be inserted/removed
for rate adaptation without altering the beginning running disparity associated with the code-group
subsequent to the inserted or removed /I2/.
In order to ensure beginning negative running disparity for each SOF, all /I2/ Idles should be
generated with RD K28.5, insuring beginning negative running disparity for the next Idle or SOF.
Per section 36.2.4.16 of IEEE 802.3, running disparity errors detected on ingress (and replaced with
10B_ERR codeword in 64B/65B encoding process), should be replaced with /V/ codeword (K30.7)
having correct disparity on egress. As an option, it is also permissible to recode received 10B_ERR
into one of the following unrecognized 10B neutral disparity codewords, depending on beginning
running disparity: 001111 0001 (RD) or 110000 1110 (RD+).
8.4

Rate adaptation in 64B/65B codes

On ingress, rate adaptation to the output payload data rate occurs in the 64B/65B encoding process.
If there is not an 8B/10B codeword available for the mapper to recode into 64B/65B block code the
mapper inserts a 65B_PAD as described in 8.1.2. Essentially, this 65B_PAD is a non-client-idle that
is used to pad 64/65B blocks for rate adaptation purposes. On egress, the demapper removes these
non-client-idle signals. Since fixed length GFP frames are used, and frames may be padded with
65B_PADs for rate adaptation, there is no need to buffer an entire GFP frame prior to inserting it
into the payload of the outgoing transport signal, thus reducing buffering and delay in the mapping
process.

28

ITU-T Rec. G.7041/Y.1303 (12/2001)

8.4.1

Egress rate adaptation procedures

There are two approaches for generating the client egress data interface clock at the GFP
client-specific sink-adaptation process. One approach is to adapt the client signal to a clock source
that is local to the GFP sink adaptation process. The other approach is to generate the client signal
egress block by deriving it from the received GFP signal and transport clock.
Should a failure occur in either the ingress client signal, or during SDH/OTN transport, a
protocol-specific local reference clock is still required at the client data egress point if the client
expects a client rate link failure signal to replace the failed client.
8.4.1.1

Rate adaptation to a local reference clock

The currently supported 8B/10B client signals specify operating frequencies with clock-offset
requirements of 100 ppm to 200 ppm, which are significantly relaxed compared to SDH or OTN.
Each of these client signals is designed to allow rate adaptation to a local reference clock, either at
repeaters or at the far-end, through client Idle (or fill-word) insertion or removal. To facilitate this
rate adaptation, each of these client signals impose minimum Inter-Packet Gap (IPG) rules, which
specify the minimum number of Idle codewords which must be inserted between data packets. Each
of these client signals also specifies the maximum data packet size. Minimum IPG rules have been
established to insure that where rate adaptation to a local clock is required, even under the worst
case condition where a fast input clock and slow output clock require some IPG Idles to be deleted,
sufficient IPG will remain between packets for successful client frame delineation.
This scheme may be employed equally well when reconstructing transparent-mapped client data on
egress. With this approach, a local reference clock is supplied at the GFP sink adaptation process.
As client data is demapped from GFP frames and recoded into 8B/10B codewords, it is rate adapted
to the local reference clock through idle insertion/removal. Client-specific processing is required to
recognize legal opportunities to insert/remove idle codewords, generate proper idle codes, and insert
such codes in the egress bit-stream. An example of a client specific parameter is the minimum and
maximum number of idles that are allowed to be inserted or removed.
Even in links containing multiple repeaters, if all "local" clocks meet the accuracy requirements for
the specific protocol, sufficient opportunities for idle insertion or removal will occur, since
aggregate timing offsets through cascaded repeaters cannot exceed worst-case clock offset
requirements.
With this approach, timing characteristics such as the jitter and wander of the reconstructed client
signal depend primarily on the quality of the local reference clock. The local reference clock is
protocol rate specific (e.g. Gigabit Ethernet, Fibre Channel, and ESCON do not share common
frequencies).
8.4.1.2

