GB2323742A

GB2323742A - Digital VTR with track identification and tracking control.

Info

Publication number: GB2323742A
Application number: GB9813320A
Authority: GB
Inventors: Masako Asamura; Nobuyoshi Okumura; Sadayuki Inoue; Tohru Inoue; Ken Onishi; Tomohiro Ueda; Taketoshi Hibi; Junko Ishimoto
Original assignee: Mitsubishi Electric Corp
Current assignee: Mitsubishi Electric Corp
Priority date: 1994-04-12
Filing date: 1995-04-10
Publication date: 1998-09-30
Anticipated expiration: 2015-04-10
Also published as: GB2323742B; GB2323240A; GB9813320D0; GB9813119D0; GB2323240B

Abstract

A digital VTR for recording recording digital video and audio signals in respective designated areas of oblique tracks in a predetermined track format, using a rotary drum on which head of two different azimuths are mounted. The VTR comprises data separating means for extracting a fast replay signal from the normal recording signal; recording means for recording the fast replay signal in one region in one track per one scanning of the head, of the regions covered by the head traces and in the tracks of identical azimuth; identification signal recording means for recording an identification signal for identifying the track; and replay means for replaying the identification signal. It may also include a facility for tracking control using the identification signal and includes an embodiment in which fast replay data is recorded over a plurality of tracks.

Description

2323742 DIGITAL VTR

BACKGROUND OF THE INVENTION

The present invention relates to a digital video tape recorder (hereinafter referred to as digital YTR) having a track format for recording digital video and audio signals in predetermined areas on oblique tracks, and relates to a digital VTR in which the digital video and audio signals are input in the form of a bit stream, and the hit stream is magnetically recorded and played back.

Fiz. 93 is a diagram showing a track pattern of a conventional, general consumer digital Y7R. Referring to the drawing, a plurality of tracks are formed on a magnetic tape 10, in a head scanning direction inclined to the tape transport direction, and digital video and audio signals are recorded therein. Each track is divided into two areas, a video area 12 for recording a digital video signal and an audio area 14 for recording a digital audio signal.

Two methods are available for recording video and audio signals on a video tape for such a consumer digital Y7R. In one of the methods, analog video and audio signals are input, and recorded, using a video and audio Ijigh-efficiency encoding means; this is called a baseband recording njethod. In the other method, the bit stream having been digitally transmitted; this method is called a transparent recording method.

For the system of recording ATV (advanced television) signals, now under consideration in the United States, the latter transparent recording method is suitable. This is because the ATV signal is digitally compressed signals, and does not require a high-efficiency encoding means or a decoding means, and because there is not degradation in the picture quality due to transmission.

The transparent recording s'steiii however is associated 1 with a problem in the picture quality in a special playback mode, such as a fast playback mode, a still mode and a slow mode. In particular, when a rotary head scans the tape obliquely to record a bit stream, almost no image is playback at the time of fast playback, if not specific measure is taken.

An improvement for the picture quality for the transparent recording system recording the ATV signal is described in an article Yanagihara, et al, "A Recording Method of ATV data on a Consumer Digital VCR", in International Workshop on HDTV, 93, October 26 to 28, 1993, Ottawa, Canada, Proceedings, Vol. II. This proposal is now explained.

In one basic specification of a prototype consumer digital VTR, in SD (standard definition) mode. when the recording rate of the digital video signal is 25 Mbps. and the field frequency is 60 Hz, two rotary heads are used for recording a digital video signal of one frame, being divided into video are as on 10 tracks. If the data rate of the ATV signal is 17 to 18 Mbps, transparent recording of the ATV signal is possible with the recording rate in this SD mode.

Fig. 94A and Fig. 94B show tracks formed in a magnetic tape using a conventional digital VTR. Fig. 94A is a diagram showing scanning traces of the rotary heads during normal playback. Fig. 94B shows scanning traces of the rotary heads during fast playback. In the example under consideration, the rotary heads are provided in opposition, 180' spaced apart on a rotary drum, and the magnetic tape is wrapped around over 180. In the drawing, adjacent tracks on the tape 10 are scanned by two rotary heads A and B having different azimuth angles, alternately and obliquely, to record digital data. In normal playback, the transport speed of the tape 10 is identical to that during recording, 2 so that the heads trace along the recorded tracks. During fast playback, the tape speed is different, so that the heads A arid B traces the magnetic tape 10 crossing seireral tracks. The arrow in Fig. 94B indicates a scanning trace by a head A at the time of five-time fast feeding. The width of arrow represents the width of the region read by the head. Fractions of digital data recorded oil tracks having a n identical azimuth angle are played back from regions ineshed in tlie drawings, within five tracks oil the magnetle tape 10.

The bit stream of the ATV signal is according to the standard of the MPEG2. ln this bit stream according to the MPEG2, on-ly the intra-frame or intra-field encoded data of the signal. i.e., the data of intra encoded block (intra encoded block) alone can be decoded independerit.ly, without reference to data of' other frame or field. Where the bit stream is recorded ill turn oil the respective tracks, the recorded data are replayed intermittently from the tracks during fast replay, arid the image nitist be reconstructed from only the intra-encoded blocks contalned in the replay data. Accordingly, the video area updated oil the screen is not continuous, aild only, the fractions of data of intra coded block are replayed, and may be scattered over the sereen. Tlie bit stream is variable-length encoded, so that it is riot ensured that all the replay data over the screen is periodically updated, arid the replay data of certain parts of the vide area may riot be updated for a long time. As a result, this type of bit stream recording system does riot provide a sufficient picture quality during fast replay in order to be accepted as a recording method for a consumer digital VTR.

Fig. 95 is a block configuration diagram showing an example of' recording systern in a conventional digital VTR. Referring to the drawing. reference numeral 16 denotes an 3 input terminal for the bit stream, 18 denotes an oulplit terminal for the bit stream, 20 denotes an HP data format circuit, 22 denotes a,;iri.11)1e- leilgtli decoder, 24 denotes a counter, 26 denotes data extractor, and 28 denotes an EOB (end of block) appending circuit.

To improve the quality of fast replay pictures, the video area on each track is divided into two types of areas. That is, the video area on each track is divided into main areas 30 for recording the bit stream of the ATV signal, and copy areas for recording important part of the bit stream which are used for recoiistruction of the image in fast i - e p lav. Only the intra-encoded blocks are effectiie (luriiig fast replay. so that they are recorded in the cop), areas. To reduce the data, further, only the low-frequency components are extracted frow all the intra-encoded blocks, aii(l recorded as HP (high priority) data.

The bit stream of MPEG2 is iiiput via the input terminal 16, and output via the output terminal 18, without modification. and sequentially recorded in the inain areas 30 on each track of the tape. The bit stream from the input terminal 16 is also input to the variable-length decoder 22, and the syntax of the bit stream of the MPEG2 is analyzed, and the intra-picture data is detected, and timing signals are generated by the counter 24, aiid the low-frequericy components of all the blocks in the intra-picture data are extracted at the data extractor 26. Furthermore, E0Bs are appended at the EOB appending circuit 28, and 11P data is constructed at an HP data format circuit, not shown. The HP data is incorporated in the recording data for one track,and recorded in the copy areas 32.

Fig. 96A and Fig. 96B show an example of replay system in a conventional digital VTR. Fig. 96A schematically shows normal replay. Fig. 96B schematically shows fast replay.

Separation of data from the magnetic tape during normal 4 replay, and fast replay, are performed respectively, in the following wa3, s. During normal replay, all the bit stream recorded in the main areas 30 is replayed, and the bit stream from the data separation means 34 is sent as the normal replay data, to an AIPEG2 decoder, provided outside the replay s3'stem. The 11P data from the copy areas 32 are discarded. During fast replay,, only the IT data from the copyr areas 32 are collected. and sent, as fast replay data, to the decoder. At the data separation means 34. the bit stream front the main areas 30 is discarded.

A metliod of fast repla.y from a track havingmain areas 30 and copy, areas 32 is next described. Fig. 97A shows a scanning trace of a liea(l. Fig. 97B shosys track regions from the replay is possible. When the tape speed is an integer njultiple of the normal speed, if phase-locking control is conducted by an ATF (automatic track following) method or the like for tracking by moving the head itself, the bead scanning is in a predetermined phase relationship with tracks having an identical azimuth.

As a result. the data replayed by the head A from the tracks recorded alternately by the heads A and B, are fixed to those from the meshed regions.

ln Fig. 97B, if the signal having an output level larger than -6dB is replayed by the heads, the data is replayed by one head from the nieshed tape regions. Th e drawing shosys an example of nine-time speed replay. if replay of the signals from the weshed regions is ensured at the ninetinie replay, the regions are used as copy areas, and the HP data are recorded in the copy areas, so that the reading of the UP data frow these regions at this speed is possible. However, reading of these signals at different speeds is not ensured. Accordingly, a plurality of areas need to be selected for the copy areas, so that the replay signals can be read at different tape speeds.

Fig. 98 shows regions where tile copy areas overlap for a plurality of different replay speeds. It shows examples of scan regions for three different tape speeds, for cases where the}lead is in synchronism with an identical-azilTilitli track. The scan regions where the reading by the llead is possible at different tape speeds overlap, at some of the regions. By selecting the regions at which the overlapping occurs as the copy areas, reading of the 11P data at different tape speeds earl be ensured. The drawings show the regions at which overlapping occurs at the fast-forward at four-time, nirie-time, and 17-time speed. Th e s e sean regions are identical to those of feed-forward at -2-time, 7-ti.nie ill(] -15-time high speeds (i.e., rewind at 2-time, 7time and 1.55-time speeds).

EveTi thotigh there are overlapping regimis for (Ii.f'fet-eiit tape speeds, it is not possible to determine a recordillg 1)att,e?-ii so tl)at identical regions are always traced at, different speeds. This js becatise tlie number of tracks crossed by the head differs depemling oil the tape speed. Moreover, it is necessary for the llead to be capable of st,ii,tiiig tracing at whichever Hentical-aziniutli track. For this reason, identical HP data is repeatedly recorded over a pltirci.l.i,ty of tracks, to solve the above problem.

Fig. 99 shows examples of scanning traces of the rotary head at differerit tape speeds. Regions 1, 2 and 3 are selected from among the overlapping regions for five-time 11P data are repeatedly ill(] nine-time speeds. If identical recorded over 9 tracks, the 11P data call be read at. either of five-time and nine-time speeds.

Fig. 100A and Fig. 100B show scanning traces at fivetime speed replay. In the illustrated example, Identical 11P data is repeatedly recorded over five consecutive tracks. As will be seen from the drawings, identical HP data is recorded over the number of tracks identical to the number 6 of t imes of' the tape speed (i. e., 5). In either of case 1 and case 2, either the bead A or B can read HP data from corresponding azimuth track. Accordingly, providing the copy areas in each track, in a number identical to the number of times of the tape speed at the fast replay, and repeatedly recording the HP data there, the copied HP data can be read at various speeds, and in either the forward or reverse direction.

In the manner described, the special replay data is recorded in t}ie copy areas, repeatedly, to improve the picture quality during tlie special replay in the transparent recording system.

Fig. 101 shows a recording format on a track in a conventional digital VTR. Main areas and copy areas are proirided in one track. In a consumer digital VTR, a video area in each track has 135 sync blocks (SB), and 97 syne blocks are assigned to main areas and 32 sync blocks are assigned to copy areas. The sync blocks at the regions corresponding to the 4-, g- and 17-time speed shown in Fig. 98 are selected for the copy areas. The data rate of the main areas is about 17.46 Mbps (97x75x8x10x30), and the data rate of the copy areas where identical data is'repeated 17 times ir, about 338.8 kbps (32 x 7.5 x 8 x 10 x 30/17).

Fig. 102A and Fig. 102B show an example of the configuration of a track containing video and audio data.

The magnetic tape of a digital VTR according to the specification (hereinafter referred to as SD specification) defined by ti)e SD mode, a video area of 149 SB arid an audio area of 1,4 SB are provided on both sides of a gal), as shown in Fig. 93, and the video and audio data are recorded in these areas, together with error correction codes. Employed as the error correction codes for the video areas in the SD specification are (85. 77, 9) code (hereinafter referred to as Cl check code) in the recording direction (right-left

7 di rection ill the drawing), and ( 149, 138, 12) Reed-Solomon code (hereinafter referred to as C2 check code) ill the vertical direction. Employed as the error correction codes for the audio areas are (85, 77, 9) Reed-Solomon code (Cl check code) in the recording direction, like the video signal, and (14, 9, 6) Reed-Solowon code (hereinafter referred to as C3 check) ill the vertical direction. Auxiliary data (VAUX data) is recorded in front of and at the back of the video data.

Fig. 103 shows an example of configuration of one sync block on the magnetic tape. As illustrated, the region of 1 SB is formed of 90 bytes, and a header consisting of syne pattern recording region 36 of two bytes, and ID signal region 38 of three bytes are formed at the head end, and recordIng region 42 for the error correction codc, (Cl check code, in the example illustrated) of 8 bytes is provided at the back of the data region 40 of 77 bytes. In Fig. 102A and Fig. 102B, the header parts are omitted.

Because the conventional VTR is configured as described above, and special replay data is repeatedly recorded in the copy areas, the recording rate for the special replay data is very low. In particular, the quality of the reconstructed pictures formed during slow replay or fast replay is low.

For instance, if the intra-frame is formed twice a second, the amourit of data of intra-eneoded blocks of the ATV signal is predicted to be about 3 Mbps. In the prior art, only 340 kbl)s can be recorded, and the quality of the reproduced picture is very degraded.

Moreover, the data for the respective fast replay speeds is recorded, being dispersed over a wide region. Accordingly, if the track is nonlinear, it is difficult to achieve accurate tracking control over the entire data region, and the replay signal from some of the regions may not be of a sufficient level.

8 Furthermore, during special replay (fast replay, sloss, replay, still replay arid the like), the rotary head crosses a plurality of recording tracks obliquely to pick up the replay data intermittently, as was described above. It is therefore riot possible to foriji error correction block (video data) shown in Fig. 102A arid Fig. 102B from the replay data during special replay. That is, during special replay, the error correction using C2 or C3 check code is not performed, but error correction using Cl check code alone is applied to the replay data.

If the error correction using the Cl check code alone is applied, if the symbol error rate 0.01, the error is 1..56 x 10-3. ans one error detection probability This me, per about 8 sync blocks is detected. Because the replay delta output is riot stable during special replay, so that the symbol error rate can often be more than 0.01. Moreover.

the recording data is variable-length encoded, so that when an error is present, the succeeding replay data cannot be used. leading to degradation in the picture quality. The rate of undetected errors is also about 7.00 x 10-8. Thus, the frequency of occurrence of undetected errors is high.

Moreover, during fast replay, the data rate is low, and only the lowfrequency components are replayed, so that tile resolution of the picture is poor.

Furthermore, it is necessary to pick up data for a plurality of fast replay regions in one scanning of the llead during fast replay, so that when the track isnon-linear, or when the scanning trace is nori-linear, the data at the fast replay region where the rion-linearity is present cannot be reproduced.

Moreover, since it is necessary to pick tip data for a plurality of fast replay regions by one scanning of the head, replay can be performed only at certain speeds. The. speed at which replay call be performed is limited. and the 9 number of the replay speeds is srijall.

Moreover,the rotary speed of the druni of the four-hea(l configuration is half that of the drum of two-head configuration, so that the angle with which the head scanning trace crosses the track is larger, and the replay with the four-head configuration druni from the fast replay region is possible only at a speed half the speed at which the replay with two-head configuration drinii is possible froni the same fast replay region.

Furtherwore, when the level of the replay signal fluctuates, the sync bit and the succeeding ID bits, and the first parity are reproducible, and the succeeding digital data is reproducible only up to its iijicldle, and the rest cannot be reproduced because of the decrease in the leve] of the replay signal. In such a situation, the errors in the digital data is not detected until the resull. of the cheek using the second parity is produced. It is therefore necessary to conduct the predefined calculation for performing the cheek, and time is spent before tl ie error detection.

Moreover, the amplitude of the replay signal varies periodically because the head crosses the recording tracks, so that burst errors frequently occur, and this cannot be detected easily nor quickly.

Moreover, the data used for fast replay is forn)ed by extracting part of the data of tile packets transparentrecorded, so that the length of data for forming a block of image is shortened. For this reason, when recording is made for the region used for transparent recording, disposing sync, ID, header, and packets in a predefined format, the fast replay signal cannot be recorded using the saiiie format. The recording signal format forming means is therefore complicated.

Moreover, the fast replay data is used in common for 1 () all the replay speeds, so that the period at which one screen of image data is reproduced and displayed during fast replay at each speed is. determined by the time for which the region in the tape longitudinal direction in which one screen for fast replay is recorded. Accordingly, the time for which one screen of image data is reproduced is inversely proportional to the speed. With higher speed, the picture changes quickly, while with lower speed, the picture changes slowly. As a result, the displayed image is easy to see for the viewer.

Furthermore, the region used for recording fast replay signal is limited to the region where reproduction is possible commonly for a plurality of fast replay speeds. Accordingly, the number of sync blocks for recording the fast replay signal is limited to the head scanning traces at the time of higghest-speed replay, and the amount of data which can be recorded is small.

Moreover, when considering the fluctuation in the position of the head scanning trace due to fluctuation in the tape transport speed or the drum rotary speed, the reeion from which the data is reproduced without fail during fast replay is further reduced. This is particularly problematical in connection with fast replay with a higher speed.

SUMMARY OF THE INVENTION

The present invention and the inventions of copending application No. 9507499.35 from which this application is divided and copending application No. which is also divided from application No. 9507499.3, have been achieved to alleviate the problems described above.

According to one aspect of the invention, there is provided a digital VTR for recording recording digital video and audio signals in 11 respective desig gnated areas of oblique tracks in a predetermined track format, using a rotary drum on which head of two different azimuths are mounted, comprising:

data separating means for extracting a fast replay signal from the normal recording signal; recording means for recording the fast replay signal in one region in one track per one scanning of the head, of the regions covered by the head traces and in the tracks of identical azimuth; identification signal recording means for recording an identification signal for identifying the track; and replay means for replaying the identification signal.

With the above arrangement, the fast replay data can be reproduced from one location in one track per one scanning of the head during fast replay, so that even when the track is non-linear or the scanning trace is nonlinear, the head can be scanned with reference to the region at said location where the fast replay data is recorded, and the data can be accurately reproduced.

