T Rec G.987.2 202302 I!!pdf e
T Rec G.987.2 202302 I!!pdf e
T Rec G.987.2 202302 I!!pdf e
Recommendation
ITU-T G.987.2 (02/2023)
Summary
Recommendation ITU-T G.987.2 describes the physical layer requirements and specifications for the
XG-PON physical media dependent (PMD) layer. Wavelength enhancement bands are described in
Recommendation ITU-T G.987.1. The transmission convergence (TC) layer is described in
Recommendation ITU-T G.987.3. The optical network unit (ONU) management and control interface
(OMCI) specifications are described in Recommendation ITU-T G.988.
Recommendation ITU-T G.987.2 describes a flexible optical fibre access network capable of supporting
the bandwidth requirements of business and residential services. The G.987 series of standards allows
for multiple upstream and downstream line rates. This Recommendation currently defines one type of
10-Gigabit-capable passive optical network (XG-PON) system with an asymmetric nominal line rate of
9.95328 Gbit/s in the downstream direction and 2.48832 Gbit/s in the upstream direction.
This Recommendation describes a system that represents an evolutionary development from the
systems described in the ITU-T G.984 series. To the greatest extent possible, this Recommendation
maintains the requirements of Recommendation ITU-T G.984.1 to ensure maximal continuity with
existing systems and optical fibre infrastructure.
Amendment 2 continues the maintenance and evolution of physical media dependent (PMD) layer
specifications for XG-PON as defined in this Recommendation. It includes technical updates and
corrections for changing references to XG-PON1 to XG-PON, replacing the mask of the eye diagram
for the ONU transmitter, updating the DD40 downstream specification, correcting the X/S tolerance
mask for ONU and updating the X/S tolerance mask for the optical trunk line (OLT).
This revision adds a new annex specifying out-of-band noise limits on XG-PON ONUs to reduce the
impact on other systems coexisting on the same PON.
History
Edition Recommendation Approval Study Group Unique ID*
1.0 ITU-T G.987.2 2010-01-13 15 11.1002/1000/10411
2.0 ITU-T G.987.2 2010-10-07 15 11.1002/1000/10889
2.1 ITU-T G.987.2 (2010) Amd. 1 2012-02-13 15 11.1002/1000/11499
3.0 ITU-T G.987.2 2016-02-26 15 11.1002/1000/12832
3.1 ITU-T G.987.2 (2016) Amd. 1 2017-08-13 15 11.1002/1000/13290
3.2 ITU-T G.987.2 (2016) Amd. 2 2020-10-29 15 11.1002/1000/14491
4.0 ITU-T G.987.2 2023-02-22 15 11.1002/1000/15124
Keywords
10-Gigabit-capable passive optical network, FTTx network, OLT, ONU, optic, optical network
terminal (ONT), passive optical network (PON), physical layer interfaces, physical layer requirements,
physical layer specification, PMD, XG-PON.
* To access the Recommendation, type the URL http://handle.itu.int/ in the address field of your web
browser, followed by the Recommendation's unique ID. For example, http://handle.itu.int/11.1002/1000/11830-en.
NOTE
In this Recommendation, the expression "Administration" is used for conciseness to indicate both a
telecommunication administration and a recognized operating agency.
Compliance with this Recommendation is voluntary. However, the Recommendation may contain certain
mandatory provisions (to ensure, e.g., interoperability or applicability) and compliance with the
Recommendation is achieved when all of these mandatory provisions are met. The words "shall" or some other
obligatory language such as "must" and the negative equivalents are used to express requirements. The use of
such words does not suggest that compliance with the Recommendation is required of any party.
© ITU 2023
All rights reserved. No part of this publication may be reproduced, by any means whatsoever, without the prior
written permission of ITU.
1 Scope
This Recommendation pertains to flexible access networks using optical fibre technology. The focus
is primarily on a network supporting services with bandwidth requirements ranging from th ose of
voice services to data services running at up to 10 Gbit/s. Also included are broadcast services.
This Recommendation describes characteristics of the physical medium dependent (PMD) layer of
an optical access network (OAN) with the capability of transporting various services between the
user-network interface and the service node interface.
The OAN described in this Recommendation enables the network operator to provide a flexible
upgrade to meet future customer requirements, in particular, in the area of the optical distribution
network (ODN). The ODN considered is based on a point-to-multipoint tree and branch option.
