ITU - Telecommunication Standardization Sector
(PLEN)
STUDY GROUP 15
Temporary Document 48-E
Original: English
Geneva, Switzerland, 12-23 October
Question: 4/15
SOURCE:
Marcos Tzannes, Editor G.996.1
Aware, Inc.
Tel: 781-276-4000
FAX: 781-276-4001
E-mail: marcos@aware.com
TITLE:
G.996.1 draft (G.test)
_______________________________________________
ABSTACT
This temporary document contains the draft for G.996.1.
1
INTERNATIONAL TELECOMMUNICATION UNION
ITU-T
Draft G.996.1
TELECOMMUNICATION
STANDARDIZATION SECTOR
OF ITU
TRANSMISSION SYSTEMS AND MEDIA
TEST PROCEDURES FOR DIGITAL
SUBSCRIBER LINE (DSL) TRANSCEIVERS
Draft ITU-T Recommendation G.996.1
(Previously ìCCITT Recommendationî)
2
(date)
FOREWORD
The ITU-T (Telecommunication Standardization Sector) is a permanent organ of the International
Telecommunication Union (ITU). The ITU-T is responsible for studying technical, operating and tariff
questions and issuing Recommendations on them with a view to standardizing telecommunications on a
worldwide basis.
The World Telecommunication Standardization Conference (WTSC), which meets every four years,
establishes the topics for study by the ITU-T Study Groups which, in their turn, produce Recommendations on
these topics.
The approval of Recommendations by the Members of the ITU-T is covered by the procedure laid down in
WTSC Resolution No. 1 (Helsinki, March 1-12, 1993).
ITU-T Recommendation G.996.1 was prepared by ITU-T Study Group 15 (1997-2000) and was approved
under the WTSC Resolution No. 1 procedure on the xxth of xx 199x.
___________________
NOTE
In this Recommendation, the expression ìAdministrationî is used for conciseness to indicate both a
telecommunication administration and a recognized operating agency.
††ITU††199x
3
All rights reserved. No part of this publication may be reproduced or utilized in any form or by any
means, electronic or mechanical, including photocopying and microfilm, without permission in writing
from the ITU.
4
Table of Contents
1.
2.
3.
4.
5.
6
7
8
9
Scope
References
Definitions
Abbreviations
Test procedures and laboratory set-up
5.1 ADSL Operation with splitters
5.1.1 Laboratory set-up
5.1.2 Crosstalk Test
5.1.3 Impulse Noise Test
5.1.4 POTS interference Test
5.2 ADSL operation without a splitter at the ATU-R
5.2.1 Laboratory Setup
5.2.2 Crosstalk Test
5.2.3 Impulse Noise Test
5.2.4 POTS interference Test
5.3 POTS QOS testing
5.3.1 Telephone Circuitry
5.3.2 Test conditions
5.3.3 Summary of Test Procedure
Test loops
6.1 Local loop configurations
6.1.1 North American Test Loops
6.1.2 European Local Test Loops
6.1.3 Test loops in an environment co-existing with TCM-ISDN DSL
6.2 In-home wiring models
Power spectral density of crosstalk disturbers
7.1 Simulated DSL PSD and induced NEXT
7.2 Simulated HDSL PSD and induced NEXT
7.3 Simulated T1 line PSD and induced NEXT
7.4 Simulated ADSL downstream PSD and induced NEXT and FEXT
7.4.1 FEXT 41
7.4.2 NEXT
7.5 Simulated ADSL upstream PSD and induced NEXT and FEXT
7.5.1 FEXT 45
7.5.2 NEXT
7.6 ETSI A, -ETSI B and Euro-K crosstalk
7.7 Power spectral density of crosstalk disturbers in an environment coexisting with
7.7.1 Simulated TCM-ISDN DSL PSD and induced NEXT and FEXT
7.7.2 Simulated HDSL PSD and induced NEXT
7.7.3 Simulated downstream ADSL PSD and induced NEXT and FEXT
7.7.4 Simulated upstream ADSL PSD and induced NEXT
Characteristics of impulse noise waveforms
Phone Model Structures for Testing DSL systems without a splitter at the ATU-R
9.1 Off phone hook model #1
9.2 Off Hook phone model #2
9.3 On hook phone model
6
6
6
6
6
7
7
8
10
11
12
12
13
13
14
14
14
16
16
17
17
17
21
25
34
36
36
37
39
40
43
45
45
47
TCM-ISDN DSL49
50
52
54
58
60
67
67
68
69
5
1. Scope
This Recommendation describes the testing procedures for ITU ADSL Recommendations. The testing
procedures described in this Recommendation include methods for testing ADSL transceivers in the presence
of crosstalk from other services, impulse noise and POTS signaling. Test loops and in-home wiring models are
specified for different regions of the world for use during ADSL performance testing. Other ADSL
Recommendations reference this document for testing procedures and configurations. This Recommendation
does not specify performance requirements for these other Recommendations. This Recommendation only
specifies the procedures for measuring the performance requirements for a particular Recommendation.
2. References
3. Definitions
downstream: ATU-C to ATU-R direction
upstream: ATU-R to ATU-C direction.
Splitter: a filter that seperates the high frequency signals (ADSL) from the voice band signals.
Voice band: The frequency band from 0-4 kHz.
Voice band services: POTS and all data services that use the voice band.
4. Abbreviations
ADC
ADSL
ATU-C
ATU-R
BER
CO
DAC
DMT:
DSL
ES
FEXT
kbit/s
NEXT
QOS
POTS
PSD
RMS
RT
⊕
XT
analog to digital converter
asymmetric digital subscriber line
ADSL Transceiver Unit, Central office end
ADSL Transceiver Unit, Remote terminal end
Bit Error Rate
central office
digital to analog converter
discrete multitone
digital subscriber line
Errored Seconds
fear end crosstalk
kilo bits per second
near end crosstalk
Quality of service
Plain old telephone service
power spectral density
Root Mean Squared
remote terminal
exclusive-or; modulo-2 addition
Crosstalk
5. Test procedures and laboratory set-up
The methods in this document test ADSL system transmission performance. These laboratory methods
evaluate a system’s ability to minimize digital bit errors caused by interference from:
6
•
•
•
•
crosstalk coupling from other systems;
background noise;
impulse noise;
POTS signaling.
These potential sources of impairment are simulated in a laboratory set-up that includes test loops, test sets,
and interference injection equipment, as well as the test system itself.
The crosstalk and impulse noise interfering signals are simulations that are derived from a consideration of
real loop conditions and measurements. The test procedure is to inject the interference into the test loops and
measure the effect on system performance by a bit error test simultaneously run on the system information
channels.
For crosstalk an initial, or reference, power level for the interference represents the expected worst case. If the
interference power can be increased without exceeding a specified error threshold, the system has a positive
performance margin. Performance margin, expressed in dB, is the difference between the interference level at
which the error threshold is reached, and the reference (or 0 dB) level.
In the case of impulse noise, an increasing interference level is similarly applied up to the error threshold, and
the estimated performance is computed from this information. Because the impulse noise characteristics of the
loop plant are not completely understood, the estimation method is based on measured data from several sites.
The estimated number of error-causing impulses is compared to a specified errored-seconds (ES) criterion. The
test procedure makes separate determinations of crosstalk margins and impulse error thresholds, although a
background crosstalk interference is applied during impulse tests.
The POTS measurement uses a number of signaling and alerting activities done with real phones and CO
lines, and also has crosstalk interference present. The BER test sets check for ES or a BER threshold.
5.1 ADSL Operation with splitters
5.1.1 Laboratory set-up
Figure 1 shows the test setup for measuring performance margins for the downstream channel in ADSL
systems with splitters. Figure 2 shows the test setup for measuring performance margins for the upstream
channel in ADSL systems with splitters. The test system consists of a central office transceiver (ATU-C), a
remote end transceiver (ATU-R), and associated splitters. The two transceivers are connected together by the
test loop. Calibrated simulated crosstalk is injected through a high impedance network across the tip and ring
of the loop at the input of one of the transceivers. Impulse noise from a waveform generator is similarly
injected. Crosstalk and impulses are injected at the ATU-R for downstream tests, and at both the ATU-C for
upstream channel tests.
For downstream testing, as shown in Figure 1, pseudo-random binary data from the transmitter of the bit error
rate (BER) test set is presented at the downstream channel input of the ATU-C. The downstream channel
output from the ATU-R is connected to the receiver of the same or a similar BER test set. The test set measures
BER or errored-seconds (ES), as needed. Similar error testing is done in the upstream direction at the rate
needed for the particular system under test (see Figure 2).
A telephone set is connected to the telephone jack of the splitter at the ATU-R end, and a working telephone
line circuit is connected to the telephone jack on the splitter at the ATU-C end.
7
Downstream
channel
error test
ATU-C
ATU-R
Splitter
Test loop
Σ
Downstream
channel
error test
Splitter
Hi-Z
Hi-Z
CO line
circuit
POTS
Set
attenuator
Simulated
Crosstalk
Noise
attenuator
Impulse
waveform
generator
Figure 1: Laboratory test set-up for measuring performance margins for the downstream channel with
splitters
Upstream
channel
error test
ATU-C
ATU-R
Splitter
Σ
Test loop
Hi-Z
Upstream
channel
error test
Splitter
Hi-Z
CO line
circuit
POTS
Set
attenuator
Simulated
Crosstalk
Noise
attenuator
Impulse
waveform
generator
Figure 2: Laboratory test set-up for measuring performance margins for the upsteam channel with
splitters
5.1.2
Crosstalk Test
Crosstalk testing is performed to evaluate the performance of DSL systems in the presence of crosstalk from
other services in the loop.
8
5.1.2.1 Crosstalk Injection
Simulated crosstalk, XT (NEXT and/or FEXT) is introduced into the test loop at the ATU-R so as to achieve
the appropriate voltage level without disturbing the impedance of the test loop or the transceiver. This is done
with a balanced series feed of high impedance. One method for both test and calibration is shown in Figure 3.
The Thevenin impedance of all noise-coupling circuits connected to the test loop shall be greater than 4000
ohms.
The simulated XT should ideally have the power and spectral density defined by the equations for PNEXT or
PFEXT in section 7. It is acknowledged, however, that if the method of generating the simulated XT is similar
to that shown in Figure 3, then its accuracy will depend on the design of the filter used to shape the white
noise. Therefore a calculated XT PSD may be defined for which a tolerance on f0 of +2% is allowed at each
null. Then the accuracy of the simulated XT shall be within +1 dB of the calculated XT for all frequencies at
which the calculated value is less than 45 dB below the peak value. The total power of the simulated XT shall
be within +0.5 dB of the specified value using the same calibration termination.
The crest factor of the simulated XT shall be at least 5.
The simulated XT PSD shall be verified using the calibration termination shown in Figure 3 and a selective
voltmeter or true RMS meter with a bandwidth of approximately 3 kHz. For DSL and HDSL crosstalk the
circuit shall be calibrated for 1.3 dB less crosstalk than specified in section 7, in order to compensate for the
use of 100 ohm terminations instead of 135 ohms.
high impedance coupling circuit
bandwidth
12kHz-2MHz
≤ 15 dBm
75 Ω unbal
<- - - - - - - - - - - - - - - - - - - - ->
5.1kΩ
transformer
NEXT
filter
1.8kΩ
0 TO 120 dB
0.5 dB steps
↔
tip
−−−
2.2kΩ
ring
75 Ω unbal
1800 Ω bal
5.1kΩ
To
test
loop -
−−−−−−−
white noise
generator
attenuator
75 Ω
Z=11.2 k Ω
>
Calibration termination
connect instead of test loop
>
R loop
100 Ω
R rec
100Ω
Figure 3: High impedance crosstalk injection circuit
Care should be taken when specifying the white noise generator in Figure 3; consideration should be given to
the following factors:
• The probability distribution of the peak amplitude: The noise shall be Gaussian within all frequency
bands;
• Crest factor: The crest factor is an indication of the number of standard deviations to which the noise
follows a Gaussian distribution; the required minimum crest factor is 5.
• The frequency spectrum: If the noise is generated using digital methods the sequence repetition rate
will affect the correlation of the samples, and hence the frequency spectrum.
5.1.2.1.1
Crosstalk Injection for TCM-ISDN environment
For testing transmission performance of an ADSL system conforming to Annex C (ADSL operating in an
environment co-existing with TCM-ISDN DSL), the calibrated simulated TCM-ISDN DSL NEXT PSD and
TCM-ISDN DSL FEXT PSD defined in section 7 shall be alternately injected in burst mode as specified in
9
Figure 4, which is the same as the TCM-ISDN DSL signal transceiver scheme. The value of X in the figure
may be fixed during testing transmission performance.
The specified error threshold and minimum performance margin shall be met for an arbitrary value of X in the
range of 6 ≤ X ≤ 40 .
800 UIDSL (= 2.5 ms)
X UIDSL
377 UIDSL
(46-X) UIDSL
377 UIDSL
period A
period C
peri od A
period B
period C
period C
period A: DSL NEXT noise injection
period B: DSL FEXT noise injection
period C: DSL XT noise non-injection
6
X
40
Note: 1 UI DSL = 3.125 µs
Figure 1.tcm: TCM-ISDN DSL crosstalk injection met
Figure 4: TCM-ISDN DSL crosstalk injection method
5.1.2.2 Crosstalk test method
Before testing, the ADSL units are trained with the required crosstalk interference present. The simulated
crosstalk power is injected at the appropriate reference level. This power level is considered the 0 dB margin
level for that type and number of disturbers. For example, the 0 dB margin level for 24-disturber DSL NEXT is
–52.6 dBm. Margin measurements are made by changing, in whole dB steps, the power level of the crosstalk
