WDM Basics

Volume 2
 Contents

1 Optical-Related
1.1 Colored Light & Grey Light /1
1.2 Fiber
1.2.1 Optical Fiber Connectors /2
1.2.2 MM Fiber & SM Fiber /3
1.2.3 Single-fiber Unidirectional & Single-fiber Bidirectional /4
1.3 Optical Module
1.3.1 eSFP SEP+XFP CFP /5
1.4 Optical Module Specifications
1.4.1 Mean Launched Optical Power /6
1.4.2 Extinction Ratio /6
1.4.3 SMSR /7
1.4.4 Transmit Eye Pattern /7
1.4.5 Receiver Sensitivity /8
1.4.6 Overload Point /9
1.5 dB,dBm,mW / 10
1.6 Transmission Performance
1.6.1 OSNR, BER, and Q Value / 11
1.6.2 Nonlinear Effect / 12
1.7 Ultra-High Speed Transmission
1.7.1 Superchannel / 13
1.7.2 Flexible grid & Fixed grid / 14
1.8 Chromatic Dispersion and Polarization Mode Dispersion / 15
2 OTN Basics
2.1 Optical and Electrical Subracks / 16
2.2 WDM Site Types / 17
2.3 Service Grooming
2.3.1 Electrical-Layer Grooming & Optical-Layer Grooming / 18
2.4 Transponder & Muxponder / 19
2.5 OTN Tributary Boards and Line Boards / 20
2.6 Service Encapsulation and Mapping
2.6.1 AMP, BMP, and GMP / 21
 Contents

2.6.2 ODUflex / 23
2.7 Protection
2.7.1 1+1 Protection and 1:1 Protection / 24
2.7.2 Unidirectional Switching and Bidirectional Switching / 25
2.8 DCN Applications / 26
3 MS-OTN
3.1 VPWS / 27
3.2 VPLS / 28
3.3 EoO Principles / 29
 WDM Basics
Colored Light & Grey Light

About Light

Sunlight

Invisible light Visible light Invisible light

Red
Yellow/Orange/
Purple/Blue/
Cyan/Green/

Intermediate
Near infrared

Microwave
infrared

Industrial
Extreme
infrared

beam
X-ray
γ-ray

UV

WDM

The light in WDM systems is in the near-infrared region and is invisible.

The wavelengths range from 1260 nm to 1611 nm.
Avoid direct eye exposure to the optical ports, preventing the laser fromhurting your eyes.

Colored Light & Grey Light

The light in WDM systems is carried over different wavelengths compliant with specific
standards. To distinguish wavelengths in different systems, the wavelengths in WDM
systems are called colored light whereas the wavelengths in common optical systems
are called grey light.

Grey light is within a certain wavelength range and does not have a standard wavelength,
for example, light at client-side optical ports of WDM devices.

Colored light is WDM-side optical signals of the OTN or line boards in a WDM system.
The signals can be directly transmitted to multiplexer devices and have standard
wavelengths. Colored light is divided into CWDM and DWDM light, depending on
wavelength division standards.

Standards compliance
Colored light: ITU-T G.694.1 (DWDM) and ITU-T G694.2 (CWDM)
Grey light: ITU-T G.957, ITU-T G.959.1, and IEEE 802.3

Application Scenario

Typical scenario
Client-side optical modules provide grey optical ports while WDM-side optical modules
provide colored optical ports.

λ1
OTU OTU
λ1-λN
λ2 MUX/ MUX/
OTU DMUX
OA OA DMUX OTU
Client-side device Client-side device
λN
OTU OTU

Grey Colored Colored Grey

optical port optical port optical port optical port

Client-side colored light

The client-side optical module of the tributary board
provides colored optical ports, which can be used to λ1 Line board
connect to a multiplexer board.
λ2 Line board
The major application scenarios are as follows:
...

1. The client-side device interconnected with the 10GE

WDM device has colored optical ports. Packet
2. The 10GE ports of the packet board are colored GE tributary
optical ports, which can be directly connected to GE board

the multiplexer board. In this situation, services GE

do not need to be processed on the universal line

board. See the figure on the right.

WDM-side grey light

Huawei WDM device
A Huawei WDM device is interconnected
OTU Line board Line board OTU
with a third-party WDM device. The OTU Line board Line board OTU
third-party WDM device, however, does OTU Line board Line board OTU
not support the tributary/line separation OTU GELine board Line board FC OTU
mode. If service grooming is required, the
Tributary board

Tributary board

Tributary board

Third-party Third-party
WDM device WDM device
line boards (grey light on the WDM side) of
Huawei WDM device can be used to receive GE GE FC
services, implementing service adding/dropping
or grooming on Huawei WDM device.

How to Distinguish Colored and Grey Optical Modules?

The transmit and receive wavelengths of colored optical modules have a nominal center
frequency and center wavelength.
The transmit and receive wavelengths of grey optical modules have a wide range and do
not have a center wavelength.

Therefore,
They can be distinguished by whether their specifications contain center wavelengths.

Wavelength stability of colored optical modules:

±6.5 nm/CWDM, 0.04 nm/DWDM 100G;
Wavelength stability of grey optical modules:
about 30 nm

-1-
 WDM Basics
Optical Fiber Connectors

Optical fiber connectors are required when fibers need to be connected to optical
modules.
The model of an optical fiber connector is determined by structure type and endface
type. Using the FC/PC connector as an example, FC indicates the outer structure
type of the optical fiber connector and PC indicates the endface type of the pin inside
the optical fiber connector.

Types of Fiber Connectors

Optical fiber connectors are classified into FC, SC, ST, and LC connectors by structure.

Type Full Spelling Description

It is a commonly-used round threaded fiber connector. The

outer reinforcement part of the connector is a ferrule and the
FC Ferrule Connector connector is fixed using a screw. Such a connector has higher
plug-resistance than plastic connectors.

It is a standard square fiber connector that uses a push-pull

latching mechanism and requires no rotation. Such a connector
SC Square Connector
is made of engineering plastic, and is heat-resisting and
oxidation-resisting.

ST Straight Connector It is a plug-in round fiber connector that is fixed using a screw.

LC Lucent Connector It is a small-sized rectangular fiber connector.

It is an LSH connector with a dust-proof cover. The cover is

E2000 - automatically closed after the fiber is removed.

It is a multi-core square fiber connector that adopts the

Multi Transmit－
MT-RJ transceiver structure. Such a connector can house two fibers at
Receive Joint
one end.
It is a multi-core square fiber connector that adopts the
MPO Multi-fiber Push On transceiver structure. Such a connector can house two fibers at
one end.

Optical fiber connectors can also be classified into PC and APC connectors by inside pin
endface type.

Type Full Spelling Description

PC Physical Contact It is a protruding polished fiber connector.

It is a protruding fiber connector that is polished at an 8-degree

Angled Physical angle. Because of the angle grinding technology, the return loss
APC in APC connector application scenarios is obviously higher than
Contact
that in PC connector application scenarios.

Common Fiber Connectors in WDM Systems

• LC/PC connector (plug-in square fiber connector/protruding polished)

• FC/PC connector (round fiber connector/protruding polished)

• SC/PC connector (square fiber connector/protruding polished)

• LSH/APC connector (connector with a dust-proof cover/protruding polished at an

8-degree angle)

• MPO/PC connector (multi-core multi-channel pluggable connector)

-2-
 WDM Basics
Fiber Basics

Mode indicates the transmission path of optical signals that enter a fiber at a certain
angular velocity. A fiber supports as many transmission modes as its diameter allows.
Fibers are classified into single-mode (SM) and multi-mode (MM) fibers based on the
number of supported transmission modes.

