Professional Documents
Culture Documents
Work?
Technical Level 2 (Engineer)
by Stuart Whitehead
Senior Product Marketing Manager
Field Test Solution
Table of Contents
GMP ......................................................................................................................................................................................... 9
Appendix A ............................................................................................................................................................................ 14
Recommended Pre-reading
For an overview and background on P-OTS (Packet-Optical Transport Systems) and how OTN (Optical Transport Network) fits
within the network, we recommend downloading the white paper OTN What is it and Why it is Important?
Background
As a result of operators expectations and needs to achieve five 9s performance as well as cost reductions, a new standard was
required and OTN is seen as the solution by many people. Although OTN was developed first for long-distance submarine
networks, it evolved and was deployed later in core networks, but its true strength is only now being capitalized on in the metro
market. The transition into the metro market was achieved with the updates finalized in the latest ITU-T (International
Telecommunication Union Telecommunication Standardization Sector) G.709 standards between 2009 and 2012.
Client Signal
OPU
ODU
OTU
OH
OH
OH
Client Signal
OPU
ODU
FEC
Digital Domain
Technology Overview
OCC
OCh OH
OMS OH
OTS OH
OTM OH
Signal OOS
Optical Domain
By reviewing the OTN frame we can get insight into how this transport technology works as well as an understanding of its benefits
and how to take advantage of them. The OTN frame is divided into five separate sections. This not only assists us in understanding
the frame but, more importantly, each section can also be associated directly with a sub-section of the network itself. An engineer
who understands the frame format can quickly identify the network section causing an alarm or concern. OTN works much like
SDH/SONET (Synchronous Digital Hierarchy/Synchronous Optical NETwork) in this regard, but as the line rate increases the frame
size remains constant.
www.anritsu.com
OTN Frame
The key sections of the OTN frame are the OTU (Optical channel Transport Unit) overhead section, ODU (Optical channel Data
Unit) overhead section, OPU (Optical channel Payload Unit) overhead section, OPU payload section and FEC (Forward Error
Correction) overhead section shown in Figure 2.
OTU Overhead
Column No#
Row No#
1
2
3
4
RES
TCM
ACT
TCM3
GCC1
7
MFAS
10
11
SM
TCM6
TCM1
12
13
GCC0
TCM5
TCM2
GCC2
14
APS/PCC
16
17
3824
3825
4080
RES
TCM4
PM
15
FTFL
OPU Payload
OTU FEC
EXP
RES
PSI
OPU Overhead
ODU Overhead
A simplified OTN network (Figure 3) can be divided into several sections with OTS as the innermost section, moving out to the
OMS section which includes the optical amplifiers. From here-on the terminology is the same as for an OTN frame shown in
Figure 2 moving further out the OCh, which is the complete OTN, followed by the OTU, ODU and OPU. It is also possible for the
higher layer to span multiple lower layers. For example, one OPU can span one or more OTU sections, which can also span
multiple networks or even operators.
We can quickly identify any major network concerns by looking at the key areas of each of the five major subsections within an
OTN frame.
Operator or
Digital ADM
Customer Edge
Optical Domain
Optical ADM
Optical ADM
Optical Amp
OTS
OTS
OMS
OCh (Channel)
OTU (Section)
OPU (Path)
OPU (Payload)
Digital Domain
Figure 3. Simplified Network
Digital ADM
Operator or
Customer Edge
OTU
As shown in Figure 4, the OTU section is composed of two main sections: the Frame Alignment section, and the Section
Monitoring (SM) section. The OTU OH (Overhead) provides the error detection correction as well as section-layer connection and
monitoring functions on the section span. The OTU OH also includes framing bytes, enabling receivers to identify frame
boundaries.
1
1
MFAS
10
SM
11
12
13
GCC0
14
RES
- BIP-8 (Bit Interleaved Parity-8) is a parity check on the full OPU overhead and
payload, which is then inserted two frames later to confirm the absence of errors.
See BIP-8 Overview in the Worth a Review section for more details.