It may be so arranged that a first recording region is provided in one track of one azimuth in which said fast replay signal is recorded, and a second recording region for recording the fast replay signal is also provided in the track of the other azimuth, and "succeeding said one track; the length of the second recording region is about half the length of the first recording region, and the center of the second recording region within the track is at about the same position as the center of the first recording region within the track.

With the above arrangement, in the case of a drum of two-head configuration, the fast replay signal in the tracks of one azimuth can be 12 reproduced from the first recording region, i,,-hile in the case of fourhead configuration. the fast replay signal in the tracks of both azimuths can be reproduced from the first and second recording regions. As a result, the total amount of fast replay data, given as the sum of the data from the heads of two different azimuths, is the same, and the screen (whole picture) can be formed from the same amount of fast replay data, regardless of the head configuration, during fast replay at the same speed.

As a result, it is possible to obtain a device with which the fast replay speed is not limited by the head configuration, and the fast replay picture quality is identical regardless of the head configuration, and the device is therefore convenient to use.

It may be so arranged that, in the upper and lower end parts of the first recording region which extend out of the region corresponding to the second recording region of the adjacent track of a different azimuth, the signal identical to those in said second recording region is recorded.

With the above arrangement, where the sub-regions formed by equally dividing the first recording region is called A I, A2, A3 and A4, in turn, the signals recorded in the regions AI and A4 are extracted, and recorded, without modification, in the second recording region, as well. In other words, the fast replay data recorded in the track of a first azimuth is divided equally and the first and fourth quarter data are recorded in the track of a succeeding, second azimuth. The data recorded in the track of the second azimuth can therefore be obtained by simple rearrangement means.

It may be so arranged that the recording means forms the fast replay signal dedicated for the particular fast replay signal for each of 1 3 the fast replay speeds, and records the fast replay signal at different positions on the magnetic recording tape.

With the above arrangement, the fast replay signals are prepared for the respective fast replay speeds, and the data is configured so that the picture is switched at an interval which facilitates watching of the reproduced picture during fast replay at each speed.

It may be so arranged that the recording means repeatedly records the fast replay signal for (M x i)-time speed replay (i = 1, 2,... n) at predetermined positions in predetermined tracks of consecutive M (M being a natural number) tracks, and repeatedly records the fast replay signal for (M x i)-time speed replay, 2 x i times, taking the M tracks as one unit for each speed.

With the above arrangement, the fast replay signal for the predetermined speed is recorded in the predetermined position in the predetermined track, of the consecutive M tracks, and the fast replay signal for the (M x n)-time speed replay is repeatedly recorded 2n times, taking the M tracks as a unit. Accordingly, during fast replay, it is C> -- sufficient if the control over the drum rotation and the tape transport speed performed in such a manner that the fast replay signal recorded at one location in the M tracks is reproduced. For instance, when the fast replay is effected at (M x n)-time speed, compared with the case in which the fast replay data is recorded at one location in M x n tracks, the amount of movement to a predetermined track in the state of transition at the time of changing the replay speed is smaller, and the reproduction of the fast replay data at the newly selected speed can be started in a shorter time.

It may be so arranged that the recording means repeatedly records the fast replay signal for 4i-time speed replay (i = 1, 2,... n) at 14 predetermined positions being a natural number) tracks, and in predetermined tracks of consecutive four (M said identification signal recording means records three types of frequency signals as pilot signal for tracking control on these four tracks, being in superimposition with the digital data.

With the above arrangement, the fast replay data is disposed taking four tracks as a unit, and the identification signals (such as the three pilot signals fO, fl. and 2 two of which (f]. and 2) may consist of two different frequency signals superimposed on the digital data signal, and the last one of which (fO) may be featured by the absence of any siggnal superimposed on the diaital data signal) for tracking control are recorded, so that, during fast replay, by the use of the identification signal, the desired track can be selected, and the fast replay data recorded in the track can be reproduced.

It may be so arranged that the digital VTR further comprises error correction code appending means for appending the error correction code formed of a predetermined number of sync bits inserted at a predetermined period in the signal sequence recorded in the magnetic recording tape, a predetermined number of ID bits succeeding said sync bits, a predetermined number of first parity bits generated from the ID bits, second parity bits generated from a predetermined number of digital data succeeding the first parity bits, third parity bits generated from a plurality of digital data extending over said sync bits, and fourth parity bits generated from the digital data and positioned at the back of said digital data; erroneous correction detection means for comparing the fourth parity bits with the first parity bits reproduced by said replay means, and detecting erroneous correction on the basis of the result of comparison.

With the above arran2ement, a fourth parity is appended only to C the digital data recorded in the sync blocks, and on the basis of the result of the fourth parity check. the burst error in which the digital data is Z continuously missing in the middle of it can be detected quickly by a relatively simple comparison means.

Moreover, on the basis of such information, the erroneous correction at the error correction decoder in a replay system at the next stage can be detected.

It may be so arranged that the error correction code appending means appends the fourth parity bits only to the fast replay signal.

With the above arrangement, errors can be detected promptly even in a fast replay in which burst errors occur frequently due to the periodical amplitude fluctuation in the replay signal. According to another aspect of the invention, there is provided a digital

VTR for recording digital video and audio signals, in designated areas on oblique tracks of a magnetic recording tape, in a predefined format, using a rotary drum on which heads of two different azimuths are mounted, and replaying from the areas, comprising:

data separiting means for extracting digital video signal (hereinafter referred to as fast replay signal) used for fast replay, from a normal recording signal; recording means for recording the fast replay signals for the respective fast replay speeds, in predefined consecutive regions in a predefined track of a group of four consecutive tracks; identification signal recording means for recording identification signal for identifying the tracks; replay means for replaying the recording signal for normal replay, or fast replay signals for +2-time speed replay, or +4N-time 16 speed replay or (-4N+2)-time speed replay (N being a positive integer); and tracking control means for performing tracking control so that said head scans the predefined regions in the predefined track of the four tracks in accordance with the identification signal.

With the above arrangement, four tracks are taken as a unit, and identical pattern is repeated every four tracks, and the data for each fast replay speed is recorded in the specific consecutive sync blocks in specific track, and during fast replay, the tracking is controlled at the specific position on the specific track. As a result, it is possible to increase the recording rate of the fast replay data.

It may be so arranged that the identification signal recording means comprises:

recording means for recording, as said identification signal, pilot signals of two different frequencies alternately, every other tracks; and said tracking control means includes comparison means for comparing the levels of the identification signals of the two different frequencies contained in the replay signal, while the head is scanning the position corresponding to the center of the area where the fast replay signal for the particular fast replay speed is recorded.

With the above arrangement, during fast replay, by comparing, at a specific timing, the levels of the identification signals of two different frequencies contained in the replay signal, and effecting tracking control on the basis of the result of the comparison, the head scans the areas where the data for the respective fast replay speed is recorded. As a result, even if there is non-linearity in the track, or the like, it is possible to accurately track the region where the necessary data is recorded.

17 It may be so arranged that the identification signal recording means comprises:

recording means for recording, as said identification signal, pilot signals of two different frequencies alternately, every other tracks; and said recording means records sync block numbers together with the fast replay signal; said tracking control means compares the levels of the identification signals of the two different frequencies contained in the replay signal, when the sync block number of the predefined sync block in the area where the fast replay speed signal for the particular fast replay speed is recorded, to achieve tracking control.

With the above arrangement, when the predefined sync block number is detected during fast replay, the levels of the identification signals of two different frequencies are compared, to detect the tracking error, and trackincy is controlled on the basis of the result of the comparison, i.e. on the basis of the detected tracking error. Accordingly, the head accurately scans the area where the fast replay data is recorded. This is, even if the position at which the fast replay data is recorded is'shifted in the longitudinal direction of the tape, the area where the necessary data is recorded can be tracked accurately.

According to another aspect of the invention, there is provided a dicrital VTR for recording digital video and audio signals, in designated areas on oblique tracks of a magnetic recording tape, in a predefined format, using a rotary drum on which heads of two different azimuths are mounted, and replaying from the areas, comprising:

data separating means for extracting digital video sig ZD nal (hereinafter referred to as fast replay signal) used for fast replay, from a normal recording signal; 18 appending means for appending sync byte, ID byte, header byte to the fast replay signal, in the same sync block configuration as said recording signal; recording means for recording the fast replay signal in areas on tracks, such that during fast replay, only one location on one track of an azimuth identical to the head is covered by the head scanning trace; identification signal recording means for recording identification signal for identifying the tracks; and g re lay means for replaying the identification signal.

p It) With the above arrangement, the areas where normal replay data is recorded, and the areas where fast replay data is recorded have an identical sync block configuration, (with identical sync, ID and header configurations) so that the appending means for appending sync byte, ID byte and header byte in the recording system, and the reading means (including the ID and header reading means) can be used in common.

The digital VTR may further comprise:

input means for inputting a password from outside; recording means for recording the password together with the digital video signal; replay means for replaying the password at the time of replay of the digital video signal; and replay inhibiting means for inhibiting display of the digital video signal unless a correct password is input at the time of replay.

With the above arrangement, it is possible to protect the program or the whole tape from unauthorized replay.

According to another aspect of the invention, there is provided a digital VTR for recording digital video and audio signals, in desigmated areas on oblique tracks of a magnetic recording tape, in a predefined 19 format. using a rotary drum on which heads of two different azimuths are mounted, and replaying from the areas, comprising: data separatin&, means for extracting digital video signal 1.

(hereinafter referred to as fast replay signal) used for fast replay, from a normal recording signal; recording means for disposing a fast replay signal for an (M x i)time speed replay (i = 1, 2,..., n) at predefined positions on predefined tracks of consecutive M tracks (M being a natural number), and repeatedly recording the fast replay signal for (M x i)-time speed replay, (2 x i) times; identification signal recording means for recording identification I signal for identifying the tracks on which the fast replay signal is recorded; and replay means for performing replay at an arbitrary replay speed which is an even-number of times the normal speed, and is lower than the (M x n)- time speed, using the fast replay signal recorded for (M x n)-time speed replay.

With the above arrangement, the data recorded for (M x n)-time speed replay can be all replayed at an even-multiple speed lower than the (M x n)-time speed, although the reproduced data may be duplicated.

According to another aspect of the invention, there is provided a digital VTR for recording digital video and audio signals, in designated areas on oblique tracks of a magnetic recording tape, in a predefined format, using a rotary drum on which heads of two different azimuths are mounted, and replaying from the areas, comprising:

ZD sync block forming means for forming sync blocks by appending sync bytes to digital signal recorded in the magnetic recording tape at a predetermined interval; data separating means for extracting a fast replay signal from the normal recording signal; recording means for sequentially and repeatedly recording n pieces of data Di (i = 1, 2,... n, n being a natural number) each of which can be recorded in one sync block, over (n + 2 x w) consecutive sync blocks Sj 0 = 1, 2,... (n + 2 x w)) at identical positions on predefined tracks; wherein n is a maximum number of sync blocks which can always be reproduced from the track regions overlapping with the head scanning traces during m-time speed replay, w is a minimum natural number which is not smaller than the maximum shift from the reference position at which the head crosses a specific track, during m-time speed replay.

With the above arrangement, the maximum amount of data a head can produce from one track at a predefined fast replay speed is recorded repeatedl in the vicinity of the head scanning trace, taking account of the head position fluctuation, the maximum amount of data which is recorded can all be reproduced during fast replay. All the data can be read during fast replay in which the effect of the head position fluctuation is large.

It may be so arranged that the recording means repeatedly records the fast replay signal in (n + 2 x w) consecutive sync blocks Sj at an identical sync block position on each track, on at least m consecutive identical-azimuth tracks.

21 With the abo,,e arrangement, the fast replay signal is repeatedly recorded at identical positions on consecutive tracks, so that the fast replay signal can be replayed whichever track the head begins scannin 9 during fast replay.

Accordingly, control over the head scanning position is simplified, and the fast replay at an arbitrary speed is possible as long as the head passes the predefined track positions.

sync block forming means for forming sync blocks by appending sync bytes to digital signal recorded in the magnetic recordin tape at a rD 9 predetermined interval; data separating means for extracting a fast replay signal from the normal recording signal; recording means for sequentially and repeatedly recording p pieces of data Di (i = 1, 2,... p, p being a natural number not more than n) each of which can be recorded in one sync block, in (p + L + 1) consecutive sync blocks Sj 0 = 1, 2,... (p + L + 1)) at the same position in each track, in at least m tracks of consecutive identical-azimuth tracks in such a manner as to satisfy ek+I = mod [{ek + p - mod(p + L + 1, p)j, p] where ek and ek+1 (integers not less than I and not more than p) are the suffixes i to the data D first recorded, where n is the maximum number of sync blocks which can always be reproduced consecutively 22 from the region of the track on the tape w.-erlapping with the head scanning trace durincr m-time speed replay.

1:1 L is the number of sync blocks which is a minimum integer not smaller than (D - B + C) where C is the difference between the starting positions of the tracks Tk and Tk+1 in the track lonoritudinal direction, D is the different between the positions, in the track longitudinal direction, at which the head crosses with the respective tracks, B is the length of the region from which the reproduction from one track is possible consecutively, during m-time speed replay, and I mod [a, b] expresses the remainder of a divided by b.

With the above arrangement, the arrangement of data repeatedly recorded on the tracks is such that the different data recorded on two identicalazimuth tracks proximate to each other and crossed by the head during one scanning are reproduced at least once without fail, so that the fast replay data can be recorded with a minimum number of repetitions. With the arrangement of data described above, even when the head scanning trace position fluctuates or the head trace phase is shifted, reading of the fast replay data is ensured, and images can be Z reproduced with a good quality, and much fast replay data can be recorded and reproduced.

data separating means for extracting intra-frame encoded image data, from an input bit stream; 2 -33 recording means for forming fast replay sign I __ als for a plurality of fast replay speeds from the image data, and recording the nl-time fast speed signal in an area therefor, at positions designated according to the corresponding position on the screen of the signals, with the signals corresponding to the edges of the screen being positioned at the ends of the recording region on the oblique track, and with the sig als gn corresponding to the position toward the center of the screen being I positioned toward the center of the recording region on the oblique track; and replay means for performing fast replay at an n2 time speed (n2 > n 1) by replaying the n I -time fast replay si n-al.

With the above arrangement, the fast replay signal of the central part of the screen is collectively recorded in the center of the area recording the n I -time fast replay signal, and replay is conducted at a fast replay speed n2, higher than nl.

Accordingly, although the areas from which the sigmal is replayed is narrowed because of the increase of the replay speed to n2, the central part of the screen can be replayed.

Embodimerits of this invention and of the inventions in the two copending applications will now be described by way of example with reference to the accompanying drawings. Embodiments I and 2 are embodiments of the invention in parent application 9507499.3; Embodiments 3 and 4 are embodiments of the invention in divisional application, and Embodiments 5 to 19 are embodiments of the present invention.

