ANSI SMPTE ST 227M-1996 Television Digital Component Recording - 19-mm Type D-1 - Helical Data and Control Records.pdf

1、SMPTE 2271 b 8357403 0002498 483 M I 4 8 ANWSMPTE 227M-1996 SMPTE STANDARD for Television Digital Component Recording - 19-mm Type D-1 - Helical Data and Control Records 1 Scope This standard specifies the content, format, and re- cording method of the data blocks forming the helical records on the

2、tape containing video, audio, and associated data in 19-mm type D-1 television digital component recording. In addition, clause 6 of this document specifies the content, format, and recording method of the longitudinal record containing tracking information for the scanning head associated with the

3、helical records. Track dimensions and locations are specified in ANSVSMPTE 224M. The standard applies to recorders operating in the 525-line television system with a frame frequency of 29.97 Hz nominal and in accord with ITU-R BT.601. One video channel and four independent audio chan- nels are recor

4、ded. Audio channels operate in accord with ANSI S4.40 at a 48-kHz sampling frequency. Figure 1 shows a block diagram of the processes involved in the recorder. 2 Normative references The following standards contain provisions which, through reference in this text, constitute provisions of this stand

5、ard. At the time of publication, the editions indicated were valid. All standards are subject to revision, and parties to agreements based on this Page 1 of 37 pages standard are encouraged to investigate the possibility of applying the most recent edition of the standards indicated below. ANSI S4.4

6、0-1992, Digital Audio Engineering - Serial Transmission Format for Two-Channel Linearly Represented Digital Audio Data ANSIEMPTE 224M-1996, Television Digital Compo- nent Recording - 19-mm Type D-1 - Tape Record ITU-R BT. 601-5, Studio Encoding Parameters of Digital Television for Standard 4:3 and W

7、ide-Screen 16:9 Aspect Ratios 3 Helical record content, format, syn- chronization, and recording method 3.1 Introduction The helical track defined mechanically in ANSVSMPTE 224M is recorded with the digital data from the video channel and the four audio channels. Data is arranged in six sectors per

8、track as shown in figure 2. Two sectors are employed for video data and four sectors each containing data from one of the four audio channels. Details of sector assignment are shown in clauses 4 and 5 of this document. Each sector is divided into the following elements: The users attention is called

9、 to the possibility that compliance with this standard may require use of an invention covered by patent rights. willingness to grant a license under these rights on reasonable and nondiscriminatory terms and conditions to applicants desiring to obtain such a license. Details may be obtained from th

10、e publisher. By publication of this standard, no position is taken with respect to the validity of this claim or of any patent rights in connection therewith. The patent holder has, however, filed a statement of No representation or warranty is made or implied that this is the only license that may

11、be required to avoid infringement in the use of this standard. CAUTION NOTICE: This Standard may be revised or withdrawn at any time. The procedures of the Standard Developer require that action be taken to reaffirm, revise, or withdraw this standard no later than five years from the date of publica

12、tion. Purchasers of standards may receive current information on all standards by calling or writing the Standard Developer. Printed in USA. Copyright O 1996 by ME SOCIEp/ OF American National Standard MOTION PICTURE AND TELEVISION ENGINEERS 595 W. Hamdale Ave., White Plains, NY ION7 (914) 761-1100

13、Approved August 14,1996 SMPTE 22711 96 = ANSVCMPTE 2271111-1995 - . AUDIO Audio Data DISTRIB. Coding p DATA Re- Audio Data 3 -h Synchronise * Pre-Process DATA Outer Error -h DIGITAL AUDIO DATA B AUDIO DATA SHUFFLE - Sync blocks containing sync pattern and an iden- tification pattern followed by a fi

14、xed length data block with error control; 23.072 Mb/s USERA (AES/EBU FORMAT) VIDEO SYNC DATA WORDS - VIDEO DIGITAL DATA VIDEO 3 DECODE 4 * DATA P? M W/s i PROGRAM TRACK- - r Video Data Video Data Video Data I SOURCE Inter- Outer Error Intra-Sector Sctor - A sector array shuffle which shuffles data a