Expanded SM Section
1
TTI
BIP-8
1
SAPI
BEI/BIAE
6
IAE
- TTI (Trail Trace Identifier) which is made up of several subsections. This single byte is
aligned to the OTU frame alignment MFAS; the TTI message is 64-byte long.
o SAPI (Source Access Point Identifier) and DAPI (Destination Access Point Identifier)
are 16-byte in size and both hold location information three characters for
international and twelve characters for national, supporting identification of the
end location points.
o There is also a 32-byte section allocated for operator specific information.
BDI
RES
DAPI
Operator
Specific
The third byte is divided into 8-bit sections that are used to carry network error information relative to the SM section. Greater
details about the cause of errors are explained in the third white paper OTN The Deep Dive into Details that Make it Tick.
The GCC0 (General Communication Channel 0) is simply a clear channel for general communications on the OTU layer.
The RES (REServed) section is allocated for future international standardization and all bits should be set to 0.
www.anritsu.com
ODU
As shown in Figure 6, the ODU section is an internal element allowing mapping or switching between different rates, which is
important in allowing operators the ability to understand how the end user pipe is transferred through to the higher network rates.
The ODU OH contains path overhead bytes allowing
the ability to monitor the performance, fault type and
location, generic communication, and six levels of
channel protection based on TCM (Tandem
Connection Monitoring) as described below.
Also with the latest updates to the ITU-T G.709
standard, user traffic can enter a network element and
may be required to be multimuxed up.
Table 1 shows the most common rates of some
interfaces often used for these rates as well as the
areas that only became part of the G.709 standard in
the newer releases from 2009 to 2012. The newer
updates allow the OTN standard to offer a much more
flexible system to carry different traffic types and
multiplex between rates. For example, customer traffic
entering the network in a 1 GigE pipe would require
muxing up to ODU1 via ODU0 before being able to be
carried over the OTN network, because the lowest
available OTN rate is twice the 1 GigE pipe.
1
2
2
RES
3
4
PM &
TCM
TCM
ACT
TCM3
GCC1
TCM6
TCM2
GCC2
11
10
12
TCM5
13
14
TCM4
TCM1
FTFL
EXP
PM
RES
APS/PCC
New
ODUflex
ODU0
ODU0 x n
1.244 160
1 GigE, MPLS
ODU1
2.498 775
STM-16/OC-48, FC-100/FC-200
ODU2
10.037 274
STM-64/OC-192, FC-400/FC-800
ODU3
40.319 219
10 GigE, STM-256/OC-768
ODU4
104.794 446
100 GigE
The PM (Path Monitoring) section shown in Figure 7 works in much the same way as the SM section explained above with the key
differences explained below.
- DMp (Path Delay Measurement) is used to indicate the RTT (Round Trip Time) delay
for the path of the ODU section. See the Delay Measurement below in the Review
of New Improvements section for more details.
TTI
BIP-8
1
SAPI
DAPI
Operator
Specific
BEI
STAT
PM &
TCM
1
DMp
Only the seventh bit is used for PM in the PM & TCM section.
Expanded PM Section
BDI
- The TTI subsection of SAPI, DAPI and Operator Specific are the same as described
above in the SM section except based on the ODU layer not the OTU layer.
- The BIP-8 completes the same OPU checking as described in the SM section above.
- The third byte is divided into 8-bit sections that are used to carry network error
information relative to the SM section. More details on the cause of these errors
are explained in the third white paper OTN The Deep Dive into Details that Make
it Tick.
The TCM ACT (TCM ACTivation Deactivation) is listed by the standard for future study.
The TCM section consists of six sections TCM1 to TCM6 each with the same details in them and able to be managed
independently. Details per layer include the number, type and direction of any errors, maintenance signal or not and a Delay
Measurement, see the third white paper OTN The Deep Dive into Details that Make it Tick for a detailed review of this section
of the frame.