Rate adaptation from the transported client signal

Client signals are provided at a smooth protocol-specific clock rate on ingress. While there may be
gaps in the client data packets themselves, these are filled with inter-packet gap (IPG) at a constant
clock rate. Transparent mapping preserves all of the client data, control, and IPG information when
recoding it using 64B/65B (assuming no client Loss of Signal or Loss of Character Synchronization
occurs). However, the recoded data is then mapped into GFP frames with 65B_PAD stuffing to
rate-adapt to the higher bandwidth transport payload channel. GFP Client Management or control
frames may also be inserted periodically or opportunistically between GFP client data frames.
Transport frames add their own overhead (Section and Path Overhead plus fixed stuff bytes in the
case of SDH). No alignment between client data, stuffing bytes or blocks, GFP frames, transport
overhead is maintained.

ITU-T Rec. G.7041/Y.1303 (12/2001)

29

On egress, clock recovery is expected to require a FIFO and desynchronizer, where the
desynchronizer would require a reference clock, PLL, and filter. Recovered clock timing would
depend on some filtered version of the FIFO fill level. The FIFO itself would be subject to fairly
dramatic changes in level under normal operating conditions due to the occurrence of large blocks
of section/transport overhead, GFP frame overhead, and GFP Client Management frames. Under
worst-case conditions, it is possible that all of the client data "gapping" mechanisms will align into
one contiguous "no client data" block. The relatively non-periodic nature of some of the gaps
combined with the relative large client data source clock frequency tolerance complicate the FIFO
and PLL design.
The advantage of this desynchronizer approach is that no protocol-specific knowledge is required to
recover client clock on egress.
The jitter and wander timing characteristics of the reconstructed client signal depend primarily on
the design of the clock recovery system. With a more complex design, a wide range of client rates
may be supported with a single design.
8.4.2

Client-specific rate adaptation aspects

On egress, transparently transported client signals must be reconstructed and output in a manner
that is compliant with the physical interface requirements specific to each protocol. Regardless of
the selected client egress timing approach, protocol-specific timing requirements must be met, as
defined in applicable standards for each client protocol. The following clauses identify key
applicable requirements, but other protocol-specific requirements may apply.
8.4.2.1

Fibre Channel payload

Fibre Channel full rate output data rate (after 8B/10B encoding) shall be 1062.5 Mbit/s 100 ppm,
as specified in ANSI X3.230, Fibre Channel Physical and Signaling Interface (FC-PH), Rev 4.3,
section 5.1. Output signal timing requirements are further specified in ANSI X3.230, sections 6.1.1
(Single-mode optical output interface), 6.2.1 (Multi-mode optical output interface), and 7 (Electrical
cable interface). Output signals will normally be generated with a minimum of six Primitive Signals
(Idles and R_RDY) between frames, as specified in ANSI X3.230, section 17.1. If rate adaptation is
performed using Fibre Channel Idle insert/removal, rate adaptation shall be applied such that the
receiving destination receives at least two Idles preceding each frame, as specified in ANSI X3.230,
section 17.1.
Rate adaptation may also be required when a continuous stream of Fibre Channel primitive
sequences is received, where primitive sequences are defined in Table 26 of ANSI X3.230. Since a
minimum of three consecutive identical primitive sequences are required to be received before the
sequence is recognized (per section 16.4.1 of ANSI X3.230), rate adaptation by inserting one
replica of the received four-character sequence, or deleting a received sequence, shall only occur
after three consecutive identical sequences have been received and retransmitted.
Depending on implementation, a continuous stream of 10B_ERR neutral disparity characters could
be generated at egress, although rate adaptation is still required here. In this case, rate adaptation
may be performed by removing or inserting a 10B_ERR neutral disparity character after 12
consecutive 10B_ERR characters have been received and retransmitted.
8.4.2.2

ESCON payload

ESCON output data rate (after 8B/10B encoding) shall be 200 Mbit/s 0.04 Mbit/s, as specified in
ANSI X3.296, Information Technology Single-Byte Command Code Sets Connection (SBCON)
Architecture, section 5.1.2. Output signal timing requirements are further specified in
ANSI X3.296, sections 5.2.1 (Multi-mode output interface) and 5.3.1 (Single-mode output
interface). Output signals will normally be generated with a minimum of four idle characters
(K28.5) between data frames, as specified in ANSI X3.296, section 6.3. If rate adaptation is
30