15、nd error correction code words within the sector, prior to being written to tape. The sector array dimensions are 32 rows by 600 columns. Each column corresponds to one outer code block and contains 30 video data bytes plus two outer correction check bytes. The sector array is further divided into 1

16、 O contiguous subarrays, each having dimensions of 32 rows by 60 columns. The 60 data bytes within a single subarray row correspond to one inner code block on tape. 4.4.1 Intraline shuffle Let the horizontal pixel index, j, be normalized to the range (O -1 79) following the intersector distribution

17、described in 4.3. For luminance component, jy = int(jy/4) For the color-difference components (CB and CR), jc = 2 int(jd8) where j indicates a normalized index. Then the sector data sequence for a given line con- tains 360 bytes as shown in table 7. Table 6 - Intersector shuffling for odd and even l

18、ines For (f + g) mod2 = O Even line numbers j = O 1 23 45 67 8 9 10 11 12 13 14 15 16 . (m. mod2 = O) ry= o2 13 o2 13 o2 13 o2 13 o rc = O 2 1 3 O 2 1 3 O Odd line numbers j= O1 23 45 67 8 9 10 11 12 13 14 15 16 . (m.mod2 = 1) ry= 13 02 13 o2 13 o2 13 o2 1 rc = 1 3 O 2 1 3 O 2 1 For (f + g) mod2 = 1

19、 Even line numbers j = o1 23 45 67 8 9 10 11 12 13 14 15 16 . (m. mod2 = O) ry= 20 31 20 31 20 31 20 31 2 rc = 2 O 3 1 2 O 3 1 2 Odd line numbers I= o1 23 45 67 8 9 10 11 12 13 14 15 16 . (m.mod2 = 1) ry= 31 20 31 20 31 20 31 20 3 rc = 3 1 2 O 3 1 2 O 3 Table 7 - Sector data sequence O 1 2 3 4 5 6 7

20、 356 357 358 byte: CBo Yo CRo Yi CB2 Y2 CR2 Y3 CBi78 Yi78 CR178 Yi79 Page 11 of 37 pages SMPTE 2271 96 W 357403 0002509 O99 W ANSVCMPTE 227M-1996 The 360 luminance and chrominance bytes are dis- tributed among 12 outer code blocks as shown in table 8. Each column represents an outer code block. The

21、last two bytes, KV1 and WO, are outer correction check bytes added by the outer coder. The byte number refers to the byte position within an outer code block. Let k be the position of a video data byte within a line of the sector data sequence, following the intersector distribution as described abo

22、ve, O5 k I 359. Let Oblk be the outer block column index of table 7, O I Oblk I 11. Let Obyt be the outer block byte number of table 7, O I ObytI 31. Then the intraline shuffle described by the following formulas is applied: The sector array shuffling is defined by algorithm 1. Tables 9 (a-j) show t

23、he result of this algorithm and figure 10 shows a conceptual block diagram of the method. The algorithm may be considered to operate as follows: The column counter is cleared at the beginning of each 50-line segment, and incremented every outer block or 12 times per Tv line. The least significant 2

24、bits of the column counter select a column within a 4-column group. The most significant 8 bits are used to address a PROM containing the column map function. The row start PROM is used to select an initial starting point for the row map sequence for each column group, except for the fourth column o

25、f the column group, which has a different initial start- ing point for the row map sequence. The row counter is loaded with the row start preset data at the beginning of each outer block and increments mod 32 every data byte. The row map PROM is used to select the actual row address where the byte i

26、s stored in the sector array. Oblk = 4int(k/l20) + (k mod 4) Obyt = int(k mod 120)/4) (For O 5 Obyt I 29) Tables 9 (a-j) explicitly list the relation between every byte in the sector array and its location in the input data stream. The array values represent normalized pixel indices, j or j, as defi