The FTFL (Fault Type and Fault Location) byte is used to indicate the presence of a forward or backward fault and degraded signal
condition. More details can be found in the third white paper OTN The Deep Dive into Details that Make it Tick.
The EXP (EXPerimental) sections are reserved for equipment vendors to use as required to carry additional information. There is no
requirement for the network equipment to forward the status of these bytes to the next network segment.
The GCC1 and GCC2 sections are the same as the SM GCC0 section except at the ODU layer.
The APS/PCC (Automatic Protection Switching/Protection Communication Channel) section supports network protection based on
the TCM layer. More details can be found in the third white paper OTN The Deep Dive into Details that Make it Tick.
The RES section is allocated for future international standardization and all bits should be set to 0.
OPU
The OPU consist of two main sections the overhead and payload as shown in
Figure 8. The overhead is used to identify and control the type and justification of the
payload.
A constant within the overhead is the PSI (Payload Structure Identifier) byte containing
the PT (Payload Type) section shown in Figure 9; the other areas of the overhead vary
depending on the PT. These other areas of the OPU OH contain the frame structure
and frequency justification information (for adapting the client signal to the payload
area). Details of the PT bytes can be found in Appendix A.
15
16
17
3824
1
2
OPU
OPU Payload
Overhead
3
4
PSI
The payload section simply contains the payload carried by the OTN frame, which
could be SDH/SONET, Fibre Channel, Ethernet, or many other types.
PT
1
CSF
RES
...
The PSI section also includes the CSF (Client Signal Fail) status of the CBR (Constant Bit
Rate) for Ethernet client signals entering the network that is then mapped to the OPUk
(Optical channel Payload Unit-k), which is set to either 1 to indicate a fail status, or to 0
for all other conditions.
255
FEC
FEC just corrects errors at the receiver end by placing information in the frame at the
sender end. The FEC section of the frame shown in Figure 10 is often considered the
most important improvement over SDH/SONET because it supports correction of bit
errors occurring due to impairments in the transmission medium.
The scope of FEC offered by OTN networks is much greater than that of SDH/SONET.
Exactly how FEC is completed can be quite complex, depending on the FEC method
used.
3825
4080
1
2
OTU FEC
3
4
www.anritsu.com
Mapping
The key advantage of OTN over many other transport technologies available today is the ability to carry almost any type of
payload or data efficiently. OTN was designed to carry many payloads but this was enhanced dramatically during a major update
to the standard from 2009 to 2012. These enhancements included adding 100 GigE (OTU4/ODU4/OPU4), GMP (Generic Mapping
Procedure), New client signals, ODUflex for CBR and GFP (Generic Framing Procedure), 1.25 Gbps tributary slots and multistage
multiplexing. With many of the above areas allowing better utilization of the earlier released ODU0, many operators are now
considering how to make better use of ODU0 and its benefits across their networks.
Figure 11 shows the full ODU mappings available today allowing the ability of any client signal to be efficiently multiplexed up to
the required OTU rate.
OTU Mapping
OTU4
x1
ODU/OPU
4
PRBS
x1
x1
ODTU4.31
(PT=21)
OTU3
ODTU4.8
(PT=21)
ODTU4.ts
(PT=21)
100 GigE
x1
ODU/OPU
3
ODTU3.ts
(PT=21)
x10
ODTU3.9
(PT=21)
x1
OTU3e1
ODU/OPU
3e1
ODU/OPU
2e/1e
x3
ODU/OPU
2f/1f
ODTU3e2.8
STM-256
OC-768
x8/ts
x1
ODU/OPU
2e
PRBS/
10 GigE
x1
Transcoding
x1
ODTU13
(PT=20/21)
PRBS/STM64/OC-192
x1
STM-64/
OC-192
x1
ODU/OPU
1
PRBS/STM-16/
OC-48
x1
x1
PRBS/Eth
MAC
GFP-F
STM-16/
OC-48
x1
ODTU3.1
(PT=21)
x32
ODTU2.1
(PT=21)
x8
ODTU01
(PT=20)
FC-200
x2
ODTU4.1
(PT=21)
x80
ODU/OPU0
x1
x1
x1
Note:
- Client signals requiring large bit-rate tolerances are mapped into an OPU payload via PT=21.