BRIEF DESCRIPTION OF THE DRAWINGS

In the drawings; 24 Fig. 1 is a block diagiparn showing a recordin system of a digital C1 9 VTR of Embodiment 1; Fig. 2A shows a transport packet of an input bit stream; Fig. 2B shows a data packet recorded on the magnetic tape; Fig. 3 shows a code configuration of the error correction block in a digital VTR of Embodiment 1; Fig. 4 shows a track configuration of a digital VTR of Embodiment 1; Fig. 5A to Fig. 5C show typical head arrangement on a rotary drum used in the SD mode, of 1 ch x 2 system, 2 ch x 1 system and 2 ch x 2 system, respectively; Fig. 6 is a table showing the number of sync blocks from which data is obtainable at each replay speed; Fig. 7A shows disposition of the special replay data recording areas in the track in an example of recording format of a digital VTR of Embodiment 1; Fig. 7B shows the data and the magnitude of the recording areas in the same example; Fig. 8 shos an example of manner of division of the error correction block in a digital VTR of Embodiment 1; Fig. 9 shows a recording format on a track in a digital VTR of Embodiment 1; Fig. 10 is a block diagram showing a replay system of a digital VTR of Embodiment 1; Fig. 11 is a flow chart showing the decoding, algorithm in the third error correction decoder; Fig. 12 is a diagram showing the rotary head scanning trace during fast replay in a 1 ch x 2 head system; Fig. 13A to Fig. 13C respectively show the tracking control point t for the rotary head at each of different replay speeds, for explaining the tracking control operation of a digital VTR of Embodiment 1; Fig. 14 is a dia am showing the head scanning trace during four- C gr time speed replay in Embodiment 2; Fig. 15A and 15B respectively show the replay signals from the respective rotary heads, and the tracking control points for explaining the tracking control operation in Embodiment 2; Fig'. 15C shows the synthesized replay data; Fig. 16 is a block diagram showing a recording system of a diaital VTR of Embodiment 3; CI Fig. 17 shows the configuration of one track in a recording format in Embodiment 3; Fig. 18 shows the track configuration in Embodiment 3; Fig. 19 is a block dia am of a replay system of a digital VTR in gr Embodiment 3; Fig. 20 is a block diagram showing a recording system in Embodiment 4; Fig. 21 shoWs digital video data of a macro block configuration; Fig. 22 shows coefficients of frequency components; Fig. 23 shows dispositions in the special replay data recording areas in tracks in Embodiment 4; a system Fig. 24A is a block diagram showing a signal processin., in a recording system of a digital VTR in Embodiment 5; Fig. 24B is a block diagram showing an example of special data forming circuit in Fig. 24A; Fig. 25 is a block diagram showing a sync block forming circuit; 26 Fig. '26A to Fig. 26F show the configurations of the special replay data recording areas in Embodiment 5; 1 Fig. 27 shows dispositions of the special replay data recording areas in tracks in Embodiment 5; Fig. 28 is a block diagram showin a modulator in front of a 9 recordin(.z amplifier; Fig. 29 shows a recording format on tracks in Embodiment 5; Fig. 3)0 is a block diagram showing a special replay data forming circuit in Embodiment 6; Fig. 31 is a block diagram showing an example of sync block forming circuit according to Embodiment 7; Fig. 32 shows an example of data packet according to Embodiment 7; Fig. 33 shows a recording format on tracks in a digital VTR according to Embodiment 8; Fig. -34 is a block diagram showing the configuration of a capstan servo system; Fig. 35 shows a specific configuration of attracting error detector in Fig. 34; Fig. 36 shows head scanning traces during +2-time speed replay in a digital VTR in Embodiment 8; Fig. 37 shows head scanning traces during +4-time speed replay in a digital VTR in Embodiment 8; Fig. 38 shows head scanning traces during +16-time speed replay in a digital VTR in Embodiment 8; Fig. 39 shows head scanning traces during +8-time speed replay in a digital VTR in Embodiment 8; 27 Fic C"). 40 shows head scanning traces during -2-time speed replay in a digital VTR in Embodiment 8; Fig. 41 shows head scanning traces during -6-time speed replay in a digital VTR in Embodiment 8; Fig. 42 shows head scanning traces during -14-time speed replay in a dieital VTR in Embodiment 8; Fig. 43 shows a specific configuration of a tracking error detector according to Embodiment 9; Fig. 44 shows head scanning traces during +4-time speed replay in a diaital VTR of a modification of Embodiments 8 and 9; Fig. 45 shows head scanning traces during +4-time speed replay in a digital VTR of another modification of Embodiments 8 and 9; Fig. 46 shows rotary head scanning traces during +4-time speed replay of special replay data of a recording format according to Embodiment 10, by means of a 1 ch x 2 system; Fig. 47 shows rotary head scanning traces during +4-time speed replay of special replay data of a recording format according to Embodiment 10, by means of a 2 ch x 1 system; Fig. 48 shovs rotary head scanning traces during +4-time speed replay of special replay data of a recording format according to Embodiment 10, by means of a 2 ch x 2 system; Fig. 49 shows rotary head scanning traces during +8-time speed replay of special replay data of a recording format according to Embodiment 10, by means of a I ch x 2 system; Fig. 50 shows rotary head scanning traces during +8-time speed replay of special replay data of a recording format according to Embodiment 10, by means of a 2 ch x I system; 28 Fig. 51 shows rotary head scanning traces during +8-time speed 1 replay of special replay data of a recordin fonnat - accordin a to 9 C> Embodiment 10, by means of a 2 eh x 2 system; Fig. 52 shows rotary head scanning traces during +16-time speed replay of special replay data of a recording format according to Embodiment 10, by means of a 1 eh x 2 system; Fig. 53 shows rotary head scanning traces during + 1 6-time speed replay of special replay data of a recording format according to In Embodiment 10, by means of a 2 eh x 1 system; Fig. 54 shows rotary head scanning traces during +16-time speed replay of special replay data of a recording format according to Embodiment 10, by means of a 2 eh x 2 system; Fia 9 a signal processor after the C. 55 is a block diagram showin error correction decoding in a replay system according to Embodiment 10; Fig. 56 is a block diagram showing a signal processor before error correction decoding in a replay system according to Embodiment t5 Fia. 57 shows an example of data packet according to Embodiment 12; Fig. 58 shows another example of data packet according to Embodiment 12; Fig. 59 is a block diagram showing a signal processor after error correction decoding in a replay system according to Embodiment 13; Fig. 60A and Fig. 60B show the configuration of a password area according to Embodiment 13; 29 Fig. 61 shows rotary head scanning traces during +6-time speed replay of 8-time speed replay data of a recording format according to Embodiment 14, by means of a 1 ch x 2 system; Fi. 62 shows rotary head scanning traces during +6-time speed 9 11 replay of 8-time speed replay data of a recording format according to Embodiment 14, by means of a 2 ch x 1 system; Fi. 63 shows rotary head scanning traces during +6-time speed 9 replay of 8-time speed replay data of a recording format according to Embodiment 14, by means of a 2 ch x 2 system; Fig. 64 shows rotary head scanning traces during + 12-time speed replay of 4-time speed replay data of a recording format according to Embodiment 15, by means of a 1 ch x 2 system; Fi2. 65 shows rotary head scanning traces during +12-time s eed 1= -- p replay of 4-time speed replay data of a recording format according to Embodiment 15, by means of a 2 ch x 1 system; Fi. 66 shows rotary head scanning traces during +12-time speed 9 replay of 4-time speed replay data of a recording format according to Embodiment 15, by means of a 2 ch x 2 system; Fig. 67A shows the configuration of 4-time speed replay data recording areas used in fast replay according to Embodiment 15; Fig. 67B shows the position on the screen which is reproduced in Embodiment 15; Fig. 68 is a block diagram showing a recording system in a digital VTR in Embodiment 16; Fia. 69 shows a rotary head scanning trace on tracks during fast replay; Fig. 70 shows a rotary head scanning trace during replay at a 56time speed; Fi. 71A shows scanning traces with which three sync blocks can 9 be reproduced; Fig. 71B and Fig. 71C show scanning traces which result with forward and backward shifts in the position; Fi. 72 shows disposition of the fast replay data according to 9 Embodiment 16; Fig. 73 shows an example of disposition of fast replay data on tracks according to Embodiment 16; Fig. 74 is a block diagram showing a replay system of a digital VTR in Embodiment 16; Fig. 75 shows the positional relationship between the scanning traces and the fast replay data according to Embodiment 17; Fig. 76 shows an example of disposition of fast replay data according to Embodiment 17; Fi(Y. 77 shows a rotary head scannin trace during fast replay at a 9 56- time speed according to Embodiment 18; Fig. 78 show sync blocks which can be reproduced when the position of the rotary head scanning trace is shifted; 31 Fig. 79 shoWs s.,iie blocks which (..ail be reproduced wher) the position of the rotary]lead scanning trace is shifted; Fig. 80 shows the positional relationship between a scanning trace and the fast replay data accor(Mig to EmbodIment 18; Fig. 81A shows a scanning trace with whIch three sync blocks can be reproduced, Fig. SIB shows a scanning trace with a shift in the position; Fig. 82 shows ail example of disposition of fast replay data according to Eviil)o(tinie.yit 18; Fig. 83 shows another example of disposition of fast replay data according to Embodiment 18; Fig. 84 shows an example of disposition of fast rep.lay data on j.(leiit- llc<-i]---izi.itiutli tracks Al and A2, during 56-time speed replay, according to Embodiment 18; Fig. 85 shows another example of disposition of fast replay, data oil identical-azimuth tracks A1 and A2, during 56-time speed replay according to Embodiment 18..

Fig. 86 shows ail example of' disposition of fast replay data according to Embodiment 18; Fig. 87 shows ail example of disposition of iii-time speed replay, data according to Embodiment 18; Fig. 88 shows ail example of disposition of fast replay data oil identical-azimuth tracks A1 and A2, during 30-time speed replay according to Embodiment 19; Fig. 89 shows another example of disposition of fast replay data oil ideritical-azimuth tracks Al and A2. during 30time speed replay, according to Embodiment 19; Fig. 90 shows an example of disposition of fast replay data according to Embodiment 19; Fig. 91 shows ail example of disposition of fast replay data oil identical-azimuth tracks Al and A2, during 56-time speed replay according to Embodiment 19; 32 Fig. 92 sll()ss,s an example of disposition of replay data on identical-azimuth tracks A1 an(l A2, (luring 44-tinle speed replay according to Embodiment 19; Fig. 93 shows a track pattern ii) a conventional common consurner digital VTR; Fig. 94A shows rotary head scanning traces (luring normal replay ill a conventional digital VTR; Fig. 94B shows a rotary head scanning trace during fast replay; Fig. 9.5 is a block diagrarn showing all example of recording system ill a conventiona-1 (Ligital. YM:

Fig. 96A shows normal replay ill an example of' replay system 1n a conventional digital V'I'R; Fi g. 96B shows fi-ist replay ill the same. examp] e of system; Fig. 97A a head scanwhig trace ill a common fast replay:

Fig. 97B shows track regioj)s from which reproduction is possible.

Fig. 98 shows over-3apping portions of the copy areas between a plurality of fast replay spee(Is; Fig. 99 shows ill example of rotary heacl seanning traces with different tape speeds; Fig. 1.00A and Fig. 100B respectively show rotary head scanning traces during five-time speed replay; Fig. 101 shows a recording format oil a track it) a conventional digital VTR; Fig. 102A and Fig. 102B show an example of' configurations of a track contairling vi(leo an(l amlio data; and Fig. 103 shows all example of the configuration of one sync block oil a magnetic tape.

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Embodiment 1

33 Fig. 1 is a block djagram showing a recording sy-stem of a digital VTR of an embodIment of the invention. In the drawing, reference numeral 50 denotes an iiiptit terminal for recei7ijig digital video arid audio signals in the form of a bit strearn, 52 denotes a packet detector for detecting packets of the video arid audio signals from the bit strean) that is received. 54 denotes a first nieniory, for storing the bit stream, and.56 denotes an intra detector for detecting intra-eneoded data in the bit stream, 58 denotes weniory for storing intra detector 56.

the intra-encoded data output Reference nUffleral 60 denotes a second f rom ti)e a fi r s t error correction encoder for appending error correction c:o(lec, to I.lie data output-. from the second nieniory Reference numeral 62 denotes a data synthesizer for s37nthesizing the data output from the first inemory 54 and the first error correction encoder 60 to form a recording bit stream, arid 64 denotes a second error correction encoder for appending error correction codes stipulated by, the SD standard, to the recording bit stream output froni the data synthesizer 62. Reference numeral 66 denotes a recording amplifier, 68 denotes a rotary drum and 70a and 70b denote rotary, heads.

Fig. 2A arid Fig. 2B show an example of configuration of a packet of the digital data. Fig. 2A shows a transport packet of the input bit stream, arid Fig. 2B shows a data packet recorded on the magnetic tape. Fig. 3 is a diagram showing the configuration of the codes of <in error correction block of' the digital WIT of the embodirnent of' the invention. Fig. 4 is a diagram sliowing a track configuration of the digital VTR of an embodiment of the invention.

Fig. SA to Fig. 5C show typical head arrangements on the rotary drum used in the SD mode. Fig. 6 is a table showing the number of sync blocks fromwhich data can be 34 obtained at each of various replay speeds. Fig. 7A and Fig. 7B show ail example of a recording format. Fig. 7A shows ail arrangement of the special replay data recording areas, arid Fig. 7B shows the data in the recording areas and their sizes.

Fig. 8 shows an example of nianner of division of the error correction block of the digital VTR of ail embodiment of the invention. Fig. 9 shows a recording format or) tracks of the digital VTR of an embodiment of the invention.

Operation during recording of Embodiment 1 will next be described ivith reference to Fig. 1 to Fig. 9. The bit stream received at the input terminal 50 coritains digital ti(lec) sigrial, the digital audio signal. and di.gital data concerning the video and audio signals. The bit stream is trall Sill j, 11 ted, being divide(] int.o packets shown in Fig. 2A. Each packet is formed of a header section 92 of 4 bytes and data section 94 of 184 bytes.

In Embodiment 1, the bit stream is detected, transport packet by transport packet. Two transport packets having been detected are converted into a recording data block of.5 sync blocks as shown ill Fig. 2B, arid recorded. Accordingly, the transport packets of the bit stream input via the input terminal 50 are detected by the packet detector 52, and are input ill the first memory 54 and the intra detector 56.

At the first memory 54, the data of the bit stream is stored Imacket by packet, arid read to form the configuration of the recording data block shown ill Fig. 2B. Ill the example shown in Fig. 2B, the data length in orie sync block is 77 hytes, arid five sync. blocks form two transport packets. Ill the drawing, 111 denotes a first header, 1.12 denotes a second header Recorded ill the first header 111 are identification data for indication the number of the sync block in the five sync blocks, arid the like. Recorded in the second header 112 are identifi cation data for indicatjri; whether the clata in the data sect-ioll is N-1(1(,,o data or audio data. II)cidelitallY, in E-niboclimmit 1. reading of' clat-a Vrom the first memory 54, the secom] memory 538, to be described later, is coticlucted in accordance with a command from the data synthesizer 62.

nie bit stream output from the packet cletector 52 is inpul. to the intra detector 536, where judgemejit is macle oil whether the clata in the transport packet is intra-encoded data or not. In the Nll-"E(;2 bit stream, when the bit stream is intra-frame or iiitra-field encoded (intra encodecl). the intra traiisport packets are transmitted consecuti%-ely. These are (letected, wid the only intra transport packets are ex t ra (1: c (I. The extracted tr;msport packelis are iiiput to the secoml memory. 58.

The iyitxi-f'rame trawsport packet data ii1put: to the second memory 58 is stored packet by packet, as at the first - the seCOTICI memory.5 so memory 34. The data is read from 8 that it Js of the recordiiig data block configuratiozi Shoml in Fig. 213, like the data from the first memory 54 That is, the data length within one S3'T)C blOCk is T7 bytes. and two transport packets are recorded over five syne 1) 1 o (- ks In the clrawiiigs, Ill denotes a first. header having a dalla longth of one byte, and 1,12 denotes a second header having a data length of 2 bytes. Recorded in the first header III are identification data for discriminating each sync block from other sync blocks iri the block, identi.fication data jiulicating special replay data. and the like. Recorded in the second header 112 are identification data indicatirig the speed of the fast replay for which tile recorded special replay (]at-. ,i is intended, and the like. In Embodiment 1, reading from the second memory 58 is also conducted according to a command from the data.13,r) tiles i zer 62.

The special replay (lata read from the secumd memory.58.

36 1 taking the five syne blocks as a unit (data length within one sync block is 77 bytes) is supplied to the first error correction encoder 60 where error correction codes are appended. The operation of the first error correction encoder GO will next be described with reference to Fig. 3.

Fig. 3 shows the code configuration of the error correction code appended to the special replay data. I n embodiment 1, (85), 77. 9) Reed-Solorrion code (Cl check code) identical to the error correction code appended to the bit stream of' the ATV signal, and (20, 15, 6) Reed-Solomon code (C4 elieck code) and having a mininjum distance. identical to that of the the error correction code for the audio signal are used in the recording direction and in the vertical direction, both as first error correction code for the special replay data.

The special replay data is read from the second memory 58, fi-ve sync blocks as a unit, and 15 sync blocks are collected at the first error correction encoder 60.

and one error correction block is formed of the 15 sync blocks. C4 cheek code is appended in the vertical direction, and the Cl check code in the recording direction is appended at the sectond first error correction encoder 64, in the same way as the ATV signal output from the first memory.54, and the error correction block of the product configuration is formed.

Because the minimum distance of the C4 check code is identical to the C3 clieck (.ode of the audio signal, tile encoder may be used in common, by simply switching the code length. The code length is 14 in the case of the audio signal, and is 20 in the case of the special replay data.

With the track configuration of the SD (of the current television system) shown in Fig. 4, 149 syne blocks are provided per track for an area 96 for recording video data, as descr.ibed in connection with the prior art example (or

37 Fig. 102A and Fig. 102B). Out of the 149 syne blocks, three blocks are used for recording VAUX data, arid eleven blocks are used for recording error correction code (C2 clieck code). One sync block is formed of 90 bytes as in the prior art example shown in Fig. 103. Out of the 90 bytes, five bytes at the head are used for recording a sync pattern and an 1D signal, arid eight bytes at tile tail are used for recording error correction code (Cl cheek code), as shown in Fig. 4. The data which earl be recorded in one sync. block is therefore 77 bytes as shown in Fig. 103 and Fig. 4.

The data synthesizing operation at the data synthesizer 62 will next be described with reference to Fig. 5A to Fig. 9.

Fig. SA to Fig. 5C, show different arrangements of the licads oil the rotary drum, and respectively shosy 1 ch x 2 system in which two heads are disposed in opposition, 2 ch x 2 system in which two heads are juxtaposed, and 2 ch x 2 system in which two sets of heads are disposed in opposition. The angle over which the magnetic tape is wrapped around the drum is 180'. In Fig. 6, the number of tile sync. blocks from which data can be obtained from one track at each of the replay speeds is shown. In the drawing, 9000 rpin system means the system having the heads as shown in Fig. 5A and Fig. 5B, and 4500 rpm system means the system having the heads as shown in Fig. 5C. The track pitch in the SD standard is 10 m m, arid tile values in the drawing show tile number of syne blocks per track which can be replayed at each of tile replay speeds. where special replay is conducted using a rotary head having a width of 10 M m. It is assumed in the calculation that the number of sync blocks per track (corresponding to 180') is 186 (see Fig. 4), arid as in the prior art example the data can be obtained from the part where the output level of the replay signal is greater than 6 dB.

1 A Cl 0 A Fig. 7A shoss's the arrangement of the special replay data recording areas in the tracks of a digital VTR of Embodiment 1 of the invention, taking account of the number of syne blocks from which data can be obtained as shown in Fig. 6. In this recording format, the special replay data recording areas are repeated at an interval of four. tracks, arid the special replay data recording areas for each of the fast replay speeds are provided oil the four tracks 98, 100, 102 and 104 forming oiie iiiter,al.. In the drawing, aal arid aa2 indicate special replay data for 2-time speed. 4-time speed and -2-time speed, and bbl arid bb2 indicate special replay data for 8-1-lime speed arid -6-time speed, and cel and c.c2 indicate special replay data for 16-time speed, and -14 time speed. Provided in the first track 98 is a recording area for Lhe replay data 1AA Provided in the second track 100 is a recording area for the special replay data bb2. Provided in the third track 102 are recording areas for the special replay data aal and ccl. Provided in the fourth track 104 are recording areas for the special replay data aa2 arid ce2.

Fig. 7B shows data (the number of sync blocks) recorded in each of the special replay data recording areas. Fig. 8 shows an example of manner of' division of an error correction block at 16-time speed(-14-time speed). In Fig. 7B, identical signals are recorded in the recording areas designated with identical reference marks. For instance. data tl in special replay data aal is recorded also as special replay data aa2. The special replay data aal and aa2 are repeatedly recorded over two tracks as shown in Fig. g. The special replay data bbl arid bb2 are repeatedly recorded over four tracks as shown in Fig. 9.

Referring to Fig. 8, twenty sync blocks of the special replay data cel and cc2 for the 16-time speed and -14-time speed form one error correction block, with the above- 39 1 A r) A mentioned error correction codes (Cl and C4 codes) being appended. which is divided into four sections, each consisting of five syne blocks. The data tSa and =9a of two upper blocks are repeatedly recorded over eight tracks, and the data #8b and #9b (ECC) of the two lower blocks are repeatedly recorded over eight tracks.

Fig. 9 shows a recording format of the special replay data for 27 tracks. Recording areas for the special replay data aal, aa2, aa3,... ' bbl, bb2, bb3...., ecl, cc2, (.c3,..., are repeated at an interval of four tracks oil the magnetic tape. The areas designated with identical reference marks are used for recording identical special replay data.