27、ned in 4.4.1. Algorithm 2 shows the de-shuffling scheme. The result is shown in table 8. The inverse mapping is given by the formula k = 120int (OblW4) + (Oblk mod 4) + (4 x Obyt) 4.4.2 Sector array shuffling 4.4.2.1 Algorithm 1, intrasector shuffling (refer- ence only) The sector array may be divid

28、ed into 150 4-column groups, ranging from O to 149. The 4 columns within a column group Ontain (cBr CR, 1 pixel data Let m designate the line number within a segment, bytes, respectively. Along a given row within a column group, CB and CR are cosited with respect to the O ImI49. source data, and cos

29、ited (or nearly so) with the first Y pixel data byte, while the second y pixel byte is horizontally offset from the first with respect to the source data. A column map, which is a permutation of the integers O to 149, is used to define the sequence in which column groups are stored in the sector arr

30、ay. A row map, which is a permutation of the integers O to 31, is used to define the sequence of rows in which data for a given column is stored in the sector array. The Let Oblk designate the outer block number within a line, as defined in 4.4.1, O I Obik I 11. Let Obyt designate the outer block by

31、te index, as defined in 4.4.1, O I Obyt I 31. Define the outer block number counting from begin- ning of the segment, Icnt, icnt = Oblk + 12 m, O I Icnt I 599 starting point of the row map is different for each column group, and, in addition, the starting point of the row map sequence for the fourth

32、 column of each Define the unpermuted 4-column group number, Igrp, igrp = int (Icnt/4), O I Igrp I 149 column group is further offset by a constant from the starting point of the row map sequence for the first 3 columns of the column group. Define the permuted 4-column group number, Jgrp, Jgrp = (41

33、 x igrp) mod 150 Page 12 of 37 pages _ SMPTE 227M 96 = 8357403 0002510 800 = 4 4 O -4 m a3 b D VI * m N 4 O ANSUSMPTE 227M-1996 Page 13 of 37 pages SMPTE 2271 96 83574011 0002511 747 ANSVCMPTE 227M-1996 ICnt (O. .599) / CP 9- I New Blk 10 New Seg CLR Define the sector array column index, Col, Define

34、 the row count starting value, Rstart, Col = 4 x Jgrp + (Icnt mod 4), O n Y Q) n im vi n E Y o al r O o alw yo Page 16 of 37 pages P E 6 5 E E Q Q r SMPTE 227M 76 8357403 0002534 456 ANSVSMPTE 227M-1996 Co E 13 Page 17 of 37 pages ANSVCMPTE 227M-1996 c3 8 rc Q E I n m QI Y a9 n im SMPTE 227M 96 m 83

35、57403 00025B.5 392 m II Page 18 of 37 pages d * L L 3 OR L * O Q m E E ici u Q) u# E ici c I n Q) Q) Y o) im v-4 FWN mmr- Pm4 “Q P-Vd d40 bnrl oow P-N mmm W mwm w-v 7-4 P-pm wnw .-I WWN WN- ri mmm WdO rl v-9 woo 7-4 nnd wmn NNP WWN drin WPN O00 wwcu aaal kLIC P P-4 ?Ud ANSVSMPTE 227M-1996 wiwwwmwwwm

36、wmovovw -omwmwommwwomomwmowwrim Sample timing: The first audio sample shall be time- coincident with line 9 of the video signal f 6 lines (20 samples); Block length: 192 bits, 24 words; Coding: See ANSI S4.40. NOTES 1 Bytes O and 1 of status data only are selected for special processing in the digit

37、al television tape recorder. The con- tents of bytes O and 1 are shown in figures 11 and 12. 2 Bytes 22 and 23 of the status data contain protection and validity information for bytes 0-21 and may be used in some source decoders. Word length: 20 + 4 bits; 5.2.3 User data Coding: Twos complement line