ITU-T Rec. G.7041/Y.1303 (12/2001)

performed using ESCON Idle insert/removal, either one or two idle characters may be added or
removed between frames, according to the rules of ANSI X3.296, section 7.2.
Rate adaptation may also be required when a continuous stream of ordered set sequences is
received, where ordered set sequences are defined in Table 15 of ANSI X3.296. Since a minimum
of eight consecutive sequences are required to be received before the sequence is recognized (per
section 6.3 of ANSI X3.296), rate adaptation by inserting a replica of the received two-character
sequence, or deleting a received sequence shall only occur after eight consecutive identical
sequences have been received and retransmitted.
Depending on implementation, a continuous stream of 10B_ERR neutral disparity characters could
be generated at egress, although rate adaptation is still required here. In this case, rate adaptation
may be performed by removing or inserting a 10B_ERR neutral disparity character after 12
consecutive 10B_ERR characters have been received and retransmitted.
8.4.2.3

FICON payload

The timing requirements for FICON are the same as those specified for Fibre Channel in ANSI
X3.230, Rev 4.3.
8.4.2.4

Full-duplex Gigabit Ethernet payload

Gigabit Ethernet (GbE) output data rate (after 8B/10B encoding) shall be 1250 Mbit/s 100 ppm,
as specified in IEEE 802.3. Output signal timing requirements are further specified in IEEE 802.3,
sections 38.5 and 38.6 (1000BASE-LX optical fiber interfaces), and 39.3.1 and 39.3.3
(1000BASE-CX (short-haul copper interface). Output signals will normally be generated with a
minimum IPG of 12 octets, per IEEE 802.3, section 4.4.2.3. GbE Idle characters are two octets, as
defined in IEEE 802.3, section 36.2.4.12. If rate adaptation is performed using full-duplex GbE Idle
insert/removal, only a single /I2/ should be removed in any IPG, and only when its removal shall
not result in no /I/ and not less than 8 octets including /T/, /R/, and /I/ remaining between frames,
for successful frame delineation according to IEEE 802.3, Figures 36-7a and 36-7b.
Rate adaptation may also be required when a continuous stream of eight-character Configuration
ordered sets (consisting of alternating /C1/C2/) is received. Since a minimum of three consecutive
/C1/C2/ Configuration ordered sets are required to be received before the Configuration set is
recognized, rate adaptation by inserting a replica of the received /C1/C2/ sequence, or deleting a
received /C1/C2/ sequence shall only occur after three consecutive identical /C1/C2/ sequences
have been received and retransmitted.
Depending on implementation, a continuous stream of 10B_ERR neutral disparity or transmission
error (/V/) characters could be generated at egress, which still requires rate adaptation. In this case,
rate adaptation may be performed by removing or replicating a 10B_ERR or /V/ character after 12
consecutive 10B_ERR or /V/ characters have been received and retransmitted.
8.5

Client-specific Signal Fail aspects

When transparent GFP mapping detects a client signal failure at ingress, it may send a "Client
Signal Fail" indication as described in 6.3.3. Client signal fail conditions include, as a minimum,
loss of 8B/10B synchronization and, in some cases, loss of signal. Other implementation-dependent
indications of a failed client signal (e.g. loss-of-clock from an interface between integrated circuits)
may be encoded as Client Signal Fail.
Since client signals are provided as a continuous serial stream of 10-bit characters, it is necessary to
find codeword alignment. Special characters containing the "comma" delimiter provide the
information necessary to achieve and maintain codeword alignment. While all 8B/10B client signals
employ the same bit alignment technique, conditions for detecting and clearing loss of 8B/10B
synchronization are protocol-specific, and are identified in following protocol-specific clauses.

