CEPT T TR 01-03 E-1986 Frame Structure at 64 kBit s for Multimedia Applications《多媒体应用64kBit 秒框架结构》.pdf

1、- L- - T- CEPT T/TR*O3-03*E b 232b434 0030345 b = Page E 1 Recommendation T/TR 01-03 (Odense 1986) FRAME STRUCTURE AT 64 KBIT/S ,FOR MULTIMEDIA APPLICATIONS Recommendation proposed by Working Group T/WG 12 cTransrnission (TR) Text of the Recommendation revised adopted by Commission “Telecommunicntio

2、ns”. “The European Conference of Posts and Telecommunications Administrations, Considering - that digital networks are now available which provide for the transparent transmission of 64 kbit/s from user to user, - that there is a need for new services making extensive use of those networks and that

3、multimedia services may be important for their development, Recognising - the need to multiplex on the same 64 kbit/s channel several flows of information relevant to a multimedia service o and to control these flows in a robust way even in a multipoint conguration and using transmission media (such

4、 as satellites) having considerable delay, - that a procedure which does not require a return link is preferable, e.g. for broadcasting applications, - that the ISDN does not prevent the use of a single 64 kbit/s channel for multimedia end-to-end services and that in the future it might provide supp

5、lementary means to transmit different flows of information in switched point-of-point calls (Note 11, - that the network does not provide any means of intenvorking of a part of the information flows relevant to a multimedia application with telecommunication services using the total channel capacity

6、, - that the subdivision of a 64 kbit/s channel is not recognised by the network and hence is a terminal matter, Recommends the use of the frame structure described hereafter for multimedia services in a single 64 kbit/s channel.” Note 1: The relevant prot Edition of September 1, 1986 cols re fc fur

7、ther study. / 1. 2. 2.1. 2.2 2.3 CEPT T/TR*01-03*E 6 2326414 O010346 T/TR 01-03 E Page 3 BASIC PRINCIPLE The 64 kbit/s channel is structured into 8-bit octets transmitted at 8 kHz. The least significant bit (bit 8) (Note 1) of each octet conveys a sub-channel of 8 kbit/s . This sub-channeI, called S

8、ervice Channel (SC), provides end-to-end signalling and consists of three parts (see Figure 1 (T/TR 01-03): - Frame Alignment Signal (FAS). This signal aliows to structure the 64 kbit/s channel into frames of 80 bytes each and multiframes (MF) of 16 frames each. Each multiframe is divided into two 8

9、-frame submultiframes (SMF). In addition to framing and multiframing information, control and alarm information may be inserted, as weli as error check information to control end-to-end error performance and to check frame alignment validity (this last point is still under study). FAS can be used to

10、 derive octet timing when it is not provided by the network. - Bitrate Allocation Signal (BAS). This signal allows the transmission of Codewords to describe the structure of the residual 56 kbit/s channel, as weil as, if necessary, the structure of the primary rate multiplex in which the basic 64 kb

11、itls channel is inserted, in the case of n x 64 kbit/s multimedia service as videoconference or videophony. Other possibilities are under study, like describing sub-multiplexing, rate adaption and low-layer protocols of the data channels. - Application Channel (AC) or Service Dedicated Information.

12、This channel allows transmission of binary information or the insertion of message-type data channel) (such as telematic information) up to 6,400 bit/s. The remaining 56 kbit/s channel, carried in bits 1-7 of each byte, may convey a variety of signals in the framework of a multimedia service, under

13、the control of the BAS and possibly the AC. These signals are carried as flows at n x 8 kbitls. Some examples follow: -Sound, coded at 56 kbit/s using truncated PCM (A-law or mu-law of CCITT Recommendation G.7111). - Sound, coded at 32 kbit/s (ADPCM according to CCITT Recommendation G.721 2) and dat

14、a at 24 kbit/s or less. - Sound, coded at 56 kbit/s with a bandwith 50-7,000 Hz (sub-band ADPCM according to CCITT Draft Recommendation G.72X). The coding algorithm is also capable to work at 48 kbit/s (bitrates of 40 or 32 kbit/s with a reduced quality are under study). Data can then be dynamically

15、 inserted at 8, 16 or 24 kbitls. - Still pictures coded at 56 kbitls. - Data at 56 kbit/s (e.g. file transfer for communicating personal computers). FRAME ALIGNMENT General An 80-byte frame length produces an 80-bit word in the Service Channel. Those 80 bits are numbered 1-80. Bits 2-8 of the Servic

16、e Channel in every other frame contain O011011 and are the Frame Alignment Word (FAW). Those bits are completed by bit 2 in every alternate frame (those not containing the FAW) to form the complete Frame Alignment Signal. So a pattern similar to the one in CCITT Recommendation G.704 3 is used (see F

17、igure 2 (T/TR 01-03). Description of the CRC procedure Depending on service and network requirements which are not yet defined, there may be a need to evaluate end-to-end quality by the means of a CRC code similar to the one used in CCITT Recommendation G.704 3. The use of bits E and Cl-C4 for compu