injected at the transceiver and monitoring the BER over the test loops. A tested system has positive margin for
a given type of crosstalk on a given loop if the system was able to operate at the specified BER with injected
crosstalk power greater than the 0 dB margin level.
The criteria for margin level determination shall include a check that the ADSL unit can train at the margin
level.
The minimum testing times to determine BERs with 95% confidence are shown in table 50.
Table 50 - Minimum test time for crosstalk
Bit Rate
Minimum test time
above 6 Mbit/s
100 seconds
1.544 Mbit/s to 6 Mbit/s
500 seconds
less than 1.544 Mbit/s
20 minutes
5.1.3
Impulse Noise Test
Impulse noise testing is performed to evaluate the performance of DSL systems in the presence impulse noise.
10
5.1.3.1 Impulse noise injection
The same coupling circuit as is used in 5.1.2.1 is used for impulse noise injection. The amplitude level of the
impulses may be measured with an oscilloscope.
5.1.3.2 Impulse noise test method
Before testing, the ADSL units are trained with the required crosstalk interference. The test procedure consists
of injecting the selected impulse wave form at varying amplitude levels and random phase. At each level the
impulse is applied 15 times with a spacing of at least one second while an error measurement is made on the
ADSL channels. The amplitude (ue) in millivolts at which half the impulses cause an error is determined for
each wave form.
Using the above amplitude determinations, the following equation gives the estimated probability that a second
will be errored:
E = 0.0037 P(u>ue1) + 0.0208 P(u>ue2)
, for 5 mV= ue = 40 mV
where: P(u>ue) =
25
2
ue
P(u>ue) =
0. 625
ue
, for ue> 40 mV
ue1 refers to waveform 1
ue2 refers to waveform 2
The resulting probability of errored seconds (ES) value shall be less than 0.14%.
5.1.4
POTS interference Test
POTS interference test is performed to evaluate the performance of DSL systems is the presence of POTS
signaling.
5.1.4.1 POTS interference injection
The interference due to POTS service on the same line is generated by use of actual telephones and central
office circuits connected in the normal way to the system under test. The following POTS signaling and
alerting activities shall be performed:
•
•
•
•
Call phone at ATU-R and allow to ring 25 times;
Pick up ringing phone at ATU-R, 25 times;
Perform off-hook and on-hook activity on phone at ATU-R, 25 times;
Perform pulse and tone dialing.
5.1.4.2 POTS interference test method
Before testing, the ADSL units are trained with the required crosstalk interference. Signaling disturbances are
created through use of the CO line connected to the splitter at the ATU-C, and the telephone set connected to
the telephone jack of the splitter at the ATU-R. During these activities the ADSL channels shall be monitored
while noting any test conditions that cause errored seconds.
11
5.2 ADSL operation without a splitter at the ATU-R
5.2.1
Laboratory Setup
Figure 5 shows the test setup for measuring performance margins for the downstream channel in ADSL
systems without a splitter at the ATU-R. Figure 6 shows the test setup for measuring performance margins for
the upstream channel in ADSL systems without a splitter at the ATU-R. The test system consists of a central
office transceiver (ATU-C), a remote end transceiver (ATU-R), and splitter at the ATU-C. The two
transceivers are connected together by the test loop and the in-home wiring model. Calibrated simulated
crosstalk is injected through a high impedance network across the tip and ring of the loop at the input of one of
the transceivers (when injecting crosstalk at the ATU-R, the high impedance network should be connected
between the test loop and the in-home wiring model). Impulse noise from a waveform generator is similarly
injected. Crosstalk and impulses are injected at the ATU-R for downstream tests, and at both the ATU-C for
upstream channel tests.
For downstream testing, pseudo-random binary data from the transmitter of the bit error rate (BER) test set is
presented at the downstream channel input of the ATU-C. The downstream channel output from the ATU-R is
connected to the receiver of the same or a similar BER test set. The test set measures BER or errored-seconds
(ES), as needed. Similar error testing is done in the upstream direction both at the rates needed for the
particular system under test.
A working telephone line circuit is connected to the telephone jack on the splitter at the ATU-C end. At the
ATU-R end telephones may be added add places in the in-home wiring model.
Downstream
channel
error test
ATU-C
ATU-R
Splitter
Test loop
Hi-Z
In-home
wiring
model
Σ
Downstream
channel
error test
Hi-Z
CO line
circuit
attenuator
Simulated
Crosstalk
Noise
attenuator
Impulse
waveform
generator
Figure 5: Laboratory test set-up for measuring performance margins for the downstream channel
without a splitter at the ATU-R
12
Upstream
channel
error test
ATU-C
ATU-R
Splitter
Σ
In-home
wiring
model
Test loop
Hi-Z
Upstream
channel
error test
Hi-Z
CO line
circuit
attenuator
Simulated
Crosstalk
Noise
attenuator
Impulse
waveform
generator
Figure 6: Laboratory test set-up for measuring performance margins for the upstream channel without
a splitter at the ATU-R
5.2.2
Crosstalk Test
Crosstalk testing is performed to evaluate the performance of DSL systems in the presence of crosstalk from
other services in the loop.
5.2.2.1 Crosstalk Injection
Simulated crosstalk, XT (NEXT and/or FEXT) is introduced into the test loop as described in 5.1.2.1.
5.2.2.1.1
Crosstalk Injection for TCM-ISDN environment
For testing transmission performance of an ADSL system conforming to Annex C (ADSL operating in an
environment co-existing with TCM-ISDN DSL), the calibrated simulated TCM-ISDN DSL NEXT PSD and
TCM-ISDN DSL FEXT PSD defined in section 7 shall be injected into the test loop as described in section
5.1.2.1.1.
5.2.2.2 Crosstalk test method
Crosstalk testing shall be performed as described in 5.1.2.2.
5.2.3
Impulse Noise Test
Impulse noise testing is performed to evaluate the performance of DSL systems in the presence impulse noise.
13
5.2.3.1 Impulse noise injection
The same coupling circuit as is used in 5.1.2.1 is used for impulse noise injection. The amplitude level of the
impulses may be measured with an oscilloscope.
5.2.3.2 Impulse noise test method
Impulse noise testing shall be performed as described in 5.1.3.2.
5.2.4
POTS interference Test
POTS interference test is performed to evaluate the performance of DSL systems is the presence of POTS
signaling.
5.2.4.1 POTS testing without a splitter at the ATU-R
The interference due to POTS service on the same line is generated by use of a telephone model structure and
central office circuits connected in the normal way to the system under test. The telephone structure is
described in section 9 and should be connected to the appropriate points in the in-home wiring model (see
section 7.2). . The following POTS signaling and alerting activities shall be performed:
•
•
•
Call ATU-R phones and allow to ring 25 times;
Pick up ringing phone “A” (see Figure 12) and hold in off-hook position;
Return phone “A” to on-hook position;
Repeat above steps 25 times.
5.2.4.2 POTS testing without a splitter at the ATU-R
Before testing, the ADSL units are trained with the required crosstalk interference. Signaling disturbances are
created through use of the CO line connected to the splitter at the ATU-C, and the telephone model structure
connected to telephone jacks in the in-home wiring model (see section 7.2). During the signaling activities
described in section 5.2.4.2 the system is tested by measuring:
•
the time required to achieve errorless transmission after a signaling event has occurred (measure fast
retrain time)
•
the data rates in the on hook and off hook states
Note that in the first on-hook to off hook transition the retrain time will be much longer since a full
initialization is required in order to generate the off-hook profile that is matched to phone “A”. Subsequent onhook to off-hook transitions should use a fast retrain and the stored profile for phone “A”.
5.3 POTS QOS testing
5.3.1
Telephone Circuitry
Figure 7 shows a simplified block diagram of a typical telephone, with signal flow direction indicated by
arrows in the connections between blocks. Signal flow on the left side of the 2W/4W hybrid is generally
bidirectional, and nonlinear distortion generated in this area can result in reflected harmonic and
intermodulation components appearing across Tip and Ring. Signal flow on the right side of the hybrid is
generally unidirectional. Nonlinear distortion of DSL signals in the speaker driver can result in audible energy
at the speaker, but will generally not result in energy seen on the local loop. The microphone interface, of
course, isn’t exposed to DSL energy in the first place, so nonlinear distortion in that area is not an issue.
14
Within the bidirectional line interface, most components are on the side of the hook switch away from Tip and
Ring. This includes the vast majority of sources of nonlinear distortion, which are isolated from the loop when
the telephone is on-hook.
Tip
Ring
Bidirectional
line
interface
(including
hook
switch)
Unidirectional
speaker
driver
2W - 4W
Hybrid
Unidirectional
microphone
interface
Figure 7: Typical telephone signal flow
Measurement Criteria
The following criteria must be incorporated when evaluating interference between DSL and POTS channels for
a splitterless DSL technology:
•
The primary configuration for measurement of channel interference shall be with one or more telephone
devices in an off-hook condition. Most of the circuitry in a typical telephone (including, in nearly all cases,
any components which cause nonlinear distortion of a DSL signal) is located on the telephone side of the
hook switch, and is isolated from DSL signals when the phone is on-hook (see Figure 7). While
measurement of energy in the POTS band during on-hook conditions is an appropriate parameter for both
splitterless and conventional ADSL systems, it is not sufficient by itself for evaluation of splitterless
systems - in fact, it is considerably less meaningful than measurements taken with phones off-hook.
•
Interference measurements must include the sound pressure output at the telephone speaker. It is not
sufficient to measure the nonlinear and intermodulation components generated across Tip and Ring, as
some or all of the distortion can take place on the four-wire side of the telephone hybrid, causing audible
noise at the speaker even in the absence of audio band energy on the line. Direct measurement of electrical
power at the speaker terminals may not be sufficient due to varying speaker conversion characteristics. For
POTS modem and facsimile devices, other measurements related to performance degradation would of
course replace a sound pressure measurement.
•
Interference measurements should also include components generated across Tip and Ring. Excessive
harmonic content on the line may result in reduced performance or even failure to connect in the
downstream direction over longer loops.
•
There is a broad range of telephone devices which must be accommodated by any splitterless DSL
solution. Test results based on a narrow sampling of devices will not accurately reflect performance in the
field. Test results presented should include the types and numbers of devices used, to provide some
indication of how representative the database may be.
15
5.3.2
Test conditions
The configuration used for testing line coding technologies for interference with off-hook telephones is shown
in Figure 19. A control telephone (with microphone disconnected) is routed through a POTS network
simulator to provide switching and signaling functions for the POTS connection. The POTS network simulator
shall be capable of introducing GSTN-band electrical noise as is commonly available in simulators designed to
test dial modems per V.56bis. The other side of the POTS network simulator is routed through a central office
POTS splitter to isolate the DSL channel from the POTS network simulator. The DSL/POTS combined port of
the CO POTS splitter is routed through a wideband local loop simulator to the local telephone and the local
(ATU-R ) line code transmitter, which transmits an upstream signal using the line code under test. The central
office line code transmitter (ATU-C) is connected to the appropriate port on the CO POTS splitter. A
condenser microphone is acoustically coupled to the telephone handset by positioning it approximately 1 mm
away from and aligned with the center axis of the handset speaker. The space between the microphone and the
speaker is left unsealed. The microphone and handset are placed in an isolated environment to minimize
ambient noise. Sound levels from the handset speaker are measured using a sound level meter, using B
weighting per IEC 651 and slow response time. The zero dB sound power reference is 20 µPascals.
A signal analyzer is used to view the signal characteristics directly across Tip and Ring using a differential
high impedance probe . This analyzer is used to set and verify signal levels from the line code transmitter and
to verify test conditions.
For each test case, a POTS connection is established and the sound output of the telephone speaker is recorded
without a DSL signal present to establish a control level. Some Recommendations specify or allow the use of
an a filter in series with the local telephone (e.g. customer premise POTS splitter, in-line filter, etc). The
position of this optional filter is shown in the figure with a dashed box.
ATU-C
Remote
Telephone