ITU-T Rec. G.7041/Y.1303 (12/2001)

31

Server layer failures, in the GFP process itself, in the 64B/65B adaptation process, or in the
transport network, may induce a CSF indication to the client adaptation process.
If the onset of CSF occurs within a GFP client data frame, the remainder of the 64B/65B blocks of
that GFP frame shall be filled with 10B_ERR codes. At the far-end these shall be decoded as errors.
At the far-end of a transport network, transparently transported client signals must still be
reconstructed and output in a manner that is compliant with the physical and coding interface
requirements specific to each protocol. The following client-specific clauses define what action
should be taken at client signal egress in response to a received far-end Client Signal Fail
indication, or any adaptation or transport defects that make it impossible to extract a client signal.
8.5.1
8.5.1.1

Fibre Channel payload

Fibre Channel Loss of Light (LOL)

Fibre Channel Loss of Signal is an implementation-dependent option. When supported, applicable

Loss of Light and Signal Detect requirements are found in sections 5.6, 6.2.3.2 and H.10 of ANSI
X3.230, Fibre Channel Physical and Signaling Interface (FC-PH), Rev 4.3.
Other implementation-dependent indications of a failed client signal (e.g. loss-of-clock from a
SerDes) may be encoded as Client Signal Fail.
8.5.1.2

Fibre Channel 8B/10B Loss of Synchronization

Fibre Channel conditions for declaring in/out of 8B/10B codeword synchronization are specified in
section 12.1 of ANSI X3.230.
8.5.1.3

Fibre Channel output due to ingress or transport Signal Fail

Since the goal of Transparent GFP Mapping is to transport client signals as transparently as
possible, it is not appropriate to initiate link initialization or link recovery procedures on egress due
to Client Signal Fail or transport failures. It is recommended that the egress Fibre Channel
transmitter continuously output the neutral disparity decoding for 10B_ERR, forcing
Loss-of-Synchronization detection and the associated action at the downstream Fibre Channel
receiver. Alternatively, the egress transmitter may generate the Not_Operational primitive per
section 16.4.2 of ANSI X3.230.
If the CSF condition persists, the client adaptation process may transmit nothing, forcing LOS
detection and associated action at the downstream Fibre Channel receiver.
8.5.2
8.5.2.1

ESCON Payload
ESCON Loss of Signal (LOS)

Optical Loss of Signal detection requirements are specified in ANSI X3.296, Information
Technology Single-Byte Command Code Sets Connection (SBCON) Architecture, sections 5.2
and 5.3 for multi-mode and single-mode interfaces, respectively.
8.5.2.2

ESCON 8B/10B Loss of Synchronization

ESCON conditions for declaring being in or out of 8B/10B codeword synchronization are specified
in ANSI X3.296, section 7.1.
8.5.2.3

ESCON output due to ingress or transport Signal Fail

Since the goal of Transparent GFP Mapping is to transport client signals as transparently as
possible, it is not appropriate to initiate link initialization or link recovery procedures on egress due
to Client Signal Fail or transport failures. It is recommended that the egress ESCON transmitter
continuously output the neutral disparity decoding for 10B_ERR, forcing Loss-of-Synchronization
detection and the associated action at the downstream ESCON receiver. Alternatively, the egress
32

ITU-T Rec. G.7041/Y.1303 (12/2001)

transmitter may generate the Not-operational sequence per section 7.4.2 of ANSI X3.296.
If the CSF condition persists, the client adaptation process may transmit nothing, forcing LOS
detection and associated action at the downstream ESCON receiver.
8.5.3

FICON Payload

The CSF handling requirements for FICON are identical to those for Fibre Channel, as specified in
ANSI X3.230, Rev 4.3.
8.5.4
8.5.4.1

Full-duplex Gigabit Ethernet payload

Gigabit Ethernet Loss of Signal

Gigabit Ethernet Physical Media Dependent (PMD) Signal Detect requirements are specified in
sections 38.2.4 and 39.2.3 of IEEE 802.3 for fiber and copper interfaces, respectively.
8.5.4.2

Gigabit Ethernet 8B/10B Loss of Synchronization

Gigabit Ethernet conditions for declaring being in or out of 8B/10B codeword synchronization are
specified in IEEE 802.3, section 36.2.5.2.6 and Figure 36-9.
8.5.4.3

Gigabit Ethernet output due to ingress or transport Signal Fail

Since the goal of Transparent GFP Mapping is to transport client signals as transparently as
possible, it is not appropriate to initiate link initialization or link recovery procedures on egress due
to Client Signal Fail or transport failures. It is recommended that the egress GbE transmitter
continuously output the /V/ ordered set per section 36.2.4.16 of IEEE 802.3, forcing
Loss-of-Synchronization detection and the associated action at the downstream GbE receiver.
If the CSF condition persists, the client adaptation process may transmit nothing, forcing LOS
detection and associated action at the downstream GbE receiver.