38、ar PCM. 5.2.2 Channel status data Bit rate: 48 kbis (nominal); Word rate: 6 kbytek; Word length: 8 bits; LS B As status data but data coding is undefined. 5.2.4 Validity data Bit rate: Coding: One bit associated with each audio word; O = sample valid; 1 = sample defective. MSB O = Consumer Use I 1 =

39、 Prof. Use O = Audio 1 = Data Pre-emphasis O -Sampling Freq. 1 Sampling Freq. O Sampling Frequency 1 = Unlocked “ = Locked Pre-emphasis 1 I I Pre-emphasis 2 Figure 11 - Channel status data - Byte O LSB MSB Chan. Mode 0- Chan. Mode 1 Chan. Mode 2 Chan. Mode 3 LU n d ef i ned Undefined Undefined Undef

40、ined Figure 12 - Channel status data - Byte 1 Page 27 of 37 pages SMPTE 227M 96 H 8357403 0002525 231 ANSVSMPTE 2271111-1 996 Table 13 - Byte status Mode O 1 O O O 1 O O 2 O O 3 O O 4 O 1 5 O 1 through through F 1 I 2 3 Undefined - 2 channel 2 channel Single channel Prim arykecondary 2 chan ne1 Ster

41、eophonic Undefined 1 5.2.5 Parity bit Bit rate: One bit associated with each audio word: Coding: audio, status, user, and validity data. Even parity of associated word including 5.3 Source processing 5.3.1 Introduction Audio data is processed in segments corresponding in duration to four helical tra

42、cks or one-fifth of a video frame. Each segment contains approximately 320 audio samples for an audio channel with associated status, user, and validity data. In addition, a number of control and user words are added to the data in the last complete block received. 5.3.2 Segment Each segment of audi

43、o data is processed into two audio blocks of 1 O x 60 bytes, each corresponding to a sector. One block contains even-numbered words and the other odd-numbered words. The data portion of the block is 7 x 60 with the balance being outer error correction words. For convenience, data is processed in fou

44、r-bit words: Audio data word: 31 8 to 322 data words with asso- ciated C, U, V, R bits (20 bits total per word) Interface control words: 6 words of four bits and 2 words of eight bits. (For security, one word, LNGH, is written four times in each block.) Processor control words: 9 words of four bits.

45、 (For security, two words, BCNT and SEQN,are written four times in each block.) User control words: 8 words of eight bits are in- cluded in each block, giving a total of 16 bytes per segment for user data. 5.3.3 Audio data word processing Input data is formed into words of twenty bits in the sequenc

46、e. 5.3.3.1 Assignment of the twenty-bit word to audio and associated data is controlled by user input in accord with table 14. The most significant bit of the audio word is in bit 19 and unused bits of lower significance are removed. The interface control word (ICW) LNGH (four bits) signals the word

47、 mode selected. 5.3.3.2 The twenty-bit words formed as in 5.3.3.1 are separated into two groups by selection of alternate words into EVEN (O, 2, 4, etc.) and ODD (1, 3, 5, etc.) beginning at the start of the sequence. 5.3.3.3 Each group of twenty-bit words is divided into 8-bit bytes as shown in fig

48、ure 13, beginning with the LSB of the first word of the word group. 5.3.3.4 Each group (ODD or EVEN) is distributed into the product block in accordance with figure 14. Word 159 (bytes 935: 9,56: 9,57) and word 160 (bytes 3,551; 3,56; 3,57) may not be present in all blocks depending on the current r

49、elation- ship between video and audio clock synchroni- zation and phasing. When not used, this space is zero filled. The processing control word (PCW) B CNT specifies the length of the block between 397/2 bytes (159 audio data words) and 402h bytes (161 audio data words). Page 28 of 37 pages SMPTE 227M b I 357403 000252b L7B I Word mode o (000) 1 (001) 2 (010) 3 (O1 1) 4 (100) 5 (101) 6 (110) 7 (111) ANSVSMPTE 227M-1996 Table 14 - Audio interface control input O 1 Bit 2 3 4 throuah 19 C C C C C v U Audio