The operation during the special replay is next described with reference to Fig. 9.

With reference to Fig. 6, in a system of 9000 rpm. data of 62 sync block can be reproduced at four-time speed, while in a system of 4500 rpm, data of only 31 sync blocks earl be reproduced. With the recording format shown in ig. 9, in a system of 9000 rpm, zill the special replay data aal recorded in one track can be reproduced, at four-time speed replay. This is because, as shown in Fig. 7B, data t-l,:t2, t,3 and 44 are 40 SBs in all, all the signals earl be reprodticed, In a system of 4.500 rpm, however, about 9 sync blocks earl be reproduced.

Accordingly, of the special replay data aal shown in Fig. 7B, data of several sync blocks at the bead of data:El, and data of' several sync blocks at the tall of' data #4 earinot be reproduced. In the digital VTR of Embodiment 1 of the present invention, auxiliary data for use in a system of 4500 rpm is recorded as the special replay data aa2. (The manner of configuring one error correction block in a system of 4500 rpm will later be described in connection with Embodiment 2.) 1 Referring agaill to Fig. 1, the data output from Lhe first memory 54 aiicl the first first error correction encoder 60 are input to the data synthesizer 62, at which the data from the first memory.54 and the first first error correction encoder 60 are synthesized, to forIT1 a predetermined track format. The operation of the data synthesizer 62 will next be described briefly.

Fi-ve sync blocks of the bit stream of the ATV signal stored in the first memory 54 form two transport packets, as shown ill Fig. 2B, and the bit stream is read from the first memory 54, one sync block as a unit. at a predetermined timing, and are disposed ill areas other thaii the special replay data recording areas ill the ATV areas (hereiriafter referred to as main areas) oil the recording tracks in Fig. 4. The data s,iii. liesizei. (32 geiieral-.es a contrul s.lgiizil for controlling the timing of reading the data from the first memory 54, and the data read out are synthesized oil the basis thereof.

The data of the 20 syric blocks having the er ror correction code appended at the first error correction encoder 60 is output to the data synthesizer 62 at a predetermined timing. Specifically, prior to the tinte (delay time) necessary for the formation of the error correction (. ode frorn the second memory 58, a control signal for reading data from the second meritory 58 is output from the data synthesizer 62. That is, the data synthesizer 62 synthesizes the data front the first memory 54 and the second memory 58, to forni a recording format shown in Fig. 9. Th e ATV signal synthesized into a predetermined format at tile data synthesizer 62, and recorded in the videoareas for oile track, and the special replay data having the C4 check code appended is ijil)ijt to the second error correction encoder 64. At the data synthesizer 62, the track format for each track is formed, so that f'otir tracks form a otie cycle. 111 41 Eijibudiment 1, the recol-ding of' the special j-el)lii, dalla repeated according to each of the replay speeds is prepared in the second memory. 58. That is, memory regions for storing data for each of the replay speeds are prepared ill the memory.58, and the data is refreshed at a predetermined period.

In the second error correction encoder 64, error correction code (C2 cheek code) is appended, in the vertical direction, to the data recorded ill the video areas synthesized at the data synthesizer 62, and the error correction code (Cl check code) is appended, ill the recording direction, thereafter. Thus, the Cl cheek code i appended to the special replay data shown in Fig. 3. at this timillg. The recording data having the error correction code are 1.o digit'll itio(lillIi..ioll, and iliil)lif,ie,(1 at the recording amplifier 66, and recorded oil the magnetic tape by means of the rotary heads 70a and 70b.

Fig. 10 shows a block diagram of a replay system of a digital YrR of' an embodiment of the invention. Ill the drawing, the rotary drum 68, the rotary, heads 70a and 70b are identleal to those ill Fig. 1. Reference numeral 72 denotes a)lead amplifier, 74 denotes a signal detector for detecting digital data from the replay signal, and 76 denotes a digital derijodulator for applying digital demodulation to the replay digital data output from the signal detector 74. Reference numeral 78 denotes a first error correction decoder for correcting or detecting errors contained in the replay signal, using the Cl check code (the error correction code in the recording direction), 80 denotes a second error correction decoder for correcting or detecting errors which have not been corrected by the Cl check code (errors detected, or not detected), using the C2 cheek code (the error correction code appended to the video signal ill the vertical direction), 82 denotes a third 42 memory, 84 denotes a third error correction decoder for correcting or detecting errors, using the error correction code (hereinafter referred to as C4 check code) in the vertical direction for the special replay data shown in Fig. 3, during replay of the ATV signal, 86 denotes a fourth memory, 88 denotes a switch, and 90 denotes a data output terminal Fig. 11 shows a decoding algorithm in the third error correcting decoder. Fig. 12 shows scanning traces of the rotary head 70a in a digital VTR at fast replays in a 1 ch x 2 head system.

The numerals "2", -4", "C, and "IC at the starting points of the arrows in the drawing indicate that the respective arrows are scanning traces for double speed replay,, four-time speed replay, eiglit-time speed replay, and 16-time speed replay are conducted with the digital VTR.

Fig. 13A to Fig. 13C are for explaining the tracking control operation in a digital VTR of an embodiment of the invention. Fig. 13A to Fig. 13C respectively show tracking control points of the rotary head at the respective replay speeds. They show the tracking control positions, and the output patterns of the replay signal output from the rotary head 70a which result when double speed replay. four-time speed replay, eight-time speed replay and 16-time speed replay are conducted in a digital VTR having a rotary head configuration shown in Fig. SA or Fig. 5B.

The operation of the replay system will next be described with reference to Fig. 10 to Fig. 13.

During normal replay, data replayed via the rotary heads 70a and 70b from the magnetic tape is amplified at the head amplifier 72, and a signal is detected at the signal detector 74, and converted into replay digital data at the digital demodulator 76. The digital-demodulated signal is subjected to error correction and detection at the first 43 error correction decoder 78, using the Cl check code appended in the recording direction (this decoding will herein after referred to as Cl decoding). The errorcorrected data is input to the second error correction decoder 80 arid the third error correction decoder 84.

At the second error correction decoder 80, error correction or decoding is conducted using the C2 check code (error correction code appended in the vertical direction) for the data which have riot been error-corrected by the Cl check code (the data for which an error has been detected, an(] the data which contains an undetected error) This error correction decoding is hereinafter referred to as C2 decoding. The data having received the C2 decoding is input to the third memory 82, where the bit stream of the ATV signal is separated from the input data, and only the bit stream is stored in the memory. The special replay data is discarded at this stage, as in the prior art example.

At the third error correction decoder 84, data replayed from the special replay data recording areas is separated from the data input to the third error correction decoder 84, to form one error correction block shown in Fig. 3. separation of the data from the special replay data recording areas is accomplished by detecting the positions of the special replay data recording areas on the track by referring to the sync block numbers recorded in the ID signals in the syne blocks, and detecting the i dentif i cation data in the header 112 in the syne blocks, arid judging whether the data is the special replay data or the bit streani of the ATV signal.

When the above-mentioned one error correction block is formed, the third error correction decoder 84 conducts error correction or detection on the data which has riot been error-corrected (the data for which an error has been detected, and the data which contains an undetected error) 44 with the Cl cheek code. using the C4 check (.ode (error correction code appended in the vertical direction of tile special replay data). This decoding is hereinafter referred to as C4 decoding. The data having received the C.4 decoding is input to the fourth memory 86.

In Embodiment 1, the minimum distance of the C4 clieck code for the special replay data, and the minimum distance for the C:3 check code for the audio data are Twide, to be ident.ical. The reason for this is as follows. The audio signal in the ATV signal. is tramsmitted. together with the digital video signal, it is recorded in the video signal - s. rather than in separate audio signal areas.

a r ea. Accordingly, during replay frOM a magnetic tape of the (ligi.l;il VTR recording the ATV signal, the error correction decoder for. 1-he audio signal is not. use(]. In Embodiment 4, by making the minimum distance of the C4 check code and the mininwin distance of the C3 check code identical, as described it)ove, the third error correction deco-der. 84 is used also as the error correction decoder for the audio signal. In this way, the size of the circuit is reduced. There is however some addition of circuits. This will be later described.

The fourt.h memory 86 stores the special replay data havinz heen subjected to the error correction. During normal. replay, the data selector 88 selects the output of the third memory 82. and the bit stream of the ATV signal restored at the third memory 82 into packet information of 188 bytes is output vi.,i the output terminal 90.

The operation in the still mode wIll next be described.

The still replay may be started by transition frovil a normal replay, or by selection in the state of halting. First-. description is made for the case where the still replay is started by transition from normal replay.

When the still mode is selected during normal replay.

the replay data is stopped, and input of data to the third memory 82 and the fourth memory 86 is interrupted. Th e selector 88 selects the output of the fourth memory 86 to output the still picture via the output terminal 90. Data shown in Fig. 213, other than Ill and H2, i.e., the data of the transport packet is stored in the third in(] fourth memories 82 and 86. The intra-encoded data having received the error-correction at the third error correction decoder 84 is stored in the fourth memory 86, so that it is only necessary to sequentially read the data stored, transport packet by transport packet. The configuration may be such that, during still replay, the data of the transport packets replayed from the special replay data recording areas for the double speed, four-time speed and -2-time speed having the most recording data amount is output. During normal replay, as the data used for still replay, the special replay data for 2-time, 4-time or -2-time speed replay may be decoded, and stored for use as the data for still replay.

Next, the situation where the still mode is selected from the state of halting is described. In the state of halting, no data is present in the third and fourth memories 82 and 86. If, in this state, the still mode is selected, it is necessary to conduct normal replay to store the data for one screen in the fourth memory 86, and stop the tape. In the case of still replay, the still mode signal is output to the decoder of the ATV, and the still picture may be formed at the memory of the ATV. Alternatively, transport packets indicating no motion compensation prediction error (i.e., the transport packets indicating a still picture) may be formed at the digital VTR and is kept output.

The operation during fast replay will next be described.

The description will be made with regard to the rotary

46 head (.oi)-1i.gtiratioii shown ill Fig. 5A. Fig. 12 shows scalillilig traces of the rotary head 70a which result when replay is made at double speed. four-time speed, eight-time. speed and 16-time speed. The scanning traces of the rotary head also result when the rotary Ilead configuration is as shown in Fig. 513. However. with regard to the head 70b, the traces will be entirely different because of the different head disposition.

First. the tracking control system during fast replay in Embodiment 1 is described with reference to Fig. 12 and Fig. 13A to Fig. 13C. During fast replay, the data is intermittently replayed, as described above. The number of syne blocks replayed from one track at the respective replay speeds is as shown ill Fig. 6. The special replay data call be obtal.ned effectively, by controlling the tracking of tile rotary head 70a so as to maximize the replay output around the areas where tile special replay data is recorded at the respective replay speeds. Fig. 13A to Fig. 13C show tile tracking control points for the rotary head 70a at the respective replay speeds. With the recording format shown in Embodiment 1, ill a system of 9000 rpm. the data of one error correction block shown ill Fig. 3 can be formed without using the data replayed via the rotary head 701). Accordingly, Fig. 12 omits showing the scanning traces of the rotary Ilead 70b.

The operation of the replay system during fast replay will next be described with reference to Fig. 10 to Fig. 13C. When a fast replay mode signal is input, the selector selects the output of the fourth memory 86. The replay data intermittently replayed via the rotary heads 70a and 70b is amplified at the head amplifier 72, arid converted to the replay digital data at the signal detector 74, and digitaldecoded at the digital decoder 76. Tile data having its sync data correctly detected at the signal detector 74 is 47 subjected to error correction using the Cl check code at the first error correction decoder 78. The Cl-decoded data is input to the third error correction decoder 78. The output of the first error correction decoder 78 is also input to the second error correction decoder 80, but as the data is intermittently replayed. C2 decoding cannot be conducted, and transport 1),i(.ket.s cannot be generated.

The operation of the third error correction decoder 84 ss-ill next he described with reference to Fig. 11 and Fig. 12.

From the data iiiptit to the third error correction decoder- 84. the data rrom the special replay data recording areas is detected, and orie error correction block shown in Fig. 3 is forme.d. The separation of the data from the replay data recording areas is icc(.)riil)li.qlle(l by replay. data recording syne block numbers bl-ocks. and judging the ATV signal or the ie in the syne of the special d e t e c. tl n g areas on the track by referring to the recorded in the ID signals in the sync whether the data is the bit stream of' replay. data by referring to t) block.

Whem orie error correction])lock is tInis Cormecl, the third error correction decoder 84 conducts (le(.ocl.itig using the C4 code according to the algorithm shown in Fig. 11. When data of one error correction block is formed, the third error correction decoder 84 judges syliether the replay mode is the one for selecting the ATV signal or riot according to the control signal output from a system controller, riot shown (stel) 106). If the replay mode is riot the one for selecting the ATV signal, the code length k for conducting the C.3 decoding is set to be "14" (step 108). If the replay mode is the orie for selecting the ATV signal, the code length k is set to be "'-10" (step 110). When the code length is set. the third error correction decoder 84 sets the the nositions 48 erasure positions detected at the time of Cl decoding, in the third error correction decoder 84 (step 112). Tben, the syndrome for the case wliere tbe code length k equals to "20" is formed on the basis of the erastire positions (step 114). For using the eircuits in common with the C3 decoding of the audio signal. it is necessary to add a selector for changing the initial value of the counter counting the code length.

When the syndrome is formed. on the basis of the result of the syndrome formation, calculation of the error position polynomial and the error value polynomial is coilducted (step 116). This part can be used in common with the C3 decoding 1)ecaiise the minimum distance is equal. In the Chien search, the error 1)o,;itioric, and error valties are determined on tile basis of the error positions and the coefficient data of the error polynomial (stel) 118). To use the eircilits in common with the C3 decoding of the audio signal, it is necessary to add a selector for altering the initial value of the Chien search, and a selector for altering the iiiitial.,,altie of the counter (.c)iiiitiiig the code length. The error correction is effected on the basis of the error positions arid the error values (stel) 120). The above steps are repeated mitil. all the data of' orie correctioii block is completed (step 1229). The C4-decoded special rel)lay data is input to the fourth memory 86. From the fotjrl.li memory 86, the ATV bit stream having heen restored into packet information of 188 bytes is output via the selector 88 and the output terminal 90.

The manner of configuring the error correction block shown in Fig. 3 will next be described. In the digital VTR of Embodiment 1, the manner of configuring one error correctiori block differs between the the losyspeed fast replay (double speed. four-time speed, 2-time speed. eighttimespeed and -6-time speed), arid high-speed fast replay (16-time speed and 14-time speed). This is because the number of the sync blocks replayed by the rotary head 70a is 49 "12" is smaller in the case of the 16-time replay. Accordingly, all the data forming one error correction block is riot replayed during one scanning by the rotary head 70a, arid the data is disposed oil the recording tracks so that one error correction block is formed by two scannings of the rotary head 70a. This is because clianging the mininium distance of the error correction code causes increase of the size of the circuit of the error correction decoder.

Accordingly, if. orily for the 16-time speed (-14-time speed), the minimuni distance were made to be identical only for tlie 16-time speed (14-time speed) replay data arid the size of the error correction block were altered, then fi,e or six syne blocks of special replay data would be obtained for five sync block of error correction code., so tliat the rate of' data collection would be low. It is for this reason that, in Embodiment 1. data is disposed oil the recording tracks stich that data of an error correction block identical to that in other fast replay speeds can be formed over two scanning periods by the rotary head 70a.

The manner of configuring one error correction block in the case of double speed, four-time speed and -2-time speed will next be described. As illustrated in Fig. 12, in the case of double speed replay, the part aal is replayed during one scanning period of the rotary head 70a. As illustrated in Fig. 7B, data of two error correction blocks is disposed in the part aal, so that the third error correction decoder 84 applies C4 decoding to each of the error correction blocks. In the case of double speed replay,, identical error correction block is replayed twice, tile decoding may be conducted only one of the error correction blocks. The control will be the same for the reverse double speed repay (-2-time speed). In the case of the four-time speed replay, the data of the part aal is replayed during one scanning period of the rotary head, so that the operation is similar 1. " - - - to that for tile double speed.

During eight-time speed replay, the data of' part bbl is replayed during one scanning period of the rotary}lead. As sliown in Fig. 7B, data of' one error correction block is disposed in the part bbl, so that the tliird error correction decoder 84 conducts C4 decoding when the data of part bbl is replayed. ln the case of -6-time speed replay,, the operation is similar, but ail identical error correction block is replayed one out of five rotations, so that this block need not be decoded. In the case of 16-time speed rel)l,iy,, as s;lioss.ii in Fig. 6, the data replayed from one track consists of 12 sync blocks, one error correction block cannot be configured from data replayed from one track oiil,. Accordingly. in Embodiment 1, the 16-time speed replay data is (1.1vided into two tracks (see Fig. 7).

In this way, the third error correction decoder 84 configures one error correction block from the data replayed over two scanning periods of the rotary heads 70a, and conducts the C4 decoding. During the first scanning period, syne blocks including the data t8a in(] tga are replayed.

and, in the next scanning period, losync blocks including the data:L8b and are replayed, and one error correction block is thereby configured.

The operation in the slow replay will next be described.

During slow replay, the speed of magnetic tape transport is lower than in normal replay, and each oblique track is scanned and replayed several times as the tape is transported. Accordingly, of the replay digital signal, the data for wIiich the sync. signals have been correctly detected at the signal detector 74, and the sync blocks have been correctly decoded at the digital decoder 76 is extracted, and is subjected to error correction using the Cl cheek code, and the replay data for double speed. four-time speed 51 an(] -4-time speed stored in the special replay, data recording areas is extracted, and output to the third error correction decoder 84. The separation of the data can be accomplished. as in normal replay, hy detecting the positions within the track. by referring to the 1D signals contained in the sync. blocks, and identifying 11)e track by referring to the header information recorded in the data areas.

The third error correction decoder 84 configures one error correction block using the above mentioned data, and conducts C4 decoding as in normal replay. The C4-decoded data is stored in the fourth memory 86. The fourth memory, 86 synthesizes a still picture, and data stored transport packet by transport packet is sequentiall.y read. T1 i e selector 88 selects the output of the fourth menion- 86.