Appendix I
Examples of functional models for GFP applications
This appendix presents some examples of functional models for GFP applications. In the absence of
layer network architectures for data layer networks (e.g. IP and Ethernet), the models presented are
for illustration purposes only.
GFP can be deployed in transport network elements (e.g. SDH) and in data network elements
(e.g. IP, Ethernet).
In the former case, a physical data interface (Ethernet or Storage Area Network type) is provided as
a tributary interface port on the transport network element. For the case where the physical data
signal is an 8B/10B coded signal, it can be transported through the transport network as a
transparent stream using GFP-T mapping (Figure I.1). For the case where only a part of the physical
interface bandwidth is carrying traffic, and only this traffic is to be transported through the transport
network, the physical data interface signal is terminated, data PDUs are extracted and forwarded via
GFP-F mapping over aVC-m-Xv, VC-n, VC-n-Xc, or VC-n-Xv signal (Figure I.2).
In the latter case, GFP processing is performed inbetween the IP Router [Ethernet Switch] fabric
and the e.g. STM-N interface port functions (Figures I.3 and I.4).

ITU-T Rec. G.7041/Y.1303 (12/2001)

33

 8B/10B (de)coding
Clock/Data Recovery

PHY-t/CBR-t

Sn-X/CBR-t
(GFP-T)

PHY-t

Sn-Xv

Physical signal

X VC-n signals

t = FC (Fibre Channel)
ES (ESCON)
FI (FICON)
GE (1000BASE)

PHY layer CHARACTERS (data, control)

encapsulated in fixed length GFP Frames
GFP stream mapped into VC-n-Xv payload

T1545480-02

Figure I.1/G.7041/Y.1303 FC/ES/FI/GE tributary interface port using

GFP-T mapping in SDH network element
Ethernet MAC Frames

Ethernet MAC
Frame extraction

Ethernet control
character termination

EthS/EthP

EthS

Ethernet Path

Ethernet-PHY line (de)coding

Clock/Data Recovery
t = Ethernet (10BASE)
Fast Ethernet (100BASE)
GigEthernet (1000BASE)
10GigEthernet (10GBASE)

PHY-t

Ethernet MAC Frames encapsulated

in GFP Frames
GFP stream mapped
into STS-n-Xv payload

Sn[-X]/EthP
(GFP-F)

Sn[-Xv]
PHY-t/EthS

Bridge or Switch function

VC-n path termination or

VC-n-Xv path termination

VC-n signal
or X VC-n signals

T1545490-02

Physical signal

Figure I.2/G.7041/Y.1303 Ethernet tributary interface port using

GFP-F mapping in SDH network element

34

ITU-T Rec. G.7041/Y.1303 (12/2001)

IP

IP Router fabric

IP frame encapsulated in PPP frame

PPP frame encapsulated in HDLC frame
PPP/HDLC frame encapsulated in GFP frame
GFP stream mapped into VC-n[-Xv/Xc] payload

VC-n[-Xv/Xc]/IP
(GFP-F)
IP
Router
fabric

Interface port

VC-n[-Xv/Xc]
VC-n[-Xv/Xc] path termination

VC-n[-Xc] signal or
X VC-n signals

T1545500-02

Figure I.3/G.7041/Y.1303 VC-n/VC-n-Xv/VC-n-Xc port on IP router,

or IP router function embedded in hybrid SDH/IP equipment

Ethernet
switch
fabric

Interface port

Ethernet path

Switch function

Sn[-X]/EthP
(GFP-F)

