1、CCITT FASCICLE 111-4 89 4862593 0503763 6 = 2 Transmultiplexer application The application on transmultiplexers for the interconnection of digitaI and analogue networks is illustrated in Supplement No. 28. Reference CCIT Recommendation Recommendations concerning translating equipments, Vol. III, Rec
2、. G.233. i Recommendation G.792 CHARACTERISTICS COMMON TO ALL TRANSMULTIPLEXING EQUIPMENTS (Geneva, 1980; further amended) The CCIT, recommends dation G.791. that the characteristics below be respected by all the transmultiplexing equipments defined in Recommen- Recommendation 0.133 contains informa
3、tion about test equipment. Account should be taken of the measurement accuracy provided by test equipment designed in accordance with that Recommendation. The following specifications are based on ideal measuring equipment. Therefore, they do not include any margin for measurement errors. To avoid l
4、evel errors produced as a result of the use of test frequencies which are sub-multiples of the PCM sampling rate, the use of integer sub-multiples of 8 kHz should be avoided. Where a nominal reference frequency of 1020 Hz is indicated (measurement of attenuation/frequency distortion and adjustment o
5、f relative levels), the actual frequency should be 1020, +2 to -7 Hz in accordance with Recommendation 0.6 18. 1 Coding law Transmultiplexers should satisfy Recommendation G.711, 0 3. 2 Sampling rate of PCM channels The nominal sampling rate of PCM channels is 8000 Hz f 50 lop6 according to Recommen
6、da- tion G.711, 8 2. 3 Amplitude limitation of PCM channels In accordance with Recommendation G.711, 4 4, the theoretical load capacity of PCM channel is +3.14 dBmO for the A-law and +3.17 dBmO for the F-law. 4 Accuracy of the analogue virtual carriers The analogue virtual carriers should satisfy th
7、e Recommendation cited in i. 5 Saturation level at the input of the analogue group The transmultiplexers should be able to accept at their analogue inputs, levels corresponding to the equivalent peak powers defined in Table 3/G.223 5 (for example, + 19 dBmO for a group and + 20.8 dBmO for a supergro
8、up). Fascicle 111.4 - Rec. G.792 575 -. CCITT FASCICLE 111.4 89 4862593 O503764 B Note - Attention is drawn to the possibility of using a transmultiplexer on the interpolated side of a digital speech interpolation (DSI) device. Given an interpolation rate of 2, this would lead to equivalent peak pow
9、ers of 19.5 dBmO for TMUX-P and 21.2 dBmO for TMUX-S (see Table 3/G.223 5). 6 Methods of measuring quality in the audio band The various possible methods of measuring quality characteristics in the audio band are indicated in Figure 1/G.792. Category A B C D E F Measuring method TMUX Analogue carrie
10、r d- frequency TMUX I-b PCM multiplexed signal Group modulator* Channel 7 modulator s -Audio frequency s-s-+ TMUX FIT TIF I I I I L,J R -I, R A * Group modulator is not required for the TMUX-P TMUX - Analogue carrier frequency PCM multiplexed signal Analogue carrier Analogue carrier multiplex freque
11、ncy signal frequency Analogue PCM multi- -+ plexed t signal frequency - FIT -+ PCM multi- carrier plexed - TIF + signal TIF TDM-to-FDM conversion FIT FDM-to-TDM conversion S Sender R Receiver FIGURE 1/G.792 Block diagrams of measuring methods for transmultiplexers Remarks Measurement in analogue car
12、rier frequency Measurement in PCM multiplexed level Measurement in audio frequency Measurement for one direction in PCM multiplexed level and analogue carrier frequency Measurement in analogue carrier frequency Measurement in PCM multiplexed level CCnT- 2781 I 576 Fascicle III.4 - Rec. G.792 CCITT F
13、ASCICLE 111.4 89 = 4862591 0503765 T When method B cannot be used because it requires digital signal generators and analyzers, which certain Administrations do not yet possess, method C can be used provisionally looping of the digital ports, use of the terminals of auxiliary analogue channels (and p
14、ossibly group modulators), assumption of the additivity of impairments and deduction of the impairments at the terminals of the channels (and possibly modulators) previously measured. Method D corresponds in fact to four possible methods, depending on whether the emission of the test signal and its
15、detection takes place on the analogue side or the digital side. Methods E and F are used for crosstalk measurements. For the sake of the convenience and precision of the measurements, it is desirable that the regulation, when included in the transmultiplexer, can be blocked with a gain equal to unit
16、y. The specifications in $9 7 to 23 assume such blocking. 7 Attenuation distortion in the voice-frequency band as a function of frequency The measuring method is method A. The variation of the attenuation of each channel of a transmultiplexer as a function of frequency must remain within the limits
17、of the mask in Figure 2/G.792. The level of emission is -10 dBmO; the reference frequency is 1020 Hz. 8 Group delay 8.1 Absolute value of the group delay The measuring method is method A. The absolute value of group delay defined as the minimum value of group delay in the speech band 300-3400 Hz sho