MM Fiber
about 50um
A fiber that has a core diameter greatly exceeding
optical wavelengths and permits tens and even
hundreds of transmission modes is called MM fiber.

Because optical signals transmitted in different modes have different rates and
phases, after long-distance transmission, issues such as time delays, broadened
spectrums,and enlarged dispersion may happen. Therefore, MM fibers suit only
low-rate and short-distance transmission.

SM Fiber
A fiber that has a core diameter in the same order
5~10um
of magnitude as optical wavelengths and permits
only one transmission mode (basic mode) is called
SM fiber.

SM fibers suit large-capacity and long-distance transmission,

and constitute the majority of fibers in the WDM system.

How are SM and MM fibers distinguished?

• MM fibers are orange and marked as MM.

• SM fibers are yellow and marked as SM.

SM and MM Optical Modules

SM and MM optical modules must be used together with SM and MM fibers respectively.
The working bands are 850 nm for MM optical modules and 1310 nm and 1550 nm for
SM optical modules.
Do not connect SM and MM optical modules.

How are SM and MM optical modules distinguished?

MM optical modules are marked as MLM, and SM optical modules

are marked as SLM.

-3-
 WDM Basics
Fiber Basics

The WDM system supports signal transmission in two modes:

single-fiber unidirectional and single-fiber bidirectional.

Single-fiber Unidirectional

In this mode, the WDM system transmits multi-wavelength

optical signals in receive and transmit directions through
separate fibers.

Feature
This mode is characterized by its simple design and low require-
ments, easy fault isolation and performance monitoring, conve-
nient O&M, and flexible networking that facilitating upgrading
and capacity expansion.

Single-fiber Bidirectional

In this mode, multi-wavelength optical signals are

transmitted through only one fiber in both receive and
transmit directions.

This mode is mainly used on the client side, imple-

mented through the filtering function of a single-fiber
bidirectional optical module. Different center wave-
lengths are used for the two directions.

Feature
This mode is characterized by a 50% savings in fiber resources compared with the
single-fiber unidirectional mode.
However, its design and O&M of the WDM system are more complicated. Protection
ranges must be configured for separating wavelengths to prevent signal interference in
two directions, which wastes spectrums.

-4-
 WDM Basics
Optical Module Types
The following are the major optical module types used by WDM equipment,
which are classified by module encapsulation types.

eSFP
The Enhanced Small Form Factor Pluggable (eSFP)
module is a type of low-speed optical module
that provides the supervisory function.
The rate of such modules is generally
less than 5 Gbit/s. The supervisory
function refers to the supervisory of
transmit and receive optical power. eSFP
Most SFP optical modules do not 125M-5G
support the supervisory function.

SFP+/XFP
In most cases, 10 G Small Form Factor
Pluggable (XFP) modules are preferentially
used to receive 10G services. Nevertheless,
the size of XFP modules is large. To resolve
XFP光模块 the density issue, Enhanced 8.5G and 10G
Small Form Factor Pluggable (SFP+) modules
are developed, which can be used to transmit
SFP+ XFP
8G-10G 8G-10G 8G fiber channel (FC) services and
10G Ethernet services.

CFP
The Centum Form Factor Pluggable (CFP) module is a type of 40G/100G pluggable
optical module that features high-rate and multi-wavelength transmission, and large
size. These modules are generally used to interconnect WDM equipment with Ethernet
switches. The commonly used interface types are 100GBASE-LR4/ER4/SR10.
CFP modules are currently driven to feature small size, low power consumption, and
high port density, for example, CFP2 and CFP4 modules that are being developed in
the industry.

CFP CFP2 CFP4

40G/100G 100G 100G

CXP
The 120 Gb/s eXtended-capability Form Factor Pluggable
(CXP) module is a type of 12-channel pluggable optical module.
Such modules use MPO interfaces and parallel cables (each
containing multiple fibers), and feature short-distance
CXP transmission. The modules are mainly used for interconnecting
WDM equipment or interconnecting WDM equipment with
supercomputers. The commonly used interface type is
100GBASE-SR10.

SR10: S indicates short wavelength, which is generally 850 nm; R indicates the Ethernet 64B/66B codec
format; 10 indicates 10 optical channels.
LR4/10: L indicates long wavelength, which is generally 1310 nm; R indicates the Ethernet 64B/66B codec
format; 4 or 10 indicates 4 or 10 optical channels.
ER4/10: E indicates extra-long wavelength, which is generally 1550 nm; R indicates the Ethernet 64B/66B
codec format; 4 or 10 indicates 4 or 10 optical channels.

-5-
 WDM Basics
Optical Module Specifications

Mean Launched Optical Power

Average optical power at a point tested
during transmission of pseudo-random
sequence signal.
Test point
Unit: dBm
TX Optical power meter
Transmitter
RX

Extinction Ratio
Eye pattern Ratio of the average optical power level at the
centre of the logical "1" to the average optical
power level at the centre of the logical "0" under
A the worst reflection conditions.

B
Extinction ratio (dB) = 10lg A
B

A: average optical power level at the centre of the logical "1"

B: average optical power level at the centre of the logical "0"

Unit: dB

This means
The larger the extinction ratio, the better the logical discrimination at the receive end.
The smaller the extinction ratio, the greater the possibility of signal interference and
increased BER.

-6-
 WDM Basics
Optical Module Specifications

Minimum Side Mode Suppression Ratio (SMSR)

Ratio of the average optical power of the main mode
to the optical power of the most prominent side mode
under the worst reflection condition.

Unit: dBm
What is the side mode?
Under ideal conditions, all signals transmitted by
optical modules are optical signals of a specified
wavelength.
However, optical signals in practice are not only carried
on this wavelength. Some signals are carried on other
wavelengths. Such signals are called the side mode.

Side mode interferes with the main mode

The larger the SMSR, the better the laser performance.

Transmit Eye Pattern

Performance evaluation of a baseband transmission system is generally
done with a lab oscilloscope to observe the waveform of received signals.
The impact that inter-bit crosstalk and noise have on system performance
is then studied in a process called eye pattern analysis.

Degree SDH/SONET Eye Pattern

An eye pattern is a visual representation
1+Y1 of the impact of inter-bit crosstalk and
Mean level of
1 centre of
noise.
1-Y1 logic “1” The size of the eye aperture corresponds
directly to the degree of inter-bit crosstalk.
0.5

Y1 Mean level of
centre of
0 logic “0”
-Y1 Time（UI）
0 X1 X2 1-X2 1-X1

Rates X1 X2 Y1
This means
STM-1 0.15 0.35 0.20 A large eye aperture and regular eye
STM-4 0.25 0.40 0.20 pattern indicate less inter-bit crosstalk,
and vice versa.

-7-
 WDM Basics
Optical Module Specifications

Receiver Sensitivity
The minimum input optical power a receiver requires
to achieve the expected BER performance.
The larger the receiver sensitivity, the poorer the receiver
performance.
Unit: dBm

What is BER?
The bit error rate (BER) measures the data transmission
precision within a specified time period.

Number of errors in transmitted bits

BER = x 100%
Total number of transmitted bits

-10
The expected BER of a system is generally less than 10 .

Factors affecting receiver sensitivity:

• OSNR, the larger the OSNR, the less the noise on the receive circuit and the less
the impact on receiver sensitivity.

• Signal waveform, which is determined by the transmit-end extinction ratio and

fiber dispersion.

• Signal transmission rate.

The larger the transmission rate, the poorer the receiver sensitivity and the
shorter the regeneration distance.

-8-
 WDM Basics
Optical Module Specifications

Overload Point
The overload point is the overload optical power,
the maximum average input optical power at the receive end
that achieves the expected BER performance.
Unit: dBm

Overload, Minimum receiver overload,

Maximum input optical power

The overload point defined in the standards means

the minimum receiver overload point.
It is a BER-related concept and indicates the maximum mean receive
optical power of the receiver to achieve the expected BER performance.

