ITU-T J 55-1990 DIGITAL TRANSMISSION OF HIGH-QUALITY SOUND-PROGRAMME SIGNALS ON DISTRIBUTION CIRCUITS USING 480 kbit s (496 kbit s) PER AUDIO CHANNEL《使用480kbit s(496kbit s)的音频通道在分配.pdf

1、INTERNATIONAL TELECOMMUNICATION UNfON ITU-T TELECOMMUNICATION STANDARDIZATION SECTOR OF ITU TELEVISION AND SOUND TRANSMISSION J .55 (ex CMTT.718) (06/90) DIGITAL TRANSMISSION OF HIGH-QUALITY SOUND-PROGRAMME SIGNALS ON DISTRIBUTION CIRCUITS USING 480 kbitls (496 kbitls) PER AUDIO CHANNEL ITU-T Recomm

2、endation J.55 (Formerly Recommendation ITU-R CMTT.718) STDmITU-T RECMN J.55-ENGL 1770 48b257L Ob30078 54I FOREWORD The ITU Telecommunication Standardization Sector (ITU-T) is a permanent organ of the International Telecommunication Union. The ITU-T is responsible for studying technical, operating an

3、d 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, established the topics for study by the ITU-T Study Groups which, in their turn, p

4、roduce Recommendations on these topics. ITU-T Recommendation J.55 (formerly Recommendation ITU-R CMTT.718) was elaborated by the former ITU-R Study Group CMTT. See Note 1 below. NOTES 1 As a consequence of a reform process within the International Telecommunication Union (ITU), the CCITT ceased to e

5、xist as of 28 February 1993. In its place, the ITU Telecommunication Standardization Sector (IT-T) was created as of 1 March 1993. Similarly, in this reform process, the CCIR and the IFRB have been replaced by the Radiocommunication Sector (ITU-R). Conforming to a joint decision by the World Telecom

6、munication Standardization Conference (Helsinki, March 1993) and the Radiocommunication Assembly (Geneva, November 1993), the ITU-R Study Group CMTT was transferred to ITU-T as Study Group 9, except for the satellite news gathering (SNG) study area which was transferred to ITU-R Study Group 4. 2 adm

7、inistration and a recognized operating agency. In this Recommendation, the expression “Administration” is used for conciseness to indicate both a telecommunication O ITU 1990 All rights reserved. No part of this publication may be reproduced or utilized in any form or by any means, electronic or mec

8、hanical, including photocopying and microfilm, without permission in writing from the ITU. - STD-ITU-T RECMN J.55-ENGL 1990 48bZ.591 Ob30079 Y80 Recommendation J.55) DIGITAL TRANSMISSION OF HIGH-QUALITY SOUND-PROGRAMME SIGNALS ON DISTRIBUTION CIRCUITS USING 480 kbit/s (496 kbit/s) PER AUDIO CHANNEL

9、(1990) The CCIR, CONSIDERING a) further audio downstream processing; b) case of DBS; c) (Recommendation 606); 4 complying with CCIR Recommendation 660; e) Recommendation 1.412, that the distribution of high-quality sound-programme signals from the studio to transmitters and users does not require th

10、at in general more than one high-quality sound-programme signal has to be conveyed over distribution circuits as in the that a sampling frequency of 32 kHz is recommended for the digital transmission of high-quality sound-programme signals that some digital broadcasting applications may require syst

11、em performances that go beyond those offered by equipment that the high-quality sound-programme signals should interface the ISDN at the H1 level as specified in CCITT UNANIMOUSLY RECOMMENDS 1. that for distribution applications where a sampling frequency of 32 kHz is used and where a dynamic range

12、corresponding to more than i4 bits is required the coding method described in 8 1 of Annex I should be used on links providing a BER of less than 10-5: 2. according to the format described in 8 2 of Annex I; that for transmission at H12 level, two stereophonic programmes or four monophonic programme