18、ting parity bits on blocks of information for this purpose is under study. The CRC may also be used to validate or invalidate frame alignment. Multiframe structure Each multiframe contains 16 consecutive frames numbered O to 15 thus making 2 submultiframes of 8 frames each. The multiframe alignment

19、signal is located in bit 1 of frames 1-3-5-7-9-11 and contains 001011. Bits 1 of frames 0-2-4-6-8-10-12-13-14-15 are reserved. Their value is provisionally ked to O (Figure 3 (T/TR 01-03). Note I: For networks where this bit 8 is needed for network purposes and therefore is not available, the second

20、 least significant bit (bit 7) might be used. Edition of September 1, 1986 _-_ CEPT T/TR*O1-03*E 6 M 2326414 0010347 T 2.4. 2.5. 3. 4. 4.1. 4.2. T/TR 01-03 E Page 4 Loss and recovery of frame alignment Frame alignment will be assumed to have been lost when three consecutive frame alignment signals h

21、ave been received with an error. Frame alignment will be assumed to have been recovered when the following sequence is detected : - for the rst time, the presence of the correct frame alignment word; - the absence of the frame alignment word in the following frame detected by verifying that bit 2 is

22、 a 1 ; - for the second time, the presence of the correct frame alignment word in the next frame. When frame alignment is lost, bit 3 (A) of the next odd frame is set to 1 in the transmit direction. Loss and recovery of multiframe alignment Multiframe Alignment is used to validate teh Bitrate Alloca

23、tion Signal (see paragraph 3.). The criteria for loss and recovery of multiframe alignment described underneath are provisional and have to be confirmed by experimentation. As the search of the Multiframe Alignment Word is located on bit 1 of each odd frame, multiframe alignment is easier. Multifram

24、e alignment will be assumed to have been lost when three consecutive multiframe alignment signals have been received with an error. It will be assumed to have been recovered when the multiframe alignment signal has been received with no error in the next multiframe. BITRATE ALLOCATION SIGNAL (BAS) T

25、he Bitrate Allocation Signal (BAS) occupies bits 9-16 of the Service Channel in every frame. It is repeated 8 times along the same submultiframe. A majority decision (5 out of 8) allows the validation of BAS. The validated value of BAS applies to the next submultiframe. A change in configuration can

26、 then occur at submultiframe rate, i.e. every 80 ms. In case of loss of frame or multiframe alignment, the BAS should keep the same value as the previously validated one, until frame and multiframe alignment are recovered. The encoding of BAS is made in accordance with the attribute method: The thre

27、e first bits (bits 9-10-1 1) represent the attribute number relevant to the description of the configura- tion. Up to now, two attributes are defined: the Audio Coding (attribute 000) and the Transfer Rate (attribute 001). Other attributes are under study. Bits 12-16 of each frame represent the valu

28、es of the attributes and describe the used configuration. Figure 4 (T/TR 01-03) gives the coding of BAS for attribute 000. Figure 5 (T/TR 01-03) gives a possible assignment of codes for attribute O01 and requires further study. APPLICATION CHANNEL (AC) It occupies bits 17-80 of the Service Channel i

29、n each frame, making a user-available bitrate of 6.4 kbit/s. According to the application, different kinds of information may be inserted herein. In particular, informa- tion concerning forward error correction or end-to-end encryption which both depend on the application, could take place in the Ap

30、plication Channel. Binary information Each bit of the Application Channel may represent a binary digit, repeated 100 times per second. If odd and even frames are identified, each bit may represent two digits, transmitted at 50 Hz each. If multiframing is used, each bit may represent 16 binary digits

31、, transmitted at 6.25 Hz. An example of this kind of information is in teleconference the use of a bit to synchronise the encoder clock on the receive clock, or to indicate the microphone number, or to signal the use of the graphics mode, etc. Synchronous message-type channel As each bit of the Appl

32、ication Channel represents a bitrate of 100 bit/s, any synchronous channel working at n x 100 bit/s may be inserted in the Application Channel. An example is, in videoconference, the Message Channel at 4 kbit/s which is used for multipoint management. Another possibility is the insertion of data cha

33、nnels at a bitrate of the hierarchy defined in CCITT Recommendation X.l 4, according to CCITT Recommendation X.30 5/1.461 6: “Support of X.21 7 and X.2lbis SI based DTEs by an ISDN”. The present frame structure is coherent with the X.30/1.461 frame structure in a double way: - It has the same length

34、 (80 bits by bearer channel at 8 kbit/s). - It needs 63 bits per frame (17 bits are used for framing information not to be transmitted), which fits into the 64 bits available in this frame structure. Edition of September I, 1986 CEQT T/TR*OL-O3*E b I 232bYL4 0030348 L I T/TR 01-03 E Page 5 4.3. Asyn

35、chronous message-type channel In case of asynchronous terminals, X.l 4 hierarchy is relevant. The existing European standard is the ECMA standard ECMA-TAxx “Bitrate adaptation for the support of synchronous and asynchronous terminal equipment using the V-series interfaces on a PSTN”. This standard a