Ethernet MAC frame encapsulated in GFP frame

GFP stream mapped into VC-n[-Xv] payload

Sn[-Xv]

VC-n signal or
X VC-n signals

VC-n path termination

or VC-n-Xv path termination

T1545510-02

Figure I.4/G.7041/Y.1303 VC-n-Xv port on Ethernet switch, or Ethernet

switch function embedded in hybrid SDH/Ethernet equipment

ITU-T Rec. G.7041/Y.1303 (12/2001)

35

Appendix II
Sample GFP payload types
Table II.1/G.7041/Y.1303 GFP Payload Types
Payload
Type
Identifier
(BIN)

Payload
FCS
Identifier
(BIN)

Extension
Header
Identifier
(BIN)

User
Payload
Identifier
(BIN)

TYPE Bits
<15:13>

TYPE Bit
<12>

TYPE Bits
<11:8>

TYPE
Bits <7:0>

000

xxxx

0000 0000

0x00

Reserved

000

xxxx

0000 0000

1x00

Reserved

000

0000

0000 0001

0001

Ethernet with Null

Extension Header &
no Payload FCS

000

0000

0000 0010

0002

PPP with Null

Extension Header &
no Payload FCS

000

0001

0000 0001

0101

Ethernet with Linear

Extension Header &
no Payload FCS

000

0001

0000 0010

0102

PPP with Linear

Extension Header &
no Payload FCS

18

GFP Frame

(HEX)

Payload Area

Length of
Extension
Headers
(# Octets)

000

0010

0000 0001

0201

Ethernet with Ring

Extension Header &
no Payload FCS

000

0010

0000 0010

0202

PPP with Ring

Extension Header &
no Payload FCS

18

1003

Transparent Fiber
Channel with Null
Extension Header &
no Payload FCS

1004

Transparent FICON
with Null Extension
Header &
no Payload FCS

1005

Transparent ESCON
with Null Extension
Header &
no Payload FCS

000

000

000

36

TYPE

0000

0000

0000

0000 0011

0000 0100

0000 0101

ITU-T Rec. G.7041/Y.1303 (12/2001)

Table II.1/G.7041/Y.1303 GFP Payload Types

Payload
Type
Identifier
(BIN)

Payload
FCS
Identifier
(BIN)

Extension
Header
Identifier
(BIN)

User
Payload
Identifier
(BIN)

TYPE Bits
<15:13>

TYPE Bit
<12>

TYPE Bits
<11:8>

TYPE
Bits <7:0>

TYPE

GFP Frame

(HEX)

Payload Area

Length of
Extension
Headers
(# Octets)

000

0000

0000 0110

1006

Transparent Gb
Ethernet with Null
Extension Header &
no Payload FCS

1xx

xxxx

xxxx xxxx

Reserved

x1x

xxxx

xxxx xxxx

Reserved

xx1

xxxx

xxxx xxxx

Reserved

Appendix III
GFP frame example illustrating transmission order and CRC calculation

Worked Example
Transmit:
User_data GFP_source adaptation scramble and DC_balance SDH
Receive:
SDH un_DC_balance and unscramble GFP_sink decapsulation client data
The following worked example shows the encapsulation of a 64-byte Ethernet frame with linear
header and FCS, before DC balancing and self-synchronous scrambling. The Ethernet data octets
are mapped to GFP octet according to Transmission Bit Order (bit 0 in IEEE 802.3, clause 3
corresponds to GFP octet bit 8, and bit 7 in IEEE 802.3, clause 3 corresponds to GFP octet bit 1).
Byte
1

Field
PLI[15:8]

2
3
4
5
6
7

PLI[7:0]
cHEC[15:8]
cHEC[7:0]
TYPE[15:8]
TYPE[7:0]
tHEC[15:8]

Value(hex)
00

4C
89
48
11
01
20

Comment______________________
; PLI = Length { Payload Header +
Payload Information Field +
Payload FCS }
; = 8 + 64 + 4 = 76 bytes
;
;
; [15:13] = '000'
(client data)
; [12] = '1'
(payload FCS enabled)
; [11:8] = '0001' (linear header)
ITU-T Rec. G.7041/Y.1303 (12/2001)