All the overload points specified in the optical module specifications of

WDM devices currently mean the minimum overload points.
A certain margin will be left for devices in actual applications.
Therefore, the measured overload point will be slightly greater than
the value specified for the minimum overload point in
optical module specifications.

Maximum input optical power indicates the maximum receive

optical power of the receiver.
If the practical optical power of a device is greater than the maximum input
optical power, the optical components on the device may be damaged.

Overload point and receiver sensitivity are both critical facets of optical power.

Receiver sensitivity and overload point

• If the input optical power is less than the receiver sensitivity, it is too
weak to ensure that equipment receives signals.

• If the input optical power exceeds the overload point,

it may be impossible to achieve the expected BER performance,
and even the device will be damaged.

-9-
 WDM Basics
Relationship Between dBm, mW, and dB

dBm and mW
Both dBm (decibel-milliwatts) and mW (milliwatts) are units of optical power.
They can be converted as follows:
dBm=10xlgP (P indicates optical power, in mW.)
For example, 1 mW can be converted into 0 dBm.

dBm and dB
dBm is the unit of optical power,
and dB is the unit of an optical power gain or attenuation.
They can be converted as follows:
dB=10xlg(P1/P2)，
Here, both P1 and P2 indicate optical power, in mW.
That is, dB=10xlgP1-10xlgP2=dBm1-dBm2.
When optical power is expressed in dBm,
dB is the unit of optical power difference.

When calculating the insertion loss of an optical component,

you only need to subtract the output optical power from the input optical power.

For example

P1 Ptotal indicates the optical power of

P total multiplexed wavelengths,
P2 and P1 and P2 indicate the optical power of
a single wavelength separately.

Ptotal (mW)=P1(mW)+P2(mW)

Assume that P1 = P2 = Psingle

Ptotal (dBm)=P
single
(dBm)+10lg2(dB)

N wavelengths

Ptotal (dBm)=Psingle(dBm)+10lgN(dB)

Fixed Network Information Dept.

- 10 -
 WDM Basics
OSNR, BER, and Q Value

OSNR
Optical signal-to-noise ratio (OSNR) is used to quantify the degree of optical noise
interference on optical signals. It is the ratio of service signal power to noise power
within a valid bandwidth.

Bm
OSNR(dB)= 10lg Pi/Ni+10lgBm/Br

In the formula:
Pi: signal power on the i channel
Bm: equivalent noise bandwidth Br
Ni: noise power introduced within the Bm
Br: reference optical bandwidth, which is usually 0.1 nm

BER
Bit error rate (BER) is the ultimate indicator for measuring transmission quality. Due to
the impact of factors such as noise, non-linear effects, and dispersion (PMD/CD), the
waveforms of optical signals coupled into fibers will be distorted when the optical signals
arrive at the end of fiber links. Bit errors are therefore present when the receiver converts
the optical signals into electrical signals.

1 1 0 1 1 1 1 1

OSNR and BER

OSNR indirectly reflects BER and can provide a warning of potential BER deterioration.

In an experimental environment where factors

Maximum correctable BER such as loss, dispersion, and non-linear effects
BER

are excluded, if the OSNR is less than the

specified threshold, the pre-FEC BER will be
excessively large and uncorrectable bit errors
will be generated. The OSNR threshold in this
case is called B2B OSNR tolerance.
OSNR（dB）
B2B OSNR tolerance

Q Value

Q value:
I 1 and σ 1 are the mean value and variance output
I −I
Q= 1 0 by Gaussian pulse 1. I and σ are the mean value
σ1 + σ 0 0 0
and variance output by Gaussian pulse 0.

Mapping between Q value and BER:

1 Q 1
exp  − Q 
2
BER = [erfc( )] ≈
2 2 2π Q  2
Q (dB):
Q (dB) = 10log10 (Q2 ) = 20log10 (Q)

The Q value and BER have a one-to-one mapping. The Q value can represent the
system tolerance in dB, just like dBm substitutes for mW to represent optical power.
A smaller pre-FEC BER indicates a larger Q value and better link performance.

- 11 -
 WDM Basics
Nonlinear Effect

What Is Nonlinear Effect?

The nonlinear effects of fibers are caused by their nonlinear polarization effects. When the
incident optical power of a fiber exceeds a certain value, the optical power density in the
fiber core will become excessively high because the effective cross-sectional area of a
fiber ranges only from 50 to 80 μm², triggering the nonlinear polarization of fiber materials.
The fiber nonlinear effect is proportional to the optical power density of signals, and
significantly affects the signal phase, pulse shape, and optical power evolution.

SPM: self-phase modulation

Common nonlinear effects:

Kerr effect XPM: cross-phase modulation

Nonlinear
effect FWM: four-way mixing

Stimulated
scattering SRS: stimulated Raman scattering

SBS: stimulated Brillouin scattering

SPM:
SPM is phase modulation of light caused by an instant change in the light intensity to
gradually broaden the signal spectrum.
XPM:
XPM is a nonlinear optical effect where one wavelength of light can affect the phase of
another wavelength of light through the optical Kerr effect.

SPM

Transmission

Compared with SPM, XPM brings more serious impacts.

A larger channel quantity indicates a smaller dispersion
coefficient but a larger phase shift caused by XPM.

FWM:
FWM is an intermodulation phenomenon in nonlinear optics, whereby interactions
between two or three wavelengths produce another two new wavelengths.

Transmission

f1 f3 f2 f f1 fFWMf3 f2 f
The FWM is obvious only when the channel spacing is small and
wavelengths are close to the zero dispersion point.

SRS and SBS:

Both SRS and SBS indicate that energy exchanging occurs between lightwave and
phonon, causing the lightwave to be scattered to different wavelengths.

SRS

p p

Output

λ λ
The phonon caused by SRS is generally optical phonon with a high frequency.
SRS can make the frequency move either upwards or downwards.

SBS

Incident spectrum Reflectance spectrum

SBS is backward. To be specific, the frequency of incident
wavelengths is moved downwards.

Impact on the WDM System

In the traditional synchronous transmission network, fibers feature linearity because the
incident optical power is small. The major factors affecting transmission are fiber loss and
dispersion. In the WDM system, however, the wavelength division multiplexing technology
makes one fiber carry dozens of or even hundreds of optical channels. In addition, with the
use of fiber amplifiers, high-power optical signals on multiple wavelengths are coupled into
one fiber, causing the nonlinearity feature in the fiber. Nonlinearity becomes a key factor
that limits the transmission performance.

Fiber Types and

Nonlinear Effect
Nonlinear Effects
Nonlinearity coefficient γ∝1/Aeff The system rate, incident optical power, number of spans, fiber
Effective cross-sectional area (unit: m ² ): loss, and channel spacing are closely related to nonlinear effects.

Channel spacing
y
alt

Fiber loss
Aeff
en

Dispersion coefficient er
rp

o w
ea

Baud rate lp s
an
in

a
ic p
l

pt
on

fs
to
N

n ro
80 72 50 Nonlinear
ide be
penalty Inc um
G.652 LEAF G.653 N

A larger optical power density, smaller channel spacing, and more spans indicate
severer nonlinear effects. The relationships between dispersion and nonlinear
effects are complex. The four-wave mixing effect becomes more obvious as the
dispersion gets closer to zero.