GSTN
Central Office
Simulator
POTS
Splitter
ATU-R
Subscriber
Loop
Signal Level
Analyzer
Filter
Local
Telephone
Sound
Level
Meter
Signal Level
Analyzer
Figure 8: Test configuration
5.3.3
Summary of Test Procedure
1
ATU-C/ATU-R transmitters are off. GSTN simulator noise is off. Remote telephone is quiet. Loop
length is zero.
2
A GSTN connection is established between remote telephone and local telephone.
3
The acoustic energy of the local telephone speaker is measured* for reference: SP1. (this establishes the
acoustic noise floor as measured at the local telephone)
4
The GSTN noise is turned on at the acceptable GSTN service dBrn level of the GSTN service
requirement**.
5
The acoustic energy of the local telephone speaker is measured: SP2***. (SP2 is thus the acoustic
equivalent of the acceptable GSTN service noise on this particular telephone.)
16
6
GSTN simulator noise is turned off. ATU-C/ATU-R transmitters are turned on at their prescribed
transmit levels.
7
The acoustic energy of the local telephone speaker is measured: SP3. . The test result is the difference
between SP2 and SP3 in dB. If the test result is positive, i.e. SP3 < SP2, then the DSL equipment does not
introduce audio noise in excess of that allowed by the GSTN service for the local telephone used.
Test Variations:
A. The GSTN requirement may have two dBrn thresholds that require testing. For example, <20 dBrnC may
represent preferred service and >30 dBrnC may be unacceptable service.
6
Test loops
6.1 Local loop configurations
6.1.1 North American Test Loops
17
13500'
T1.601 Loop #7
ATU-C
ATU-R
26 AWG
1500
26 AWG
3000 ft
T1.601 Loop #9
ATU-C
1500
26 AWG
6000
26 AWG
1500
26 AWG
1500
ATU-R
26 AWG
26 AWG
1500'
26 AWG
T1.601 Loop #13
ATU-C
9000'
2000'
26 AWG
24 AWG
ATU-R
500'
24 AWG
400
26 AWG
CSA Loop #4
ATU-C
CSA Loop #6
ATU-C
800
26 AWG
550 ft
6250
26 AWG
26 AWG
800
ATU-R
26 AWG
9000
ATU-R
26 AWG
800'
24 AWG
10700'
CSA Loop #7
ATU-C
ATU-R
24 AWG
12000
CSA Loop #8
ATU-C
ATU-R
24 AWG
6000
mid-CSA Loop
ATU-C
ATU-R
26 AWG
< 10'
CSA Loop #0
ATU-C
ATU-R
26 AWG
18
T1.601 Loop #1
16,500’
1,500’
26 AWG
24 AWG
ATU-C
ATU-R
1,500’
24 AWG
3,000’
13,500’
T1.601 Loop #2
ATU-C
ATU-R
24 AWG
26 AWG
6,000’
9,000’
T1.601 Loop #5
ATU-C
ATU-R
24 AWG
26 AWG
Shortened
T1.601 Loop #7
1,500’
26 AWG
11000'
ATU-C
ATU-R
26 AWG
Figure 9: North American Test loop
NOTES:
1.
Lengths are in feet (‘)
2.
AWG = American Wire Gauge
Resistance and insertion loss
Table 1, Table 2 and Table 3 provide the calculated resistance and insertion loss between 100 ohm
terminations of the loops shown in
.
Table 1: Resistance and insertion loss values for test loops at 700 F
Loop #
Resistance
(ohms)
Frequency (kHz)
Insertion loss (dB)
20
40
T1.601 # 7 1127
29.8 36.7
T1.601 # 9 877
27.6 36.4
T1.601 # 13 909
26.6 34.1
CSA # 4
634
17.6 22.0
CSA # 6
751
20.0 24.4
CSA # 7
562
17.3 20.9
CSA # 8
630
19.2 22.8
Mid-CSA
501
13.3 16.2
100
45.2
52.5
47.9
29.6
30.1
26.8
27.7
20.0
200
52.8
47.5
48.3
39.6
35.2
33.6
34.4
23.4
260
57.3
55.7
55.7
40.1
38.2
37.8
38.3
25.4
300
60.2
62.0
61.3
42.5
40.2
38.6
40.8
26.8
400
67.7
60.3
62.2
49.2
45.1
43.1
46.9
30.1
500
74.8
71.5
71.4
50.2
49.9
49.9
52.4
33.2
600
81.7
72.2
74.1
53.8
54.4
57.9
57.4
36.3
780
93.0
82.7
85.3
55.7
62.0
60.2
65.4
41.3
1100
110
96.2
100
70.7
73.6
72.7
77.8
49.1
780
1100
Table 2: Resistance and insertion loss in dB for test loops at 900 F
Loop #
Resistance
(ohms)
Frequency (kHz)
Insertion loss (dB)
20
40
100
200
260
300
400
500
600
19
T1.601 # 7
T1.601 # 9
T1.601 # 13
CSA # 4
CSA # 6
CSA # 7
CSA # 8
Mid-CSA
20
1176
915
948
658
784
586
657
523
30.6
28.4
27.4
18.0
20.5
17.9
19.8
13.8
37.9
37.5
35.2
22.6
25.2
21.6
23.6
16.7
46.9
53.4
49.0
30.4
31.2
27.7
28.7
20.7
54.6
49.1
49.9
40.3
36.4
34.0
35.4
24.2
59.1
57.2
57.2
41.0
39.4
38.7
39.3
26.2
62.1
63.1
62.5
43.5
41.4
39.5
41.8
27.6
69.6
61.9
63.7
50.0
46.4
44.1
47.9
30.9
76.6
72.8
72.8
50.9
51.1
50.9
53.5
34.0
83.4
73.6
75.6
54.3
55.6
58.8
58.6
37.1
95.0
84.2
87.0
56.6
63.3
61.4
66.8
42.2
113
98.1
102
71.6
75.2
74.0
79.4
50.1
Table 3: Resistance and insertion loss in dB for test loops at 1200 F
Loop #
Resistance
(ohms)
Frequency (kHz)
Insertion loss (dB)
20
40
T1.601 # 7 1250
31.9 39.6
T1.601 # 9 972
29.5 39.1
T1.601 # 13 1008
28.5 36.8
CSA # 4
704
18.9 23.8
CSA # 6
833
21.4 26.3
CSA # 7
623
18.7 22.6
CSA #8
699
20.7 24.8
Mid-CSA
555
14.4 17.5
100
49.4
54.7
50.7
32.2
32.8
29.1
30.2
21.8
200
57.4
51.5
52.3
41.9
38.2
36.2
36.9
25.5
260
61.8
59.5
59.5
42.8
41.2
40.0
40.8
27.5
300
64.8
65.0
64.5
45.2
43.2
43.2
43.3
28.8
400
72.3
64.1
66.0
51.5
48.2
45.5
49.4
32.1
500
79.3
74.4
74.9
52.8
52.9
52.5
55.1
35.1
600
86.1
75.7
77.9
56.0
57.4
60.2
60.4
38.3
780
97.9
86.4
89.4
58.7
65.3
63.2
68.8
43.5
1100
116
101
105
74.1
77.5
76.0
81.7
51.6
Primary constants
The primary constants R, L, C, and G of both polyethylene insulated cable (PIC) and pulp insulated cable, at 0°
F, 70° F, and 120° F, are specified in Annex G of T1.601-1992.
The variation of R and L with frequency can be accurately modeled as follows
R = (roc^4 +ac.f^2)^0.25,
L = (L0 + L8 .xb)/(1 + xb),
xb = (f/ /fm )^b
and
where
The six coefficients for 24 and 26 AWG PIC cable (per kft at 70° F) are given in Table 4.
Table 4: Coefficients for calculation of R and L (24 and 26 AWG PIC at 70° F)
24 AWG
26 AWG
roc
0.0537
.0836
ac
.000386
.001034
L0
.1873
.1867
L8
.1292
.1343
Fm
.6973
.8696
b
.8188
.8472
For PIC cable the C values are constant with frequency (15.72 nF/kft), and the G values are negligibly small.
6.1.2
European Local Test Loops
The variation of the primary line constants (R, L and C) with frequency for the different reference cable types
are given in Table 6 and Table 7. Note that G is assumed zero. The RLC values are quoted per km at a
temperature of 20°C and are measured values that have been smoothed.
The variation of R and L with frequency can be accurately modeled as follows
and
where
R = (roc^4 +ac.f^2)^0.25,
L = (L0 + L8 .xb)/(1 + xb),
xb = (f/ /fm )^b
The six coefficients for the cable types referenced in
Figure 10 are given in Table 8.
Note also that there are adjustable sections (marked 'X') in
21
Figure 10. The nominal lengths of these sections are shown Table 5. The lengths of the sections are based on
the reference RLC values for each cable type shown in tables Table 6 and Table 7. For repeatability of
measurement results, however, the lengths of these section length shall be adjusted to give the overall insertion
loss given in tables H.5 to H.8. Insertion loss is measured at 300 kHz with 100 Ω (balanced resistive) source
and termination impedances
ATU-R (note 1)
ATU-C (note 1)
0 km; null loop
loop # 0
loop # 1
X km
0.4 mm
X km
loop # 2
0.5 mm
1.5 km
loop # 3
X km
•
0.4 mm
0.5 mm
loop # 4
0.5 km
•
0.5 km
•
0.9 mm
loop # 6
•
4.0 km
•
0m
•
.75 km
0.32 mm
0.4 mm
0.5 km BT
0.4 mm
X km
•
0.4 mm
0.5 mm
1.25 km
X km
•
0.4 mm
X km
•
0.4 mm
1.1 km
0.2 km
•
0.5 mm
0.9 mm
loop # 8
•
0.63 mm
0.5 km
X km
0.4 mm
0.5 km
0.63 mm
loop # 7
•
0.5 mm
0.63 mm
loop # 5
1.5 km
0.2 km
0.32 mm
X km
•
0.4 mm
0.5 km BT
0.4 mm
Notes: 1. These test loops are shown with the ATU-Rs on the left; this is the European
convention, which is in contrast to fig. 38, where the ATU-Rs are on the right.
2. All cable is Polyethylene insulated;
3. 1 km = 3.28 kft;
4. BT = Bridged tap (i.e., section of unterminated cable)
Figure 10: ETSI Test loops
22
Loop #
1
2
3
4
5
6
7
8
Loop insertion loss
36dB @ 300kHz
Nominal value of
adjustable length 'X'
(km)
2,55
3,40
1,40
0,90
1,45
1,30
0,60
0,75
Loop insertion loss
36dB @ 300kHz
Nominal value of
adjustable length 'X'
(km)
2,55
3,40
1,40
0,90
1,45
1,30
0,60
0,75
Loop insertion loss
51dB @ 300kHz
Nominal value of
adjustable length 'X'
(km)
3,60
4,80
2,45
1,90
2,50
2,35
1,60
1,80
Loop insertion loss
61dB @ 300kHz
Nominal value of
adjustable length 'X'
(km)
4,30
5,70
3,15
2,65
3,20
3,05
2,20
2,50
Table 5: Examples of Loop set insertion loss, nominal lengths for European testing
Table 6: RLC values for 0.32, 0.4, and 0.5 mm PE cables
Freq
kHz
0.
2.5
10.
20.
30.
40.
50.
100.
150.
200.
250.
300.
350.
400.
450.
500.
550.
600.
650.
700.
750.
800.
850.
900.
950.
1000.
1050.
1100.
0.32 mm
C=40 nf/km
R
ohm/km
409.000
409.009
409.140
409.557
410.251
411.216
412.447
422.302
437.337
456.086
477.229
499.757
522.967
546.395
569.748
592.843
615.576
637.885
659.743
681.138
702.072
722.556
742.601
762.224
781.442
800.272
818.731
836.837
L
mH/km
607.639
607.639
607.639
607.639
607.639
607.639
607.639
607.631
607.570
607.327
606.639
605.074
602.046
596.934
589.337
579.376
567.822
555.867
544.657
534.942
526.991
520.732
515.919
512.264
509.503
507.415
505.831
504.623
0.4 mm
C=50 nF/km
R
ohm/km
280.000
280.007
280.110
280.440
280.988
281.748
282.718
290.433
302.070
316.393
332.348
349.167
366.345
383.562
400.626
417.427
433.904
450.027
465.785
481.180
496.218
510.912
525.274
539.320
553.064
566.521
579.705
592.628
L
mH/km
587.132
587.075
586.738
586.099
585.322
584.443
583.483
577.878
571.525
564.889
558.233
551.714
545.431
539.437
533.759
528.409
523.385
518.677
514.272
510.153
506.304
502.707
499.343
496.197
493.252
490.494
487.908
485.481
0.5 mm
C=50 nF/km
R
ohm/km
179.000
179.015
179.244
179.970
181.161
182.790
184.822
199.608
218.721
239.132
259.461
279.173
298.103
316.230
333.591
350.243
366.246
381.657
396.528
410.907
424.835
438.348
451.480
464.258
476.71
488.857
500.720
512.317
L
mH/km
673.574
673.466
672.923
671.980
670.896
669.716
668.468
661.677
654.622
647.735
641.208
635.119
629.489
624.309
619.557
615.202
611.211
607.552
604.192
601.104
598.261
595.639
593.217
590.975
588.896
586.966
585.169
583.495
23
NOTE: G = 0 at all frequencies.
Table 7: RLC values for 0.63 and 0.9 mm PE cables
Freq
kHz
0.
2.5
10.
20.
30.
40.
50.
100.
150.
200.
250.
300.
350.
400.
450.
500.
550.
600.
650.
700.
750.
800.
850.
900.
950.
1000.
1050.
1100.
0.63 mm
C = 45 nF/km
R
ohm/km
113.000
113.028
113.442
114.737
116.803
119.523
122.768
143.115
164.938
185.689
204.996
222.961
239.764
255.575
270.533
284.753
298.330
311.339
323.844
335.897
347.542
358.819
369.758
380.388
390.734
400.816
410.654
420.264
L
mH/km
699.258
697.943
693.361
687.008
680.714
674.593
668.690
642.718
622.050
605.496
592.048
580.960
571.691
563.845
557.129
551.323
546.260
541.809
537.868
534.358
531.212
528.378
525.813
523.480
521.352
519.402
517.609
515.956
0.9 mm
C = 40 nF/km
R
ohm/km
55.000
55.088
56.361
59.941
64.777
70.127
75.586
100.769
121.866
140.075
156.273
170.987
184.556
197.208
209.104
220.365
231.081
241.326
251.155
260.615
269.745
278.577
287.138
295.452
303.538
311.416
319.099
326.602
L
mH/km
750.796
745.504
731.961
716.775
703.875
692.707
682.914
647.496
625.140
609.652
598.256
589.504
582.563
576.919
572.237
568.287
564.910
561.988
559.435
557.183
555.183
553.394
551.784
550.327
549.002
547.793
546.683
545.663
NOTE – G = 0 at all frequencies.
Table 8: Coefficients for calculation of R and L
0.32 mm
0.4 mm
0.5 mm
0.63 mm
0.9 mm
24
roc
.4090
.2800
.1792
.113
.0551
ac
.3822
.0969
.0561
.0257
.0094
L0
.6075
.5873
.6746
.6994
.7509
L8
.5
.4260
.5327
.4772
.5205
fm
.6090
.7459
.6647
.2658
.1238
b
5.269
1.385
1.195
1.0956
.9604
6.1.3
Test loops in an environment co-existing with TCM-ISDN DSL
To test the performance of the G.dmt/G.lite system, the test loops specified in Figure 11 shall be used. Note
that there are adjustable sections (marked “X”) in the figure. The nominal lengths of these sections
corresponding to each loop coverage are shown in Table 9. The lengths of the sections are based on the
reference RLC values for each cable type shown below. The calculated overall loop insertion loss at 300 kHz
and dc loop resistance are also provided in Table 9.
Customer
Customer
Loop TCM #1 Premises
Central
Office
0 km
null loop
Loop TCM # 0 Premises
5.0 km
0.9 mm
polyethylene
insulated
X km
0.65 mm
paper insulated
Central
Office
0.1 or 0.3 or 0.5 km BT
0.65 mm polyethylene insulated
Customer
Loop TCM #2 Premises
1.5 km
0.65 mm
polyethylene insulated
Central
Office
X km
0.4 mm
paper insulated
all combinations of three BT lengths (ten configurations including non-BT case)
{0, 0} {0.1, 0.1} {0.3, 0.3} {0.5, 0.5}{0.1, 0.3} {0.1, 0.5} {0.3, 0.5}{0.3, 0.1} {0.5, 0.1} {0.5, 0.3}
Customer
Loop TCM #3 Premises
Central
1.0 km
Office
0.32 mm
polyethylene insulated
Customer
X km
0.65 mm
paper insulated
Central
Office
Customer
X km
0.4 mm
paper insulated
Central
Office
Loop TCM # 4 Premises
Loop TCM # 5 Premises
i
X km
0.4 mm
polyethylene insulated
: Test loops in an environment co-existing with TCM-ISDN DSL
Figure 11: Test loop in an enviroment co-existing with TCM-ISDN DSL
25
Table 9: Loop-set insertion loss and noimnal lengths
Table 1.tcm: Loop-set insertion loss and nominal lengths
Loop insertion loss Loop
Nominal value Loop insertion loss d.c. loop
(160 kHz)
TCM # of "X"
(300 kHz)
resistance
26.0 dB
1
0.42 km
36.1 dB
314 ohms
(60% coverage)
2
1.32 km
34.6 dB
518 ohms
3
0.78 km
31.9 dB
642 ohms
4
3.62 km
37.0 dB
376 ohms
5
2.07 km
34.0 dB
567 ohms
37.0 dB
1
1.95 km
51.8 dB
474 ohms
(90% coverage)
2
2.19 km
49.0 dB
758 ohms
3
1.75 km
45.6 dB
910 ohms
4
5.15 km
52.7 dB
536 ohms
5
2.94 km
48.3 dB
807 ohms
50.0 dB
1
3.76 km
70.3 dB
662 ohms
(99% coverage)
2
3.23 km
66.0 dB
1041 ohms
3
2.91 km
61.7 dB
1227 ohms
4
6.97 km
71.2 dB
724 ohms
5
3.97 km
65.3 dB
1090 ohms
65.0 dB
1
5.85 km
91.7 dB
879 ohms
(99.9% coverage)
2
4.42 km
85.6 dB
1368 ohms
(for G.lite only)
3
---------------------------4
9.06 km
92.6 dB
941 ohms
5
5.16 km
84.9 dB
1417 ohms
Note-1: The values in the table for Loop TCM # 2 are given by considering only the
section terminated by Customer Premises and Central Office, on condition of detaching
the bridge taps(BTs); The ef fect on theoverall insertion loss characteristics caused by
the BTs isnot taken into account.
Note-2: Test loopsof the 60-99% coverage are f or the use of G.dmt performance