18、uld remain less than 3 ms for all the channels of a transmultiplexer. Note - When the transmultiplexer is used for satellite digital communication at the earth station the minimum value of the group propagation time in the audiofrequency band may be increased from 3 ms to 6.5 ms. In all other cases,
19、 the value of 3 ms should be complied with. 2001 400 600 1020 2000 2400 3000 3400 3600 4000 Hz 300 T161011Od9 Frequency (f) FIGURE 2E.792 Attenuation distortion mask as a function of the frequency to be observed for all channels of a transmultiplexer Fascicle III.4 - Rec. G.792 577 r -. CCITT FASCIC
20、LE 111.4 89 m 4862593 05037bb 3 m 8.2 Group-delay distortion The measuring method is method A. The group-delay distortion should not exceed the limits of the mask in Figure 3/G.792. The minimum group delay is taken as a reference; the power level at the input is O dBmO. O 500 600 1000 2000 2600 2800
21、 Hz T1810120d9 Frequency (f) FIGURE 3/G.792 Mask of group-delay distortion as a function of frequency 9 Noise 9.1 Idle channel noise, with all channels-idle The measuring method is method B. When a PCM signal corresponding to amplitude O for the plaw and the number 1 for the A-law in all channels of
22、 the transmultiplexer is applied to the digital input of the transmultiplexer, the psophometric noise measured over any channel at the digital output should not exceed -65 dBmOp. The measurement is conducted in the presence of pilots. 9.2 Channel noise, with all channels loaded except the one measur
23、ed The measuring method is method A. In this case an intermodulation measuring set-up using the white noise method is employed, as described in the Recommendation cited in 6. The level of emission of the noise signal being equal to the conventional load of the FDM signal considered (the Recommendati
24、on cited in 7: 3.3 dBmO for the group, 6.1 dBmO for the supergroup), the noise measured in any given measuring slot should not exceed -62.5 dBmOp (Le., -60 dBmO in a 3100 Hz band). The centre frequencies of the specified measuring slots (CCIT Recommendation G.230 8 and CCIR Recommendation 482 9) app
25、licable to the transmultiplexers are: - - This measurement is carried out without emitting pilots or out-of-band signalling. Note - Attention is drawn to the possibility of using a transmultiplexer on the interpolated side of a digital speech interpolation (DSI) device. Given an interpolation rate o
26、f 2, this would lead to conventional loads of 4.5 dBmO for TMUX-P and 7.3 dBmO for TMUX-S (see Table 2/G.223, 7). for the base group: 70 and 98 kHz for the base supergroup: 394 and 534 kHz. 578 Fascicle 111.4 - Rec. 6.792 CCITT FASCICLE 111.4 89 m 4862593 0503767 3 m 9.3 Single frequency noise outsi
27、de the band 300-3400 Hz The measuring method is B. When a PCM signal corresponding to amplitude O for the p-law and amplitude 1 for the A-law in all channels is applied to the digital input of the transmultiplexer, the noise over any frequency should not exceed -50 dBmO with the exception of the fre
28、quency of 80 Hz where it should not exceed -40 dBmO. 9.4 Idle noise in the PCM - FDM direction all channels idle The measuring method is method D. A PCM signal, amplitude O for the p-law and 1 for the A-law is applied at the digital inpuf of the transmultiplexer in all channels. The power of the noi
29、se measured at the analogue output in any channel must be less than -70 dBmOp. Note - White noise is assumed, and to take account of the psophometric weighting, the measurement can be made in a band of 1740 Hz, centred on the odd multiples of 2 kHz. The measurement may be difficult in certain channe
30、ls due to the presence of pilots. 10 Intermodulation The measuring method is method A. If two sine-wave signals of different frequencies fi and fi belonging to the band 300-3400 Hz of the channel considered, having no harmonic relation and of equivalent levels in the -4 to -21 dBmO range, are applie
31、d simultaneously to the analogue ports of the transmultiplexer, there should be no intermodulation product of the type 2J; - fi of a level higher than -35 dB with respect to the level of one of the two input signals. 11 Total distortion including quantizing distortion The measuring method is method
32、B (or provisionally method C). If method B is used, the test signal is generated digitally and is therefore affected by theoretical quantizing A choice between the two following methods is recommended: distortion. Method 1 The signal-to-total distortion ratio measured according to method 1 described
33、 in 0 8 of Recommenda- tion G.712 should respect the mask of Figure 4/G.792. The mask is to be complied with by all channels of the transmultiplexer. Method 2 With a sine-wave signal at a frequency between 700 and 1100 Hz or 350 and 550 Hz (e.g. 420 f 20 Hz) (except for submultiples of 8 kHz) being