37

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
39
40
41
42
43
44
45
46
47
48
38

tHEC[7:0]
EHDR[15:8]
EHDR [7:0]
eHEC[15:8]
eHEC[7:0]
DATA
DATA
DATA
DATA
DATA
DATA
DATA
DATA
DATA
DATA
DATA
DATA
DATA
DATA
DATA
DATA
DATA
DATA
DATA
DATA
DATA
DATA
DATA
DATA
DATA
DATA
DATA
DATA
DATA
DATA
DATA
DATA
DATA
DATA
DATA
DATA

63
80
00
1B
98
FF
FF
FF
FF
FF
FF
06
05
04
03
02
01
00
2E
00
01
02
03
04
05
06
07
08
09
0A
0B
0C
0D
0E
0F
10
11
12
13
14
15

; [7:0] = '00000001' (Ethernet)

; CID[07:00] = 0x8000 the value is just an example
; SPARE[7:0]
; eHEC calculated over CID,SPARE
; End extension header
; 1d Ethernet DA = 0xFFFFFFFFFFFF
; 2d
; 3d
; 4d
; 5d
; 6d
; 7d Ethernet SA = 0x060504030201
; 8d
; 9d
; 10d
; 11d
; 12d
; 13d Ethernet TYPE/LENGTH
; 14d
; 15d Ethernet payload
; 16d
; 17d
; 18d
; 19d
; 20d
; 21d
; 22d
; 23d
; 24d
; 25d
; 26d
; 27d
; 28d
; 29d
; 30d
; 31d
; 32d
; 33d
; 34d
; 35d
; 36d

ITU-T Rec. G.7041/Y.1303 (12/2001)

49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80

DATA
DATA
DATA
DATA
DATA
DATA
DATA
DATA
DATA
DATA
DATA
DATA
DATA
DATA
DATA
DATA
DATA
DATA
DATA
DATA
DATA
DATA
DATA
DATA
DATA
DATA
DATA
DATA
FCS[31:24]
FCS[23:16]
FCS[15:8]
FCS[7:0]

16
17
18
19
1A
1B
1C
1D
1E
1F
20
21
22
23
24
25
26
27
28
29
2A
2B
2C
2D
DE
E1
90
D0
56
CF
2B
B0

; 37d
; 38d
; 39d
; 40d
; 41d
; 42d
; 43d
; 44d
; 45d
; 46d
; 47d
; 48d
; 49d
; 50d
; 51d
; 52d
; 53d
; 54d
; 55d
; 56d
; 57d
; 58d
; 59d
; 60d
; 61d Ethernet FCS computed over 60 bytes
; 62d
; 63d
; 64d
; First byte of optional GFP payload FCS
; Covers only payload information field, excludes
; the extension header (i.e. 64 bytes)
; Last byte of optional GFP FCS

The core header is XORed with the DC Barker code, the rest of the GFP frame is unchanged.
Byte
1
2
3
4
5
...

Field
PLI[15:8]
PLI[7:0]
cHEC[15:8]
cHEC[7:0]

Value(hex)
B6
E7
B8
A8

Comment
; 00 xor B6
; 4C xor AB
; 89 xor 31
; 48 xor E0

ITU-T Rec. G.7041/Y.1303 (12/2001)

39

The following example shows the calculation of the cHEC for PLI[15:0] = 0x004C. The polynomial
16
12
5
is G(x) = x + x + x + 1. The PLI is shifted into the CRC-16 calculator with PLI[15:8] first, then
PLI[7:0], most significant bit first for each octet.
15

0000000000000000 <- CRC-16 initial state

Input bit

0001000000100001 <- CRC-16 after input bit

0010000001000010

0100000010000100

1001000100101001

0010001001010010

0100010010100100

1000100101001000

Transmit the CRC-16 starting from x

15

gives the GFP octets cHEC[15:0] = 0x8948.

43

The GFP frame is input to the x + 1 scrambler in network bit order (most significant bit first).
Starting with the first byte of the TYPE field (the core header is not scrambled):
Bit #1 TYPE[15]
Bit #2 TYPE[14]
Bit #3 TYPE[13]
...