testing. Test loops of the 60-99.9% coverage are for the use of G.lite performance
testing. The performance objective f or G.dmt/G.lite testing is10-7 BER with 6 dB
(G.dmt) and 4 dB (G.lite) margins with -140 dBm/Hz background noise and with X T
noises defined in section 6.3.1. It isnot necessarily required to comply with
NEX T/FEXT PSL values (NPSL n/FPSLn) def ined in section 8.tcm f or test loopsof the
99.9% coverage. The resultant net data rates may be given asa f unction of
NPSLn/FPSLn values for thiscase.
Note-3: TCM #3 f or 99.9% coverage isremoved from the list becausethe d.c. loop
resistance isgreater than 1500 ohms.
Primary line constants
The primary line constants are R, L, C, and G. The variation of R, L , and G with frequency can be accurately
calculated by the equations shown below. The coefficients for the calculation of R, L, and G are given in Table
10. The equations below give the value of R in ohm/m, L in H/m, G in mho/m, C in F/m, and f in Hz by using
the coefficient values shown in the table. Note that the capacitance C is assumed constant with frequency (C =
50 pF/m).
The primary line constants (R, L, G, and C) calculated by the equations below for the cable types referenced in
Figure 11 are shown in Table 12, Table 13, Table 14, Table 15, Table 16, as a function of frequency, where the
values are indicated R in ohm/km, L in µH/km, G in µmho/km, C in nF/km, f in kHz, and at a temperature of
20ºC.
26
R = 2( Ri + Rn + Rns )
L = 2( La + Li + Ln + Lns)
[ohm / m]
[ H / m]
G = ω C tan δ
[mho / m]
C = 50 × 10 − 12
[ F / m]
1
(R)
Ri = 2 f i
πr σ
1+
J0(
1
1+ j
δ
= 2 Re[
r
1+
2δ
πr σ
J1 (
δ
j
j
r)
]
r)
1+ j
J1 (
r)
1
1
1
+
j
( R)
δ
=
Re[
−
]
Rn =
f
r
n
1+ j
δ
πσd 2
πσd 2
J0(
r)
δ
1+ j
J1 (
r)
1
4
1+ j
( R)
δ
]
4 fn =
Re[−
Rns =
r
1+ j
δ
πσd 2
πσd 2
J0(
r)
δ
d
µ
La = 0 ln( )
r
2π
1+ j
J0 (
r)
µ r µ0 ( L) µ
δ
δ
Li =
fi =
Re[ −
]
2π
2π
(1 + j )r J (1 + j r )
1
δ
1+ j
J2(
r)
µ0 r 2 ( L)
µ0 r 2
δ
Ln = −
( ) fn = −
( ) Re[−
]
1+ j
2π d
2π d
J0 (
r)
δ
1+ j
J2(
r)
µ0 r 2
µ0 r 2
( L)
δ
Lns = −
( ) 4 fn = −
( ) 4 Re[−
]
1+ j
2π d
2π d
J0 (
r)
δ
where
ω = 2 πf ; angular frequency
r ; the radius of a conductor
σ ; conductivi ty of copper (conductor)
δ =
2
; skin depth
ω σ µ r µ0
d = 2 2 (r + co) ; the distance between the conductor centers of a pair
co ; the insulator thickness of a wire
µ r ; relative permeabili ty of copper (conductor)
µ 0 ; permeabili ty of vacuum
27
Table 10: Coefficients for calculation of R and L
Coefficient
Paper
0.4mm
0.5mm
-3
r(m)
0.2 x 10
co(m)
0.09 x 10-3
polyethylen
0.65mm
0.9mm
-3
0.325 x 10
0.11 x 10-3
0.17 x 10-3
0.25 x 10
-3
0.32mm
0.45 x 10
-3
0.16 x 10
0.24 x 10-3
µr(H/m)
0.4mm
-3
0.2 x 10
0.05 x 10-3
0.5mm
-3
0.9mm
0.325 x 10
0.15 x 10-3
0.20 x 10-3
0.25 x 10
0.13 x 10-3
0.65mm
-3
-3
0.45 x 10-3
0.27 x 10-3
1
µ0(H/m)
4π x 10
σ(mho/m)
-7
5.8 x 107
tan δ
2.5 x 10
-2
4.0 x 10-4
5.0 x 10-4
The variant of R and L with frequency can be accurately modeled as follows. The five coefficients for the cable
types referenced in Figure 11 are given in Table 11. The approximate equations below give the values of R in
ohm/m, L in H/m, G in mho/m, C in F/m, and f in Hz by using the coefficient values shown in the table.
R = (roc + ac × f )
4
L = xa + xb × f
G = ωC tan δ
2
1
2
1
4
[ohm / m]
+ xc × f
C = 50 × 10 −12
1
3
[ H / m]
[mho / m]
[ F / m]
Table 11: Coefficients calculation of R and L (at 20C)
T a b le 3 .t c m : C o e ff ici e n t s c a lc u la t i o n o f R a n d L (a t 2 0 º C )
C ab l e ty p e
P aper
ac
xa
0 .4
mm
2 .6 8 8 x 1 0 - 1
ro c
2 .2 6 7 x 1 0 -1 3
6 .8 3 4 x 1 0 -7
-2 . 0 9 4 x 1 0 -1 0
7 .2 0 5 x 1 0 -1 0
0 .5
mm
1 .7 2 4 x 1 0 - 1
9 .3 7 4 x 1 0 -1 4
7 .3 5 1 x 1 0 -7
1 . 9 3 0 x 1 0 -1 1
-2 . 3 3 0 x 1 0 -9
0 .6 5 m m
1 .0 4 1 x 1 0 - 1
2 .7 8 7 x 1 0 -1 4
8 .0 0 6 x 1 0 -7
xc
2 .6 9 6 x 1 0 -1 0
- 5 .3 4 0 x 1 0 - 9
mm
5 .5 8 9 x 1 0 -2
7 .1 8 0 x 1 0 -1 5
8 .3 0 4 x 1 0 - 7
5 .1 1 1 x 1 0 -1 0
- 8 .1 6 1 x 1 0 - 9
0 .3 2 m m
4 .1 7 5 x 1 0 -1
6 .9 9 8 x 1 0 -1 3
6 .0 0 3 x 1 0 - 7
- 3 .9 1 9 x 1 0 -1 0
3 .0 8 1 x 1 0 - 9
0 .4
2 .7 1 4 x 1 0 -1
1 .7 0 5 x 1 0 -1 3
7 .2 5 7 x 1 0 - 7
- 2 .0 5 9 x 1 0 -1 0
9 .6 7 8 x 1 0 - 1 0
0 .9
mm
mm
1 .7 4 2 x 1 0 -1
7 .3 4 6 x 1 0 -1 4
7 .6 1 8 x 1 0 - 7
- 1 .5 4 7 x 1 0 -1 1
- 1 .6 5 6 x 1 0 - 9
0 .6 5 m m
1 .0 4 8 x 1 0 -1
2 .4 3 6 x 1 0 -1 4
8 .1 3 9 x 1 0 - 7
2 .3 5 4 x 1 0 -1 0
- 4 .8 0 1 x 1 0 - 9
0 .9
5 .6 3 0 x 1 0 -2
6 .4 8 6 x 1 0 -1 5
8 .4 0 7 x 1 0 - 7
4 .8 1 6 x 1 0 -1 0
- 7 .7 2 1 x 1 0 - 9
P o ly e t h y le n e 0 . 5
28
xb
mm
Table 12: RLC values for 0.4 and 0.5 mm paper insulated cables
Table 4.tc m: RLC values for 0.4 and 0.5 mm paper insulate d c abl es
Frequenc y (kHz )
0.4mm
C=50nF/km
R(ohm/ km)
L(µ H /km )
0.5mm
C=50nF/km
R(ohm/km)
L(µ H /km )
0.00
274.41
664.51
175.62
661.75
2.50
274.42
664.51
175.64
661.73
10.00
274.62
664.42
175.97
661.52
20.00
275.28
664.15
176.99
660.85
30.00
276.36
663.69
178.68
659.74
40.00
277.87
663.07
180.99
658.24
50.00
279.78
662.27
183.88
656.37
100.00
294.75
656.09
204.86
643.13
150.00
316.56
647.33
231.49
627.35
200.00
342.22
637.42
258.98
612.46
250.00
369.40
627.46
285.41
599.57
300.00
396.69
618.03
310.31
588.68
350.00
423.38
609.41
333.73
579.47
400.00
449.19
601.65
355.79
571.61
450.00
474.02
594.68
376.61
564.83
500.00
497.90
588.42
396.34
558.92
550.00
520.87
582.79
415.08
553.73
600.00
542.99
577.69
432.93
549.13
650.00
564.30
573.06
449.98
545.03
700.00
584.86
568.84
466.34
541.34
750.00
604.72
564.98
482.06
538.02
800.00
623.92
561.42
497.22
535.00
850.00
642.50
558.15
511.88
532.24
900.00
660.52
555.11
526.09
529.71
950.00
678.00
552.30
539.88
527.38
1000.00
694.99
549.69
553.30
525.23
1050.00
711.52
547.25
566.38
523.23
1100.00
727.62
544.96
579.14
521.37
29
Table 13: RLC values for 0.65 and 0.9 mm paper insulated cables
Table 5.tcm: RLC values for 0.65 and 0.9mm papaer insulated cables
Frequency (kHz)
30
0.65mm
C=50nF/km
R(ohm/km)
L(µH/km)
0.9mm
C=50nF/km
R(ohm/km)
L(µH/km)
0.00
103.92
684.18
54.20
686.87
2.50
103.95
684.15
54.27
686.73
10.00
104.45
683.60
55.20
684.79
20.00
106.02
681.89
57.98
679.07
30.00
108.52
679.19
62.04
670.97
40.00
111.83
675.66
66.81
661.83
50.00
115.78
671.51
71.88
652.64
100.00
140.25
647.61
96.11
616.45
150.00
165.42
626.69
116.50
594.37
200.00
188.55
610.67
133.77
579.83
250.00
209.58
598.33
148.81
569.58
300.00
228.77
588.57
162.27
561.93
350.00
246.38
580.67
174.61
555.98
400.00
262.68
574.16
186.08
551.17
450.00
277.89
568.69
196.86
547.19
500.00
292.19
564.04
207.06
543.81
550.00
305.73
560.02
216.77
540.90
600.00
318.64
556.51
226.06
538.36
650.00
331.01
553.41
234.97
536.11
700.00
342.90
550.65
243.55
534.10
750.00
354.37
548.17
251.83
532.30
800.00
365.47
545.92
259.84
530.67
850.00
376.23
543.88
267.60
529.18
900.00
386.68
542.00
275.14
527.82
950.00
396.85
540.27
282.47
526.57
1000.00
406.77
538.67
289.61
525.41
1050.00
416.44
537.19
296.58
524.33
1100.00
425.88
535.81
303.38
523.33
Table 14: RLC values for 0.32, 04 and 0.5 mm polyethelene insulated cables
Table 6.tcm: RLC values for 0.32, 0.4mm and 0.5mm polyethylene insulated cables
0.32mm
C=50nF/km
L(µH/km)
0.4mm
R(ohm/km)
C=50nF/km
L(µH/km)
0.5mm
R(ohm/km)
C=50nF/km
Freq (kHz)
R(ohm/km)
L(µH/km)
0.00
428.76
624.66
274.41
716.20
175.62
703.89
2.50
428.77
624.66
274.42
716.19
175.64
703.88
10.00
428.92
624.62
274.59
716.13
175.91
703.70
20.00
429.41
624.48
275.12
715.91
176.79
703.14
30.00
430.23
624.26
276.02
715.55
178.23
702.23
40.00
431.37
623.94
277.25
715.05
180.20
700.98
50.00
432.83
623.54
278.83
714.42
182.66
699.43
100.00
444.67
620.31
291.19
709.50
200.68
688.42
150.00
463.20
615.30
309.35
702.50
223.85
675.19
200.00
486.98
608.97
330.92
694.53
248.16
662.55
250.00
514.44
601.81
354.05
686.44
271.91
651.46
300.00
544.17
594.24
377.58
678.73
294.59
641.96
350.00
575.05
586.61
400.91
671.59
316.14
633.82
400.00
606.26
579.16
423.72
665.08
336.56
626.79
450.00
637.24
572.03
445.91
659.18
355.93
620.66
500.00
667.67
565.31
467.42
653.82
374.32
615.28
550.00
697.37
559.03
488.25
648.94
391.79
610.53
600.00
726.25
553.18
508.41
644.49
408.45
606.30
650.00
754.29
547.75
527.90
640.42
424.36
602.52
700.00
781.52
542.72
546.76
636.68
439.60
599.12
750.00
807.95
538.05
565.01
633.23
454.24
596.04
800.00
833.64
533.70
582.67
630.05
468.35
593.25
850.00
858.61
529.65
599.77
627.10
481.98
590.69
900.00
882.92
525.88
616.36
624.36
495.18
588.35
950.00
906.60
522.34
632.44
621.82
508.00
586.20
1000.00
929.68
519.03
648.07
619.45
520.46
584.20
1050.00
952.21
515.92
663.27
617.23
532.60
582.35
1100.00
974.20
513.00
678.07
615.15
544.44
580.62
31
Table 15: RLC values for 065 mm and 0.9 mm polyethelyne cables
Table 7.tcm: RLC values for 0.65mm and 0.9mm polyethylene insulated cables
Frequency (kHz)
32
0.65mm
C=50nF/km
0.9mm
C=50nF/km
R(ohm/km)
L(µH/km)
R(ohm/km)
L(µH/km)
0.00
103.92
707.72
54.20
703.89
2.50
103.95
707.69
54.26
703.77
10.00
104.41
707.19
55.14
701.96
20.00
105.84
705.66
57.75
696.66
30.00
108.14
703.22
61.57
689.13
40.00
111.17
700.03
66.08
680.63
50.00
114.81
696.28
70.89
672.04
100.00
137.50
674.59
94.16
637.90
150.00
161.20
655.37
113.97
616.72
200.00
183.24
640.46
130.78
602.67
250.00
203.43
628.83
145.41
592.72
300.00
221.91
619.57
158.49
585.30
350.00
238.88
612.03
170.48
579.52
400.00
254.58
605.79
181.62
574.85
450.00
269.22
600.55
192.09
570.98
500.00
282.98
596.08
202.00
567.70
550.00
296.00
592.23
211.44
564.88
600.00
308.41
588.86
220.46
562.41
650.00
320.29
585.88
229.12
560.22
700.00
331.71
583.23
237.45
558.28
750.00
342.73
580.85
245.50
556.53
800.00
353.40
578.69
253.28
554.94
850.00
363.73
576.73
260.82
553.50
900.00
373.78
574.93
268.14
552.18
950.00
383.55
573.27
275.27
550.96
1000.00
393.07
571.73
282.20
549.83
1050.00
402.36
570.31
288.97
548.79
1100.00
411.44
568.98
295.58
547.81
Table 16:G values
Table 8.tcm: G values
Frequency (kHz)
0.4, 0.5, 0.65, 0.9 mm 0.32mm
0.4, 0.5, 0.65, 0.9 mm
polyethylene
(µmho/km)
polyethylene (µmho/km)
paper (µmho/km)
0.00
0.00
0.00
0.00
2.50
19.64
0.31
0.39
10.00
78.54
1.26
1.57
20.00
157.08
2.51
3.14
30.00
235.62
3.77
4.71
40.00
314.16
5.03
6.28
50.00
392.70
6.28
7.85
100.00
785.40
12.57
15.71
150.00
1178.10
18.85
23.56
200.00
1570.80
25.13
31.42
250.00
1963.50
31.42
39.27
300.00
2356.19
37.70
47.12
350.00
2748.89
43.98
54.98
400.00
3141.59
50.27
62.83
450.00
3534.29
56.55
70.69
500.00
3926.99
62.83
78.54
550.00
4319.69
69.12
86.39
600.00
4712.39
75.40
94.25
650.00
5105.09
81.68
102.10
700.00
5497.79
87.96
109.96
750.00
5890.49
94.25
117.81
800.00
6283.19
100.53
125.66
850.00
6675.88
106.81
133.52
900.00
7068.58
113.10
141.37
950.00
7461.28
119.38
149.23
1000.00
7853.98
125.66
157.08
1050.00
8246.68
131.95
164.93
1100.00
8639.38
138.23
172.79
33
6.2 In-home wiring models
Phone B
10 feet
100 feet
NID
10 feet
ATU-R
60 feet
100 feet
Phone C
Phone A
Figure 12: Customer premise in-home wiring model #1
Note that phone “A” will be used for off-hook testing.
34
Area
Coverage
House (Detached and Apartment houses)
(Note-1 and -2)
60%
5m
Office building
(N ote-1 and -2)
89 m
Phone A
Phone A
ATU -R
Typically 0.5 mm untwisted single-pair wire
(N ote-3)
Typically 0.4 mm twisted pair wire
(non quad configuration) (Note-3)
Connecting cord (Typically 3 m) (Note-4)
Connecting cord (Typically 3 m) (Note-4)
NID
90%
ATU-R
NID
RJ-11
2m
4m
Phone A
RJ-11
110 m
70 m
70 m
3m
ATU-R
ATU-R
Phone B
Phone B
5m
Phone A
ATU-R
Phone B
2m
180 m
Phone A
180 m
4m
99%
Phone A
5m
3m
7m
Phone B
Phone A
216 m
84 m
3m
Phone A
84 m
ATU-R
Phone B
ATU-R
ATU-R
ATU-R
Phone B
Phone B
10 m
Phone A
300 m
5m
Phone B
ATU-R
5m
Phone A
300 m
Phone B
ATU-R
Phone B
Note-1: In-home wiring modes in the figure is to be substituted for the box labeled “customer premises” in the
test loops (Figure 2.tcm) for G.lite system performance test, where system performance should be tested with one
phone (Phone A) off-hook. It is not required for G .dmt system performance test to comply with in-home wiring
models defined in the figure.
Note-2: A linear phone model is provided for testing purpose of G.lite system performance, w here only steadystate impedances are considered and transients are not considered. An off-hook impedance model for Phone A is
shown in Figures 3.1.tcm, 3.2.tcm, and 3.3.tcm. A n on-hook impedance model for Phone A and Phone B is shown in
Figures 3.4.tcm and 3.5.tcm. Notice that the phone impedance models are moderate ones; are not typical or
representative or average or worst. They are specified to provide a unified test condition so as to validate test results
comparison. Test results do not guarantee G.lite system performance for any cases of splitterless operation.
Note-3: The equations to calculate primary line constants R, L, C, and G with frequency for in-home wiring
cables are shown in section 7.1.3. The coefficients for the calculation are given in Table 8.1.tcm. Notice that the inhome wiring cables defined in Figure 3.tcm do not adopt a quad structure despite that the cables defined in the Figure
2.tcm adopt a quad structure, so the distance between the conductor centers of a pair “d” used in the equations is to
be “d = 2 (r+co)”; Sqrt[2] is not multiplied as in the case of a quad-structured cable.
Note-4: The insertion loss characteristics of a 3 m connecting cord is shown in Figure 3.6.tcm.
Figure 3.tcm: In-home wiring models in an environment co-existing with TCM-ISDN DSL
35
7 Power spectral density of crosstalk disturbers
Crosstalk margin measurements were made for four types of disturbers, DSLs, HDSLs, T1s and ADSL lines.
DSL, HDSL, and ADSL crosstalk is from pairs in the same binder group; T1 crosstalk is from pairs in an
adjacent binder group
7.1 Simulated DSL PSD and induced NEXT
The power spectral density (PSD) of Basic Access DSL disturbers is expressed as:
PSD DSL-Disturber = K DSL ×
2
×
fo
  πf   2
sin   
  fo 
 πf 
 
