SMPTE ST 247M-2003 Television Digital Recording - 19-mm Type D-2 Composite Format - Helical Data and Control Records.pdf

1、SMPTE 2471111-2003 Revision of ANSIEMPTE 1993 SMPTE STANDARD for Television 19-mm Type Helical Data Digital Recording - D-2 Composite Format - and Control Records Page 1 of 25 pages 1 Scope 1.1 This standard specifies the content, format, and recording method of the data blocks forming the helical r

2、ecords on the tape containing video, audio, and ancillary data in the 19-rnm type D-2 helical-scan television recorder. In addition, clause 4 of this standard specifies the content, format, and recording method of the longitudinal record containing tracking information for the scanning head associat

3、ed with the helical records. Track dimensions and locations are specified in ANSVSMPTE 245M. 1.2 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 SMPTE 244M. One video channel and four independent audio channe

4、ls are recorded. Audio channels operate in accord with ANSI S4.40 at a nominal 48-kHz sampling frequency. 1.3 Figures 1 and 2 show a block diagram of the processes involved in the recorder. 2 Normative references The following standards contain provisions which, through reference in this text, const

5、itute provisions of this standard. At the time of publication, the editions indicated were valid. All standards are subject to revision, and parties to agreements based on this standard are encouraged to investigate the possibility of applying the most recent edition of the standards indicated below

6、. ANSI S4.40-1992 (RI 998), Digital Audio Engineering - Serial Transmission Format for Two-Channel Linearly Represented Digital Audio Data ITU-R BT.470-6 (1 1/98), Conventional Television Systems 3 Helical record content 3.1 Introduction Six helical tracks are used to record each TV field. The helic

7、al track is recorded with the digital data from the video channel and the four audio channels. The audio data is contained in four recorded sectors per track, two at the beginning of the track and two at the end of the track. The audio data is recorded twice. The video data is recorded in a sector i

8、n the middle part of each track. An edit gap between sectors accommodates timing errors during editing. Figure 3 shows the arrangement of video and audio sectors on the tape. Copyright 92003 by THE SOCIEPI OF MOTION PICTURE AND TELEVISION ENGINEERS 595 W. Hartsdaie Ave., White Plains, NY 10607 (914)

9、 761-1100 Approved Februaiy 7,2003 SMPTE 247M-2003 VIDEO CHANNEL OUTER INTRA- .) SECTOR SWITCH CODER SHUFFLE ANALOG/ MUX I) ECC (ANALOG)* DIGITAL * t INTERFACE (DIGITAL) RECORD DRIVER .C AND HEAD HELICAL TRACK CHANNEL + INNER SYNC/ ID GEN. ECC CODER CODER Figure 1 - Block diagram - Record AUDIO AUDI

10、O OUTER AUDIO (ANALOG) DIGITAL ERROR 4 ECC e DATA (DIGITAL) - ANALOG/ INTERFACE CONCEAL DECODER DESHUFFLE -,-. VIDEO - ANALOG/ VIDEO CHANNEL OUTER INTRA- (ANALOG) DIGITAL 4 ERROR 4 DEMUX ECC 4 SECTOR 4 BUFFER (DIGITAL) INTERFACE CONCEAL SWITCH DECODER DESHUFFLE A 1 A CONTROL TRACK I P.B. CONTROL TRA

11、CK HEADS AND PLAYBACK INTERFACE TIME CODE CUUREF P.B. AMP Figure 2 - Block diagram - Reproduce Page 2 of 25 pages CMPTE 247M-2003 HEAD - - EDIT GAP 0 E AUDIO SECTOR EDIT GAP 1 AUDIO SECTOR (6 SYNC BLOCKS) - - - (6 SYNC BLOCKS) - - -i I i VIDEO SECTOR (204 SYNC BLOCKS) I L- E EDIT GAP E - - * - E AUD

12、IO SECTOR (6 SYNC BLOCKS) L, AUDIO SECTOR (6 SYNC BLOCKS) NOTES 1 T = Track preamble (62 bytes) 2 E = in-track preamble (28 bytes) 3 P = Postamble (6 bytes) 4 Sync block = 190 bytes 5 Edit gap = 156 bytes nominal Figure 3 - Sector arrangement on helical track Each sector (audio or video) is divided