As described in connection with the prior-art example, during special replay (slow replay, fast replay, etc.), the rotary head crosses tbe recording tracks obliquely, so that the replay signal obtained from the tracks is intermittent. As a result, the error correction block (video data) shown in Fig. 102A cannot be obtained as in the prior art example However, in Embodiment 1, one error correction block for special replay shown in Fig. 3 is formed and recorded, so that it is possible to conduct error correction using the C4 check code for the data for which error correction using the Cl check code was not conducted. As a consequence, in the case where the symbol error rate is 0.01, the error detection rate will be 1.54 x -13 10 10, and the error detection rate is improved by 10 ' so that it is a level which is practically satisfactory. The undetected error rate is also 2.38 x 10- 16, which is practically satisfactory.

In addition, as described in connection with tile prior art example, it often happens that the symbol. error rate is

52 0.01 or more during special replay. However, with regard to the result of calculation, the error rate is of' the practically satisfactory level when the above code configilration is used, so that satisfactory special pictures can be obtained. Embodiment 2 In Embodiment 2, description is made of the operation of a system of 4. 500 rpm shown in Fig. 5C. It is assumed that the recording format is the same as in Embodiment 1. The operation during normal replay, still replay. and slow replay is identical to that ill Embodiment 1, so its description is omitted, ill(] the description is made only ill connection with the fast replay.

F1g. 14 shows the scanning traces of the rotary head at the time of four-time speed replay ill Embodiment 2..111) the drawing. the scanning 1-.races of the heads 70a all(] 701) are shown by arrows. The method of tracking control during fast replay in Embodiment 2 is similar to that ill Embodiment 1, and the tracking of' the rotary)lead 70a is controlled so that the replay output is maximum around the areas 15.1lere the special replay data is recorded.

The operation of the replay system of Embodiment 2 will next be described referring also to Fig. 10. 151hen the fast replay mode signal is input. the selector 88 selects the output of the memory 86. The replay data obtained intermittently via the rotary heads 70a and 70b is amplified at the head amplifier 72, and converted into replay digital data at the signal detector 74, and digital decoded at the digital decoder 76. The data for which the sync data is correctly detected at the signal detector 74 is subjected to error correction using the Cl check code at the first error correction decoder 78. The Cl-decoded data is input to the third error correction decoder 84. In the system of 4500 rpm shown in Fig. SC, the same number of replay signal 53 s-sl,.elTI.G ( from the head amplifier 72 to the f i rst error correcti on decoder -7 8) as the number of the channels ( i. e. two) are provided, although not shown as such, Is it does not relate to the essential feature of Embodiment 2.

With regard to the data input to the third error correction decoder 84, the data from the special repla.y data recording areas is detected, and one error correction block shown in Fig. 3 is formed. In a system of 4500 rpm. the number of' blocks from one track during four time speed replay is 31 as shown in Fig. G. 11-1 is therefore not to configure one error correction block from the data by- the rotary, head 70a. Thal, is, data to f'orm an of' one frame is not.

r e 1) 1 a y e. (1.

F.Jg. 15A to Fig. 1.5C are I'or expJaining the tracking control operation in Embodiment 2. Fig. 15A and Fig. 15B show the replay, signal rel)l.ared by the respective rotary, heads. and the tracking control points. Fig. 15C shows the s37ritlie.,,jzc,(] repla.y data. In the drawing, in the parts designated with identical reference marks (the parts designated by and "t4"), identical data is recorded.

In the- s),stem of 4500 rpm, auxiliary data rel)la,eci 1)3, the head 701) is used to form data of one error correctioll block. That is, during fourtime speed replay, a first error correction block is formed by combining the data l#1 replayed by, the rotary head 70b and the data #2 repla3,ed 1).y the rotary head 70a, and a second error correction block is formed by, combining the data:t.3 replay,ed by the rotary head 70a and the data #4 replayed by the rotary head 701). Fig. 15C shows two error correction blocks configured in the above described manner. The separation of the data from the special replay data recording areas is accomplished by detecting the positions of the special replay data recording areas by, referring to the sync block numbers recorded In the 54 ID signalc; ill the sYlIC 1)1()(1cl, and judging whe.ther the data is from the bit stream of the ATV signal or the special replay data by referriiig to the headers in the s.yil(,- blocks.

When the data of oiie error correction block is configured, the third error correction decoder 84 conducts the decoding using the C4 code according to the algorithm shown ill Fig. 11. The operation of the C4 decoding is similar to that ill Embodiment 1, so that its detailed description is omitted. The C4-decoded special replay data is input to the fourth memory 86. The ATV bit stream having been restored into the packet, informatioll of 188 bytes in the fourth memory 86 is output via the selector 88 and the Output termillal 90.

J.ii Embodimerit 2, description is made of' the case of four-time speed replay. However, Cast rep.lay cati be similarly effected at the double speed, 2-1.iiiie speed, 8time speed, 6-time speed, 16-time speed, or -14time, as in Embodimeiit 1. Moreover, by using the special replay auxiliary data reproduced by the rotary head 701). orle error correction block can be formed ill the system of 4500 rpm, like Embodiment 1. That is, data necessary for forming ail intra picture of one frame earl be reproduced. With regard to 16-time speed and -14-time speed replay, oy)e error correction block is formed by two scaimings of the rotary heads 70a and 701).

For special replay (slow replay, fast replay), the rotary head crosses the recording tracks obliquely, so that the replay signal is ol)tii.iied intermittently from the respective tracks. Accordingly, error correction blocks (video data) shown in Fig. 102A is not formed in this embodiment, like the prior art example. However, one error correction block shown ill Fig. 3 call be formed by the use of the special replay auxiliary data reproduced by the rotary head 701) ill the system of 4500 rpm described in connection with Embodiment 2. It is therefore possible to apply error correction using C4 clieck code on the data whose errors were not corrected by the error correction using the Cl cheek code. The error detection probability, for the symbol error rate of 0.01 is about 1.54 x 10-13, and the error detection probability is improved by 1010 times, and practically satisfactory results are obtained. Undetected error rate will be about 2. 38 x 10-16, which is practically, Satisfactory.

As described in connection with the prior art example, the symbel error rate is often more than 0.01 during special replay, but as far as the result of calculation concerning the error rate, practically satisfactory levels are attained with the above code configuration. and special replay pictures with good qualities are obtained. That is, the recording formats described in connection with Embodiment 1 is also suitable for all the rotary head arrangements shown in Fig. SA to Fig. 5C.

In Embodiment 1 and Embodiment 2, sync block special replay data recording areas are disposed on the recording tracks such that an error correction block is formed by one scanning of the rotary head 70a at the low-speed special replay speed (still replay. slow replay, and doxible. fourtime and eight-time speed replay). Accordingly, the storage capacity of the memory in the third error correction decoder 84 for forming one error correction block can be reduced. Moreover, the timings for control over writing of replay data into the memory and reading from the memory. and starting the error correction are synchronized with the rotation of the rotary head 70a, and the control over the memory and the error correction decoder is simplified, and the size of the circuit can be reduced.

In Embodiment 1 and Embodiment 2, where special replay is conducted at predetermined replay speeds, the special 56 replay data recording areils for the respective replay speeds are disposed collecti-,.ely at predetermined positions on the tracks, as shown in Fig. 7A and Fig. 7B or Fi.g..9. This is for the following reason. During fast replay, the tracking control is effected at the central parts of the special replay data recording area, as described above, so that if they were disposed over a plurality of tracks, it could happen that the predetermined areas cannot be replayed because of the non- linearity inherent to a WR.

If the special replay data for the respective replay speed is collectively recorded, the special replay data can be repI Jy ed without being influenced by the non-linearity of the tracks so much, and a special replay picture with a good quality can be obtained.

In Embodiment 1 and Eiril)c)(].liiiejil-. 2, the iiiiiii.iiitjiTi distance of the error correction (.ode appended at the error correction appending means is identical to the minimum distance of the error correction code appended to the digital audio signal. With this feature, by slightly modifying the error correction circuit for the digital video signal or the digital audio signal, error correction decoding can be achieved without adding a separate error correction circuit, and the size of the circuit (.an be reduced.

In particular, in Embodiment 1, the minimum distance of the error correction code appended at the error correction appending means is identical to the minimum distance of the C3 code for the audio signal. It is sufficient, in connection with the decoding, to add a circuit for setting the tralue of the counter counting the code length of the syndrome forming circuit, and a circuit for setting the value of the counter counting the number of tinies of Chien search. In Embodiment 1, the minimum distance of the error correction code appended at the error correction appending 57 -ile minimum distance of' t lic. C3 code.

Illealls is Iderillcal to 11 for the audio siglial. The Invention is riot limited to this, alid it may be identical to the minimum distance or the Cl code (the Cl decoder decodes only the special replay data during special replay, so that it has time to spare), or of the C2 code (C2 decoding is riot conducted during special decoding), and yet similar effects are obtained.

In Embodiment 1 arid Embodiment 2, the error correction blocks are so formed that the size of one error correction block is identical for the respective replay speeds, so that the (lecocliji,,," of the special replay data cari be decoded at an identical error correction circuit. As a result. the size of the circuit cari be reduced.

Where the block size of the error correction block i.,:; changed for the respective replay speeds, it is so arranged that the minimuni distance of the error correction code within one error correction block is made to be identical for the respective replay speed. With such an airangemerit. the error correction decoder can be used in common. by simply.,altie of syndrome and the initial value of the code lerigth the time of' Chien search. The effects similar to those described (such as the reduction in the circilit size) cari also be obtained.

In Embodiment 1 and Embodiment 2, the predetermi tied replay speeds are those corresponding to positive and negative tape transport speeds having the same absolute,va 1 u e. ln this connection, it should be noted that + ii-tinie eed and -(ri-2)-tinie replay speed (ii being an larger than 1) correspond to positive and nsport speeds having the same ahsolute the predetermined replay speeds are set as adding a selector circuit for setting an initial the code length setting counter at the time of formation, and the initial values of the registers setting counter at replay sI arbitrarv number negative tape tra value. Because 58 described above, it is possible to use the special replay data recording area for the positive and negative symmetrical speeds for wbich the data amount (tbe number of' sync blocks) reproduced from one track at the replay speeds corresponding to positive and iiegiil-.i,e tape transport speeds having the same absolute value, and the maximum use can be made of the special replay data recording areas to forIT) one error correction block. In particular, in the ease of a high-speed fast replay, the number of sync blocks replayed from one track is very small, as shown in Fig. 6.

Accordi ngly, the special replay speeds are set to be values cot-responding to and negative tape transport speeds having the same absolute value. and the special replay data recording areas are so disposed on the recor.ding tracks olle block is formed by two scannings of the rotary head. so that It is riot necessary to repeat special. replay data more thall necessary. Moreover, the size of one error correction block for the respective replay speeds can be nuide to 1) e identical, mid the circuit size can be reduced.

In connection with Embodiment 1 and Embodiment 2, description is made with respect to the cases where the replay speed is 2-time, 4-time, 8- time, 16-time, -2-tinte, 6-time, and -14-time speed. In the digital VTR having a recording format shown in Fig. 7A and Fig. 7B, satisfactory special replay can be achieved at any arbitrary speed of from -14-time to 14-time speed, and the effects similar to those described above (including the reduction of the circuit size) can also be. achieved.

In Embodiment 1 and Embodiment 2, description is made of the digital VTR having the recording format shown in Fig. 9. However, the invention is riot limited to this. Simi lar effects, are obtained with any other recording format as long as it can be used for recording a special replay signal with

59 new error correction code appended to it. The error correction code configuration is riot limited to that shown in Fig. 3. Embodiment 3 Fig. 16 is a block diagram showing an example of a recording system of a digital VTR of Embodiment 3 of the invention. In Fig. 16, reference numeral 50 denotes an input terminal for receiving digital video and audio signals in the form of a bit stream, 52 denotes a packet detector for detecting packets of video and audio signals from the bit stream, 54 denotes a first memory for storing the bit stream, 130 denotes a third error correction encoder for forming video areas arid conducting error correction encoding, 56 denotes an intra detector for detecting intra encoded data from the bit stream, 58 denotes a second memory for storing the intra encoded data. 132 denotes a fourth error correcting encoder for forming audio areas arid conducting error correction encoding, 134 denotes a digital modulator for conversion into data suitable for recording on the magnetic tape, 66 denotes a recording amplifier, 68 denotes a rotary drum, arid 70a and 70b denote magnetic heads.

Fig. 17 shows the recording format on the tracks in Embodiment 3. Fig. 17(A) shows the configuration of the entire track, Fig. 17(B) is an enlarged view of the audio area. Fig. 17(C) shows the configuration of a sync block (SB.m.0) in the data area, and Fig. 17(D) shows the configuration of another sync block (SB 41,13).

Fig. 18 shows the track configuration in Embodiment 3, and shows the data format of the audio area 136 and the video area 138.

The operation during recording will next be described with reference to Fig. 16 to Fig. 18, as well as Fig. 2A arid Fig. 2B.

Referring in particular to Fig. 16. the bit strearil received at the input terminal 50 contains digital video and audio signals, and digital data concerning the video and alidio signals. It is transmitted, being partitioned into transport packets as shown in Fig. 2A. The packet is formed of a header 532 of four bytes and data section 94 of 184 bytes.

In Embudiment 3, the bit stream is detected transport packet bY transport packet an(] the packets of' intra encoded data are recorded in the audio areas. Transport packets are therefore detected at the packet detector 52 from the bit stream received at the input terminal 50, and input to the fj,r.c,t memory.54 and the intra detector 5G.

The data of the bit stream is stored in the first memory 54, packet by packet. and is read so as to form the data of' the recording data blocks shown in Fig. 2B. Fig. 2B shows the example in which five sync blocks form two transport 1)ackets. where the data length of the one sync block is 77 bytes, as described above. In the draiving, 111 denotes a first header, 112 denotes a second header. Data recorded in the first header 111 include identification data indicating the sync block number of each sync block within the five sync. blocks (which of the five sync blocks each syne block is), and data recorded in the second header H2 include identification data for indicating whether the data is video data or audio data.

The data of the transport packet read from the first memory.54 is supplied to the third error correction encoder 130, where first and second headers 111 and H2 are appended to form a configuration as shown in Fig. 2B. and then error correction encoding for the video area 138 is effected, and the data is then. supplied to the digital modulator 134.

The bit stream output from the packet detector 52 is also supplied to the intra detector 56, where judgement is 61 made whether the data in the transport packet is intraencoded data or riot. As described in connection with the prior art, in the MPEG2 bit stream, if the data is intraframe or intra-field encoded (intra-encoded), intra transport packets are consecutively transmitted. By detecting such transport packets coiisecuti,.e.'ly transmitted, the intra transport packets are extracted, and the extracted transport packets are written in the second memory 58.

When the intra-encoded transport packet is read from the second memory.58 in the form shown in Fig. 2B, arid input to the fourth error correction encoder 132, where headers 111 arid H2 are appended, arid error correction encoding for the audio area 136 is effected, and the data is then supplied to the digital modulator 134.

The data configuration in the audio area 1.36 is next described.

Referring to Fig. 17(A) to Fig. 17(D), one track consists at least of a video area 138 arid an audio area 136. The audio area 136 is formed of data tO to t13 of 14 sync blocks (SBs), arid each sync block is 90 bytes long (Fig. 17(B)).

As shown in Fig. 17(C), one sync block is formed of a header section 140 of 5 bytes, data (C2 check code) section 142 of 77 bytes, and Cl check code section of 8 bytes. The header section 140 is formed of a sync pattern of 2 bytes, and identification (ID) code of 3 bytes. As shown in Fig. 18. nine sync blocks are allotted to the data region, and five sync blocks are allotted to the C2 cheek code region, and the data section of 77 bytes is divided into an auxiliary data (AAUX data) and the audio data.

* The recording data packets formed as shown in Fig. 2B are disposed as the AAUX data and audio data in Fig. 18, i.e., data section 142 (Fig. 17(D)). Each recording data packet is formed of five sync blocks. The data section 142 62 in the audio area 136 is formed of nine sync blocks, so that one recording data packet is recorded over a plurality of tracks.

As in the prior art example, in the digital VTR recording one frame of video over ten tracks, the rate of data recorded in the audio area is about 1.8 Mps. and if the ATV signal rate is about 18 Mbps, the number of bits per intra-frame is predicted to be about 1.5 Mbps, so that about one picture can be recorded per second.

The output of the fourth error correction encoder 132 and the output of the third error correction encoder 130 are input to the digital modulator 134, where digital modulation such as interleaved NRZI in the data format of Fig. 17(A) to Fig. 17(D) and Fig. 18 is conducted. The modulated data is passed via the recording amplifier 66, and recorded or) oblique tracks, shown in Fig. 93, formed on the magnetic tape by means of the rotary heads 70a and 70b.

Fig. 19 is a block diagram showing a replay system of the digital VTR of Embodiment 3. In the drawing, the rotary drum 68, the rotary heads 70a and 70b are identical to those in Fig. 1. Reference numeral 72 denotes a head amplifier, 74 denotes a signal detector for detection digital data from the replay signal. 76 denotes a digital demodulator for performing digital demodulation on the replay digital data. 146 denotes a third error correction decoder for correcting errors in the replay signal, 148 denotes a fourth error correction decoder for correcting errors in the replay signal. 82 denotes a third memory, 86 denotes a fourth memory, 88 denotes a selector and 90 denotes a data output terminal.

The operation of the replay system will next be described. Still replay is started either by selection of the still mode during normal replay, or by selection in the state of halting. First, the situation where the still mode 63 is selected during normal replay is described.

During normal replay, the data replayed by tlie rotary heads 70a and 70b from the magnetic tape is amplified by the replay amplifier 72, and supplied to the signal detector 74 where signal detection is performed to produce the original digital data. At the digital demodulator 76, interleaved NRZI demodulation is effected, and the replay data from the,,, -ideo areas 138 in Fig. 17(A) is supplied to the third error correction decoder 146 and the replay data from the audio areas is supplied to the fourth error correction decoder 148. The third error correction decoder 146 and the fourth error correction decoder 148 respectively correct errors during replay, and the error corrected data from the third error correction decoder 146 is written in the third memory 82 and the error correeted data rrom the fourtli error correction decoder 148 is written in the fourth memory 8G. The data selector 88 selectively outputs eitlier the output of the third memory 82 or the output of the fourtli memory 86, to the output terminal 90. During normal replay, the data selector 88 selects the output of the third memory 82, and the data identical to the bit stream input via the input terminal 50 is output via the output terminal 90.