How to Compensate?
The hazards of fiber nonlinear effects to the WDM system vary depending
on their origins.
In general, the following measures can be taken for compensation:

Decrease 1. Use fibers with a larger effective cross-sectional area to reduce the
the optical optical power per unit area.
power density. 2. Reduce the incident optical power. To increase the OSNR of signals
in a fiber, large incident optical power may be required. However, if
the WDM system uses Raman amplifiers with low noise, the OSNR
can be increased only by inputting optical signals with low optical
power, reducing the impact of fiber nonlinear effects.

1. Reserve a certain amount of dispersion in the operating band to

Manage reduce the four-wave mixing effect.
dispersion. 2. Reduce the dispersion slope of the fiber to expand the operating
wavelength range of the WDM system and increase the
wavelength spacing.
3. Decrease the polarization mode dispersion of the fiber and reduce
the dispersion on the operating band to adapt to the continuous
increase of the single-channel rate.

Use a new Amplitude modulation: Use the RZ modulation format.

modulation Phase modulation: For example, use quadrature phase shift keying
format. (QPSK), which is multi-level phase modulation. The bit rate under
this modulation format is twice the baud rate. Therefore, a phase
modulation format is applicable to transmission systems with a more
than 40G rate. QPSK modulation features narrow spectral width and
smooth spectrum. It can effectively suppress various nonlinear
effects in fibers.

The following methods can be used to suppress various nonlinear effects:

Nonlinear
Suppression Method
Effect
SPM Decrease the single-wavelength optical power.

Consider both incident optical power and dispersion

XPM
compensation, and use Raman amplifiers.

FWM Reserve a certain amount of dispersion in the

operating band to avoid zero dispersion.
Decrease the single-wavelength optical power and
SRS
total optical power for equalization.
Decrease the single-wavelength optical power, and
SBS
add a low-frequency scrambling signal.

In conclusion, nonlinear distortion limits the incident

optical power and therefore shortens the transmission
distance. It affects the system dispersion tolerance,
making the dispersion compensation scheme of the
system more complex.
The technology for suppressing fiber nonlinear effects is
crucial to the ultra-long-haul transmission WDM system.

- 12 -
 WDM Basics
Superchannel

Background

A superchannel is an optical channel with a larger bandwidth that binds multiple subcarriers on the
WDM side for unified transmission.
The superchannel technology is currently used in scenarios of 37.5 GHz wavelength spacing, under
which the cascading of multiple ROADM boards brings a large filtering penalty and therefore guard
bands need to be added for normal signal transmission.

Introduction

The superchannel technology uses the multi-subcarrier mode to convert a high-speed baseband
data signal into multiple low-speed signals, and to modulate these signals into multiple subcarriers.
These subcarriers are then combined into a larger-bandwidth optical channel on the WDM side for
unified signal transmission.

400 Gbit/s 1 Tbit/s 16QAM 1 Tbit/s QPSK

superchannel superchannel superchannel

λ1 λ2 λ1 λ2 λ3 λ4 λ5 λ1 λ2 λ3 λ4 λ5

Advantage
Using the Flexible Grid technology, the superchannel improves the spectral efficiency, increases the
system transmission capacity, and reduces the penalty in filter cascading.

Application Example

When the spacing between Flexible Grid wavelengths is 37.5 GHz, the cascading of multiple ROADM
boards introduces a large filtering penalty. Therefore, guard bands need to be added for normal
signal transmission.
Furthermore, if wavelengths are separately used for signal transmission, guard bands need to be
added for each and every wavelength.
With the superchannel technology, five wavelengths can be transmitted over a total bandwidth of 200
GHz. However, when the wavelengths are separately used, only four wavelengths can be transmitted
simultaneously.

The filtering penalty is too large, and wavelengths cannot be used.

Without guard bands

Guard bands need to be added on the two sides of each wavelength.

Only four wavelengths can be transmitted.
Separate wavelength

Guard bands need to be added on the two sides of each superchannel.

Five wavelengths can be transmitted.

…
Superchannel
…

One superchannel
Adjacent superchannel Adjacent superchannel

Guard band 37.5 GHz spacing

Related Terms
• Subcarrier: a signal with a single-wavelength spectrum width of 37.5 GHz, 50 GHz, or other values.
• Guard band: an idle spectrum added outside of an edge subcarrier to reduce the filtering penalty of
the edge subcarrier. The guard band size varies depending on the line modulation formats.

• The subcarriers of a superchannel must share the same modulation format,

source node, sink node, and path.
• Guard bands need to be added on both sides of a superchannel. If there are
adjacent supperchannels, guard bands also need to be added for the adjacent
supperchannels. Otherwise, the attenuation of the guard bands cannot be adjusted.

- 13 -
 WDM Basics
Fixed Grid and Flexible Grid

Fixed Grid
The traditional 40- or 80-wavelength DWDM system uses the Fixed Grid (fixed spectrum)
mode, which features a fixed center frequency and fixed wavelength spacing of 50 GHz
or 100 GHz.
However, in the Fixed Grid mode, bandwidths cannot be adjusted flexibly.

100 GHz 100 GHz 100 GHz 100 GHz 100 GHz 100 GHz
DWDM ring

100G

400G

400G

100G
40G …

40G
Wavelength

Spectrums wasted

With the system signal rate increase, the spectrum of optical signals is broadened. If
signals of multiple rates are transmitted on a network, the maximum spectral bandwidth
required by these signal rates is selected as the fixed frequency spacing, greatly wasting
spectrums.

Flexible Grid
As the network traffic increases sharply, the transmission rate increases to 400 Gbit/s or
even 1 Tbit/s, broadening the spectrum of optical signals.
To implement flexible spectrum allocation and improve spectrum usage, the Flexible Grid
technology is utilized.
The Flexible Grid technology can provide 37.5 GHz to 400 GHz wavelength spacing.
37.5 37.5
GHz 75 GHz 150 GHz GHz 75 GHz
DWDM ring
200G

200G

400G
400G

1T

Wavelength

It divides spectrums into slices with smaller widths, such as 6.25 GHz slices or 12.5 GHz
slices. A high-speed signal can occupy multiple spectrum slices, implementing flexible
bandwidth adjustment and improving network-wide spectrum usage.
Advantages:
1. Adjusts bandwidths flexibly.
Spectrum slices can be selected for
signals randomly so that the signals can n x 12.5 GHz slice
occupy different bandwidths, and the f = 192.11875 f = 192.18125 f = 192.25 f = 192.3125
center frequency of wavelengths can be
flexibly adjusted. 12.5 GHz

2. Improves bandwidth usage.

Transmission of multi-rate signals on a Channel 1 Channel 2 Channel 3 Channel 4

network reduces the spacing between

bands, improves bandwidth usage, and
satisfies requirements for large-capacity
bandwidths.

Example
When dual-carrier 400 Gbit/s signals are transmitted using the 16QAM code pattern,
each 200G wavelength requires a 37.5 GHz bandwidth.
If the 50 GHz Fixed Grid mode is used, a spectral width of 100 GHz (50 GHz x 2) will be
occupied.
If the Flexible Grid technology with 12.5 GHz slices is used, a spectral width of only 75
GHz (12.5 GHz x 6) will be occupied, reducing a 25 GHz bandwidth and improving
spectrum usage.
Fixed Grid Flexible Grid
50 GHz 50 GHz 2 x 50 GHz 75 GHz (16QAM) 150 GHz (QPSK)
6 x 12.5 GHz 12 x 12.5 GHz
…
40G 100G 400G
400G 400G
λ1 50GHz
ch1
λ2 50GHz VS
ch2 m*12.5GHz
λ3 50GHz

λ4 50GHz ch3 n*12.5GHz

…

λ80 50GHz chn

- 14 -
 WDM Basics
Chromatic Dispersion and Polarization Mode Dispersion

Chromatic Dispersion (CD)

A form of dispersion where optical pulses are

spread because different wavelengths are Distance L(km)

transmitted at different speeds in optical fibers

and the periods for different wavelengths to λ1 λ1
traverse the same distance are different. λ2 λ2
λ3 λ3
... ...