2 References
The following ITU-T Recommendations and other references contain provisions which, through
reference 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. The reference to a document within this
Recommendation does not give it, as a stand-alone document, the status of a Recommendation.
[ITU-T G.652] Recommendation ITU-T G.652 (2016), Characteristics of a single-mode
optical fibre and cable.
[ITU-T G.657] Recommendation ITU-T G.657 (2016), Characteristics of a bending loss
insensitive single-mode optical fibre and cable.
[ITU-T G.783] Recommendation ITU-T G.783 (2006), Characteristics of synchronous
digital hierarchy (SDH) equipment functional blocks.
[ITU-T G.825] Recommendation ITU-T G.825 (2000), The control of jitter and wander
within digital networks which are based on the synchronous digital hierarchy
(SDH).
[ITU-T G.957] Recommendation ITU-T G.957 (2006), Optical interfaces for equipments and
systems relating to the synchronous digital hierarchy.
[ITU-T G.959.1] Recommendation ITU-T G.959.1 (2016), Optical transport network physical
layer interfaces.
[ITU-T G.982] Recommendation ITU-T G.982 (1996), Optical access networks to support
services up to the ISDN primary rate or equivalent bit rates.
[ITU-T G.984.1] Recommendation ITU-T G.984.1 (2008), Gigabit-capable passive optical
networks (GPON): General characteristics.
[ITU-T G.984.2] Recommendation ITU-T G.984.2 (2003, as amended), Gigabit-capable
passive optical networks (G-PON): Physical media dependent (PMD) layer
specification.
3 Definitions
5 Conventions
See clause 5 of [ITU-T G.987].
Table 6-1 – Classes for optical path loss defined in this Recommendation
Nominal1 class Nominal2 class Extended1 class Extended2 class
(N1 class) (N2 class) (E1 class) (E2 class)
Minimum loss 14 dB 16 dB 18 dB 20 dB
Maximum loss 29 dB 31 dB 33 dB 35 dB
DD20 DD40
Maximum differential distance 20 km 40 km
For DD40 ODNs, due to the intrinsic wavelength dependence of the optical fibre loss [ITU-T G.652],
the fibre attenuation coefficient at the downstream wavelength (1577 nm) is lower than that at the
upstream wavelength (1270 nm) resulting in a downstream loss margin. This margin is expected to
be at least 1 dB for a fibre length greater than 20 km. The additional maximum optical path penalty
allowed for DD40 is therefore fully compensated by the lower fibre loss at the downstream
wavelength. Thus, the other PMD values (except optical path penalty (OPP)) for DD40 do not change
from those PMD values specified for DD20.
In DD40 deployments, the actual fibre cable loss characteristics should be assessed, to make sure
there is the required extra margin of 1dB with respect to the maximum OPL class loss at the
downstream wavelength for ODN branches with a maximum OLT-ONU fibre length greater than
20 km.
7 Services
See clause 7 of [ITU-T G.987.1].
Figure 9-1 – Relationship between ONU power levels and burst times
9.2.7.4 Minimum extinction ratio
The extinction ratio (ER) is defined as:
ER = 10 log10 (A/B)
where A is the average optical power level at the centre of the binary 1 and B is the average optical
power level at the centre of the binary 0.
Table 9-5 – Mask of the eye diagram for OLT transmitter – Numeric values
9.95328 Gbit/s
x3 − x2 (Note 1) 0.2
y1 0.25
y2 0.75
y3 0.25
y4 0.25
NOTE 1 – x2 and x3 of the rectangular eye mask need not be equidistant with respect to
the vertical axes at 0 UI and 1 UI.
NOTE 2 – The values are taken from [ITU-T G.959.1], clause 7.2.2.14.
Table 9-6 – Mask of the eye diagram for ONU transmitter – Numeric values
Figure 9-5 – Test set-up for mask of the eye diagram for ONU transmitter
The jitter transfer function of an ONU shall be under the curve given in Figure 9-6, when input
sinusoidal jitter up to the mask level in Figure 9-7 is applied, with the parameters specified in this
figure for each line rate.
Figure 9-7 – High-band portion of sinusoidal jitter mask for jitter transfer
Figure 10-1 – X/S tolerance mask for ONU (Versatile WDM configuration)
S: Received power.
X: Maximum total power of additional services received in the blocking wavelength range.