 fo 
2
×
1
 f 
1+ 

 f3dB 
4 ,
f 3dB = 80 kHz, 0 ≤ f < ∞
2
where f o = 80 kHz, KDSL =
5 Vp
×
, V p = 2. 50 Volts and R = 135 Ohms
9 R
This equation gives the single-sided PSD; thjat is, the integral of PSD, with respect to f, from 0 to infinity,
gives the power in Watts. PSDDSL-Disturber is the PSD of an 80 kbaud 2B1Q signal with random
equiprobable levels, with full-baud square-topped pulses and with 2nd order Butterworth filtering ( f3dB = 80
kHz).
The PSD of the DSL NEXT can be expressed as:
3

PSDDSL− NEXT = PSDDSL− Disturber x n f 2 




0 ≤ f < ∞, n = 1, 10, 24, 49
1
14
1.134 × 10
1
x4 =
14
4.950 × 10
1
x10 =
13
2.849 × 10
1
x24 =
13
1.677 × 10
where: x1 =
The integration of PSD DSL-Disturber and PSD DSL-NEXT over various frequency ranges of interest are
presented in Table 17.
36
Table 17: DSL transmit and DSL-induced NEXT power
Frequency Range
Transmit Power
dBm
0–0.16 MHz
0–0.32 MHz
0–1.544 MHz
0–3 MHz
0–10 MHz
13.60
13.60
13.60
13.60
13.60
NEXT Power
dBm
24 disturber
–52.62
–52.62
–52.62
–52.62
–52.62
-90
-100
-110
-120
-130
-140
-150
0
100000
200000
300000
400000
500000
600000
Frequency (Hz)
Figure 13: 24-disturber DSL NEXT
7.2 Simulated HDSL PSD and induced NEXT
The PSD of HDSL disturbers is expressed as:
2
PSD HDSL- Disturber = KHDSL ×
×
fo
  πf  
sin 

  fo 
 πf 


 fo 
2
2
×
1
 f 
1+ 

 f 3dB 
8
,
f 3dB = 196 kHz, 0 ≤ f < ∞
2
where f o = 392kHz, K HDSL =
5 Vp
, V p = 2. 70 Volts, and R = 135 Ohms
×
9
R
This equation gives the single-sided PSD; that is, the integral of PSD, with respect to f, from 0 to infinity,
gives the power in Watts. PSDHDSL-Disturber is the PSD of a 392 kbaud 2B1Q signal with random
37
equiprobable levels, with full-band sguare-topped pulses and with 4-th order Butterworth filtering ( f3dB = 196
kHz ).
The PSD of the HDSL NEXT can be expressed as:
3


PSDHDSL− NEXT = PSDHDSL− Disturber  x n f 2  ,


0 ≤ f < ∞, n = 1, 10, 24, 49
The integration of PSDHDSL-Disturber over various frequency ranges of interest is presented in Table 18
along with the induced NEXT power.
Table 18: HDSL transmit and induced NEXT power
Frequency Range
Transmit Power
dBm
0–0.196 MHz
0–0.392 MHz
0–0.784 MHz
0–1.544 MHz
0–1.568 MHz
0–3 MHz
13.44
13.60
13.60
13.60
13.60
13.60
NEXT Power
10 disturbers
dBm
–46.9
–46.3
–46.3
–46.3
–46.3
–46.3
-90
-95
-100
-105
-110
-115
-120
-125
-130
-135
-140
-145
-150
0
100000
200000
300000
400000
500000
600000
Frequency (Hz)
Figure 14: 10–disturber HDSL NEXT
38
700000
800000
7.3 Simulated T1 line PSD and induced NEXT
The PSD of the T1 line disturber is assumed to be the 50% duty–cycle random Alternate Mark Inversion
(AMI) code at 1.544 Mbit/s. The single–sided PSD has the following expression.
  πf  
 sin   
V 2P 2   f o  
×
PSDT1−Disturber =
RL f o   πf  
   
  fo 
2
2  πf 
sin 
×
 2 fo 
1
 f 6
1+ 

 f 3dB 
×
2
f
,
f 2 + f3 dB 2
0 ≤ f <∞
The total power of the transmit T 1 signal is computed by:
PT1−total =
2
1 Vp
4 RL
It is assumed that the transmitted pulse passes through a low–pass shaping filter. The shaping filter is chosen
as a third order low–pass Butterworth filter with 3–dB point at 3.0 MHz. The filter magnitude squared
transfer function is:
Hshaping( f ) =
2
1
6
 f 

1+ 
 f 3dB 
In addition, the coupling transformer is modeled as a high–pass filter with 3–dB point at 40 kHz as:
2
f2
H Transformer( f ) = 2
f + f3dB2
Furthermore, it is assumed that Vp = 3.6 Volts, RL = 100 Ohms, and fo = 1.544 MHz.
The PSD of the T1 NEXT can be expressed as:
3