- Client signals requiring narrow bit-rate tolerance are mapped into an OPU payload via PT=20.
- Client signals of a rate less than 2.5 Gbps are mapped into an OPU payload via PT=21.
PRBS/Eth
MAC
GFP-F
x1
x4
x16
OTU1
Internal Mapping x1
PRBS/
10 GigE
FC-1200
x1
ODTU12
(PT=20/21)
Bit-synchronous
Mapping Procedure
Generic Mapping
Procedure
Asynchronous
Mapping Procedure
PRBS/Eth
MAC
GFP-F
ODU/OPU
2
x40
40 GigE
PRBS/FC-400/
FC-800
x4
x4
OTU2
ODTU4.2
(PT=21)
Transcoding
ODU/OPU
Flex
ODTU2.ts
(PT=21)
x4
x1
PRBS/Eth
MAC
GFP-F
x32/ts
x80/ts
ODU/OPU
3e2
PRBS/STM256/OC-768
x1
x1
x10
OTU3e2
OTU2f/1f
GFP-F
x2
ODTU23
(PT=20/21)
OTU2e/1e
PRBS/Eth
MAC
PRBS
GFP-F
PRBS/Eth
MAC
GFP-T
Transcoding
GigE
x1
STM-4/STM-1/
OC-12/OC-3/FC-100
GMP
AMP (Asynchronous Mapping Procedure) can efficiently support the original standardized rates which all are multiples of the base
OPU1 payload. This supported both direct mapping to any OPU as well as ensured no more than a single step to the highest order
(OPU3 at the time). A single mapping step simplifies the operator network management tasks. With implementation of the newer
rates (including 100 Gbps), finding a way to carry the different payloads efficiently became a concern that GMP was developed to
address.
GMP offers the ability to support different line rates across all layers efficiently as well
as a more flexible client ppm offset range. To support different line rates, GMP only
populates the part of the OPU it is being mapped to but can also be mapped across
multiple frames.
Supporting a changing client speed requires the transmission side (mapper) to monitor
the incoming client stream clock constantly and compare it to the server clock: the
resultant calculation Cn is then transmitted in the OH. At the receiver side
(De-mapper), Cn is extracted from the OH, allowing recovery of the client signal
clock speed. The ability of Cn to adjust quickly to the changing client signal speed is
very important; it must be able to adapt not only to the client signal presence or
absence (replaced by AIS (Alarm Indication Signal)) but also to variations in the client or
OPU speed itself. See Figure 12 for details.
GMP Timing
Mapper
OH
Payload
area
OH
Payload
area
Payload
area
OH
Determine C n
Insert C n into OH
Insert C n client data
De-mapper
OH
Payload
area
OH
Payload
area
Payload
area
OH
Extract C n from OH
Extract C n client data
This total flexibility makes GMP a very important change for both the higher data rates like 100 Gbps and also for mapping the new
lower bitrates to the lower-order OPUs.
Of course, the GMP mappings simply compliment the earlier released AMP and BMP (Bit-synchronous Mapping Procedure), which
should still be used as described in the original G.709 standard release.
www.anritsu.com
ODUflex Rates
FC-400
(4GFC)
To allow for mapping in non-rate sensitive payloads such as Ethernet, the ODUflex GFP is utilized allowing any multiple of 1.25 Gbps
(lowest possible rate) directly mapped into the relative OPU2/3/4 rate. ODUflex rates are only implemented when there isnt a
currently defined mapping structure.
ODUflex allows for mapping of both time-sensitive payloads as well as rate-sensitive 20 ppm and non-rate-sensitive 100 ppm
payload structures, but all ODUflex rates are implemented at 100 ppm regardless of the payload.