13、s should be multiplexed 3. method and multiplexing format described in Annexe II should be used. North America) shall be made using Recommendation 660. that for cases where a higher ancillary data capacity is required and dedicated 2048 kbits links are available, the coding Note - Exchange of intern

14、ational digital sound-programme signals on networks with other hierarchical rates (1544 kbit/s in ) Formerly Recommendation ITU-R CMTT.718. Recommendation J.55 (06/90) - STD-ITU-T RECMN J.55-ENGL 1990 98b2591 Ob30L00 T22 W ANNEX 1 DIGITAL TRANSMISSION OF HIGH-QUALITY SOUND-PROGRAMME SIGNALS ON DISTR

15、IBUTION CIRCUITS USING 480 kbiffs AUDIO CHANNEL 1. Coding characteristics 1.1 Sampling fiequency The sampling frequency shall be 32 kHz. The sampling frequency tolerance shall be f5 x as specified in CCITT Recommendations G.732 and (3.733 for primary PCM multiplex equipment. This sampling frequency

16、is consistent with that indicated in CCIR Recommendation 606. 1.2 Coding method 16/14-bit floating point companding with a 2 ms coding block length (Le. 64 consecutive samples per block) and a 3-bit scale factor (transmitted by signalling in parity). The 16-bit samples of the sound signal are repres

17、ented in a 2 s complement format. The first bit of each word is the MSB (sign bit, O - +), and the last the LSB. Using a floating point system, the 16-bit samples are converted into 14-bit code words for transmission. A 3-bit scale factor applying to a block of 64 samples indicates how many of the b

18、its (O . . . 7) following the sign bit (yi) in all sampled words have the same value as the sign bit (Fig. la). The redundancy indicated by the scale factor does not need to be transmitted. Instead, the samples and their relevant information must be shifted towards the sign bits (floating-point syst

19、em). This allows the 15th and 16th bits of the source code words to be transmitted in the case of low signal amplitudes. The bits marked Zl to Z5 have not yet been assigned (Fig. 1 b). At the receiving end the scale factor is used to shift the bits of the samples back to their original value. This y

20、ields 16-bit samples and limits the effects of unrecognized bit errors to the amplitude range indicated by the scale factor. 1.3 Sample error-protection After having applied the floating-point technique for reducing the bits per sample from 16 to 14, a parity bit is calculated on the seven most sign

21、ificant bits of each sample, such that the number of “1”s in the group of the seven protected bits and the parity bit is odd. Thus, the best protection against clicks due to bit errors is guaranteed provided concealment in the form of the arithmetic mean value of the samples adjacent to the faulty s

22、ample is performed at the receiving end. If the concealment is done after the 14/16 bit reconversion, errors in least significant bits can also be concealed in an optimum way. 1.4 Ancillary data A data capacity of 4 kbit/s per channel is transmitted by signalling in parity. 1.5 Signalling in parity

23、Chambers, 19851 Signalling in parity is achieved by transmitting the parity bits of an odd number of successive samples without inversion or after inversion, depending on the bit to be signalled. Inversion has to be done, if the bit to be transmitted is 1. The signalling technique used for the scale

24、 factor, its related parity bit, and the ancillary data is based on majority-decision logic at the receiver: for each channel, it processes twelve groups of five consecutive samples (three for the scale factor, one for its parity bit and eight for the ancillary data) to recognize simultaneously the

25、odd or even parity of the sample and the data signalled in parity (see Fig. 2). A similar process is used to recognize the synchronization ofthe 2 ms frames through one group of four consecutive samples. 1.6 Total bit rate With the parameters mentioned above the total bit rate required for one monop

26、honic channel is 480 kbit/s 32 kHz x (14 + 1) bit. 2 Recommendation J.55 (06/90) In N *i N o -0- o - o - - o cor o .- 03: o or c o o - F - c O0 -c 00, u O000 - r - - c - Fro O0 oui. U - VI D B a .- a 5 YI u 51 01 . . . dol-se Recommendation 5.55 (06/90) 3 STD=ITU-T RECMN J-55-ENGL I770 iBb257I Ob301