13、into the following elements: - Preamble containing clock run-up sequence, sync pattern, and identification pattern; - Sync blocks containing sync pattern and identification pattern, followed by a fixed length data block with error control; - Postamble containing sync pattern and identification patte

14、rn. 3.2 Labelling conventions for audio and video data In this standard, the least significant bit is shown on the lef and is the first recorded to tape. The lowest numbered byte is shown at the left/top and is the first encountered in the input data stream. Byte values are expressed in hexadecimal

15、notation unless otherwise noted. 3.3 Sync block The sync block format is common to both audio and video sectors. Each sync block contains a sync pattern (2 bytes), and two inner code blocks. Each inner code block contains 85 data bytes (outer check bytes are considered data) plus 8 inner check bytes

16、. Inner code block O includes and protects the two bytes of an identification pattern. Figure 4 shows the sync block format. Page 3 of 25 pages SMPTE 247M-2003 3.3.1 Sync pattern (a) Length: 16 bits (2 bytes) (b) Pattern: 30 F5 (in hexadecimal notation) LSB MSB ByteO- O O O O 1 1 O O Byte 1 - 101011

17、11 (c) Protection: None 3.3.2 Identification pattern The first byte of the identification pattern identifies a particular sync block of a helical track. The second byte of the identification pattern identifies a particular track. Figure 5 shows the format of the identification pattern. The first and

18、 second audio copies have the same segment numbers. The audio segment number located at the start of a track (second copy) is the same as the video segment number on the same track. Each audio field is composed of 3 segments, the first segment number is Oh. (a) Length: 16 bits (2 bytes) (b) Arrangem

19、ent: The sync block number (byte 2) follows a coded sequence along the track. Figure 6 shows the sequence of sync block numbers. The sector ID (byte 3) identifies a particular sector. The V/A bit distinguishes between audio and video sectors. The T bit distinguishes between two tracks corresponding

20、to channels O and 1. The segment count is modulo 3, the field count for video sectors is modulo 4, and the field count for audio sectors is modulo 5. (See figure 7.) Page 4 of 25 pages SMPTE 247M-2003 c) The field address FO, FI (bits 4 and 5 of the sector ID for video sync blocks) shall identify th

21、e 4-field color sequence as defined in ITU-R BT.470-6, figure 5(c), and has the following values: Fo Fi Color frame A Field I O O Color frame A Field II 1 O Color frame 6 Field Ill O 1 Color frame 6 Field IV 1 1 (d) Protection: The identification pattern is protected by inner code block O. O ARRANGE

22、MENT . 1 I BY TE 2 BYTE 3 1 I I SYNC BLOCK NUMBER SECTOR ID OSECTORITRACK ID FOR AUDIO SYNC BLOCKS BYTE 3 LSB MSB I - d SEGMENT FIELD I =O TRACK 0 SECTORITRACK ID FOR VIDEO SYNC BLOCKS BYTE 3 LSB MSB O 1 2 3 4 5 6 7 VIA I TI so 1 SI I Fo I F1 1 01 MIC 1 I 1 I = O FOR COLOR = 1 FOR MONOCHROME I - SEG

23、MENT FIELD I -1 TRACK Figure 5 - Sync block identification format Page 5 of 25 pages SMPTE 247M-2003 HEAD TRACK PREAMBLE 3 AUDIO SECTOR 02 03 -FI o4 EDIT GAP AUDIO 1 1: 1 SECTOR 43 -w POSTAMBLE 1 IN-TRACK PREAMBLE VIDEO SECTOR POSTAMBLE EDIT GAP EDIT GAP I Figure 6 - Sync block ID number TAPEMOTION

24、NOTES 1 T = Track number (O, 1) 2 S = Segment number (O . 2) 3 F = Field number (O . 3) 4 Audio sectors not shown Figure 7 - Track, segment and field numbers Page 6 of 25 pages SMPTE 247M-2003 3.3.3 Sync block data fielderror correction coding The sync block format is common to both audio and video