Multistage Multiplexing
In the earlier G.709 standard releases, to simplify management and control of an OTN, it recommended only single stage
multiplexing was allowed. With the implementation of ODU/OPU0 and ODUflex rate this was no longer possible. A major benefit
of multistage mapping as shown in Figure 14, is of course being able to have lower rates such as ODU0 or ODUflex traverse the
network and exit the network at any level such as OTU2, OTU3 or OTU4. This ability allows extremely flexible mapping options for
the operator. The ability of being able to map to the lower rates and switching the traffic within the network at different OPU levels
allows the operator to offer end-to-end OTN solutions to their end users. Thus, although multistage mapping could be seen as
adding complexity for the operator to manage the network, it can also remove the requirement for multiple management systems.
Without the requirement for management systems having to span multiple technologies, management of the network is vastly
simplified for the operator.
OTU2
Non Multiplexed
ODU/OPU
2
ODTU12
(PT=20/21)
x4
x1
PRBS/STM-64/
OC-192
Single Stage
Multiplexed
x1
ODU/OPU
1
PRBS/STM-16/
OC-48
ODTU01
(PT=20)
Multistage
Multiplexed
x2
ODU/OPU0
x1
STM-1/STM-4/
OC-12/OC-3/FC-100
Delay Measurement
With GMP now supporting time-sensitive technologies, such as Fibre Channel, the importance of understanding the delay across a
link has increased. G.709 standard now supports monitoring of the delay on multiple levels at the PM section as well as the TCMi
(i = layers 1 to 6) section. This allows great flexibility not only in monitoring the delay but also in allowing the ability to understand
delay in a diverse route, ensuring the diverse route will not cause issues for time-sensitive traffic if an APS occurs.
The delay measurement is implemented by placing a contiguous string of 0 s or 1 s in the DMp (PM section) or DMt (TCM
section); the changing state of the string indicates the measurement start point which is calculated by measuring the number of
frame periods upon receiving the change of state back at the origin. Network equipment designed based on G.709 standard
released after 2008 should loopback the DMp and DMt within approximately 100 s; equipment designed before 2008 can ignore
this bit, because it was designated as reserved for future international standardization and should be set to 0.
10
Worth a Review
TCM Update
As mentioned above, the G.709 standard allows for six TCM layers from 1 to 6 and the OIF (Optical Internetworking Forum) offers
the Guidelines for Application of OTN TCM network. TCM was originally designed and implemented for SDH/SONET and used
by one operator to monitor the quality of another operator which his traffic is routed via, it completes this quality check utilizing the
N bytes in the frame OH. TCM was often overlooked in SDH/SONET and was only capable of a single layer, but with OTN TCM, a
6-layer method offers a much more valuable tool if implemented correctly by the engineer. A more detailed description of TCM is
available in the white Paper OTN The Deep Dive into Details that Make it Tick.
BIP-8 Overview
BIP-8 is a simple error-checking method that divides the incoming stream into 8-bit segments and is
used today in SDH/SONET networks. The transmitting network element adds the BIP-8 parity check,
which is then confirmed to be correct at the receiving end. The incoming bit stream is arranged in
the number of rows corresponding to the size of the OPU overhead and OPU itself divided by 8-bit
(1 byte) per row. With OTN the total OPU including overhead is used, so bytes 15 to 3824 = 3810 4
rows = 15240-byte (or rows used to calculate the BIP-8 value). The parity check is completed down
each of the 8 columns and if the additive of the 1 s and 0 s in the column is even, a 0 is added to the
BIP-8, or if the additive is odd, a 1 is added to the BIP-8. For example, for the following bit stream
from the transmitter (first bit) 1011 0010 0100 1101 1011 0011 1110 0101 (last bit), the resultant
calculation matrix is shown in Figure 15, at the receiver it will confirm the BIP-8 is equal to the relative
stream, if equal, no BIP-8 error will be declared.
8-bit
1011 0010
0100 1101
1011 0011
1110
0101
Maintenance Signals
The OTN protocol has many different alarm and error indications, and several of these are referred to as maintenance signals.