27、02 8T5 W - PPPP PPPPP PPPPP PPPPP PPPPP PPPPP P PPPPP PPPPP PPPPP Sync Scale factor Ancillary data - - - r Bit So Bit SI Bit s2 Parity bit Bit I Bit 8 O101 00000 11111 00000 11111 00000 o 11111 00000 00000 - - e u- - PPPP PPPPP PPPPP PPPPP PPPPP PPPPP P PPPPP PPPPP PPPPP 2 rns 4 1 5 10 15 20 25 30 5

28、0 55 60 64 FIGURE 2 - Signalling in parity coding for one audio channel a) Parity bits (P denotes the parity bit associated to each individual audio sample). 6) Sync pattern, scale factor, parity bit for the scale factor and ancillary data bits. The parity bit for the scale factor should be c) Modif

29、ied parity bits. even as shown in the example. do2-sc 2. Frame structure for transmission Within the ISDN, a usable bit rate of 1920 kbit/s for one multiplex frame is available according to CCITT Recommendation G.704. In order to ensure the compatibility between the signals transmitted at 480 kbit/s

30、 according to this Recommendation with those transmitted at 384 kbits according to CCITT Recommendation G.737, the bits of each audio channel should be allocated according to the present proposal depicted in Fig. 3. The bits of each audio channel should be transmitted as a group of 30 bits within a

31、half-frame. Moreover, the bits of each sample are interleaved such that the most significant bit (MSB) is followed by the least significant bit (LSB). etc. (see Fig. 3). This arrangement of bits has proven to be a good protection against double errors, which otherwise cannot be recognized by a singl

32、e parity bit and thus cannot be concealed. On this basis, channels using 480 kbit/s and 384 kbit/s can be combined according to Table I: TABLE I Number of channels I II1 I 2 2 I I I l 5 4 Recommendation J.55 (06/90) P b5 b7 b9 bll bl bO bl b2 b3 b4 b5 b6 b7 b8 b9 bll b13 i !i!, 1 1 1 I , I I I I ,b1

33、b11bZ,b3,b4, ,b6, ,b8 1 , 9 p 3 I , Interleaved 3 6 9 12 15 19 22 25 28 31 FGURE 3 A, F3, C, D: monophonic channel 480 kbitls a, b. c, d, e: monophonic channel 384 kbitls MSB: most signiftcaut bit LSB least significant bit : not usable for sound transmission d03-sc Two monophonic channels of the sam

34、e coding scheme can be combined to form a stereophonic channel. In some realized systems for 384 kbit/s channels only the combinations a, b and c, d are possible to form stereophonic channels. REFERENCES CHAMBERS, J. P. 1985 Signalling in parity: a brief history. British Broadcasting Corporation, BB

35、C RD 1985/15. Recommendation J.55 (06190) 5 ANNEX II DIGITAL TRANSMISSION OF HIGH-QUALITY SOUND-PROGRAMME SIGNAL ON DISTRIBUTION CIRCUITS USING 496 kbit/s PER AUDIO CHANNEL 1. Coding characteristics 1.1 Samplingj%equency See Annex I. 1.2 Coding method See Annex I. 1.3 Sample error-protection See Ann

36、ex I. 1.4 Ancillary data 12 kbits per channel are available for transmission of ancillary data. 1.5 Signalling in parity Chambers, 19851 Signalling in parity is achieved by transmitting the parity bits of an odd number of successive samples without inversion or after inversion, depending on the bit

37、to be signalled. Inversion has to be done, if the bit to be transmitted is “1”. The signalling technique used for the scale factor is based on majority-decision logic at the receiver: for each channel, it processes three groups of twenty-one consecutive samples to recognize simultaneously the odd or