34、applied in the channel concerned at the digital input of the transmulti- plexer, the ratio of signal-to-total distortion power, measured with appropriate noise weighting (see the Recom- . mendation cited in io), should be below the limits of the mask represented in Figure VG.792. The mask is to be c
35、omplied with by all the channels of the transmultiplexer. 12 In-band spurious signals The measuring method is method A. The transmultiplexers must meet the provisions of Recommendation G.712, 0 9. 13 Variation of gain with the input level The measuring method is method A, the pilots being present at
36、 the analogue input. With a sine-wave signal at a frequency between 700 and 1100 Hz (except for submultiples of 8 kHz) and a level between -55 and +3 dBmO being applied in the channel concerned at the analogue input of the transmultiplexer, the variation of gain with respect to its value for an inpu
37、t level of - 10 dBmO should remain between the limits of the mask shown in Figure 6/G.792. The mask is to be complied with by all channels of a transmultiplexer. Fascicle 111.4 - Rec. 6.792 579 :- CCITT FASCICLE 111.4 89 W 4862593 0503768 5 W dB 33,9 33.2 1 30.2 1- .- o 26.3 E 25.0 5 O .- .- x U m 4
38、- O Y O - ; 10.0 C m v, O -55 dB 32,s 31.8 29,l 24.2 0 24.9 L C O O U 3 O Y O Y m c m Yj .- t: 4- In .- - - 9.2 o -55 -40 -34 -27 Input level al Method C -40 -34 -6 -3 Input level 61 Method B FIGURE 4lG.192 Signai-to-total distortion ratio as a function of the input level according to method 1 (Reco
39、mmendation G.712, p 8) CClTT-27W1 580 Fascicle 111.4 - Rec. 6.792 dB 33. c 32,3 .e 25.5 I 2 20.5 .- ti w C 5 U m O Y Y - 4- - m m v) .- O d 3F! 9 24.9 e C O O U m t8 e 19.9 4- .- - ri! O *: m rn v) - .- O Input level al Method C dem0 -45 -40 -30 Input level bl Method B FIGURE 5/G.792 Signai-to-total
40、 distortion ratio as a function of the input level according to method 2 (Recommendation G.712, 5 8) O dBmO CCilT-27841 Fascicle III.4 - Rec. 6.792 581 I CCITT FASCICLE 111.4 89 48b259L 0503770 3 W d0 +3 +l Q 4 - +0,5 c O .- .- Po i? L m -0.5 -1 .- -3 V/ /4 fi 4 fi 4 4 fi /fi fi /A 7 I b -55-50 -40
41、-10 +3 dBmO Input level (S) CCITT-27850 FIGURE 6/G.792 Gain variation as a function of input level S, Method 2 in Recommendation G.712, $j 10 (sinusoidal test signal) 14 Crosstalk For measuring crosstalk, two transmultiplexers must be connected back-to-back (methods E and F). There - - - - are two p
42、ossible configurations and four possible measurements (see Figure 7/G.792): far-end crosstalk digital to digital (see Note 1) near-end crosstalk digital to digital (see Note 1) far-end crosstalk analogue to analogue near-end crosstalk analogue to analogue 14.1 Intelligible crosstalk When a sine-wave
43、 signal between 700 and 1100 Hz and with a level of O dBmO is injected in any channel on the digital or analogue side of the transmultiplexer, the crosstalk ratio between the signal channel and any other channel must be greater than 65 dB for any of the four crosstalk contributions identified above
44、(see Note 2). 14.2 Unintelligible crosstalk When a conventional telephone signal according to Recommendation G.227 i i, is injected in any channel on the digital or analogue side of the transmultiplexer, at a level of O dBmO, the level of crosstalk measured in any other channel for any of the four c
45、rosstalk contributions identified above must be below -60 dBmOp (see Note 3). Note 1 - In this configuration, the two transmultiplexers are connected at the level of the analogue FDM signal and there will generally be a problem of level adaptation between the send and the receive sides. This can be
46、solved with the use of attenuators or amplifiers of appropriate gain. Attention must be given on the risk of introduction of additional crosstalk in these complementary devices. It should be desirable to include the level adaptation facilities in the transmultiplexer itself. r - 582 Fascicle 111.4 -
47、 Rec. 6.792 CCITT FASCICLE 111.4 89 48b2.591 0503771 5 W Note 2 - In order to overcome fundamental gain enhancement effects associated with PCM encoders, which can mask the true crosstalk, measuring methods using activating signals based on those defined in Recommendation G.712 can be used. Note 3 -
48、 Recognizing the difficulty of generating conventional telephone signals according to Recom- mendation G.227 in a suitable format for insertion into either the analogue or digital input to the transmulti- plexer, it shall be adequate to demonstrate, via suitable single frequency crosstalk tests, tha
49、t the intent of the above specification is met, without actually using a conventional telephone signal. - Perturbating signal b TMUX2 + Measure ? Act vat ing -* signal TMUXl a) Far-end crosstalk measurement Pertubating signal - b Activating 4 signal TMUX 1 TMUX2 4 Measure 4 CClIl49U bI Near-end crosstalk measurement FIGURE 7/G.792 Measurements of crosstalk with methods E and F 15 Go-to-retth crosstalk For measuring go-to-return crosstalk, two transmultiplexers must be connected back-to-back (methods E - - When a sine-wave signal between 3