PSDT1− NEXT = PSDT1− Disturber x n f 2  ,




0 ≤ f < ∞, n = 1, 10, 24, 49
The T1 transmit and T1–induced NEXT power using n–crosstalk models (Xn) are presented in Table 19.
Table 19: T1 transmit and induced NEXT power with shaping and coupling transformer
Frequency Range
0–1.544 MHz
0–3 MHz
Transmit Power
dBm
14.1
14.57
NEXT Power
4 disturbers
dBm
–34.7
–32.8
NEXT Power
24 disturbers
dBm
–30.0
–28.1
39
.
-85
-90
24-disturber
-95
-100
4-disturber
-105
-110
-115
-120
-125
-130
-135
-140
-145
-150
0
200000
400000
600000
800000
1000000
1200000
1400000
Frequency (Hz)
Figure 15: 4 and 24–disturber T1 NEXT
For testing, the T1 interferer power and the PSD curve are adjusted downward by a total of 15.5 dB to allow
for the adjacent binder effect and an averaging factor which accounts for non–collocation of the T1 and
ADSL terminals.
7.4 Simulated ADSL downstream PSD and induced NEXT and FEXT
The PSD of ADSL disturbers is expressed as:
PSD
ADSL-Disturber
= K ADSL ×
2
×
fo
  πf  

sin
  f o  
 πf 
 
 fo 
2
2
× |LPF ( f )|2 × | HPF( f )|2 ,
0≤ f <∞
where f o = 2.208 × 10 6 Hz, K ADSL = 0.1104Watts,
This equation gives the single sided PSD, where KADSL is the total transmitted power in Watts for the
downstream ADSL transmitter before shaping filters, and is set such that the ADSL PSD will not exceed the
maximum allowed PSD. fo is the sampling frequency in Hz and
2
LPF ( f ) =
40
f hα
36
, f h = 1.104 × 10 6 Hz , α =
= 11.96
α
α
f + fh
10 log(2)
is a fourth order low pass filter with a 3 dB point at 1104 kHz, and
f α + f lα
, f l = 4000 Hz , f h = 25875 Hz , α =
HPF ( f ) = α
f + f hα
2
57.5
f
10 log h
fl
= 7.09
is a fourth order high pass filter with a 3 dB point at 20 kHz, separating ADSL from POTS. With this set of
parameters the PSDADSL is the PSD of a downstream transmitter that uses all the channels.
7.4.1
FEXT
The FEXT loss model is:
| HFEXT ( f )|2 = | Hchannel ( f )|2 × k × l × f 2
where Hchannel (f) is the channel transfer function, k is the coupling constant and is 3.083x10–20 for 10, 1%
worst-case disturbers, l is the coupling path length in feet and equals 9000 ft for CSA #6, and f is in Hz. The
FEXT noise PSD is therefore:
PSDADSL − FEXT = PSDADSL × | HFEXT ( f )| 2
The integration of PSDADSL and PSDADSL–FEXT over the various frequency ranges is presented Table 20.
Figure 16 shows the theoretical PSD of 10 disturber downstream ADSL FEXT on CSA loop #6 using the PSD
model described above.
Figure 17 shows the theoretical PSD of 49 disturber downstream ADSL FEXT on T1.601 loop 7 using the PSD
mask from G.lite Annex A
Table 20: ADSL and ADSL-FEXT power with shaping and coupling transformer
Frequency Range
0–1.104 MHz
0–2.204 MHz
0–4.416 MHz
Transmit Power
DBm
18.99
19.15
19.15
FEXT Power
10 disturbers
dBm
–69.61
–69.61
–69.61
41
-90
-95
-100
-105
-110
-115
-120
-125
-130
-135
-140
-145
-150
0
200000
400000
600000
800000
1000000
Fequency (Hz)
Figure 16: 10–disturber downstream ADSL FEXT on CSA 6 using the PSD model from section 7.4
42
Figure 17: 49-disturber downstream ADSL FEXT on T1.601 Loop 7 using the PSD mask from G.lite Annex
A.
7.4.2
NEXT
The PSD of the ADSL NEXT into the upstream can be expressed as
PSDADSL-NEXT = PSDADSL-Disturber × (x (n)f 1.5 ) for n = 1,10,24,49
where x(n) is defined in section 7.1. The integration of the induced NEXT over the
band from 0 to 1.104 Mhz for n = 49 is –25.4 dBm.
Figure 18 shows the theoretical PSD of 49 disturber downstream ADSL NEXT for the non-overlapped ADSL
mask.
43
Figure 18: 49-disturber downstream ADSL NEXT into the upstream using the PSD mask from G.lite Annex
A
44
7.5 Simulated ADSL upstream PSD and induced NEXT and FEXT
7.5.1
FEXT
The FEXT loss model is:
H FEXT ( f ) = H channel ( f ) × k × l × f
2
2
2
where Hchannel (f) is the channel transfer function, k is the coupling constant and is
8 × 10-20× (n/49)0.6
for n<50 or 3.083 × 10-20 for 10, 1% worst-case disturbers, l is the coupling path length in feet, f is in Hz.
The FEXT noise PSD is therefore:
PSD ADSL− FEXT = PSD ADSL− Disturber × H FEXT ( f )
2
The integration of the induced FEXT over the band from 0 to 1.104 MHz for n=49 and using T1.601 loop 7 is
–81.8 dBm, and -83.7 for n=24.
Figure 19 shows the theoretical PSD of 49-disturber upstream ADSL FEXT on T1.601 loop #7.
Figure 19: Theoretical 49-disturber upstream ADSL FEXT on T1.601 Loop 7 using the PSD mask from G.lite
Annex A.
7.5.2
NEXT
45
The upstream ADSL signal nominally occupies the band from 25 to 138 kHz, but the upper sidelobes of the
pass–band signal beyond 138 kHz may also contribute to the NEXT into the downstream signal. Their effect
will depend on the method of anti–aliassing used in the remote transmitter. The PSD of the upstream ADSL
NEXT can be expressed as:
3


PSDADSL − NEXT = PSDADSL,us − Disturber  xn f 2  ,


0≤ f <∞
where x10 = 3.3982 10 –14 and f is in Hz. PSDADSL,us – disturber is difficult to define precisely because
of the various sidelobes of the passband signals
The PSD of ADSL disturbers is expressed as:
PSDADSL -Disturber = K ADSL ×
2
×
fo
where f o = 276 × 10 3 Hz, K ADSL
  πf
sin 
  fo
2



× | LPF ( f ) | 2 × | HPF ( f ) | 2 ,
2
 πf 
 
 fo 
= 0.0437 Watts ,
0≤ f <∞
This equation gives the single sided PSD, where KADSL is the total transmitted power in Watts for the
upstream ADSL transmitter before shaping filters, and is set such that the ADSL PSD will not exceed the
maximum allowed PSD. f0 is the sampling frequency in Hz and
2
LPF ( f ) =
f hα
24
, f h = 138 × 10 3 Hz , α =
= 20.32
α
α
10 log(181.125 / 138)
f + fh
is a low pass filter with a 3 dB point at 138 kHz and 24 dB attenuation at 181.125 kHz, and
2
HPF ( f ) =
f α + f lα
, f l = 4000 Hz , f h = 25875 Hz , α =
f α + f hα
59.5
f
10 log h
fl
= 7.34
is a high pass filter with 3 dB points at 4 and 25.875 kHz and 59.5 dB attenuation in the voice band, separating
ADSL from POTS. With this set of parameters the PSDADSL-Disturber is the PSD of an upstream transmitter
that uses all the sub-carriers.
Figure 20 shows the theoretical PSD of 10–disturber upstream ADSL NEXT into the downstream.
46
-90
-95
-100
-105
-110
-115
-120
-125
-130
-135
-140
-145
-150
0
50000
100000
150000
200000
250000
300000
Frequency(Hz)
Figure 20: 10–disturber upstream ADSL NEXT into the downstream using the PSD model from section
7.5.2.
7.6 ETSI A, -ETSI B and Euro-K crosstalk
The power spectral density of the crosstalk noise sources used for performance testing is given in Figure 21 for
-ETSI A, and in Figure 22 for ETSI B. ETSI A includes discrete tones, which represent radio frequency
interference that is commonly observed, especially on wire pairs routed above ground. Further details of the
specification of these noise models are shown in Table 21, Table 22 and Table 23.
The resulting wideband noise power over the frequency range 1 kHz to 1.5 MHz for model A is –49.4±0.5
dBm and for model B is –43.0±0.5 dBm.
The noise probability density function should be approximately Gaussian with a crest factor = 5.
The accuracy of the power spectral density shall be within ±1 dB over the frequency range 1 kHz to 1.5 MHz,
when measured with a resolution bandwidth of 1 kHz.
47
10 discrete tones
PSD (dBm/Hz)
−100
−110
−120
−40 dB/decade slope
−130
−140
−150
1
10
100
1000
Freq. (kHz)
Figure 21: Single sided noise power spectral density into 100 Ω for ETSI A
Table 21: Co-ordinates for noise ETSI A
Freq.
kHz
1
79.5
795
1500
PSD
dBm/Hz
-100
-100
-140
-140
Table 22: Tone frequencies and powers for noise ETSI A
Freq.
kHz
99
207
333
387
531
603
711
801
909
981
48
Power
dBm
-70
-70
-70
-70
-70
-70
-70
-70
-70
-70
PSD (dBm/Hz)
−70
−80
−20 dB/decade slope
−90
−100
−40 dB/decade slope
−110
−120
−130
1
10
100
1000 kHz
Figure 22: Single sided noise power spectral density into 100 Ω for ETSI B
Table 23: Co-ordinates for noise ETSI B
Freq.
kHz
1
10
300
711
1500
PSD
dBm/Hz
–80
–100
–100
–115
–115
PSD
µV/√Hz
31.62
3.16
3.16
0.56
0.56
The EURO-K model has a floor at -140dBm/Hz, its maximum at -94dBm/Hz between 135kHz and 260kHz.
Below 135kHz it rolls off towards the floor at +7dB/ decade. Above 260kHz it rolls off towards the floor at 75dB/ decade.
Note that Euro-K is based on a re-calculation of ETSI A crosstalk interference at the ATU-C end. It basically
removes the secondary HDSL NEXT effects which are exploited at the ATU-R end by noise ETSI A but which
do not apply at the ATU-C end.
7.7 Power spectral density of crosstalk disturbers in an environment coexisting with
TCM-ISDN DSL
Crosstalk margin measurements are made for three types of disturbers from pairs in the same binder group;
TCM-ISDN DSL, HDSL and ADSL lines.
Note: The term “ADSL” is used to indicate both G.dmt and G.lite, unless otherwise stated.
49
7.7.1
Simulated TCM-ISDN DSL PSD and induced NEXT and FEXT
The PSD of Basic Rate Access (BRA) DSL disturbers using a Time Completion Multiplex (TCM) method is
expressed as:
PSDDSL - Distuerber = KDSL ×
2
f
× [sin(π )]2 ×
f0
f0
f 2
)]
1
2f 0
×
f 2
f 4
(π
)
1+ (
)
f 3 dB
2f 0
[sin(π
where f in Hz , f 0 = 320 × 103 Hz , f 3dB = 2 × f 0 , KDSL =
(0 ≤ f < ∞)
V 0 P2
, V 0 P = 6.00 volts and R = 110 ohms
4R
The equation PSDDSL-Disturber gives the single-sided PSD of a 320 kbaud Alternate Mark Inversion (AMI) signal
with random continuous symbol sequence, with 50% duty-cycle pulses and with 2nd order low-pass
Butterworth filtering (f3dB = 640 kHz).
7.7.1.1 DSL NEXT
The PSD of the DSL NEXT can be expressed as:
3
 − NPSLn
− 
PSD DSL − NEXT = PSD DSL − Distuerber × 10 10 × f NXT 2  × f


where f in Hz, f NXT = 160 × 10 3 Hz
3
2
(0 ≤ f < ∞)
The crosstalk loss as the result of multiple interfering pairs is known as the Power Sum Loss (PSL), and 99%
cumulative (1% worst case) values are adopted for ADSL performance test. PSL values vary depending on the
number of disturber pairs, the insulation material of the pair conductors (paper, polyethylene), and line
conditioning states (pair selection conditions) for DSL installation into a cable pair in the same binder.
NEXT Power Sum Loss for the disturber pair number of “n” is abbreviated to “NPSLn”. The values of NPSLn
are defined in the following.
The NPSL24 value of 47.0 dB is for the case of the paper insulated cable and intra-quad line conditioning state.
The NPSL24 value of 52.5 dB is for the case of the paper insulated cable and inter-quad line conditioning state.
The value of 52.5 dB is for the case that interfering and interfered transmission systems are different; from
DSL to ADSL and vice versa.
The NPSL24 value of 49.5 dB is for the case of the polyethylene insulated cable and intra-quad line
conditioning state.
The NPSL24 value of 53.5 dB is for the case of the polyethylene insulated cable and inter-quad line
conditioning state. The value of 53.5 dB is for the case that interfering and interfered transmission systems are
different; from DSL to ADSL and vice versa.
The integration of PSDDSL-Disturber and PSDDSL-NEXT over various frequency ranges are presented in Table 24 for
the case of the paper insulated cable with intra-quad line conditioning state.
50
Table 24: TCM-ISDN DSL transmit and induced NEXT power
Table 9.tcm: TCM-ISDN DSL transmit and induced NEXT power
Frequency Range
Transmit Power
NEXT Power
NPSL24= 47.0 dB
0 ~ 640 kHz
0 ~ 1280 kHz
0 ~ 1920 kHz
0 ~ 2560 kHz
0 ~ 3200 kHz
18.6 dBm
18.6 dBm
18.6 dBm
18.6 dBm
18.6 dBm
-27.3 dBm
-26.8 dBm
-26.7 dBm
-26.7 dBm
-26.7 dBm
PSD [dBm/Hz]
-30
-40
PSDDSL-Disturber
-50
-60
-70
-80
PSD DSL -NEXT
(NPSL24 = 47.0 dB)
-90
-100
-110
-120
0
320
640
960 1280 1600 1920 2240 2560
Frequency (kHz)
Figure 4.tcm: 24-distur ber TCM-I SDN DSL NEXT
Figure 23: 24 disturber TCM-IDSN DSL NEXT
7.7.1.2 DSL FEXT
The PSD of the DSL FEXT can be expressed as:
 − FPSLn
−1
−2 
PSD DSL − FEXT = PSD DSL − Distuerber × H channel ( f , d ) × 10 10 × d FXT × f FXT  × d × f