Polarization Mode Dispersion (PMD)

Delay difference A form of dispersion where optical pulsesare

Fast axis spread because optical signals in different
phase status are transmitted at different
Distance L(km)
speeds due to the random birefringence of
optical fibers.

Slow axis

Impact of Dispersion on the System

The spreading of optical pulses in the time domain caused by CD and PMD will lead to
distortion of signals and inter-code crosstalk, thereby causing bit errors.

Distance L(km) Distance 2L(km) ...

The dispersion is accumulated as the transmission distance is prolonged, and the impact
of dispersion on the system also increases. As a result, the transmission distance is limited.

CD (ps/nm) = Transmission distance (km) x CD coefficient (ps / nm ⋅ km)

PMD (ps) = Transmission distance (km) x PMD coefficient (ps / km )

As the transmission speed increases, the pulse width is reduced, the impact of dispersion
on the system becomes more serious. Therefore, for 100G and higher rate systems, the
coherent technology and optical signal processing technology must be used to reduce the
impact of dispersion on the system and improve the dispersion tolerance of the equipment.

- 15 -
 WDM Basics
Optical and Electrical Subracks

Electrical Subrack Optical Subrack

An electrical subrack houses only cross- An optical subrack houses only OADM boards,
connect boards, OTU boards, tributary multiplexer boards, demultiplexer boards,
boards, line boards, or protection boards. optical amplifier boards, OSC boards, optical
spectrum analyzer boards, OLP boards used
Electrical subracks include universal
for optical line protection, regeneration boards,
platform subracks that house only
or OTU boards in regeneration mode.
OTU boards in non-regeneration
mode, and optical/electrical hybrid A subrack equipped with only protection
subracks. boards is an optical subrack.

An electrical subrack converts client-

side signals into standard-wavelength An optical subrack grooms and
optical signals for grooming manages optical signals on an OTN.
implemented by an optical subrack.

Line board
ROADM board
demultiplexer board
Tributary board

ROADM board
ROADM board
Multiplexer/
Line board

Cross-connect board

Line board ROADM board

An NE consisting of only optical subracks or electrical subracks is called an optical

or electrical NE, respectively. A hybrid NE is composed of both optical and electrical
subracks.
Currently, optical and electrical NEs are separated for site deployment to achieve
centralized grooming, low power consumption, and easy management.

- 16 -
 WDM Basics
WDM Site Types

WDM sites are divided into the OTM, OLA, OADM, and REG types by function.

OTM
OTM: optical terminal multiplexer

An OTM site adds service signals to the lines of the WDM system through the
multiplexer unit. It also drops the service signals from the lines of the WDM system
through the demultiplexer unit.

OTU
MUX

OTU Even
wavelength
OA

OTU
MUX

OTU Odd
wavelength
Client side ITL OSC FIU Line side

OTU
DEMUX
MUX

OTU Even
wavelength
OA

OTU
DEMUX
MUX

OTU
Odd
wavelength

OLA
OLA: optical line amplifier

An OLA site amplifies bidirectionally transmitted optical signals to extend the

transmission distance.

DCM
OA

West line side FIU OSC FIU East line side

OA

DCM

OADM
OADM: optical add/drop multiplexer
The main function of an OADM site is to add or drop one or more wavelengths from a
multi-wavelength channel during transmission.
There are two OADM sites available: fixed OADM (FOADM) and reconfigurable OADM
(ROADM).
A FOADM site drops optical signals with a specific wavelength from multiplexed
signals. It transmits the signals to an OTU or line board, and multiplexes the optical
signals over an ITU-T-compliant WDM wavelength transmitted by the OTU or line
board into the multiplexed signals.

A ROADM site adds or drops single-wavelength or multi-wavelength signals to

achieve dynamic wavelength grooming in multiple directions. It can also perform
the reverse process.
ROADM is the mainstream OADM site type. A ROADM site can add or drop any
wavelength, and wavelength pass-through at a ROADM site does not require manual
fiber connections.
ROADM site configuration model:
Direction B Direction C

S WS
WS S

Direction D
Direction A
WSS

WS
S

Colorless A&D Colored A&D Colored A&D Colored A&D

λ* λ* λ1 λn λ1 λn λ1 λn

Configuration example of a 4-dimensional ROADM site:

Direction 2 Direction 3

OA OA
OA OA
Direction 1 AM4 DM4
Direction 4
DM4 AM4
IN OUT IN OUT
ROADM ROADM ROADM
DM1 AM1 DM4 AM4 DM1 AM1 DM4 AM4 IN OA
OA IN ROADM
FIU FIU
OA OUT OUT OA
AM1 DM1
DM1 AM1
AM1 DM1 AM2 DM2 AM3 DM3 AM4 DM4
ROADM

IN OUT

WSM9 WSD9

WSM9 WSD9
OTU

OTU

OTU

OTU

The IN/OUT ports are used to add and drop services.

Pass-through services are groomed to other directions
through the AM/DM ports.

REG
REG: regeneration station
In long-haul transmission, if one or more factors such as dispersion, optical power, noise,
and nonlinear effects limit line extension, an REG site can be configured to regenerate
electrical signals, improving signal quality.

OSC
DEMUX

Line/OTU
MUX

Even Even
wavelength wavelength
Line/OTU
XUMED
MUX

OA OA
FIU ITL ITL FIU
DEMUX

MUX

Line/OTU
OA OA
Odd Odd
wavelength wavelength
MD

Line/OTU
MUX

UE
MUX

XUXM

Example
Service signals A and B are transmitted from city A to cities E and D respectively.
1 The signals are added to a line at the OTM site.→ 2 The signals reach
city B. Since the optical power weakens, the signals need to be amplified at the OLA
site. → 3 The signals reach city C. The signal quality seriously downgrades, and
the signals need to be regenerated at the REG site. → 4 The signals reach city
D. Signal B reaches its destination and is dropped at the OADM site. → 5 Signal
A reaches city E and is dropped at the OTM site.

City A City B City C City D City E

OTM OA REG OADM OTM

- 17 -
 WDM Basics
Electrical-Layer Grooming & Optical-Layer Grooming

Background
I n a t radit iona l WD M s y s t e m , t o ch a n g e a fixe d sig n a l tr a n sm ission path,
mai n t enanc e p e r s o n n e l h a v e t o visit th e site to in sta ll r e la te d h a rdw are
a nd co nnec t o p t i c a l f i b e r s , w h i ch a r e tim e - a n d la b o r - co n su m in g . The
O TN net work u s e s e l e c t r i c a l - l a ye r g r o o m in g a n d o p tica l- la ye r g r oomi ng
t o addres s t he s e i s s u e s . W i t h th e h e lp o f th e two g r o o m in g te ch nol ogi es,
mai n t enanc e p e r s o n n e l c a n r e m o te ly a d ju st th e tr a n sm issio n p a ths of
si gnals f rom t h e c o n t r o l c e n t e r.

How do electrical-layer grooming and optical-layer grooming

work respectively?

Electrical-Layer Grooming
G room s elec t ri c a l s i g n a l s a t a gr a n u la r ity o f ODUk( k = 1 , 2 ( e ) , 3 , 4, or
f l e x)
T h e c ore unit f o r e l e c t r i c a l - l a y er g r o o m in g is th e cr o ss- co n n e ct board.