When still mode is selected during normal replay, the replay data is stopped, and data is no longer input to the third and fourth memories 82 and 86. The input of the data selector 88 is then switched to select the output of the fourth memory 86. In this way, the still picture (.an be output via the output terminal 90. The data written in the third and fourth memories 82 and 86 include the data shown in Fig. 2B except the headers 111 an(] 112. i.e., the data of the transport packets shown in Fig. 2A. Only the intraencoded data in the audio area 136 is %Yritten in the fourth memory 86, so that it is sufficient to sequentially write the data transport packet by transport packet. 64 1 - The situation where the still mode is selected froill the state of

halting will next be described. In the halting state, no correct data is stored in the third and fourth memories 82 and 86. and if the still mode is selected from this state. norinal replay is conducted once, arid one frame of data is stored in the fourth memory 86, and then the tape is stopped.

Next, the operation of the slow replay is described. During slow replay,, the magnetic tape transport speed is lower than in the normal replay, so that the same track is repeatedly crossed and data is replayed from the same track for a certain riumber of times. By extracting the sync blocks which are correctly demodulated by the digital derijodulator 76, and inputting them into the fourth error correction decoder 148, a still picture can be obtained. 1 particular. at the tape speed of one-half the normal or less, all the data recorded in the audio area 136 can be replayed. Embodiment 4 Description is next made of another entbodiment with which deterioration in the picture quality is small even during special replay, such as fast replay. Fig. 20 is a block diagram showing a recording system of Embodiment 4. In Embodiment 4, the special replay data is recorded, being divid.ed into thevideo areas arid audio areas.

In the drawings, reference numeral 150 denotes a fifth memory for receiving the bit stream via the input terminal.50, and special replay data, 152 denotes a special replay data generator receiving the intraencoded transport packets and generating special replay data, and 154 denotes a sixth memory for receiving the special replay data.

The special replay data generator 152 extracts the lowfrequency component from the packets of the intra-encoded data that have been detected, arid supplies low-frequency comporient to the ri uh memory 150, and the s"hsvquvnL highrrequnnm component to the sixth memory 154. In the prior art example. the same data is recorded 17 Limes in the copy areas or about 5.8 mbps. so that the data rate or the speciaL replay data is 340 kbps. In this cmhodiment, the special replny data is also recorded in the nudto aren of nbout 1.8 Mhps, resulting in the copy arens or 7.0 Mhps. I r the same dAn is recorded 17 times, the data rate or the spec[al replay data AU he about 450 khps.

The special replay data generator 152 therefore encodes so that 1 Ls output is about 450 khps, and the data ror 340 khps is supp] led to the ri rLh memory 150 and the data for Lhe rrma 1 n 1 ng I 10 khps 1 s supp] led to the s 1 x Lh memory 154 To enable replay or the special replay data at a Ugher specd, A is necessary to record the data macro block hy macro block.

Fig. 21 shows the digital video data or the macro block configuration in Embodiment 4. Each block is formed or 8 pixels by 8 pixels 1" horizontal and verLicni directions oil the screen. i.e., G4 pixels, and rour blocks or a luminnnee signal (YO. Yl. Y2. Y3). nnd two blocks or a chrommance signal (Ch. Cr) (the pixel density of the chrominance signal. being 1/2 in each or the horizontal and vertical directions, compared with the pixel density or the luminance sig"W), i.e., six blocks in all farm a video daLa of one macro block.

Fig. 22 shows coefficient of the frequency components in Embodiment 4. The pixel data of each block shown in Fig. 21 is subjected to orthogonal transform such as DCT, and deCOMpOsOd into tAlf3 f'requency component's as shown ill Fig. 22. The respective r7requency components are sequentially scanned. starti.ng with a DC component, and in a so-called z1g-zag scanning, to perform variable-length encoding. 03, controlling the variable-length encod0l; so that the data 66 rate of the special replay data is about 4.50 khps, the special replay data can be generate(].

It is necessary that the special replay dal.a is encoded macro block by macro block and partitioned irito sync blocks. This is because in a fast replay in which tracks are crossed for the scanning for replay, data is replayed sync block by sylic block.

Fig. 23 shows the disposition of the special replay data rectording areas in the tracks in Embodiment 4. During replay, a process reverse to that for recording is followed to form special replay data. Fig. 23 shows the positions at which the special. replay data is recorded in a predetermined 'I is recorded in track pattern. Since special replay data the audio areas, and special replay data #42 and t3 is recorded in the video arens, by replaying data f't-otii the audio area, special replay data of a higher data rate call be obtained. Even if the special replay data ?''2 an(] 4-43 only are reproduced, special replay data having the same quality as in the prjor art can be obtained. This mearis even if the VTR cannot pick up data from the audio areas, special replay data can be replayed.

In Embodiment 3, description is made of the case where the data is intraemeoded frarne by frarne, or field by field. The data may alternatively be encoded macro block by macro block. In this case, the recording packets shown in Fig. 2B can be reconstructed for each unit of intra- encoding. and the data may be recorded transport packet by transport packet.

In Embodiment 4, the special replay data is recorded in both of thevideo areas arid audio areas. The intra-encoded data may be recorded as is in both the areas. In this case. it is possible to record a great many still pictures for the still and slow replay. For instance, five pictures per second can be recorded with the special replay data rate of 67 about 7.6 M1)ps in Etiil)o(li.iiie.iit 4. Embodiment 5 Fig. 24A is a block diagram showing a recording system of a digital VTR of Embodiment 5. In the drawing, reference numeral 160 denotes a bit stream input terminal. 162 denotes an output terminal for a bit stream for main areas. 164 denotes a low-speed fast replay replay data output terminal, 166 denotes a middle-speed fast replay data output terminal, and 168 denotes a high-speed fast replay data output terminal. Reference numeral 170 denotes a TP header analyzer for analyzing transport headers and outputting transport packets containing a transport header and intra data, 172 denotes a TP)leader modifying circuit for modifying the transport headers having been separated, and 174 denotes a depacketing circuit for converting transport packets into a bit stream, 176 denotes a header analyzer for analyzin g headers such as sequence headers and picture headers contained in the bit stream and outputting the headers and intra data, and 178 denotes a special replay data generator for generating special replay data for the respective replay speeds from the intra bit strearti arid outputting it.

Reference numeral 180 denotes a header appending circuit for appending, to the low-speed fast replay data, those of the headers extracted at the header analyzer 176 which are necessary, 182 denotes a packeting circuit for packeting the data into tile size of a transport packet, 184 denotes a modified TP header appending circuit for appending the modified transport headers. and 186 denotes a low-speed fast replay data generator formed of the TP header modifying circuit 172. the header appending circuit 180, the packeting circuit 182 and the modified TP header appending circuit 184. Reference numeral 188 denotes a middle-speed fast replay data generator 188. Reference numeral 190 denotes a 68 1 A 11 - high-speed fast replay data generator. The middie-speed fast replay data gerierator 188 and tile high-speed fast replay data generator 1.90 have a corifiguratiori similar to that of the low-speed fast replay data generator 186.

The operation will next be described. The bit stream received at the input terminal 160 is output via the output terminal 162 for the bit stream for the main areas, as the data for tile main areas. It is also supplied to the TP header analyzer 170, where headers of the transport packets are detected from the input bit strearri, and the headers are analyzed. and if data is contained in the succeeding bit stream, the trajisporl". packet is supplied to I.he depacketing circuit 174, and the transport header is supplied to the TP header modifyirig circuit 172.

The depacketing circuit 174 depackets the input transport packet, and supplies the resultant bit stream to the header analyzer 176, where headers such as sequence headers and picture headers in the bit stream are analyzed, and only the intra data is supplied to the special replay data generator 178 arid tile headers are output to the header appeiidiiig circuit 180.

The special replay data generator 178 generates special replay data for low-speed fast replay, special replay data for middle-speed fast replay and special replay data for high-speed fast replay, from the input intra data. The subsequent data is identical for the respective replay speeds, so that description is made only in corinection with the low-speed fast replay data. The low-speed fast replay data output from the special replay data generator 178 is input to the low-speed fast replay data generator 186. Th e low-speed fast replay data is input to the header appending circuit 180, where those of the input headers tfict are necessary are appended. The output of the header appending circuit 180 is supplied to the packeting circuit 182, where

69 the low-speed fast replay data with the necessary Ileaders having been appended is packeted, dividing the data into tile size of the transport packet. The packeted low-speed fast replay data is supplied to the modified TP header appending circuit 184, where modified transport headers are appended, and then output. The modified transport headers are formed by modifying the transport headers separated at the TP header analyzer 170, into a suitable form. In this way,, the low-speed fast replay data is converted into the form of transport packets, and is then output via the low-speed fast replay data output terminal 164.

The description has been made of the formation of transport packets from the low-speed fast replay data. Similar processings are applied to the middle-speed fast repla-v data and the high-speed fast replay data. Th e middle-speed fast replay data and the high-speed fast replay, data output from the special replay data generator 178 are respectively input to the middle-speed fast replay data generator 188 and the high-speed fast replay dat a generator 190, and headers and modified headers are appended, and output in the form of transport packets via the middle-speed fast replay, data output terminal 166 and the high-speed fast replay data output terminal 168.

Further description of the special replay data generator 178 will next be given.

Fig. 24B is a block diagram showing an example of the special replay data generator 178. In the drawing. reference numeral 192 denotes an input termir)al for receiving a bit stream of intra data, 194 denotes a variable-length decoder for forming low-speed special replay data, 196 denotes a variable- length decoder for forming middle-speed special replay data, and 198 denotes denotes a variable-length decoder for forming high-speed special replay data. Reference numerals 200, 202 and 204 denote c oun t e r s - Reference numerals 206, 208 and 210 denote data extractors for low-speed fast replay data, middle-speed fast replay data, and high-speed fast replay data, respectively.

Reference numeral 21-2 denotes an EOB appending circuit for appending EOB (end of block) code to the low-speed fast replay data, 214 denotes an COB appending circuit for appending EOB code to the middle-speed fast replay data, in(] 216 denotes an EOB appending circuit for appending EOB code to the high-speed fast replay data. Reference nutrieral 218 denotes an output terminal for low-speed fast replay. data, 220 denotes an output terminal for middle-speed fast. replay data, and 222 denotes an output terminal for high-speed fast replay data.

The operation of the special replay data generator 1.78 (F.ig. 2413) next The decoder 194 variable-length decodes the input bit stream. On the basis of the decoding, the counter 200 counts the number of the decoded DCT coefficients, and outputs the result to Clata extractor 206 extracts the bits stream correspond!rig to the predetermined number of DCT coefficients, from the input bit stream, at a predetermined timing, on the basis of the input from the counter 200. The counter 202 arid the data extractor 208, zinc] the counter 204 arid the data extractor 210 perform similar operation. The data extractor 206 extracts the low-speed fast replay data from the input bit stream, the data extractor 208 extracts the middle-speed fast. replay data from the input bit stream, and the data extractor 210 extracts the high-speed fast replay data frori) the input bit stream. The extracted low-speed fast replay data is supplied to the EOB appending circuit 212 where EOB codes are appended, arid then output as the low-speed fast replay data via the output terminal 218. The extracted middle-speed fast replay data is supplied to the EOB appending circuit 214 where EOB codes are appended, and then 71 output as the middle-speed f'ast replay data via tlip outplit termimil 220. The extracted high-speed fast replay data is supplied to the EOD appending circuit 216 where EOB (.odes are appended, and then output as tlie high-speed fast replay data via the output terminal 222.

The timings at which the data is extracted at the respective data extractors may be identical to each other, or may be different. If they are different, the number of DCT coefficients within one video block to be recorded (the imit with which the orthogonal. transform is performed at the encoding means) differs. Since the special replay area where special replay data is recorded is limited as will be described later, if the special replay area is of t)ie same areas (size), increase in the number of the DCT coefficients of' one vj.(](o block requires more special. replay areas ror recording. and the refreshing of the screen during replay is S l osv However, the picture quality is good. Decision oil the timing for extraction is therefore made by trade-off between the delay in refreshing and the picture quality.

Fig. 2.5 is a block diagram showing a syric block forming circuit.

In Fig. 25, reference numeral 224 denotes an input terminal for a bit stream for main areas, 226 denotes an input terminal for low-speed fast special replay data, 228 denotes ail input terminal for middle-speed fast special replay data, 2.30 denotes ail input terminal for high-speed fast special replay data. The input terminals 224. 226, 228 and 2:30 are respectively connected to the output terminals 162. 164, 166 and 168 in Fig. 24A. Reference numeral 232 denotes ail SB format circuit for converting the input data and the bit stream into a sync block format. Reference numeral 234 denotes ail SB output terminal for outputting SB data.

The synthesis of the bit stream for the main areas and 72 the special replay data for the respective fast-. iiext be described with reference to Fig. 2.5. The data and the bit - am received at the iill)tit termimils 224 to 2.30 are input stre, to the SB format circuit 2.32, where data to be recorded ill the respective sync block are selected for each track and for each sync block. A header is apperided to Pach sync block of data, atid the syric blocks within a track are formed to thereby form the predetermined pattern as described later. and the resultant data is output via the SB output termimil 234.

The operation of the SB format circuit 232 will next be described. Ill this embodiment, the drum may be of ally of 1 ch x 2, 29 ch x 1 and 2 ch x 2 configurations. llc)s%,e,.,er, two azimuth angles are provided, and the head 13aviiig oiie azirliuth angle Is called ill A-chatmel head, aild Llie 1jead tlic other azimuth angle is called B-chaimel head.

Fig. 26A to Fig. 26F are diagrams showiiig the configurations of the special replay data recording areas a c c.

ording to Embodiment 5. Ill the drawing. refer'enee immeral 242 cleiic)tes A-chanriel low-speed fast replay recoi.cliiig area for i-ec.ot- (liilg low-speed fast replay data by nie,-iiis of;in A-charmel head, 244 denotes a B-chanriel lowspeed fast replay recording area for recordirig loss-speed fast replay data by mpans of a B-charmel head, 246 denotes A- ellaTMel middle-speed fast replay recording area for recording low-speed fast replay data by means of an Achailnel. head, 248 denotes a B-channel low-speed fast replay recording area for recording low-speed fast replay data by mearls of a B-chamiel Ilead, 250 denotes A-chaimel Iiigi)-speeci fast replay recording area for recording low-speed fast replay data by means of all A-chamiel head, and 252 denotes a B-chaimel high-speed fast replay recording area for recording low-speed fast replay data by means of' a B-channel head. The B-channel data is that obtained mIlen 2 ell x 2 73 drurl, ('onfiguratiori is used Compared with other drum collf'i gurat ions, in the case of the 2 ch x 2 drum configuration (assumifig that the replay speed is identical) tile number of times the head crosses the track is larger, arid tile number of sync blocks reproduced per track is small.

As a result, it is necessary to supplement the data of special replay data recording areas from which the data i riot produced by the A-charinel head.

The special replay data recording areas for the Bchannel head are provided for the above-described reason. The B-chatmel low-speed fast replay data recording areas 244 supplement the A-charinel low-speed fast replay data recording areas 242, tile B-channel middle-speed fast replay data recordiTlg areas 248 supplement the A-CIMT)Tlel low-speed fast replay data recordin.,; areas 246, aiid the 13-c-hannel high-speed fast replay data recordiTIg areas 2.50 supplement the A-charmel high-speed fast replay data recordiTig areas 252. With regard to the size of the respective areas, since the same size of the areas for the A- and B- channels can be used in the 2 ch x 2 drum configuration, and data can be replayed from about double the areas by means of the Achannel head in other drum configurations, the ratio of the A-chaTmel special replay area to the B-channel special replay area is 2:1.

The numbers 1 to 14 allotted to the respective blocks in Fig. 26A to Fig. 26F indicate the content of the data. That is identical number deTlOtes identical data. Tile data at the upper arid lower ends of tile A-channel special replay areas and also form the data of the B-clialMel special replay areas. The reason is as explained above. Each block is formed of rri syr)c blocks (m being a natural number).

Fig. 27 shows the disposition of the special replay data recording areas in the tracks. In this recording format, is in Embodiment 1, the special replay data 74 recording areas are repeated at a period of fotir tracks. The special rel)lay data recording areas corresponding to the respective replay speeds are provided in fotir tracks 98. 100, 102 and 104 of one period. In the drawing, the track 98 is a first track recorded by an A-channel Ilead, the track 100 is a second track recorded by a Bchannel head, the track 102 is a third track recorded by the A-channel head, the track 104 is a foxirtil track recorded by the 13-charinel head. The first to fourth tracks 98 to 104 form a tinit.

fO, fl and f2 represent pilot signals for identifying the the respective tracks. The pilot signal fl is a signal of a frequency. denoted by fl. superimposed with tile digital signal recorded on the track. The pilot signal f2 is a signal of' another 1'recltietic,, denoted by f2, different from fl, and superimposed with the digital signal recorded 01) the track. The pilot signal fO is aettially in the form of absence of stiperiniposed signals fl and f2. The areas other than the areas 242 to 2552 are used as main areas for recording data for norinal replay. Data from the areas for special replay can be reprodticed by one scan of a head whatever is the configtiration of the drum. In the case of the 2 ch x 1 or 1 ch x 2 drilin configin-ation, the special replay data in concentrated areas on one track can he rel)rc)(lij(...e(l by one scan of a head. In the case of' the 2 ch x 2 druin c.orifigijratioii, the special replay data cari be formed frorn adjacent tracks. By recording the special replay data collectively, in concentrated areas as shown in Fig. 27. it is possible 1. o remove. the effeets of non-linearity of the tracks.

The pilot signals can be superimposed on the digital data at a modulator 502, shown in Fig. 28, provided in front of a recording amplifier 503, frow which recording signals are supplied to a recording head 504 for recording the signals on the magnetic tape 505. The superilTIP0 S it l 011 ca n 1) e ach.leved 1)y (11,,.i(l.iilg the co(le sequence into imi ts of' 24 bits, and adding one bit to the head of each unit of 24 bits, and selectively setting the additional bit to "0" or "l" to therehy vary the DSV (digital sum variation).

It should be noted, the system shown in Fig. 1, Fig. 16 and Fig. 20 also is provided with a modulator ill front of the recording amplifiers 66, but such a modulator is not shown for simplicity of illustration.