3
3
where f in Hz , d in meter , f FXT = 160 × 10 Hz , d FXT = 1.0 × 10 meters
2
2
(0 ≤ f < ∞ )
Hchannel(f,d) is the channel transfer function, “d” is the coupling path distance in meter, and depends on test
loops defined in section 7.1. The power sum of PSDDSL-FEXT is required for the test loops of mixed type
connection.
FEXT Power Sum Loss for the disturber pair number of “n” is abbreviated to “FPSLn”. The 99% cumulative
values of FPSLn are defined in the following.
51
The FPSL24 value of 45.0 dB is for the case of the paper insulated cable and both intra-quad and inter-quad
line conditioning states. The value of 45.0 dB is for the case that interfering and interfered transmission
systems are different; from DSL to ADSL and vice versa.
The FPSL24 value of 51.0 dB is for the case of the polyethylene insulated cable and both intra-quad and interquad line conditioning states. The value of 51.0 dB is for the case that interfering and interfered transmission
systems are different; from DSL to ADSL and vice versa.
The integration of PSDDSL-Disturber and PSDDSL-FEXT over various frequency ranges are presented in Table 25 for
the case of the 0.4 mm gauge paper insulated cable with intra-quad line conditioning state and with d = 2.07 x
103, 2.94 x 103, 3.97 x 103, and 5.16 x 103 meters.
Table 25: TCM-ISDN DSL transmit and induced FEXT power
Table 10.tcm: TCM-ISDN DSL transmit and induced FEXT power
Frequency Transmit FEXT Power (FPSL24 = 45.0 dB)
Range
0 ~ 640 kHz
0 ~ 1280 kHz
0 ~ 1920 kHz
0 ~ 2560 kHz
0 ~ 3200 kHz
PSD [dBm/Hz]
-100
-110
-120
-130
-140
-150
-160
-170
-180
-190
-200
0
Power
18.6 dBm
18.6 dBm
18.6 dBm
18.6 dBm
18.6 dBm
d= 2.07 km
-50.3 dBm
-50.3 dBm
-50.3 dBm
-50.3 dBm
-50.3 dBm
d = 2.07 km
d= 2.94 km
-59.5 dBm
-59.5 dBm
-59.5 dBm
-59.5 dBm
-59.5 dBm
d= 3.97 km
-70.4 dBm
-70.4 dBm
-70.4 dBm
-70.4 dBm
-70.4 dBm
PSDDSL-FEXT
(FPSL24 = 45.0 dB)
d = 2.94 km
d = 3.97 km
d = 5.16 km
320
640
960
1280
Frequency (kHz)
Figure 5.tcm: 24-disturber TCM-ISDN DSL FEXT
Figure 24: 24 distrurber TCM-ISDN DSL FEXT
7.7.2
Simulated HDSL PSD and induced NEXT
The PSD of HDSL disturbers is expressed as:
52
d= 5.16 km
-82.7 dBm
-82.7 dBm
-82.7 dBm
-82.7 dBm
-82.7 dBm
PSDHDSL - Distuerber = KHDSL ×
2
×
f0
f 2
)]
1
f0
×
f
f 8
(π ) 2
1+ (
)
f0
f 3 dB
[sin(π
where f in Hz , f 0 = 392 × 103 Hz , f 3dB =
(0 ≤ f < ∞)
f0
5 V 0 P2
, KHDSL = ×
, V 0 P = 2.70 volts and R = 135 ohms
2
9
R
The equation PSDHDSL-Disturber gives the single-sided PSD of a 392 kbaud 2B1Q signal with random continuous
equiprobable level symbol sequence, with 100% duty-cycle pulses and with 4-th order low-pass Butterworth
filtering (f3dB = 196 kHz).
The PSD of HDSL NEXT can be expressed as:
3
 − NPSLn
− 
PSD HDSL− NEXT = PSD HDSL − Distuerber × 10 10 × f NXT 2  × f


where f in Hz, f NXT = 160 × 10 3 Hz
3
2
(0 ≤ f < ∞ )
The 99% cumulative values of NPSLn are defined in the following.
The NPSL24 value of 47.0 dB is for the case of the paper insulated cable and intra-quad line conditioning state.
The NPSL24 value of 52.5 dB is for the case of the paper insulated cable and inter-quad line conditioning state.
The value of 52.5 dB is for the case that interfering and interfered transmission systems are different; from
HDSL to ADSL and vice versa.
The NPSL24 value of 49.5 dB is for the case of the polyethylene insulated cable and intra-quad line
conditioning state.
The NPSL24 value of 53.5 dB is for the case of the polyethylene insulated cable and inter-quad line
conditioning state. The value of 53.5 dB is for the case that interfering and interfered transmission systems are
different; from HDSL to ADSL and vice versa.
The integration of PSDHDSL-Disturber and PSDHDSL-NEXT over various frequency ranges are presented in Table 26
for the case of the paper insulated cable with intra-quad line conditioning state.
Table 26: HDSL transmit and induced NEXT power
Table 11.tcm: HDSL transmit and induced NEXT power
Frequency Range Transmit Power NEXT Power (NPSL24= 47.0dB)
0 ~ 196 kHz
0 ~ 392 kHz
0 ~ 784 kHz
0 ~ 1568 kHz
0 ~ 3136 kHz
13.4 dBm
13.6 dBm
13.6 dBm
13.6 dBm
13.6 dBm
-37.4 dBm
-36.8 dBm
-36.8 dBm
-36.8 dBm
-36.8 dBm
53
PSD [dBm/Hz]
-30
-40
-50
-60
-70
-80
-90
-100
-110
-120
-130
-140
-150
-160
-170
-180
PSDHDSL-Disturber
PSDHDSL-NEXT
(NPSL24 = 47.0 dB)
0
588
784
980
Frequency (kHz)
Figure 6.tcm: 24-disturber HDSL NEXT
196
392
1176
Figure 25: 24 disturber HDSL NEXT
7.7.3
Simulated downstream ADSL PSD and induced NEXT and FEXT
The PSD of the downstream G.dmt/G.lite disturbers is expressed as:
2
  f 

sin ð
1
2   f 0 
1
×
×
PSD ADSL,ds-Distuerber =K ADSL ,ds × ×
N
12
2
f0
 f 
 f 
 f HP 3 dB 
ð



1 + 
1 + 
 f 
 f0 
 f LP 3 dB 
f
f
where f in Hz, f 0 = 2.208 × 10 6 Hz, f LP 3 dB= 0 ( for G.dmt ) , f LP 3 dB= 0 ( for G.lite),
2
4
3
f HP 3 dB = 25.875 × 10 Hz , N = 8 ( for EC ) ,
(0 ≤ f < ∞ )
f HP 3 dB = 138 × 10 3 Hz , N = 16 ( for FDM ) ,
K ADSL,ds = 0.1104 watts (corresponding to the maximum pass − band PSD of − 40dB / Hz )
The equation PSDADSL,ds-Disturber gives the single-sided PSD with 6-th order low-pass Butterworth filtering
(fLP3dB = 1104 kHz (G.dmt) or 552 kHz (G.lite)), rejecting stop-band PSD and with N-th order high-pass
filtering (fHP3dB = 25.875 kHz, N = 8 (EC) or 138 kHz, N= 16 (FDM)), separating ADSL signal from POTS
signal for EC, or separating ADSL downstream signal from ADSL upstream signal for FDM.
7.7.3.1 Downstream ADSL NEXT into the upstream
The PSD of the downstream G.dmt/G.lite NEXT into the upstream can be expressed as:
54
3
 − NPSLn
− 
PSD ADSL,ds − NEXT = PSD ADSL, ds − Distuerber × 10 10 × f NXT 2  × f


3
2
(0 ≤ f < ∞ )
where f in Hz, f NXT = 160 × 10 3 Hz
The 99% cumulative values of NPSLn are defined in the following.
The NPSL24 value of 47.0 dB is for the case of the paper insulated cable and intra-quad line conditioning state.
The NPSL24 value of 53.5 dB is for the case of the paper insulated cable and inter-quad line conditioning state.
The value of 53.5 dB is for the case that interfering and interfered transmission systems are identical; from
ADSL to ADSL.
The NPSL24 value of 49.5 dB is for the case of the polyethylene insulated cable and intra-quad line
conditioning state.
The NPSL24 value of 55.5 dB is for the case of the polyethylene insulated cable and inter-quad line
conditioning state. The value of 55.5 dB is for the case that interfering and interfered transmission systems are
identical; from ADSL to ADSL.
The integration of PSDADSL,ds-NEXT over various frequency ranges are presented in Table 27 (G.dmt, FDM) and
Table 28 (G.lite, FDM) for the case of the paper insulated cable with intra-quad line conditioning state.
Table 27: Downstream G.dmt (FDM) transmit and induced NEXT power
Table 12.tcm: Downstream G.dmt (FDM) transmit and induced NEXT power
Frequency Range Transmit Power
NEXT Power (NPSL24 = 47.0 dB)
0 ~ 1104 kHz
0 ~ 2208 kHz
0 ~ 4416 kHz
18.4 dBm
18.5 dBm
18.5 dBm
-20.4 dBm
-20.0 dBm
-20.0 dBm
Table 28: Downstream G.lite (FDM) transmit and induced NEXT power
Table 13.tcm: Downstream G.lite (FDM) transmit and induced NEXT power
Frequency Range Transmit Power
NEXT Power (NPSL24 = 47.0 dB)
0 ~ 552 kHz
0 ~ 1104 kHz
0 ~ 2208 kHz
15.5 dBm
15.8 dBm
15.8 dBm
-26.7 dBm
-25.9 dBm
-25.9 dBm
55
PSD [dBm / Hz]
-30
-40
P SD A DS L ,d s-D is tu rber
-50
-60
-70
P SD A DS L ,d s-N EX T (NPSL 2 4 = 47.0 dB )
-80
-90
-100
-110
-120
-130
-140
-150
0
276
552 828 1104 1380 1656 1932 2208
Frequency (kHz)
Figure 7.tcm: Downstream 24-disturber G.dmt (FDM) NEXT
Figure 26: Downstream 24 disturber G.dmt (FDM) NEXT
PSD [dBm/Hz]
-30
-40
PSDADSL,ds-Disturber
-50
-60
-70
PSDADSL,ds-NEXT (NPSL 24 = 47.0 dB)
-80
-90
-100
-110
-120
-130
-140
-150
0
276 552 828 1104 1380 1656 1932 2208
Frequency (kHz)
Figure 8.tcm: Downstream 24-disturber G.lite (FDM) NEXT
Figure 27: Downstream 24 disturber G.lite (FDM) NEXT
7.7.3.2 Downstream ADSL FEXT into the downstream
The PSD of the downstream G.dmt/G.lite FEXT into the downstream can be expressed as:
56
 − FPSLn
2
−1
−2 
PSD ADSL, ds − FEXT = PSD ADSL, ds − Distuerber × H channel ( f , d ) × 10 10 × d FXT × f FXT  × d × f


where f in Hz , d in meter , f FXT = 160 × 10 3 Hz , d FXT = 1.0 × 10 3 meters
2
(0 ≤ f < ∞ )
The 99% cumulative values of FPSLn are defined in the following.
The FPSL24 value of 45.0 dB is for the case of the paper insulated cable and intra-quad line conditioning states.
The FPSL24 value of 45.5 dB is for the case of the paper insulated cable and inter-quad line conditioning states.
The value of 45.5 dB is for the case that interfering and interfered transmission systems are identical; from
ADSL to ADSL.
The FPSL24 value of 51.0 dB is for the case of the polyethylene insulated cable and intra-quad line
conditioning states.
The FPSL24 value of 53.0 dB is for the case of the polyethylene insulated cable and inter-quad line
conditioning states. The value of 53.0 dB is for the case that interfering and interfered transmission systems
are identical; from ADSL to ADSL.
The integration of PSDADSL,ds-Disturber and PSDADSL,ds-FEXT over various frequency ranges are presented in Table
29 (G.dmt, FDM) and Table 30 (G.lite, FDM) for the case of the 0.4 mm gauge paper insulated cable with
intra-quad line conditioning state and with d = 2.07 x 103, 2.94 x 103, 3.97 x 103, and 5.16 x 103 meters.
Table 29: Downstream G.dmt (FDM) transmit and induced FEXT power
Table 14.tcm: Downstream G.dmt (FDM) transmit and induced FEXT power
Frequency
Transmit FEXT Power (FPSL24 = 45.0 dB)
Range
Power
d= 2.07 km d= 2.94 km
d= 3.97 km
d= 5.16 k m
0 ~ 1104 kH z
18.4 dB m -54.9 dBm
-66.5 dB m
-79.5 dB m
-94.2 dBm
0 ~ 2208 kH z
18.5 dB m -54.9 dBm
-66.5 dB m
-79.5 dB m
-94.2 dBm
0 ~ 4416 kH z
18.5 dB m -54.9 dBm
-66.5 dB m
-79.5 dB m
-94.2 dBm
Table 30: Downstream G.lite (FDM) transmit and induced FEXT power
Table 15.tcm: Downstream G.lite (FDM) transmit and induced FEXT power
Frequency
Transmit FEXT Power (FPSL24 = 45.0 dB)
Range
Power
d= 2.07 km d= 2.94 km
0 ~ 552 kHz 15.5 dBm -55.2 dBm
-66.5 dBm
0 ~ 1104 kHz 15.8 dBm -55.1 dBm
-66.5 dBm
0 ~ 2208 kHz 15.8 dBm -55.1 dBm
-66.5 dBm
d= 3.97 km
-79.5 dBm
-79.5 dBm
-79.5 dBm
d= 5.16 km
-94.2 dBm
-94.2 dBm
-94.2 dBm
57
PSD [dBm/Hz]
-100
-110
-120
-130
PSDADSL,ds-FEXT (FPSL 24 = 45.0 dB)
d= 2.07 km
-140
-150
-160
-170
-180
d= 2.94 km
-190
-200
d= 5.16 km
d= 3.97 km
0
276
552
828
1104
Frequency (kHz)
Figure 9.tcm: Downstream 24-disturber G.dmt (FDM)FEXT
Figure 28: Downstream 24 disturber G.dmt (FDM) FEXT
PSD [dBm/Hz]
-100
-110
-120
-130
-140
-150
-160
-170
-180
-190
-200
0
PSDADSL,ds-FEXT (FPSL24 = 45.0 dB)
d= 2.07 km
d= 2.94 km
d= 3.97 km
d= 5.16 km
276
552
828
1104
Frequency (kHz)
Figure 10.tcm: Downstream 24-disturber G.lite (FDM) FEXT
Figure 29: Downstream 24 disturber G.lite (FDM) FEXT
7.7.4
Simulated upstream ADSL PSD and induced NEXT
The single-sided PSD of the upstream G.dmt/G.lite disturbers can be assumed as:
58
PSD ADSL,us-Distuerber =K ADSL ,us
2
  f 

sin ð
1
2   f 0 
×
× ×
2
f0
 f
 f 

ð
1 + 
 f LP 3 dB
 f0 
×
1
16
f


1 +  HP 3dB
 f

where f in Hz, f 0 = 276 × 10 3 Hz, f LP 3 dB=138 × 10 3 , f HP 3 dB=25.875 × 10 3 Hz ,