Pa c k e t Tri b u ta ry bo a rd
Line boar d OT U

STM -N Tri b u ta ry bo a rd
Line boar d OT U
FC Tri b u ta ry bo a rd
Line boar d OT U
10GE Tri b u ta ry bo a rd

Line boar d OT U
GE Tri b u ta ry bo a rd

 Se rv ic es are t r a n s m i t t e d t o tr ib u ta r y b o a r d s a n d g r o o m e d to line
b oards t hro u g h t h e c r o s s - c o n n e ct b o a r d , e n ca p su la te d a n d m apped
to O TU s ign a l s o n t h e l i n e b oa r d s, a n d th e n tr a n sm itte d to th e WD M
si d e.
 T hrough elec t r i c a l - l a y e r g r o o m in g , se r vice s o f d iffe r e n t g r a n u lari ti es
a re groom ed a n d e n c a p s u l a t ed in to o n e wa ve le n g th a n d o u tp u t to
th e W DM s id e . T h i s e n a b l e s m u ltip le se r vice s to sh a r e th e b a n dw i dth,
g reat ly impro v i n g b a n d w i d t h u tiliza tio n .

Optical-Layer Grooming
G room s opt ic a l s i g n a l s a t a g r an u la r ity o f wa ve le n g th ( λ) b y fle xibl y
se l e cting t rans m i s s i o n p a t h s .
T h e c ore unit f o r o p t i c a l - l a y e r gr o o m in g is th e ROADM b o a r d .

N S
ROADM ROADM

W E
N
R OA D M ROADM
E
W
S

ROADM

OTU

 Afte r rec eiv in g O T U o p t i c a l s i g n a ls, th e ROADM b o a r d cr e a te s opti cal

cros s -c onnect i o n p a t h s i n t e r na lly a n d o u tp u ts th e sig n a ls to sp eci fi ed
e gres s es . E a c h e g r e s s c o r r e sp o n d s to a sp e cific p a th .
 Op erat ion pe r s o n n e l c a n r e m o te ly co n tr o l th e tr a n sm issio n p a ths of
o pt i cal s igna l s b y c r e a t i n g a n d a d ju stin g cr o ss- co n n e ctio n p a th s on
th e NM S .
 Op t ic al-lay er g r o o m i n g u s e d to g e th e r with th e ASON te ch n o lo g y can
i mplement au t o m a t i c f a u l t d e te ctio n a n d lin e a d ju stm e n t to e n sure
n ormal t ransm i s s i o n o f s e r v i c es.

E lec t ric a l - l a y e r g r o o m i n g a n d o p tica l- la ye r g r o o m in g fo r m the

groom in g c e n t e r o f a n O T N, so th a t th e tr a n sm issio n p a th s of
s ignals c a n b e c o n t r o l l e d r e m o te ly, r e a lizin g fle xib le g r o o mi ng of
s ignals .

- 18 -
 WDM Basics
Transponder & Muxponder

Background
A n Op t ic al Tran s p o n d e r U n i t ( O TU) b o a r d co n ve r ts clie n t- sid e se rvi ces
in to opt ic al s ig n a l s c a r r i e d o v e r IT U- T – co m p lia n t WDM wa ve le n g ths after
pe rf orm ing m ap p i n g , m u l t i p l e x i ng , co n ve r g e n ce , a n d o th e r o p e r a tions.
OTU boards c a n b e d i v i d e d i n t o Tr a n sp o n d e r a n d M u xp o n d e r b y functi on,
wh i ch are abbr e v i a t e d a s T P a n d M P r e sp e ctive ly.

What is the difference between Transponder and Muxponder?

Transponder
N x l o w-s peed c l i e n t - s i d e s e r v i c e s to N x o p tica l sig n a ls ca r r ie d over
IT U -T– c om plian t WD M w a v e l e n gth s

N-I n N - O u t N- In N- Ou t

RX1 OUT1 IN1 TX1

MUX/DMUX
MUX/DMUX

TX1 IN1 OUT1 RX1

N O

N O

N O

N O

Client Client
×DUk

×TUk

×TUk

×DUk

services services
RXn OUTn INn TXn

TXn INn OUTn RXn

Muxponder
N x l o w-s peed c l i e n t - s i d e s e r v i c e s to o n e h ig h - sp e e d o p tica l sig n al
c arri e d ov er I T U - T – c o m p l i a n t W DM wa ve le n g th s

N-I n O n e - O u t N- In On e - Ou t

RX1 TX1
MUX/DMUX
MUX/DMUX

TX1 RX1
N O

1 O

1 O

1 O

1 O

N O

OUT IN
×DUk

×DUk

×TUk

×TUk

×DUk

×DUk

Client Client
services IN OUT services
RXn TXn

TXn RXn

 C l i ent s erv ic e s a r e S D H , S O N ET, OT N, SAN, Eth e r n e t, vid e o , and other

serv ic es .

- 19 -
 WDM Basics
OTN Tributary Boards and Line Boards

Huawei OTN product series support the use of separate tributary and line boards.
Tributary and line boards work with cross-connect boards.
A tributary board plus a line board together performs the functions of an OTU board.
Different from an OTU board, the tributary and line boards achieve more flexible and
fine-grained grooming of electrical services and offers a higher bandwidth utilization
by working with a cross-connect board.

OTN Tributary Boards

Function: OTN tributary boards are used mainly to locally add/drop client services
from the WDM side.
Specifically, an OTN tributary board receives a client service, converts the service
into an electrical signal, maps the signal into an appropriate ODUk signal,
and sends the ODUk signal to the cross-connect board.

Positions of OTN tributary boards in a WDM system

Client-side service ODUk ODUk WDM-side service

OTN tributary
Line board
board
OM

OA
Client-side equipment

OTN tributary
Line-side

Line board
board
FIU

SC1

OTN tributary Line board

board
OA
OD

OTN tributary
Line board
board

OTN Line Boards

Function:
Allows for local add/drop of client services on the WDM side
by working with an OTN tributary board.
Specifically, the OTN tributary board receives a client service, converts the
service into an electrical signal, maps the signal into an appropriate
ODUk signal, and sends the ODUk signal to the cross-connect board for
grooming. When the OTN line board receives the ODUk signal from
the cross-connect board, it maps and multiplexes the signal and converts
the signal into an OTUk signal carried over an ITU-T G.694.1-compliant
DWDM wavelength. The reverse conversion is similar.

Transparently transmits WDM-side services by working with another

OTN line board.
When receiving a west WDM-side service, this OTN line board performs
E/O conversion, demapping, and demultiplexing to convert the service
into an ODUk signal. Then this OTN line board sends the ODUk signal to
the cross-connect board for grooming.
When the other OTN line board receives the ODUk signal from the
cross-connect board, it converts the ODUk signal into an OTUk signal
carried over an ITU-T G.694.1-compliant DWDM wavelength by
performing mapping, multiplexing, and E/O conversion, and sends the
OTUk signal to the east. The reverse process is similar.

Positions of OTN line boards in a WDM system

Local client-side
services added
WDM-side service ODUk ODUk WDM-side service

OTN tributary OTN line

board board
OD
OA

OM

OA

OTN line OTN line

East line-side ODF
West line-side ODF

board board

SC1 SC1
OTN line OTN line
board board
OM
OA

OA
OD

OTN tributary OTN line

FIU

FIU

board board

Local client-side
services dropped

Why the tributary/line separated architecture

is needed?
OTU boards can be used to add client-side services to the WDM side.
Why the tributary/line separated architecture is needed?

OTU architecture

OTU
OM

OA

OTU

SC1
FIU

OTU
OA
OD

OTU

Tributary/Line separated architecture

OTN tributary OTN line

board ODUk board
OM

OA

ODUk
OTN tributary OTN line
board board
SC1
FIU

OTN tributary OTN line

board board
OA
OD

OTN tributary OTN line

board board

As shown in the preceding figures, the major difference between the tributary/line
separated architecture and the OTU architecture lies that centralized cross-connect
boards are introduced into the tributary/line separated architecture for grooming
of ODUk signals at different levels. To be specific, under the tributary/line separated
architecture, client-side services are not encapsulated into WDM-side services (OTUk)
and added to the line side using one board. This architecture improves electrical-layer
grooming flexibility and bandwidth usage.