X/S: In the mask (hatching area) should not cause the XG-PON receiver to fail to meet its sensitivity requirements.
The purpose of this annex is to specify a set of additional requirements on the ONU transmitter which
should be applied in multisystem coexistence deployments. These requirements are stricter than those
in the base text of this Recommendation.
The requirements specified in this annex are optional. Implementations of this Recommendation
should not be assumed to meet the requirements of this annex. If the implementer declares the support
of this annex, the implementation shall meet all the requirements specified herein.
Coexistence with other PON systems (e.g., ITU-T G.984.x Series, ITU-T G.989.x Series, ITU-T
G.9804.3, etc.) on the same ODN is a frequent requirement for PON system deployments. In such
cases, light from ONU transmitters of one PON type may fall into the upstream operating wavelength
band of a coexisting PON system. If the optical power of this out-of-band (OOB) light is too high,
then this may cause harmful crosstalk penalties on the upstream transmissions of the coexisting PON
systems. Such problems are exacerbated in PON ODNs where the differential optical loss is high, and
the interfering ONU is on a lower loss path.
To ensure successful operation in PON coexistence use cases, the out-of-band power spectral density
(OOB-PSD) must be below a certain level. The OOB-PSD mask for the ONU transmitter at the R/S
interface is shown in Figure A.1 and the values are given in Tables A.1 and A.2. To meet the OOB-
PSD mask requirement, the ONU PSD remains below the dashed mask line under all operating
conditions. The in-band power limit (Y) is the maximum ONU launched power and the points at the
ONU operating band edge (Y-SMSR) represent the power limit given by the ONU transmitter side-
mode-suppression ratio (SMSR). The out-of-band PSD (X) is the power outside of the ONU operating
wavelength band that may fall into coexisting PON upstream operating wavelength bands. The OOB-
PSD spectral interval is 0.1 nm according to the industry convention for optical signal-to-noise
measurement.
Wavelength (nm)
λ1 Not applicable
λ2 1260
λ3 1280
λ4 1284
See Appendix II of [b-ITU-T G.984.5] for a generic consideration of wavelength allocation for
XG-PON, G-PON and video distribution services.
The physical layer overhead time (Tplo) is used to accommodate five physical processes in the PON.
These are: laser on/off time, timing drift tolerance, level recovery, clock recovery and start of burst
delimitation. The exact division of the physical layer time to all these functions is determined partly
by constraint equations and partly by implementation choices. This appendix reviews the constraints
that the OLT must comply with, and suggests values for the discretionary values.
Tplo can be divided into three sections with respect to what ONU data pattern is desired. For simplicity,
these times can be referred to as the guard time (T g), the preamble time (Tp) and the delimiter time
(Td). During Tg , the ONU transmit levels are not specified, provided they comply with the PMD
Tables 9-3 and 9-4. During the Tx disable bit period immediately following the assigned burst, the
maximum launch power level allowed is the zero-level corresponding to the minimum ER specified
in Tables 9-3 and 9-4. During Tp, the ONU will transmit a preamble pattern that provides the desired
transition density and signal pattern for fast level and clock recovery functions. Lastly, during Td , the
ONU will transmit a special data pattern with optimal autocorrelation properties that enable the OLT
to find the beginning of the burst. Table III.2 gives recommended values for T g, Tp, Td and Tplo . Figure
III.1 shows the timing relationship between the various physical layer overhead times.
Figure III.1 – Timing relationship between the various physical layer overhead times
An additional parameter of the control logic on the PON is the total peak-to-peak timing uncertainty
(Tu). This uncertainty arises from variations of the time of flight caused by the fibre and component
variations with temperature and other environmental factors.
The constraint equations with which the OLT must comply are then:
𝑇𝑔 > 𝑇𝑜𝑛 + 𝑇𝑢 and 𝑇𝑔 > 𝑇𝑜𝑓𝑓 + 𝑇𝑢
These equations can be explained as follows. The first equation makes sure that the following burst's
laser on ramp-up does not fall on top of the last burst's data. The second equation makes sure that the
last burst's laser off tail-off does not fall on top of the following burst's preamble.