36、lso uses the same 80-bit frame structure as X.30 5/1.461 6 presented above. The Application Channel will therefore allow adoption of this ECMA standard if needed. 4.4. Error correction and encryption When needed, forward error correction and encryption information may take place in the Application C

37、hannel. The bitrate and the protocol to be used will depend of the application. 5. ACCESS TO DATA IN THE 56 KBlT/S CHAMVEL The 56 kbit/s channel, when used with sound coded according to CCITT Recommendation G.721 2 or Draft Recommendation G.72X, allows the static or dynamic allocation of data channe

38、ls at n x 8 kbitls. The access to those channels could be done according to standardised procedures (e.g. 1.461 6, 1.462 9, 1.463 lo or ECMA-T Axx.). The 56 kbit/s channel may also be used to transmit data only. In some applications, it may be desirable to merge the Application Channel with the data

39、 channel in order to have a single user-data path. 6. CONCLUSION A frame structure for multimedia applications in a single 64 kbit/s channel has been defined which makes the best use of the characteristics and properties of the audio compression algorithm, of the transmission framing structure and o

40、f the existing CCITT Recommendations. It offers several advantages : - It takes into acount CCITT Recommendations as G.704 3, X.30 5/1.461 6, etc., and may make use - It is simple, economic and flexible. It may be implemented on a simple microprocessor with very well - It is a synchronous procedure.

41、 The exact time of a configuration change is the same in the transmitter - It needs no return link, since a configuration is signalled by a continuously transmitted codeword. - It is very secure in case of transmission errors, since the BAS is repeated 8 times with a majority logic - It allows the c

42、ontrol of a higher multiplex configuration, into which the basic 64 kbit/s channel is inserted - It can be used to derive byte sync in IDNs when it is not available. - It can be used in multipoint, where no dialogue is needed to negotiate the use of a data channel. - It provides a variety of data bi

43、trates (from 6.25 bit/s up to 64 kbitls) to the user. of existing hardware or software. known hardware principles. and the receiver. Configurations can be changed at 80 ms intervals. decision. (case of n x 64 kbit/s multimedia services as videoconference). Edition of September 1, 1986 r- I CEPT T/TR

44、*O1-03*E 86 m 2326414 0010349 3 m - TITR 01-03 E Page 6 34 ss bb cc hh aa nn nn ee 11 # 34 uu - Bit number 1 Mi Note 1 1 2 3 4 5 6 I 8 O O 1 1 O 1 1 Frame Alignment Word S b C h a n n e 1 # 1 U - Mi Note 1 2 1 A E c1 c2 c3 c4 Note 2 Note 3 Note 4 S b C h a n n e 1 # 2 U - 5 S b C h a n n e 1 # 5 u -

45、 .6 I ss bb cc hh aa nn nn ee 11 # 61 uu - - 8 FAS BAS AC Octet number Figure 1 (T/TR 01-03). Frame structure. FAS: Frame Alignment Signal (Note I). BAS: Bitrate Allocation Signal. AC: Application Channel. Note 1: The block termed as FAS contains also other information than for frame alignment purpo

46、ses. Successive frames Even frames (those containing FAW) Odd frames Figure 2 (T/TR 01-03). Assignment of bits 1-8 of the Service Channel in each frame. Note 1; Mi - Bits reserved for multiframing. Note 2: Bit used to avoid simulation of FAW by a frame-repetitive pattern. Note 3: A - Remote Alarm In

47、dication. This bit is set to 1 to signal the following fault conditions: - Failure of power supply. - Failure of equipment (to be further studied). - Loss of frame alignment. Note 4: The use of bits E and Cl-C4 is under study (see 2.2.). Edition of September 1, 1986 CEPT T/TR*OL-O3*E b m 232b4L4 OOL

48、0350 T m TJTR 01-03 E Page 7 Multiframe Sub-Multiframe (SMFI SMF 1 SMF 2 Frame O 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 1 R1 O R2 O R3 1 R4 O R5 1 R6 1 R7 R9 R8 R10 Bits 1 to 8 of 2 O 1 O 1 O 1 O 1 O 1 O 1 O 1 O 1 3 O A O A O A O A O A O A O A O A ie Service Channel in every frame 4 1 E 1 E 1 E 1 E 1 E

49、 1 E 1 E 1 E 5 1 c1 1 c1 1 c1 1 c1 1 c1 1 CI 1 CI 1 CI 6 O c2 O c2 O c2 O c2 O c2 O c2 O c2 O c2 I 1 c3 1 c3 1 c3 1 c3 1 c3 1 c3 1 c3 1 c3 8 1 c4 1 c4 1 c4 1 c4 1 c4 1 c4 1 c4 1 c4 Figure 3 (T/TR 01-03). Assignment of bits 1-8 of the Service Channel in each frame in a multiframe. R1-R10: Reserved (provisionnally set to O). A, E, Cl-C4: As in Figure 2 (T/TR 01-03). Edition of September 1, 1986 t. T TJTR 01-03 E Page 8 Attribute Bits 9-11 O00 Audio coding Attribute value Bits 12-16 00000 soo1 O 1 soo01 O0 1 O0 o1 10 11 O00 O0 1 O10