- 20 -
 WDM Basics:
AMP, BMP, and GMP

Overview

Different board types support various encapsulation modes.

Key modes are:
• Asynchronous Mapping Procedure (AMP)
• Bit-synchronous Mapping Procedure (BMP)
• Generic Mapping Procedure (GMP)
AMP and BMP map ODU1 signals to ODU2/ODU3 and ODU2 signals to ODU3.
Both modes feature simple mapping with a 2.5 Gbit/s tributary timeslot.
GMP maps ODU0, ODU2e, ODU4, ODUflex, and various other client-side signals.
This mode features more flexible mapping with 1.25 Gbit/s and 2.5 Gbit/s
tributary timeslots.

AMP
This mode adjusts NJO and PJO bytes so that the signal rates match on both
server and client sides.

• At the transmit end, values of the NJO, PJO, and JC bytes are determined
by the client-side signal rate.
• At the receive end, justification bytes are classified as stuff bytes or
client-side signals depending on the value of the JC bytes.

Example
TN52TOM: STM-1 <—> OPU1 (AMP) <—> ODU1
15 16 17 18 3824

1 RES JC

2 RES JC

3 RES JC

4 RES NJO PJO

JC: justification control NJO: negative justification opportunity PJO: positive justification opportunity

Signal Rate Comparison JC NJO PJO

Client-side signal = Server-side signal 00 Justification byte Data byte
Client-side signal > Server-side signal 01 Data byte Data byte
Client-side signal < Server-side signal 11 Justification byte Justification byte

Note
The two signals are distinguished as client-side and server-side by encapsulation direction.
When A is encapsulated into B, A would be the client-side signal and B the server-side
signal. For example, when STM-16 is encapsulated into OPU1, STM-16 is the client-side
and OPU1 the server-side signal respectively.

- 21 -
 BMP
This mode is used when the server-side signal rate matches the client-side signal
rate and the signal frequencies of both sides do not deviate.
In this mode:
• The value of JC bytes is 00.
• NJO bytes are used as justification bytes and their values are 0.
• PJO bytes are used as data bytes.
Example
TN52TDX: 10GE LAN <—> BMP <—> OPU2e <—> ODU2e

GMP

This mode only allows signal transmission when the server-side signal rate
exceeds the client-side signal rate.

Example
TN11LOA: CPRI Option3 <—> GMP <—> OPU1 <—> ODU1

In the GMP mode, the Sigma/Delta algorithm is used to stuff client-side signals
with bytes so that the two signal rates match.

Payload Area
The number of bits provided by the
server layer is a fixed value 3808 x 4 x 8.
Client data
Therefore, the transmit-end equipment
OH only needs to send a Cn(t)-specified
Stuff
number of bits to the receive-end
equipment. The receive end can then
separate client-side signals using the
Sigma/Delta algorithm.
Service frame or multi-frame

Sigma/Delta Algorithm

• Client data (C) if ( j*Cn(t)) mod Pserver < Cn(t)

• Stuff (S) if ( j*Cn(t)) mod Pserver ≥ Cn(t)

j: a bit sequence number within the range of [1..Pserver].

Cn(t): the number of bits for client-side signals in one frame.
Pserver: the number of bits server-side signals can provide in one frame.
For example, server-side signals can provide 3808 x 4 x 8 bits in one OTU
frame.

Assume that server-side signals provide eight bits and client-side signals
require six bits.

Bit sequence number 1 2 3 4 5 6 7 8

Multiplied by Cn(6) 6 12 18 24 30 36 42 48
Remainder from mode Pserver(8) 6 4 2 0 6 4 2 0
Final result S C C C S C C C

- 22 -
 WDM Basics
ODUflex

ODUflex is a flexible optical channel data unit.

It has a bandwidth range of 1.25G to 100G and two main features:

1. Efficient transmission:
Flexible rate adaptation mechanisms ODU2
enable users to configure capacities for
containers based on service traffic volume. FC4G
Bandwidth use is now more efficient,
with reduced transmission cost per bit.
VS
2. Good compatibility:
ODUflex applies to various types of services, ODUflex
FC4G
including video, storage, and data services,
and works with transmission requirements
of future IP services.

At present, the two types of ODUflex as defined by ITU-T G.709 are ODUflex(CBR)
based on constant bit rate services, and ODUflex(GFP) based on packet services.

ODUflex (CBR)
Bit rate = 239/238 × client signal bit rate
Only client signals with a bit rate greater FC400/FC800… Pipe FC400/FC800…
than 2.488 Gbit/s can be encapsulated
into ODUflex signals in BMP mode.

ODUflex (GFP)

Any bit rate works for an ODUflex(GFP) signal.

To maximize efficiency, ODUflex(GFP) signals should fill
an integer number of tributary timeslots in the smallest
HO ODUk channel carrying the signals.
The bit rate is equal to n×ODUk.ts (k=2,3,4)，
approximately n×1.25G, where n is the number of
tributary timeslots occupied by ODUflex(GFP).
Packet service signals are encapsulated into ODUFlex
signals in GFP-F mode.

Application Example

Four timeslots Four timeslots

(discontinuous slots allowed) (discontinuous slots allowed)
1 1
2 2
3 3
BMP 4 OTN OTU2
4
FC4G ODUflex ODUflex FC4G
5 network
5
6 6
7 7
8 8
ODU2 ODU2

Encapsulation path: FC4G->ODUflex->ODU2 ODUflex signals are encapsulated into

the four ODU2 timeslots.
The remaining timeslots can carry other service signals, achieving 100% bandwidth
usage.

- 23 -
 WDM Basics
1+1 Protection and 1:1 Protection

1+1 Protection
Site A Site B
Each working channel has a dedi-
cated protection channel. At the Working channel

transmit end, services are transmit- Protected

service Protection channel
ted to both the working and protec-
tion channels. In normal cases, the
Dual feeding Before
receive end receives services from at the source switching
Selective receiving
at the sink
the working channel.

Site A Site B
If the working channel becomes Working channel
abnormal, the receive end selective-
Protected
ly receives services from the protec- service Protection channel
tion channel based on channel
status and external commands.
Dual feeding After Selective receiving
at the source switching at the sink

1:1 Protection

1:1 protection: Each working channel corresponds to a protection channel, and they
protect each other.

Site A Site B Services, however, are not concurrently

Working channel transmitted to both the working and pro-
Protected
service tection channels. Instead, the transmit
Protection channel end sends primary services to the work-
Extra service
ing channel and extra services (low-level
Before services) to the protection channel. Then,
switching the receive end receives primary services
from the working channel and extra ser-
vices from the protection channel.

Site A Site B When the working channel is faulty, the

Protected Working channel transmit end sends primary services to
service the protection channel and the receive
Protection channel end receives primary services from the
Extra service
protection channel. At this time, the extra
After services are terminated to ensure normal
switching transmission of the primary services.

M:N protection: M protection channels provide protection for N working channels.

Generally, M is equal to or less than N.

- 24 -
 WDM Basics
Unidirectional Switching and Bidirectional Switching

Unidirectional switching
When the channel in one direction is faulty, only services in this direction are switched
and services in the other direction are still received from the original channel. Service
switching in each direction is independent and has no impact on service switching in
the other direction.