Appendix IV
Number of superblocks used in Transparent GFP
This appendix is intended to provide guidance for choosing the minimum number of superblocks to
be used in a Transparent GFP frame (Nmin). The minimum values of N for several known client bit
rates in their minimum-sized SDH channels is given in Table IV.1. The formula for determining
Nmin is as follows:
N min = [(CSBWmax )(GFPOH ) / (512)(ChRmin ) (536)(CSBWmax )]
where:
CSBWmax = the worst case client signal bandwidth (i.e. fastest clock tolerance).
GFPOH = the number of overhead bits in the GFP frame.
CMFBW
is the bandwidth available for sending client management frames.
ChRmin

is the worst case channel rate (i.e. slowest channel clock tolerance).

Client data frame length = (GFP overhead) + (N)(536 bits)

40

ITU-T Rec. G.7041/Y.1303 (12/2001)

Table IV.1/G.7041/Y.1303 SDH Path capacity and number of superblocks

per Transparent GFP frame
Client un-encoded data
rate

Example client signal

VC Path Size

Min. number of 65B

blocks/GFP frame

160 Mbit/s

ESCON

VC-3-4v

425 Mbit/s

Fibre Channel

VC-4-3v

13

850 Mbit/s

Fibre Channel / FICON

VC-4-6v

13

1000 Mbit/s

Gbit Ethernet

VC-4-7v

95

1700 Mbit/s

Fibre Channel

VC-4-12v

13

NOTE The minimum number of superblocks shown here assumes a Null Extension Header and no
optional payload FCS.

ITU-T Rec. G.7041/Y.1303 (12/2001)

41

ITU-T Y-SERIES RECOMMENDATIONS

GLOBAL INFORMATION INFRASTRUCTURE AND INTERNET PROTOCOL ASPECTS
GLOBAL INFORMATION INFRASTRUCTURE
General
Services, applications and middleware
Network aspects
Interfaces and protocols
Numbering, addressing and naming
Operation, administration and maintenance
Security
Performances
INTERNET PROTOCOL ASPECTS
General
Services and applications
Architecture, access, network capabilities and resource management
Transport
Interworking
Quality of service and network performance
Signalling
Operation, administration and maintenance
Charging
For further details, please refer to the list of ITU-T Recommendations.

Y.100Y.199
Y.200Y.299
Y.300Y.399
Y.400Y.499
Y.500Y.599
Y.600Y.699
Y.700Y.799
Y.800Y.899
Y.1000Y.1099
Y.1100Y.1199
Y.1200Y.1299
Y.1300Y.1399
Y.1400Y.1499
Y.1500Y.1599
Y.1600Y.1699
Y.1700Y.1799
Y.1800Y.1899

SERIES OF ITU-T RECOMMENDATIONS

Series A

Organization of the work of ITU-T

Series B

Means of expression: definitions, symbols, classification

Series C

General telecommunication statistics

Series D

General tariff principles

Series E

Overall network operation, telephone service, service operation and human factors

Series F

Non-telephone telecommunication services

Series G

Transmission systems and media, digital systems and networks

Series H

Audiovisual and multimedia systems

Series I

Integrated services digital network

Series J

Cable networks and transmission of television, sound programme and other multimedia signals

Series K

Protection against interference

Series L

Construction, installation and protection of cables and other elements of outside plant

Series M

TMN and network maintenance: international transmission systems, telephone circuits,

telegraphy, facsimile and leased circuits

Series N

Maintenance: international sound programme and television transmission circuits

Series O

Specifications of measuring equipment

Series P

Telephone transmission quality, telephone installations, local line networks

Series Q

Switching and signalling

Series R

Telegraph transmission

Series S

Telegraph services terminal equipment

Series T

Terminals for telematic services

Series U

Telegraph switching

Series V

Data communication over the telephone network

Series X

Data networks and open system communications

Series Y

Global information infrastructure and Internet protocol aspects

Series Z

Languages and general software aspects for telecommunication systems

Printed in Switzerland
Geneva, 2002