Fig. 29 is a djagram showing a recording format oil tracks, ill Friffludiment 5. The unit of four tracks shown in Fig. 27 is repeated. and recording is made oil the repeated units, to form the recording pattern. Ill the recording pattern shown in Fig. 29, four-time speed is set as the low speed fasi. replay specd, ci.glil,-1.Itiie speed is Sel, as the fast replay speed, and speed js set. as the high-speed fast replay speed. The data for foxir-time speed is repeatedly recorded over two units of four tracks, the data for the eight-time speed is repeatedly recorded over four units of four tracks, all(] the data for the 16-tinle speed replay is repeatedly recorded over eight tinits of four tracks. To generalize, the data for the (bl x i)-time speed replay is repeatedly recorded over 2 x i units of four tracks, where M is four in the illustrate(] example, and i = 1. 2.. - - n.

By forming the recording pattern as described above, the effects of any non-linearity of tracks can be minimized. Moreover, because dedicated areas are provided for the respecl-live fast replay speeds, the refreshing ancl the picture quality call be set for the respective fast replay speeds. Embodiment 6 Embodiment 6 relates to a different configuration of a special replay data generator 178. The special replay data generator 178 ill Embodiment 5 was in the form shown in Fig.

76 1 '---- 24B. The invention is not 1111lited to such a configuration, but the configuration shown in Fig. 30 may be used.

Referring to Fig. 30, the differences from the special replay data generator 178 shown in Fig. 24B will be described.

Reference numeral 260 denotes a variable length decoder for variablelength decoding the input bit stream, 262 denotes a counter, 264 denotes a data extractor for extracting low-speed data, 266 denotes a data extractor for extracting middlespeed data, and 268 denotes a data extractor for extracting high-speed data. Ref erence numerals 192, 212, 214, 216, 218, 220 and 222 in Fig. 30 denote Tnembers identical to those in Fig. 24B.

The operation of the special replay data generator in Embodiment 6 next be deseribed. The intra data received at the input terminal 3-92 is input. to the Irariablelength decoder 260, and the data extractors 264, 266 and 268. The vai-iable- length decoder 260 performs variablelength decoding on the bit stream. The counter 262 counts the number of DCT coefficients obtained as a result of the decoding, and the count value is supplied to the data extractors 264. 266 and 268. The data extractor 262 extracts the data at a timing predetermined according to the input. Similarly, the data extractor 266 and the data extractor 268 respectively extract data at timings predetermined independently of' each other. The extracted data is supplied to the EOB appending eircuits 21.2, 214 and 216, where EOB code is appended, and then output via the output terminals 218, 220 and 222. By forming the circuit as described above, the special replay data similar to that of Fig. 24B can be generated. Embodiment 7 Fig. 31 is a block diagram shoisring an example of sync block forming eircuit of' Embodiment 7. In Embodiment 7, a 77 configuration different from the sync block forming circuit of Embodiment. 5 (Fig. 25) is used to synthesize the bit stream for the ntain areas and the special replay data for the respective fast. replay speeds.

ln Fig. 31, reference numeral 224 denotes ail input terminal for receiving the main area bit stream, 226 denotes ail input terminal for receiving special replay data for lowspeed. 228 denotes ail input terminal for receiving special replay data for middle-speed, and 230 denotes ail input terminal for receiving special replay data for high- speed. Reference numeral 232 denotes ail SB format circuit for converting 1-.lie input data and bit stream into the format of sync blocks, 270 denotes ail error correction encoder, and 234 denotes ail output terminal for outputting the SB data.

Referring to Fig. 31, the operation for synthesizing the main stream bit stream and the special replay data for the respective fast replay speeds will next be described. The data and bit stream input via the input terminals 224 to 230 are input to the SB format circuit 232, where the data to be recorded in the respective sync blocks is selected, frorit the respective inputs, for each of the tracks and for each of the sync blocks. A header is appended to each of the sync block of data, and the sync blocks ill each track are configured so as to form the predetermined pattern, to be described later. and a second parity (Cl code) formed of digital data, and a third parity (C2 code) formed of a plurality, of items of digital data extending across a sync bit, are appended, and the result is output via the SB output terminal 234.

Tile configuration of the special replay data recording areas, and the disposition of the special replay data recording areas, and the recording format on the tracks may be identical to those described with reference to Fig. 26A to Fig. 29 in connection with Embodiment 5.

78 Description ivill next be made as to in isliall fortmat, the transport packets are recorded in fixed areas. sueli as sync blocks.

Fig. 32 is a diagram sl)owiiig an example of data packet according to Embodiment 7. It shows all example for the case where two transport packets are recorded over five sync blocks. In the drawings, reference numeral 300 denotes a sync of a sync block 0 (SBO), 301 denotes a sync of a sync block 1 (SBI), 302 denotes a sync of a sync block 2 (SB2), 303 denotes a sync of a sync block 3 (SB3), and 304 denotes a sync of a sync block 4 (SB4). Reference numeral 305 denotes ID of SBO, 306 denotes ID of SB1, 307 denotes ID of SB2, 308 denotes ID of S133, and 309 denotes TD of SB4. Reference numeral 310 denotes a)leader appended to SBO, 311 denotes a licader appended to SBI, 31.2 denotes a lleader appended to SB2..31,3 denotes a header;ipl)eiide.(] to SB3, and 314 denotes a 1)ei(ler apperided to SB4. Reference nximeral 315 denotes a transport header of the transport packet A, 316 denotes data of the transport packet A. 317 denotes a transport header B of the transport packet B, and 318 denotes data of the transport packet B. Ref erence numeral 319 denotes a dummy area.

Reference numeral 320 denotes a syne parity generated from the digital data succeeding ID 305. Reference numeral 321 denotes a syne parity generated from the digital data succeeding ID 306. Reference numeral 322 denotes a sync parity generated from the digital data succeeding ID 307. Reference numeral 323 denotes a sync.. parity generated from the digital data succeeding ID 308. Reference numeral 324 denotes a sync parity generated from tile digital data succeeding ID 309. Reference numeral 330 denotes a Cl check code which is a second parity appended at the error correction encoder 270. Reference numeral 3.31 denotes a C2 cheek code which is a third parity appended at the error 79 correction eyleoder 270.

Description is made of SBO. ID 305 and header 310 contain an address foridentifying the particular sylic block within the five sync blocks, a sigiial indicating whether normal replay data or sl)ecial replay data is recorded, a signal for identifying the speed where the special replay data is recorded, a signal for indicating the iclei)tity of data for several units needed since identical sl)ecial repla, data is recorded for several units am] discriminating frofil the special replay data recorded iri the succeeding several units, and a signal for identifying the assembly of the five sync blocks, for each unit of the five blocks, arid a signal indicating whether the central part of the screen (pieture) of an i.iit.rl-friiiie or intra-field.

1n Embodiment 7, address identifying each syric. block within the group of five sync blocks arid a signal indicating wliether iiormal replay data or special replay (Inta is contained are recorded in ID 305, arid the rernaiii.der is recorded iTi the Ileader 310 disposed after the ID, for each sylic block. The ID 30.5 records the necessary signals among the signals stipulated by the SI) specification.

That is. ID 30.5 contains a parity of tile ID signal which is a first parity. This parity is for c.lieckiiig whether the II) signal containing the parity is correct, an(] its size is mie hyte. Tile Cl check code 330 whIch is the second parity is of eight bytes, arid the C2 cheek code which is the third parity is of 11 bytes. The fourth 1)arity is the syric parity 320 which is of one byte.

SB1. SB2, SB3 arid SB4 record an ID arid a header, like SBO. In this embodiment, the size of the syiic block is 82 bytes (excluding the Cl area), the size of each sync is 2 bytes, the size of each ID is 3 bytes, arid the size of each header is one byte. The size of each sync parity is one hyte. The size of the tratisport packet is 187 bytes (as the signal of one hyte which (.;in be appended at.. t.lie time of replay is reriio,,,eci from the transport licader at the time of recording). Accordingly, two transport packets (187 x 2 374 bytes) can be recorded in the data regions of five sync blocks (76 x 5 = 300 bytes). The remaining one byte forms the dummy area 319 in Fig. 32. In this way, two transport packets can be recorded in five sync blocks. 133, recording, at the tail of the sync block, syne parities generated front the digital data contained in the sync block, it is possible to provide a format permitting detection of wliether the digital data contained in the sync block is cr-roneous. Embo(lilTient 8 Fig. 33 is diagram showing a recording format on tracks of a digital VTR according to Embodillient 8. In the drawings, four tracks form one unit, and a pattern formed of four tracks is repeated.

Tliat is, of the 135 sync blocks of sync block Nos. 21 to 155 of the respectiie tracks. the data for +8-time speed replay and -6-time speed replay is recorded in the area bO formed of' the sync blocks Nos. 96 to 115 in the first track of the group of four tracks, and the area bl formed of the sync blocks Nos. 96 to 106 in the second track of the group of four tracks. The data for +2-tirrie speed replay, +4-time speed replay and 2-time speed replay is recorded in the area aO formed of the sync blocks Nos. 104 to 143 in the third track of the group of four tracks, and the area al formed of the sync blocks Nos. 109 to 128 in the fourth track of the group of four tracks. The data for +16-time speed replay and -14-time speed replay is recorded in the area cO formed of the sync blocks Nos. 72 to 81 in the third track of the group of four tracks, and the area cl formed of the sync blocks Nos. 70 to 79 in the fourth track of the group of four tracks.

The data recorded in the areas al, bl and cl are 81 1.

iclelll-.j cal to the data recorded in the botli end parts of the areas aO, bo arid cO, resl)ectivel,, arid is used to supplement when the data at the end parts of the;ireas;10, bO and co is not obtained. Witl) regard to the data for +2-ti.itie speed replay, +4-time speed replay, arid -2- time speed replay, identical data is recorded in two tracks. Witli regard to the data for +8-time speed replay and -6-time speed replay, identical data is recorded in four tracks. With regard to the data for +16-time speed replay and -14-time speed replayr, identical data is recorded in eight tracks. In the remaining video areas, normal replay data is recorded. arid the s?iic block number is recorded in each sy,iic block. As in the SD mode, pilot signals for tracking control are recorded in the respective tracks, in the order of fO, fl, fO and f2. in superimposition with the digital signal. Accordingly, the pilot signal fo is recorded in the first arid third tracks, the pilot signal fl is recorded in the second track, and the pilot signal f2 is recorded in the fourth track.

The configuration of the head used for tl)e recording or replay may for example be as shown in Fig. SA to Fig. 5C, in which one head each is disposed at opposite positions 180 apart on the drum, two heads are disposed at positions close to each other on the (]rum, or two heads each are disposed at positions opposite positions 180' apart on the drum. In the following description, the 2ch x 1 configuration in which two heads are disposed at positions close to each otlier on the drum will be taken as an example. The head having the same azimuth as the first arid third tracks in which the pilot signal fO is recorded is called a first head, while the head having the same azimuth as the second arid fourth tracks in which the pilot signals fl- arid f2 are recorded is called a second head.

During fast replay, the specific scanning trace is followed depending on the the replay speed to reproduce the 82 1 --- - desired replay. data. The metliod of tracking will be described.

Fig. 34 is a schematic. block diagram slios,itig the configuration of the capstan servo system. In the drawing, reference numeral 340 denotes a capstan motor, 342 denotes a FG (fi-ecitieii(.y generalor) section for generating FG signal of a frequency corresponding to the rotary speed of the capstan motor 340,.344 demotes a speed detector for detectIng the speed error of the capstan motor 340, by detecting the period of the FG signal, 346 denotes a trackirig error detector for detectirig the tracking error, 348 denotes an adder for addirig the outputs of the speed deleetor.344 and the tracking error detector 346. atid 3.50 deriotes a driver for driving 11)e capstan motor in accordance with the output or the adder- 348.

Fig..3.5 is a diagram showing;in example of Configuration of the tracking error detector.346 in Fig. 34. In the drawing. reference mimeral.3.52 denotes a first head, 354 denotes a head amplifler. 356 aricl 3.58 re. spect lvely denote BPFs (bandpass filters) having central frequencies fl and f2, respectively..360 and.362 denote detectors, 364 and 3G6 denote sample-hold eircuits, and.368 deriot.es a sampling pulse generator for generating sampling pulses for the sample-liold circuits 364 arid 3G6. Reference numerals 370 and 372 denotes selectors for selecting the outputs of the sample-hold circuits 364 arid 366. Reference numeral 374 denotes a controller for coritrolliii,,, the selectors.370 at)d.372. Reference numeral 376 denotes a subtractor for performing subtraction on the outputs of the selectors 370 and 372.

The replay operation of the digital VTR of Embodiment 8 will next be described with reference to Fig. 36 to Fig. 42.

Fig. 36 shows the head scanning traces. During +2-time speed replay, the target speed for the speed detector 344 is 83 set at tivice the speed during recording, and the tape speed is controlled to be the double speed. By the function of the tracking error detector 346, the tracking control is effected. The signal reproduced by the first head 352 at the +2-time speed is amplified by the head amplifier 354, and the frequency components of the pilot signals fl and f2 are extracted by the BPFs 3.56 and 3.58. The amplitudes of the fl and f2 components are substantially proportional to the artiount over which the first head.352 is on the track. Th e pilot signals extracted by the BPFs 3.56 and 358 are envelope-detected at the detectors 360 and 362, and then sample-held at the sample-hold circuits 364 and 366. The timing for the sample-holding is determined by the samplehold pulse from the sampling pulse generator 368.

In the case of +2-tinie speed replay, the sampling pulse generator 368 is made to generate one pulse per drunt rotation such that the sampling takes place when the first head is at about the sync block No. 124 at the center of the area aO formed of the syric block Nos. 104 to 143 where + 2time speed replay data is recorded. In the case of +2-tifite speed replay, the selector 370 is made to select the output of the sample-hold circuit 364 and the selector 372 is made to select the output of the sample-hold circuit 366. on the basis of the control signal from the controller.374. Accordingly, the output of the sample-hold circuit 364 is inpull to the "+" input terminal of the subtractor 376, while the output of the sample-hold circuit 366 is iriput to the 11 input terminal of the subtractor 376. The output of the subtractor.376 is a tracking error signal corresponding to the {(pilot signal fl component) - (pilot signal f2 component) If' the head is toward the fourth track (of the four tracks), rather than the third track in the lateral direction of the tracks when the first head.352 is at about 84 1 ' - - the s3,nc block No. 124 in the longitudinal direction of the track, the pilot signal f2 component is larger than fl component, and the tracking error signal will be small. This tracking error signal is output from the tracking error detector 346, and added to the output of the speed detector 344 at the adder 348. By the resultant output of the driver 350, the capstan motor 340 is decelerated, to retard the tracking phase. Conversely, if the head is toward the second track (of the four tracks). rather than the third track in the lateral direction of the tracks when the first head 352 is at about the sync block No. 124 in the - longitudinal direction of the track, the pilot signal f2 component is smaller than fl component, and the tracking error signal will be large. This tracking error signal is output from the tracking error- detect.or 346. and added to the output of the speed detector 344 at the adder 348. By the resultant output of the driver 350, the capstan motor 340 is accele rated, to advance the tracking phase.

In this way, the tracking is so controlled that the pilot signal fl an(] f2 components are equal so that the first head scans the center, in the lateral direction of the tracks, of the third track (of the four tracks) when the first head.352 is at about the syne block No. 124 in the longitudinal direction of' the track. In the center of the first track (of the four tracks) the pilot signal fl arid f2 components are equal to each other, but as the front-rear relationship between fl and f'2 is opposite, the polarity of the tracking error signal will be opposite, and the tracking is not stabilized in this position, but is pulled into the center of the third track. That is, if the tracking is shifted toward the fourth track (of the precedijig group of four track), the output of the driver 350 will decelerate the capstan motor 340 to retard the tracking phase, to thereby bring the head to the third track in the preceding group of four tracks. while if the tracking is shifted toward the second track (next to the first group (of the same group of four tracks), the output of the driver 350 will further accelerate the capstan motor 340 to advance the tracking phase, to thereby bring the head to the third track (of the same group of four tracks). When the first head 352 scans the center of the third track, the second heads scans the center of the fourth track. In this way, the +2-time speed replay data in the areas aO and al in every group of four tracks is replayed.

Fig. 37 and Fig. 38 respectively show head scanning traces at the time of +4-time speed replay and +16-time speed replay. respectively. During +4time speed replay and +16-time speed replay, the target speed of the speed detector:344 is set rour-1-Imes atid 16 times the recording speed, respectively, and by the function of the speed control system. the tape speed is controlled to be the +4time speed and +16-time speed. respectively. Thp operation for producing the tracking error signal is similar to that described in connection with the case of +2-time speed replay. That is, in the case of +4-time speed replay, the tracking error detector 346 outputs the tracking error signal corresponding to the ((pilot signal A component) (pilot signal f2 component)l when the first head is at about the sync block No. 124 at the center of the area aO formed of the sync blocks Nos. 104 to 143 where the +4-time speed replay data is recorded. In the case of +16-time speed replay, the tracking error detector 346 outputs the tracking error signal corresponding to the (pilot signal fl component) - (pilot signal f2 component) when the first head is at about the sync block No. 77 at the center of the area cO formed of the sync blocks Nos. 72 to 81 where the +16time speed replay data is recorded.

In accordance with this tracking error signal, the 86 tracking is so control.ted that the pilot signal. fl arid f2 components are equal so that the first head scans the center, in the lateral direction of the tracks, of the third track (of the four tracks) when the first head.352 is at about the eenter, in the longitudinal direction of the track. of the area where the fast replay data for the respective speeds is recorded. In this way. +4-tiriie speed replay data in the area aO and the area al of every eight tracks, or +16-time speed replay data in the area cO arid the area cl. of every 32 tracks is reproduced. In the case of +4-time speed replay, the head scanning trace follow either of two patterns, but as the same data is recorded on two tracks, the sarite data is reproduced which ever of the scanning trace patterns is followed. This also applies to other replay speeds.