Advantage: Site A Site B

Working channel

The implementation mechanism of

unidirectional switching is simple and
Protection channel
does not require any protocols, and
service switching is quick.
Unidirectional Switching

Bidirectional switching
When the channel in one direction is faulty, services in both directions are switched
regardless of whether the channel in the other direction is faulty.

Advantage:

In bidirectional switching mode, services in both directions pass through the same
devices and services are easy to manage.
Site A Site B
If a fault occurs on the network, services Working channel
will not be transmitted in the fault area
and the fault can be rectified without
triggering excessive service switching. Protection channel

The latencies in both directions are the

same. Bidirectional Switching

- 25 -
 WDM Basics
DCN Applications

A data communication network (DCN) helps remotely manage and maintain NEs
by providing communication channels for centrally managed NEs.

NMS Standby NMS

External DCN
An external DCN is usually a LAN or WAN
External DCN that provides communications between
NMSs and between the NMS server and
NEs.

Internal DCN
Internal DCN usually refers to the
communication network between NEs.
Huawei WDM equipment uses Huawei
Embedded Control Channel (HWECC) or
TCP/IP for communication and uses OSCs
Internal DCN and ESCs to carry these communication
protocols.
Gateway NEs (GNEs) are
Network cable GNE required to forward information
between the NMS server and
Optical fiber NE non-GNEs.

OSC
WDM equipment provides dedicated optical supervisory channels, which use OSC
units to transmit supervisory and management information between NEs.
OTM1 OLA OTM2

O O
OTU OA OA OA OA OTU
M D
F F F F
OSC I I OSC I I OSC
U U U U
O O
OTU OA OA OA OA OTU
D M

Supervisory channel signal Service signal

ESC
ESCs transmit supervisory information of NEs using overhead bytes (DCCs/GCCs)
in the fixed frame structures of services.
OTM1 OTM2

O O
OTU OA OA OTU
M D
F F
I I
U U
O O
OTU OA OA OTU
D M

The following describes two typical DCN solutions using Huawei WDM equipment.

HWECC Solution

NMS

External DCN

Running HWECC

Network cable Optical fiber GNE NE

If a network has only WDM equipment that supports the HWECC protocol stack,
HWECC is preferred. This solution is easy and convenient to configure and apply.

IP over DCC Solution

Third-party
Huawei WDM equipment Huawei WDM equipment
NMS equipment

External DCN

Running TCP/IP

Network cable Optical fiber GNE NE

If a network has both Huawei WDM equipment and third-party equipment that supports
the TCP/IP protocol stack, IP over DCC is preferred. This solution resolves management
issues for a network consisting of both Huawei equipment and third-party equipment.

- 26 -
 WDM Basics
VPWS Basics

Overview
Virtual private wire service (VPWS) is a point-to-point Ethernet service model
defined by the IETF. When this model is used, service streams on the client side
can be divided by VLAN and carried using MPLS. In MS-OTN, the VPWS is used to
transmit VBR services.
• VPWS (Virtual Private Wire Service)

Service Model

Carrying both single-VLAN

PW1 and multi-VLAN services
Tunnel ODUk λ means powerful service
PW2 bearing and convenient
access.

Service

Typical Networking

Service1
Port1
CVLAN100
Service1 NE2
Branch1
Port1 PW1 Service2
CVLAN100 Port2
PW2 CVLAN200

Tunnel
HQ NE3 Branch2
Service2 NE1
PW3
Port2
CVLAN200 NE4
Service3 Service3
Port3 Port3 Branch3
CVLAN300
CVLAN300

P2MP transmission is adopted between the HQ and its branches:

On the user side, port+CVLAN differentiates three services configured for the HQ.
On the network side, the services enter three separate PWs and are transmitted to
three branches over different tunnels.

- 27 -
 WDM Basics
VPLS Basics

Overview

VPLS is a multipoint-to-multipoint Ethernet service model defined by the IETF.

When this model is used, the LAN networks that are geographically isolated are
connected through the MPLS-TP network to provide multipoint-to-multipoint and
point-to-multipoint services.
• VPLS (Virtual Private LAN Service)

VPLS implements the isolation of transmission tunnels using the MPLS technology.
A virtual switching instance (VSI) is separately created on PE1 and PE2 and a
bidirectional PW is created between them. The bidirectional PW is used as the
virtual Ethernet interface of the VSIs. In addition, a V-UNI interface is also created
for each VSI on the client side.

Basic Concepts

Split Horizon
Here, a split horizon group is created on NE1 with all PWs as group members.
Such a group prevents packet forwarding between PWs to avoid the occurrence of
a network loop or broadcast storm.

MAC Address Learning and Flooding

Each VSI maintains a MAC address table to learn, flood, and forward MAC
addresses, so that packets are sent to the correct PEs and finally forwarded to the
correct users.

Typical Networking

PE1 to PE5 form a multipoint-to-multipoint network where the VPLS technology is

used to implement multipoint-to-multipoint communication. The networks that are
geographically isolated are connected through the network and users share band-
widths based on isolated PWs.

- 28 -
 WDM Basics
EoO Principles

Introduction
Ethernet over OTN (EoO): a WDM channel-based transmission technology that
processes, encapsulates, and maps Ethernet signals for transmission in an OTN
system.
Traditional OTN carries Ethernet private lines and does not support rate limiting and
aggregation scenarios. The EoO technology is introduced to support L2 functions,
such as LAG, QoS, and OAM, and implement port-based rate limiting, service
aggregation, bandwidth sharing, and traffic monitoring.

EoO Board Principles

ODU
ODUk
RX1/TX1
ODUk ODUk
ODUk
GE/10GE LAN/FE OTN line
board
Packet ODU
process Mapper
RXn/TXn
L2 ODUk

EoO

L2 transparent
1 transmission
PORT1 Trunk
PORT2 VLA
N10
0(se
PORT3 rvic
e1)
VLAN20
0(servic
e2)
2 L2 aggregation
Trunk
3)
PORTn 0 (serv e
ic
VLAN30 ODUk
GFP-F
EoO board
4)
service flow 0(service Trunk
VLAN40
PORTk VLAN500(service5) 3 External port sharing
VLA Trunk
N60
0(se
GE/FE/ rvice Trunk
6)
10GE LAN
OTN processing
L2 module
Trunk indicates an internal virtual port

A Trunk port is mapped to an ODUk (k = 0, 1, 2, 3, 4, flex) through

GFP-F
Based on actual traffic, bind a proper number of ODUflex timeslots and
assign suitable bandwidths to Trunk ports to effectively carry services.
Fixed Grid
During board interconnection, use the same the mapping mode.

EoO Features
Compared with Traditional OTN Tributary Boards

VLAN-based rate limiting

10GE 10GE
10GE EoO Trunk Trunk EoO 10GE
GE GE

Port-based rate limiting

10GE 10GE
EoO Trunk Trunk EoO
10GE 10GE
GE GE

1.EoO boards support port-based rate limiting and VLAN-based rate limiting and
implement charging by bandwidth and bandwidth monetization.
2.EoO boards support service convergence and pipe sharing. On the national and
provincial backbone networks with scarce resources and high pipe rentals, EoO boards
can reduce costs and simplify management.
3.EoO boards support simple configuration and implement bandwidth expansion, fast
capacity expansion, and service continuity

Compared with Packet Boards

1.The cross-connect granularity of the cross-connect boards connected to EoO boards is

ODUk, and that of the cross-connect boards connected to packet boards is PKT.
2.EoO boards only require common line boards, and packet boards must work with
universal line boards.

Impact on the live network: Packet boards need

Impact on the live network: Only EoO boards need
to be added at network edges, and universal line
to be added.
boards need to be added at intermediate nodes.
Universal
line board
Universal Universal
line board line board

Universal Universal
line board Universal line board

line board

- 29 -