25、data, and the associated inner ECC code blocks. (a) Length: 2 inner code blocks. Inner code block O contains 95 bytes consisting of two identification pattern bytes, 85 data bytes (outer ECC check bytes are considered data), plus 8 inner ECC check bytes. Inner code block 1 contains 93 bytes consisti

26、ng of 85 data bytes plus 8 inner ECC check bytes. (b) Arrangement: See figure 4 (c) Interleaving: None (d) Protection: (Inner ECC code) Type: Reed Solomon Galois field: GF(256) Field generator polynomial: x8 + x4 + x3 + x2 + 1, where xi are place-keeping variables in GF(2), the binary field. Order o

27、f use: Left-most term is most significant. Code generator polynomial GF(256): G(x) = (x + 1) (x + a) (x + a2) (x + a3) (x + a4) (x + as) (x + as) (x + a7), where a is given by (02) in GF(256). Check characters are K7, K6, K5, K4, K3, K2, Ki, KO in K7x7 + Ksx6 + K5x5 + K4x4 + K3x3 + ex2 + Kix + KO ob

28、tained as the remainder after dividing xsD(x) by G(x), where for Inner code block O: D(x) = IDox86 + IDix85 + B84xM + . + B2x2 + Blx + Bo Inner code block 1: D(x) = Ba4x84 + Ba3x83 + . + B2x2 + Bix + Bo Equation of full inner code block O: IDox94 + ID1x93 + Ba4x92 + B83x91 + . + Bixg + Box8 + K7x7 +

29、 K + . + K2x2 + Kix + Ko Equation of full inner code block 1: B84x92 + B83x91 + . + Blxg + Boxa + K7x7 + Ksx6 + . + K2x2 + Kix + KO 3.4 Preamble All sectors are preceded by a preamble consisting of a clock run-up sequence, sync pattern (2 bytes), identification pattern (2 bytes), and fill pattern (4

30、 bytes). The clock run-up sequence varies in length and pattern depending on the sector. The remaining elements of the preamble have the same format for all sectors. When a sector is edited, the appropriate preamble, including run-up sequence, shall be recorded. 3.4.1 Track preamble for start of fie

31、ld track pair This preamble precedes the first sector of the first pair of tracks of every field (segment O). The run-up sequence is 54 bytes long and consists of 18 repetitions of the three-byte pattern BH, DH, DBH. (a) Length: 62 bytes (b) Arrangements: See figure 8A (c) Run-up pattern: BH, DH, DB

32、H Page 7 of 25 pages CMPTE 247M-2003 LSB MSB ByteO- O 1 1 O 1 1 O 1 Byte 1 - 10110110 Byte 1 - 11011011 (d) Protection: None 3.4.2 Track preamble This preamble precedes the first sector of every track other than the first pair of tracks of every field. The run- up sequence is 54 bytes long and conta

33、ins AAH. Length: 62 bytes Arrangement: See figure 8B Run-up pattern: AAH LSB MSB 01010101 Protection: None 3.4.3 In-track preamble This preamble precedes every sector which is not the first sector of a track. The run-up sequence is 20 bytes long and contains AAH. (a) Length: 28 bytes (b) Arrangement

34、: See figure 8C (c) Run-up pattern: AAH LSB MSB 01010101 (d) Protection: None. 3.5 Postamble All sectors are followed by a postamble containing a sync pattern (2 bytes), identification pattern (2 bytes), and fill pattern (2 bytes). When a sector is edited, the postamble shall be recorded together wi

35、th the new data. (a) Length: 6 bytes (b) Arrangement: See figure 8D (c) Protection: None Page 8 of 25 pages SMPTE 247M-2003 TRACK PREAMBLE FOR START OF FIELD TRACK PAIR BYTE o12345 51 52 53 54 55 56 57 58 59 60 61 I B6I 6D I BD 186 I6D I BDI + IB6I6DIBDl SYNC 1 ID AAIAAIAAIAA 0 TRACK PREAMBLE BYTE o

36、123 51 52 53 54 55 56 57 58 59 60 61 AAAA AA I AA tf IAAAAAA SYNC I ID IAAIAAIAAJAA 0 IN-TRACK PREAMBLE BYTE O 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 AAI AA 1 AA I AA IAAIAA/AAI SYNC I ID IAAIAAIAAIAA POSTAMBLE BYTE o12345 SYNC I ID IAAAA Figure 8 - Sector preamble a