8
(0 ≤ f < ∞ )
K ADSL ,us = 0.02187 watts (corresponding to the maximum pass − band PSD of − 38dB / Hz )
The PSD of the upstream G.dmt/G.lite NEXT into the downstream can be expressed as:
n
3
 − NPSL
− 
PSD ADSL,us − NEXT = PSD ADSL ,us − Distuerber × 10 10 × f NXT 2  × f


where f in Hz, f NXT = 160 × 10 3 Hz
3
2
(0 ≤ f < ∞ )
The 99% cumulative values of NPSLn are defined in the following.
The NPSL24 value of 47.0 dB is for the case of the paper insulated cable and intra-quad line conditioning state.
The NPSL24 value of 53.5 dB is for the case of the paper insulated cable and inter-quad line conditioning state.
The value of 53.5 dB is for the case that interfering and interfered transmission systems are identical; from
ADSL to ADSL.
The NPSL24 value of 49.5 dB is for the case of the polyethylene insulated cable and intra-quad line
conditioning state.
The NPSL24 value of 55.5 dB is for the case of the polyethylene insulated cable and inter-quad line
conditioning state. The value of 55.5 dB is for the case that interfering and interfered transmission systems are
identical; from ADSL to ADSL.
The integration of PSDADSL,us-Disturber and PSDADSL,us-NEXT over various frequency ranges are presented in Table
31 for the case of a paper insulated cable with intra-quad line conditioning state.
Table 31: Upstream G.dmt/G.lite transmit and induced NEXT power
Table 16.tcm: Upstream G.dmt/G.lite transmit and induced NEXT power
Frequency Range Transmit Power NEXT Power (NPSL24= 47.0dB)
0 ~ 138.000 kHz
0 ~ 181.125 kHz
0 ~ 224.250 kHz
10.9 dBm
11.0 dBm
11.0 dBm
-40.9 dBm
-40.6 dBm
-40.6 dBm
59
PSD [dBm/Hz]
-30
-40
-50
PSDADSL,us-Disturber
-60
-70
-80
PSDADSL,us-NEXT (NPSL24 = 47.0dB)
-90
-100
-110
-120
-130
-140
-150
0 25.875
138
181.125
224.25 267.375 310.5
Frequency (kHz)
Figure 11.tcm: Upstream 24-disturber G.dmt/G.lite NEXT
Figure 30: Upstream 24 disturber G.dmt/G.lite NEXT
8 Characteristics of impulse noise waveforms
The two test impulse waveforms are shown in Figure 31 and Figure 32. Table 32 and Table 33 contain the
impulse wave amplitude given in millivolts at 160 nanosecond time intervals. The specific means of
generating these waveforms for test purposes is left to the implementor.
60
20
10
A
m
p
l
0
i
t
u
d
e
(mV) - 10
- 20
0.125
0.130
0.135
0.140
0.145
0.150
Time (msec)
Figure 31: Test impulse 1
40
A
20
m
p
l
i
0
t
u
d
e - 20
(mV)
- 40
0.125
0.130
0.135
0.140
0.145
0.150
Time (msec)
Figure 32: Test impulse 2
61
Table 32: Impulse number 1
Interval #
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
62
Amplitude mV
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.9590
–0.4795
–1.2787
–1.1188
–1.4385
–1.5984
–2.2377
–1.4385
7.6721
6.7131
–16.6229
–12.9467
18.7008
9.5902
–13.5861
–5.2746
–6.3934
–1.9180
23.0164
3.9959
–23.4959
–3.1967
4.3156
–3.0369
10.7090
2.2377
–12.9467
3.1967
1.9180
–9.9098
5.5943
5.9139
–6.7131
2.3975
1.2787
–8.4713
2.5574
2.8771
–6.0738
2.2377
1.7582
Interval #
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
Amplitude m
–6.3934
1.7582
2.2377
–4.9549
2.2377
1.7582
–5.5943
1.4385
2.3975
–3.6762
1.4385
0.4795
–5.7541
–0.4795
0.3197
–3.3566
2.3975
2.3975
–3.1967
0.7992
0.6393
–3.5164
1.1188
1.7582
–2.3975
1.2787
0.9590
–3.3566
0.0000
0.1598
–3.0369
1.1188
1.5984
–2.0779
0.1598
0.3197
–2.5574
0.1598
0.1598
–2.0779
0.6393
0.9590
–1.7582
–0.1598
–0.6393
–3.0369
–0.3197
0.4795
–1.4385
0.4795
Interval #
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
Amplitude mV
0.1598
–1.7582
0.1598
0.4795
–1.2787
0.7992
1.2787
–0.7992
0.0000
–0.3197
–2.2377
–1.1188
–0.7992
–1.5984
0.1598
0.4795
–0.9590
0.0000
–0.3197
–1.5984
0.0000
0.4795
–0.7992
0.4795
0.7992
–0.9590
–0.9590
–0.4795
–0.6393
0.4795
1.1188
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
Table 33: Impulse number 2
Interval #
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
Amplitude mV
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
–0.6404
0.9606
0.1601
–5.4433
–12.3276
–12.1675
0.0000
5.2832
0.1601
–20.8128
–45.3078
–46.7487
–28.9778
–13.4483
0.6404
0.9606
–14.4089
–13.7685
–9.4458
–17.4507
–2.5616
26.5763
16.1699
–17.7709
–17.1305
13.6084
27.0566
18.0911
14.2488
5.6034
–8.1650
12.4877
37.3029
9.6059
–18.8916
5.1231
22.2537
1.1207
–0.9606
20.4926
14.2488
Interval #
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
Amplitude mV
0.6404
15.5295
18.8916
–3.8424
–3.0419
11.6872
–0.3202
–7.5246
13.4483
18.4113
–0.4803
–3.0419
9.7660
11.2069
4.0025
0.6404
0.6404
1.7611
3.3621
5.6034
7.8448
2.5616
–4.6428
0.6404
10.7266
8.3251
1.9212
3.6823
4.3227
0.3202
2.7217
7.2044
3.2020
–2.7217
–1.4409
1.2808
1.4409
0.8005
0.1601
0.0000
1.1207
1.1207
0.6404
1.1207
0.6404
–1.1207
–0.8005
0.1601
–1.2808
–1.4409
Interval #
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
Amplitude mV
0.6404
0.6404
–0.4803
–0.3202
–0.9606
–2.8818
–2.5616
–0.8005
–0.4803
–0.8005
–0.4803
–0.9606
–1.1207
–0.6404
–0.4803
–0.9606
–1.4409
–1.6010
–1.2808
–0.9606
–0.9606
–1.2808
–1.1207
–1.1207
–1.4409
–1.4409
–1.4409
–2.0813
–2.4015
–1.9212
–1.4409
–1.1207
–1.2808
–1.9212
–2.2414
–2.2414
–2.5616
–3.0419
–3.0419
–2.5616
–1.2808
–0.1601
–0.6404
–2.5616
–3.2020
–3.0419
–2.5616
–2.0813
–1.4409
–1.6010
63
Table 33: Impulse number 2 (continued)
Interval #
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
64
Amplitude mV
–1.9212
–1.9212
–2.0813
–2.4015
–2.5616
–2.5616
–1.9212
–1.6010
–1.6010
–1.9212
–1.9212
–2.0813
–2.2414
–2.5616
–2.7217
–2.2414
–1.2808
–1.2808
–2.2414
–3.0419
–2.8818
–2.5616
–2.2414
–1.9212
–1.9212
–2.2414
–2.5616
–2.7217
–2.5616
–2.4015
–2.2414
–2.0813
–1.7611
–1.6010
–1.7611
–2.2414
–3.0419
–3.2020
–2.7217
–1.9212
–1.2808
–0.9606
–1.1207
–2.0813
–2.8818
–3.0419
–2.7217
–2.7217
–2.0813
–1.4409
Interval #
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
Amplitude mV
–0.8005
–0.9606
–1.6010
–2.4015
–2.5616
–2.8818
–2.7217
–1.9212
–1.1207
–0.9606
–1.1207
–1.4409
–1.7611
–2.4015
–2.5616
–2.2414
–1.7611
–1.7611
–1.4409
–0.9606
–0.8005
–0.9606
–1.6010
–2.2414
–2.4015
–2.2414
–1.9212
–1.4409
–0.4803
0.0000
–0.6404
–1.6010
–1.7611
–1.6010
–1.9212
–1.9212
–1.4409
–0.4803
0.0000
0.0000
–0.6404
–1.6010
–2.4015
–1.9212
–0.9606
–0.4803
–0.1601
–0.1601
0.0000
–0.8005
Interval #
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
Amplitude mV
–1.2808
–1.6010
–1.6010
–1.4409
–0.4803
0.4803
0.4803
–0.4803
–0.9606
–1.1207
–1.4409
–1.2808
–0.1601
0.3202
0.0000
–0.4803
–0.4803
–0.4803
–0.6404
–0.4803
–0.1601
0.0000
0.0000
–0.1601
–0.1601
–0.4803
–0.6404
–0.3202
0.1601
0.4803
0.3202
–0.1601
–0.3202
–0.4803
–0.6404
–0.4803
0.1601
0.6404
0.6404
0.4803
0.0000
–0.6404
–0.6404
–0.4803
–0.1601
0.4803
0.6404
0.4803
0.6404
0.4803
Table 33: Impulse number 2 (continued)
Interval #
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
Amplitude mV
–0.1601
–0.9606
–0.9606
–0.1601
0.6404
0.8005
0.8005
0.4803
0.1601
–0.1601
–0.3202
–0.1601
0.0000
0.1601
0.6404
0.8005
0.6404
0.4803
0.0000
–0.4803
–0.4803
0.1601
0.8005
0.8005
0.6404
0.1601
0.4803
0.4803
0.3202
–0.3202
–0.4803
0.0000
0.6404
1.1207
1.2808
0.6404
0.1601
–0.1601
0.0000
0.0000
0.1601
0.3202
0.8005
1.2808
1.2808
0.9606
0.1601
–0.8005
–0.9606
–0.1601
Interval #
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
Amplitude mV
0.8005
1.4409
1.6010
1.2808
0.6404
0.0000
–0.4803
–0.6404
0.0000
0.8005
1.4409
1.6010
1.2808
0.6404
0.0000
–0.4803
–0.1601
0.1601
0.9606
1.4409
1.6010
1.1207
0.3202
–0.4803
–0.4803
0.1601
0.8005
1.1207
1.1207
0.9606
0.6404
0.1601
0.0000
0.1601
0.6404
1.1207
0.9606
0.6404
0.6404
0.6404
0.3202
0.0000
0.4803
1.1207
1.1207
0.6404
0.1601
0.0000
0.1601
0.8005
Interval #
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
Amplitude mV
0.9606
0.6404
0.4803
0.6404
0.6404
0.4803
0.3202
0.1601
0.3202
0.4803
0.9606
1.2808
0.9606
0.1601
–0.1601
0.0000
0.4803
0.8005
0.6404
0.4803
0.8005
0.8005
0.4803
0.1601
0.0000
0.0000
0.1601
0.3202
0.6404
0.9606
0.8005
0.3202
0.1601
0.0000
0.1601
0.1601
0.1601
0.1601
0.6404
1.1207
0.9606
0.4803
0.0000
–0.3202
–0.3202
0.0000
0.1601
0.6404
0.9606
0.8005
65
Table 33: Impulse number 2 (continued)
Interval #
451
452
453
454
455
456
457
458
459
460
66
Amplitude mV
0.6404
0.0000
–0.8005
–0.8005
0.0000
0.4803
0.6404
0.6404
0.8005
0.6404
Interval #
461
462
463
464
465
466
467
468
469
470
Amplitude mV
0.0000
–0.9606
–1.1207
–0.4803
0.4803
1.1207
1.1207
0.6404
0.0000
0.0000
Interval #
471
472
473
474
475
476
477
478
479
480
Amplitude mV
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
9 Phone Model Structures for Testing DSL systems without a splitter at
the ATU-R
9.1 Off phone hook model #1
Off hook phone model #1 was generated for testing in the TCM-ISDN noise environment. It could be used for
testing in other noise environments as well.
67
9.2 Off Hook phone model #2
Off hook phone model #2 was generated for testing in the splitterless North American noise environment.
It could be used for testing in other noise environments as well.
R2
R1
C
Figure 33: Simple Model of an Off-Hook Telephone
(R1 = 720Ω, R2 = 183Ω, C = 4300pF)
68
1600Ω
1400Ω
1200Ω
1000Ω
20 mA
800Ω
600Ω
400Ω
Model, 35 mA
200Ω
0Ω
10KHz
100KHz
1MHz
Figure 34: Off-Hook Impedance of Telephone “B” Compared with Model
9.3 On hook phone model
This on hook phone model was generated for testing in the TCM-ISDN noise environment. It could be used for
testing in other noise environments as well.
9.4
Zon
C3
C3 = 500 [pF]
Figure 3.5.tcm: On-hook impedance model
69
Table 8.1.tcm: Coefficients for calculation of primary line constants
for in-home wiring cables
Coeff i ci ent
0.5 mm untwi sted si ngl e- pai r wi re 0.4 mm twi sted pai r w ir e
r (m)
0.25 x 10
-3
0.2
x 10
-3
co (m)
0.75 x 10
-3
0.15 x 10
-3
µ r (H/m)
1
µ 0 (H /m)
4 π x 10
σ (mho/m)
5.8 x 107
tan δ
5.0 x 10
C ( F/m)
55 x 10
-7
-4
-1 2
Note: d = 2 ( r + co)
Loss [dB/ 3 m]
0.3
0.2
0.1
0.0
30
100
1000
Frequency [kHz]
Figure 3.6.tcm: Insertion loss characteristics of 3 m connecting cord
70
71