38、 even parity of the sample and the data signalled in parity (see Fig. 4). 2 mo r 1 2 3 4 5 6 7 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 . . 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 PPPPPPPPPPPPPPPPP PPPPPPPPPPPPPPPPPPPPPPPPPPP 1 2 3 4 5 6 7 8 9 . . 18 19 20 21 0) I 1 b/ iiio

39、ioiiaioiioiioioiio 01101101101101100 1 I FIGURE 4 - Signalling in parity coding for one audio channel a) Parity bits (P denotes the parity bit associated to each individual audio sample b) Scale factor bits c) Modified parity bits 1.6 Synchronization 4 kbit/s are used for synchronization of the floa

40、ting point companding block. 1.7 Total bit rate With the parameters mentioned above the total bit rate required for one monophonic channel is 496 kbits 32 kHz x (14 +i) bits + 12 kbit/s + 4 kbits. 6 Recommendation J.55 (06/90) STD-ITU-T RECMN J-55-ENGL 1770 48b2571 Ob30105 504 2. Frame structure for

41、 transmission The frame structure for transmission is based on an interface operating at 1024 kbitis, Two of the 496 kbit/s channels are combined with a frame alignment signal to form a multiplexed signal of 1024 kbits, which constitutes the bit rate of the interface. The structure of the frame is a

42、lmost the same as that specified in CCITT Recommendation G.704 for the primary hierarchical level of 2048 kbit/s. It should be noted that the frame repetition rate is 4 kHz instead of 8 kHz. The frame format is shown in Fig. 5. Left channel Right channel Bit O Z most significant bit Bit 13 I: least

43、significant bit P 1 parity bit Synchronization and data block 8 30 30 30 30 8 30 30 30 30 J I I I I I I I I l, - 22 Left Right channel channel EE mm -5-5 FIGURE 5 - Fmmefor the 1024 kbWs signal with a length of256 bits and a 4 Wz repetition mte SL, SR: synchronization signai for oang point Compandin

44、g dffisc For each channel, the synchronization of the floating point companding block (64 samples carried in 8 frames) is achieved by means of an 8-bit synchronization word. Two synchronization words are defined: sy = 00011011 SI . . .sg and the inverted form - ;sr = 11100100 - - ;SI . . . . . . ;Sg

45、 For a stereophonic pair the synchronization word used is Sy for the left channel and -;Sy for the right channel. Recommendation 5.55 (06/90) 7 STD-ITU-T RECMN J-55-ENGL 1990 W YBb259L Ob301Ob L(L(0 For monophonic applications, the synchronization word used for each channel is alternatively S, and -

46、;SY (see Fig. 6). Note - The relation between SL and SR should be as shown. Transmission order I Note - For monophonic application there is no need for a fixed relation between SL and SR. FIGURE 6 - Tmnsrnission ofthe synchronization words d06sc By way of synchronous insertion into a 2048 kbis frame

47、 structure, it is not necessary to transmit the frame alignment signal (FA) which is contained in the frame structure shown in Fig. 5. Thus, for one channel the net bit rate for transmission is still 496 kbit/s. In cases where dedicated 2048 kbis links are available, up to four 496 kbits channels ca

48、n be combined in one multiplex frame according to CCITT Recommendation G.704. In order to ensure compatibility between the signals transmitted at 496 kbit/s according to this Annex II with those transmitted at 384 kbits according to CCITT Recommendation G.737, the available transmission capacity sho

49、uld be allocated as shown in Fig. 7 CCIR, 1982-86al. In Fig. 7 each block L I, R I, L II, R II comprises 60 bits corresponding to a capacity equivalent to 4 samples. Each block results from bit- and sample-interleaving over two consecutive Recommendation G.704 frames. 8 Recommendation J.55 (06190) STD*ITU-T RECMN J.55-ENGL 1940 Li8b2571 Ob30107 387 W 1) O 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 IS 19 20 21 222324 25 28 27 20 2930 31 11111111111111 1111111l111111 Mixed allocation