37、nd postamble 3.6 Edit gaps The space between sectors on a track, exclusive of postamble and preamble, is nominally 156 bytes long and is used to accommodate timing errors during editing. In an original recording, the edit gap shall contain the pattern AAH repeated 156 times. During an edit, the edit

38、 gap may be partially rewritten with AAH, provided that the preamble and postamble of adjacent unedited sectors are not overwritten. 3.7 Channel code The channel code shall be the Miller-squared code which is defined by the following code rules: (1) The data stream is divided into the following type

39、s of sequences: (a) Any number of consecutive ones; (b) Two zeros separated by either no ones or any odd number of ones; (c) One zero followed by any even number of ones. Note that a sequence of type (c) cannot be followed by a sequence of type (a). (2) Sequences of types (a) and (b) are encoded acc

40、ording to the Miller code (equivalent to modified FM (MFM) rules. That is, data ones are encoded as transitions in the middle of the bit cell, isolated data zeros are ignored, and transitions are inserted at the boundary of a bit cell between adjacent data zeros. Page 9 of 25 pages SMPTE 247M-2003 (

41、3) Sequences of type (c) are encoded according to the Miller code rules except that the transition associated with the last bit of the sequence is suppressed. 3.8 Magnetization The recorder shall operate in reproduce without regard to the polarity of data flux during recording on the helical tracks.

42、 The record current will be constant for all recorded frequencies involved in the Miller-squared spectrum. The record magnetization shall be optimized for best signal-to-noise ratio at a frequency of one-half the maximum channel data rate. 3.9 Video processing 3.9.1 Sampling Signals are sampled at 4

43、fsc (14.31818 MHz), using 8-bit linear quantization from 01 to FFH inclusive. The sample value of (001-1) shall not be recorded on tape nor should it occur at the interface. (See SMPTE 244M.) 3.9.2 Recorded data Information received during the horizontal blanking interval and vertical sync interval

44、is not recorded on tape. 3.9.2.1 Recorded samples of the television line 768 samples per line are recorded, centered about the active picture. Figure 9 shows the relationship between video signals in the analog and digital domains together with the address numbers of the digitized samples for zero-d

45、egree ScH phase of the incoming signal. Under this condition, sample number 785 occurs 44.2 ns (57“ of color subcarrier) after the 50-percent point of the leading edge of the horizontal sync pulse. The first active sample to be recorded at address location O (decimal) of line 10 of field 1 of color

46、frame A, as defined in ITU-R BT.470-6, is the I sample. 3.9.2.2 Recorded lines of the television frame From each field, 255 consecutive lines are recorded (3 segments of 85 lines each). The first recorded line of each field varies over a four-field sequence as follows, with the line numbers defined

47、as in ITU-R BT.470-6 (figure 5c), except the line numbers repeat on a television frame basis: - From field I of color frame A, the first recorded line is number 10; - From field II of color frame A, the first recorded line is number 272 (line 9 of field II); - From field III of color frame B, the fi

48、rst recorded line is number 9 (line 9 of field Ill); - From field IV of color frame B, the first recorded line is number 271 (line 8 of field IV). 3.9.3 Channel distribution of samples The samples are distributed between 2 channels in a checkerboard pattern which alternates from line to line. Figure

49、 10 shows the distribution of samples. In figure 10, the channel number (O or 1) coincides with the track number as defined in 3.3.2(b) and figure 7. Page 10 of 25 pages CMPTE 247M-2003 DIGITAL ACTIVE LINE DIGITAL BLANKING 768 WORDS 142 WORDS (0-767) (768-904) 4 *t b TOTAL LINE 910 WORDS (0-909) + b 310101 . o1 101010 101010 . 10 010101 o1 101010 10 . NOTE - Zero-degree ScH. ._i Figure 9 - Horizontal sync relationship SEGMENT O SEGMENT 1 SEGMENT 2 768 SAMPLEYLINE 4 101010 10101 SEGMENT 010101 o1 .I 101010 -1