OTN operates in much the same way as SDH/SONET; when receiving an error it notifies the sending node. The feedback
mechanism is discussed in more detail in the white paper OTN The Deep Dive into Details that Make it Tick.
www.anritsu.com
11
Benefits of OTN
With the new (2009 to 2012) enhancements to the G.709 standard, it has clearly moved from a core (originally submarine) market
into all areas of the network. The ability of the operator to extend OTN connections all the way to customers via ODUflex with
better utilization of ODU0, allows for a single end-to-end management system with detailed fault analysis. This detailed fault
analysis allows the operator via TCM to troubleshoot faults deep into a OTN including a GMP 100 Gbps network looking to isolate
an end-user traffic concern or confirm whether the issue is within the network or the network of a third party.
The ability to support a diverse range of new rates, support and manage legacy and newer technologies (via GMP) and end-to-end
customer connections/monitoring has clearly made OTN the only true total end-to-end network solution for a modern operator.
Further Reading
White Papers in this series:
OTN The Deep Dive into Details that Make it Tick (Coming soon)
Technical Level 3 (OTN Engineer)
ll you need to know about OAM, TCM, FTFL and FEC written for OTN engineers, but readable by anyone with a keen
A
interest in OTN technology.
12
References
ITU-T G.709 (Interfaces for the Optical Transport Network)
http://www.itu.int/rec/T-REC-G.709
ITU-T G.872 (Architecture of Optical Transport Networks)
http://www.itu.int/rec/T-REC-G.872
Standards mentioned above are publicly available and are definitive reference but, more importantly, they are very readable.
List of Acronyms
Acronym
Definition
Acronym
Definition
ACT
Activation
OCh
Optical Channel
AIS
ODU
AMP
OH
Overhead
APS
OMS
ATM
OOS
BDI
OPU
BEI
OPUk
BIAE
OTM
BIP-8
OTN
BMP
OTS
CBR
OTU
Cn
PCC
CSF
PM
Path Monitoring
DAPI
P-OTS
dB
decibels
PRBS
DMp
PSI
DMt
PT
Payload Type
EXP
EXPerimental
RES
REServed
FAS
RS
Reed-Solomon
FC
Fibre Channel
RTT
FEC
SAPI
FTFL
SDH
GCCx
SM
Section Monitoring
GFEC
SONET
GFP
STAT
Status
GMP
STM
IAE
TCM
MFAS
Multi FAS
TCM ACT
OC
Optical Carrier
TTI
OCC
www.anritsu.com
13
Appendix A
PT Details
Hex
Binary
02
0000 0010
PT Description
Asynchronous CBR mapping
03
0000 0011
04
0000 0100
ATM mapping
05
0000 0101
GFP mapping
06
0000 0110
07
0000 0111
14
08
0000 1000
09
0000 1001
0A
0000 1010
0B
0000 1011
0C
0000 1100
0D
0000 1101
0E
0000 1110
0F
0000 1111
10
0001 0000
11
0001 0001
12
0001 0010
13
0001 0011
14
0001 0100
15
0001 0101
16
0001 0110
17
0001 0111
18
0001 1000
19
0001 1001
1A
0001 1010
1B
0001 1011
1C
0001 1100
20
0010 0000
21
0010 0001
FD
1111 1101
FE
1111 1110
Note:
www.anritsu.com
15
United States
Italy
Singapore
Anritsu Company
Canada
Anritsu S.r.l.
Sweden
Anritsu AB
Finland
Anritsu AB
Brazil
Denmark
Mexico
United Kingdom
France
Anritsu S.A.
12 avenue du Qubec, Btiment Iris 1- Silic 612,
91140 VILLEBON SUR YVETTE, France
Phone: +33-1-60-92-15-50
Fax: +33-1-64-46-10-65
Anritsu A/S
Russia
Japan
Anritsu Corporation
Korea
Germany
India
Anritsu GmbH
Australia
Taiwan
1404
Please Contact:
Printed in Japan
07/JUL/2014 ddcm/CDT