Tp must be sufficient for the physical layer to recover the signal level (essentially, setting the decision
threshold) and the signal clock phase. There are many diverse design approaches to these two
problems, each with its own benefits and costs. Some designs are very fast, but require an external
trigger signal and produce sub-optimal error performance. Other designs are slower, but do not
require a reset signal and produce bit errors that are normally distributed. In addition, each of these
designs may have special requirements on the data pattern used for the preamble. Some designs prefer
a maximum transition density pattern, while others prefer a pattern with a balance of transitions and
controlled runs of identical digits.
N Pseb
8 2.8 × 10–7
12 2.2 × 10–10
16 1.8 × 10–13
20 1.5 × 10–16
24 1.3 × 10–19
32 1.1 × 10–25
64 4.9 × 10–50
With these considerations taken into account, the worst case and objective allocations of the physical
layer overhead are given in Table III.2. This table also lists the values for the ONU Tx enable time
and Tx disable time, and the total physical layer overhead time for reference. The worst-case values
are intended to provide a reasonable bound for easy implementation, and the objective values are
intended to be the design target for more efficient implementation with optimized components. These
values are for a simple ODN without reach extenders. Reach extenders may require their own guard
and preamble time allowances, making the total overhead larger.
Figure IV.1 – The place of the jitter budget in the overall jitter specification
The jitter budget specifies jitter components that are not covered by the low frequency jitter
specifications. All jitter components accounted for in the jitter budget are integrated over the
frequency domain that starts from the upper corner frequency (ft) of the jitter tolerance mask
(see clause 9.2.9.7.2).
The jitter budget is based on the dual Gaussian jitter model. In this model, the jitter components are
classified into deterministic jitter (DJ) and random jitter (RJ). The DJ components are modelled as a
bimodal distribution and the RJ components as a Gaussian distribution. In addition, the duty-cycle
distortion (DCD), which is a DC component, is included into the DJ specification of the jitter budget.
The basic assumptions of the model are as follows:
1) Jitter is represented assuming the DJ has an equiprobable bimodal distribution, and the RJ
has a Gaussian distribution.
2) All sources of random jitter are assumed independent; therefore, RJ RMS values can be added
by squares.
3) All sources of DJ are assumed to be correlated (this is a worst-case assumption, meaning that
all DJ components are either together at maximum value or together at minimum value, with
equal probability for the minimum and the maximum to occur).
One of the major challenges for the burst mode clock and data recovery (CDR) implementation is the
need to cope with the transient effects causing additional eye closure at the beginning of the burst.
The burst mode CDR has to acquire the phase information exactly on the preamble portion of the
incoming data stream, hence it would be reasonable to require the optical module to preserve signal
good quality. The better the performance of the optics, the shorter the preamble required for correct
system operation.
We define the burst mode eye opening as the opening of the eye pattern that is collected starting from
an offset X from the beginning of the burst. The burst mode acquisition time is defined as the shortest
offset X from the beginning of the burst that will render a compliant eye pattern. A compliant eye
pattern means an eye opening that meets the jitter budget and correct logical signal levels.
A set-up for measuring burst mode acquisition time and burst mode eye opening occurring at the
beginning of each burst is presented in Figure V.1.
The following is a list of recommendations related to measurement of burst mode acquisition time
and burst mode eye opening compliance:
− Use a real time sampling scope capable of 40 GS/s or more and a memory depth that can
cover at least 125 μs (5M samples) for capturing the eye pattern.
− In order to separate the eye pattern that corresponds to a single given ONU, the oscilloscope
is triggered on RSSI-strobe signal [b-ITU-T G Suppl. 48]. The timing diagram for this signal
is presented in Figure V.2.
The test set-up shown in Figure V.1 can be used to determine the burst mode acquisition time of a
given set of optics. The eye pattern measurement is repeated for different values of the parameter X
(defined in Figure V.3) and the minimum setting of X for which the (horizontal) burst mode eye
opening at TPd is better than 0.38 UI determines the burst mode acquisition time of the system. Note
that 0.38 UI is equal to (1 − TJ) for test point TPd in Table IV.2.
The system will perform correctly for settings of the preamble length that are longer than the burst
mode acquisition time.
Series J Cable networks and transmission of television, sound programme and other multimedia
signals
Series K Protection against interference
Series L Environment and ICTs, climate change, e-waste, energy efficiency; construction, installation
and protection of cables and other elements of outside plant
Series M Telecommunication management, including TMN and network maintenance
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
Published in Switzerland
Geneva, 2023