Tile +16-time speed replay will next be described. Fig. 39 shows head scanning traces at tile time of +8-tirile speed replay. During +8-time speed replay, the target speed of the speed detector 344 is set at eight times the recording speed, arid by the function of the speed control system, the tape speed is controlled to be the eight-time speed. By the function of the tracking error detector 346, the tracking is controlled. At +8-time speed, the signal picked up by the first head 352 is amplified by the head amplifier 354, arid pilot signal fl arid f2 components are extracted by the BPFs 356 arid 358, respectively, arid envelope-detected by the detectors 360 arid 362, respectively, arid then sample-held by the saritple-hold circuits 364 and 366, respectively. The sampling timing is determined by the sampling pulses from the sampling pulse generator.368. In the case of the +8-time speed replay, the sampling pulse generator 368 is made to generate one pulse per drum rotation such that the sampling takes place when the first head is at about the sync block No. 106 at the 87 center of the area bo formed of the syne block Nos. 96 to 115 where +8time speed replay data is recorded. In the case of +8-time speed replay, the selector 370 is made to select the output of the sample-hold circuit. 36G and the selector.372 is made to select the output of the sample-hold circuit 364, oil the basis of the control signal from the controller 374. Accordingly, the output of the sample-hold circuit 364 is input to the "" input terminal of the subtractor 376, while the output of the samplehold circuit 366 is input to the "+" input terminal of the subtractor 376. The output of the subtractor 376 is a tracking error signal corresponding to the {(pilot signal f2 component) (pi.1011 signal fl.

oil 1111e basis of this trackiiig error signal, the trackinK is so controlled that the pilot sigrial fl and ú2 components are equal, and the first head 3.52 scans the center, in the lateral direction of the tracks, of the first track (of the four tracks) when the first head 3.52 is at about the sync block No. 106 in tile longitudinal direction of the track. In this way, the +8-time speed replay data in the area bO and bl of every 16 tracks is reproduced.

Fig. 40, Fig. 41 and Fig. 42 respectively show head scanning traces at the time of -2-time speed replay, -6-time speed replay. and -14-time speed replay. In the case of reverse fast replay. the tape is transported in the reverse direction at the respective fast replay speeds, and the tracking control in -2-time speed replay, -6-tinte speed replay, and -14- time speed replay, is effected in the same way as in +4-time speed replay (for the -2time speed replay), ill +8-time speed replay (for the -6-time speed replay), and in +16-tinie speed replay (for the -14-tinte speed replay), respectively. However. since, in the reverse fast replay, the tape transport direction is opposite to forward fast replay, it is necessary to reverse the polarity 88 of the tracking error signal (as compared with the case for forward fast replay), arid the positions of the selectors 370 and 372 are opposite to those for the corresponding forward fast replay. That is, tile positions of the selectors 370 and 372 in the -2-time speed replay is opposite to positions of the selectors in tile +4-time speed replay-, the positioris of the selectors 370 arid 372 in tile -6-time speed replay is opposite to positions of the selectors in the +8-time speed replay; and the positions of the selectors 370 and 372 in the -14-time speed replay is opposite to positions of the selectors in the +16-time speed replay. Embodiment 9 In Embodiment 8, the sampling timing pulses for the sample-hold circuits 364 and 366 are generated at the sampling pulse generator 368 in accordance with tile signal indicative of tile drum rotation phase. The accuracy of the sampling timing can be improved if the sync block number in the replay signal. This method will next be described.

Fig. 43 shows an example of configuration of a tracking error detector. The schematic illustration of the capstan servo system is identical to that illustrate(] in Fig. 34. The configuration of the tracking error detector shown in Fig. 43 is similar to that of Fig.:34, but a second sampling pulse generator 382 and a selector 380 are added. The selector 380 selectively connects the outputs of the sampling pulse generators 368 and 382 to the sample-hold circuits 364 and 366. The second sampling pulse generator:382 processes the replay signal output from the head amplifier 3554 and detects the sync block number. The second sampling pulse generator 354 generates a sampling pulse when it detects the sync block riumber of the sync block at the center of the area formed of tile syric blocks where the fast data for the selected replay speed is recorded, i.e., at the sync block No. 124 at the (-enter of the syne block Nos. 104 89 1. 4 to 143 (aO) where +2-time speed replay data and +4-time speed replay data are recorded, during +2-time speed replay or +4-time speed replay, the sync block No. 106 at the center of the sync block Nos. 96 to 115 00) where +8-time speed replay data is recorded, during +8-time speed replay, and the sync block No. 77 at the center of the sync block Nos. 72 to 81 where +16-time speed replay data is recorded, during +16-time speed replay.

The operation for detecting the tracking error will next be described. Except that the manner of generating the sampling pulses is different, the operation is identical to that of Embodiment 8. Accordingly, the manner of sampling pulse generation will be described. When the fast replay is just started, and the tracking control is not in pull-in state, the selector 380 selects the output of the first sampling pulse generator 368. and the sampling pulse generated from the signal indicative of the drum rotation phase is supplied to the sample-hold circuits 364 and 366. When the system is brought into a state in which the tracking control is nearly in pull-in state, and the replay signal from the areas where the required fast replay signal is recorded is obtained, and sampling pulses are generated from the second sampling pulse generator 388, then the selector 380 selects the output of the second sampling pulse generator 382, and each time the predefined sync block is detected, a sampling pulse is supplied to the sample-hold circuits 364 and 366.

The switching operation of the selector 380 can be controlled by a control means, such as a microcomputer, not shown. For instance, judgement is made whether a sampling pulse is output from the second sampling pulse generator 382 is generated during each ration of the drum, and if a sampling pulse is output from the second sampling pulse generator 382 during each rotation of the drum, then the selector 380 is switched to select the output of the second sampling pulse generator 382. Otherwise, the selector is switched to select the output of the first sampling pulse generator 368.

In the description of Embodiments 8 and 9, the drum configuration is 2 ell x 1 type in which two heads provided at positions close to each other. The drum configuration may alternatively be of the 1 ell x 2 type in which two heads are at opposite positions, 180 apart on the drum, as described riext.

Fig. 44 shows head scanning traces during +4-time speed replay in a modification of Embodiments 8 and 9. In this case, the head scanning traces of the first head is identical to those of the 2ch x 1 configuratioii, while the head sciiriri.itig traces oC the second head is dlf'fereiit. Tile fast replay is effected using only the data in the areas aO, bO or cO picked up by the first llead.

The above-described tracking control can be achieved even in the 2ch x 2 system configuration in which two heads each ire disposed at opposite positions, 180 apart. Fig. 45 shows head scanning traces during +4-time speed replay in such a system configuration. In comparison with the case where two heads are at positions oil the drum close to each other. the arigle of inclination of the head scanning traces is different, but the data at both ends of the areas aO, bO and cO which caririot be reproduced is supplemented by the data at the areas al. bl and cl, so that tile fast replay signal can be obtained in a similar manner.

In connection with Embodiments 8 and 9. description is made for the case of +2-, +4-, +8-, +16-, -2-, -G- and -14time speed replays. But the replay may be at any of +4N- or (-4N+2)-time speed (N being a positive integer), and the the fast replay data may be recorded in positions other thail those shown. as far as the data is collectively recorded.

91 -log ', In Embodiments 8 and 9, the pilot signals fl and Q of two different frequencies and fO where none of them is recorded, are used as the pilot signals for tracking. Alternatively, four types of pilot signals may be used, as in 8mm VTR, for tracking control, and yet similar result will be obtained. Embodiment 10 In Embodiment 10, replay of a magnetic tape (Fig. 29) having been recorded as in Embodiment 5 will be described. In Embodiment 5, the low- speed fast replay speed was set at a four-time speed, the middle-speed fast replay speed was set at a eight-time speed and the high-speed fast replay speed was set at a 10-time speed. In Embodiment 10. the replay at the respective fast replay speeds will be described.

Fig 46 shows the rotary head scanning traces followed during four-time speed fast replay of the special replay data in the recording format, using a I ch x 2 head system, according to Embodiment 10. The arrows indicate the head scanning traces. The servo is locked in the the area where four-time speed replay data is recorded. Since two units of the four-time speed replay data is recorded repeatedly, one of the two units is scanned by the A-channel head, while the In this way, it is possible to reproduce the four-time speed replay data recorded using the A-channel head.

Fig. 47 shows the rotary head scanning traces followed during four-time speed fast replay of the special replay data in the recording format, using a 2 ch x I head system, according to Embodiment 10. The arrows indicate the head scanning traces. The servo is locked in the the area where four-time speed replay data is recorded. Since two units of the four-time speed,replay data is recorded repeatedly, one of the two units is scanned by either of the 2 ch heads.

other is scanned by the B-channel head.

92 in tliis way, it is possible to reproduce the four-time Speed replay data recorded using the A-chanliel Ilead.

Fig. 48 shows tl)e rotary head scanning traces followed during four-time speed fast replay of the special replay data in the recording format, using a 2 ch x 2 head system, accordijig to Embodiment 10. The arrows indicate the head scanning traces. I'lie servo is locked in the the area where four-time speed replay data is recorded. Since two units of the four-time speed replay data is recorded repeatedly, one of tI)e two units is scanned by eitl)er of the 2 ch heads. However, for tl)e reason described in connection wil,l) Embodiment 5, not all the four-time speed replay data can be reproduced by the A-channel head alone. However, by synthesis with the four-time speed replay data recorded by tl)e Bchayinel liead and picked up tlie 13-channe.). Ilead, the replay is possible.

Fig. 49 shows the rotary head scanning traces followed during eight-time speed fast replay of the special replay data in the recording format, using a 1 ch x 2 head system, according to Embodiment 10. The arrows indicate the head scanning traces. The servo is locked in the the area where eight-time speed replay data is recorded. Since four units of the eight-time speed replay data is recorded repeatedly, one of the four units is scanned by the A-charmel head. while another of the four units is scanned by tile B-cliarinel head. In this way, it is possible to reproduce the eighttime speed replay data recorded using the A-cliannel head.

Fig. 50 shows tlie rotary head scanning traces followed during eight-time speed fast replay of' the special replay data in the recording format, using a 2 ch x 1 liead system, according to Embodiment 10. The arrows indicate the head scanning traces. Tlie servo is locked in the the area where eight-time speed replay data is recorded. Since four units of the eight-time speed replay data is recorded repeatedly, 93 one of the four units is scanned by, either of the 2 ch lleads. In this s. a., it is possible to reproduce the eighttime speed replay data recorded using tile A-channel head.

Fig. 51 shows the rotary head scanning traces followed during eight-time speed fast replay of the special replay data in the recording format, using a 2 ch x 2 head system, according to Embodiment 10. The arrows indicate the Ilead scanning traces. The servo is locked in the the area where eight-time speed replay, data is recorded. Since four units of the eight-time speed replay data is recorded rel)eitecll-y7, one of tlip four units is scanned by either of' tile. 2 ch heads. llove,,,er, for the reason described in connection with 1 E speed replay, data can 1111)o(lillielit 5, llot all the be reproduced t), the A-channel liend alone. 11 o w e. v e, r. bys v Tl I- is-it'll the speed replay- datn recorded 1)- the B-channel)lead and picked ill) the 13-chamiel head, tile. repla)? is possible.

Fig. 52 shows the rotary head scanning traces followed (Ixjrj.iig 16-time speed fast replay of the special replay data in the recording format, usIng a 1 eli x 2 liend s.sl.eiii.

according to Embodiment 10. The arrows indicate. 1-Ale head Scalilling traces. The set-,%,o is locked in the the nrea ssrliere 1C3-1lime speed replay data is recorded. Since eight. units of the 16-time speed replay, data is recorded ovie of the eight units is scanned by the A-channel head, while another of the eight units is scanned by the 13-eliannel head.

Ill this way, it is possible to reproduce the 16-time speed replay data recorded using the A-channel head.

Fig. 53 shows tile rotary head scanning traces followed during 16-time speed fast replay of' tile special replay data in tile recording format,using a 2 ch x 1 head system, according to Embodiment 10. The arrows indicate the head scanning traces. The servo is locked in the area where 16time speed replay data is recorded. Since eight units of 94 the 16-time speed replay dat.a is recorded repeatedly, olie of the eight units is scaimod by either of the 2 ch lieads. In this way, it is possible to reproduce the 16-time speed replay data recorded using the A- channel head.

Fig. 54 shows the rotary head scanning traces followed during 16-time speed fast replay of the special replay data in the recording format, using a 2 ch x 2 head system, according to Embodiment 10. The arrows indicate the head scanning traces. The servo is locked in the area where 16time speed replay data is recorded. Since eight units of the 16-time speed replay data is recorded repeatedly, one of the eight units is scanned by either of the 2 ch heads. However. for the reason described in connection with Embodiment 5, not all the 16-time speed replay data can be reproduced by the A-channel head alone. However, by synthesis with the 16-time speed replay data recorded by the B-chanrlel head and picked up the B-channel head, the replay is possible.

The processing during replay will next be described.

Fig. 55 shows a circuit for signal processing after the error correction decoding in the replay system according to Embodiment 10. Reference numeral denotes 390 a replay data itiput terminal for input of replay data, 392 denotes a mode signal it)l)tjt termirial for input of a mode signal from a system controller or the like, 394 denotes all ID amilyzer for amilyzing the ID of the syiie block and selectilig the replay data, 396 denotes an SB header analyzer for analyzing the header appended for each sync block and selecting the replay data, 398 denotes an SB/TP converter for converting the replayed sync block into transport packets, arid 400 deriotes a replay SB output terminal.

The replay operation of the signal proeessiiig circuit will next be described. The replay data received at the replay data 1riput terminal (having received error correction decoding of the SD specification), is input to the ID analyzer 394. A signal indicating the replay ITIOde is also input mode input terminal 392, and theri to the ID analyzer 394. Ori the basis of the mode signal, the ID analyzer 394 judges whether the normal replay or the special replay is selected, and outputs the normal replay data recorded ill the main areas, to the next stage. If the special replay is selected, the data recorded ill the special replay areas is output, sync block by sync. block, to the next stage. Ill ench of the replay mode, data for the other replay Mode is discarded. Whether each syric bl.ock is from the main areas ot. Crom the specia.] rephiy areas Is (letermined from the ID or the header appended for each sync block.

The (lata selected and output by the ID analyzer 394 is iriput. to the SB header atial.yzer 396. On the basis of the replay mo(le signal, the SB header analyzer 396 is informecl of the spee(l of the fast replay and outputs the sync blocks corresponding to the speed of the fast replay. The data from the special replay areas which (1o not correspond to the replay mode signal Is disearded. Dur.ing normal replay, the irlpilt ffiltil is 011tpUt ilS iS The discrimination is made oil the basis oC the ID or the [leader appended for each sync. block.

The data output from the SB header analyzer 396 is input to the SB/TP converter 398, which converts the sync blocks into transport packets, arid output via replay SB output terminal 400.

Ill this way, only the data recorded ill the main areas is used during normal replay, while only the data recorded in the special replay areas is used in the special replay at various replay speeds. Both the normal replay arid special replay at various speeds earl thus be achieved. Embodiment 11 96 Ill Embodiment 11, replay of a magnetic tape having beeli recorded as ill Embodimeiit 7 will be described. Ill Embodiment 7, like Embodiment 5, the low-speed fast replay speed was set at a four-time speed, the middle- speed fast replay speed was set at all eight-time speed and the highspeed fast replay speed was set at a 16-time speed. Ill Embodiment 11, the replay at the respective fast replay speeds is performed ill tile same way as ill Embodimerit 10.

The processing during replay will first be described. Fig. 56 is a circuit for processing after the error correction decoding ill tile replay system of Embodiment 11. In the drawing, reference numeral 402 denotes a replay data input terminal, 404 deriotes an ID check circuit for checking whether the IDs are corrected reproduced, 406 deriotes a syiic parity cireijit for cliec.kiiig tile digital data withiti sync. block after the ID, 408 denotes a replay data output terminal, and 410 denotes a flag output termirial.

The operatiori of the signal processing circuit will next be described. The replay data received at the replay data input terminal 402 is supplied to the ID check circuit 404, which checks the ID of the sync block of tile replay data. If the ID is correctly reproduced, the data of the syllc block is output via the replay data output ternlirlal 408. The replay data received at the replay data j.vil)ut terminal is also supplied to the sync parity cheek circuit 406, which cheeks the digital data within the sync block and output a flag indicating tile result of the clieck, via tile flag output terminal 410. If it is found, as a result of the check of the digital data using the sync parity that all error is contained, the flag via the flag output terminal 410 indicates to the error correctiori decoder ill the next stage, that the data being output via tile replay data output terminal 408 may contain an error. Ill this way. it is possible to promptly detects input of replay data containing 97 a Imr-st. error to the error correction decoder. and to detect. erroneous correction at the error correction decoder.

File error correction decoder performs error correction using the cl code 330 arid C2 code 3,31 shown in Fig. 32. The processing of data output from the error correction decoder is similar to the processing after the error correction decoding (Fig. 55) described in connection with Embodiment 10.

In this way, only the data recorded in the main areas is used during the normal replay, while orily the data recorded in special replay areas is used during special replay at various replay speeds, and normal replay arid special replay at various speeds can be achieved.

Tri Embodiment 11, the flag is output to the. error correction decoder. As m alterriative, a gate cIrcult may be provided, arid decision may be made as to whether or riot the replay data should be supplied to the error correction decoder based oil the flag. With such an arrarigement, the data containing a burst error can be detected promptly. Embodiment 12 In Embodiment 12, description is made of the format ill which tile transport packets are recorded ill fixed areas such as syric blocks.

Fig. 57 shows an example of data packet according to Embodiment 12. This data packet format is basically identical to the format in which five syric blocks are recorded in two transport packets according to Embodiment 7. In the drawing, reference numeral 300 denotes a sync of a sync block 0 (SBO),.301 denotes a sync of a sync block 1 (SB1), 302 denotes a sync of a sync block 2 (SB2), 303 denotes a sync of a sync block 3 (SB3), arid 304 denotes a sync of' a sync block 4 (SB4). Reference ritinteral 305 denotes ID of SBO, 306 denotes ID of SB1, 307 denotes ID of SB2,.308 denotes ID of SB3, and 309 denotes ID of SB4. Reference 98 1 ' - numeral 310 denotes a header appended to SBO, 311 denotes a header appended to SBI, 312 denotes a header appended to SB2, 31.3 denotes a Ileader appended to SB3, and.314 denotes a header appended to S34. Reference numeral 315 denotes a transport header of the transport packet A, 316 denotes data of the transport packet A, 317 denotes a transport header B of the transport packet B, arid 318 denotes data of the transport packet B. Reference numerals 319a and 319b denote dummy areas.

Description is made of SBO. ID 305 and Ileader 310 contain an address for identifying the particular sync block within the five sync blocks, a signal indicating whether normal replay data or special replay data is recorded, a signal for identifying the speed where the special replay data is recorded, a signal for indicating the identity of data for several units needed since identical special replay data is recorded for several units arid discriminating from the special replay data recorded in the succeeding several units, arid a signal for identifying the assembly of the five sync blocks, for each unit of the five blocks, and a signal indicating whether the central part of the screen (picture) of an intra-frame or intra-field. In this embodiment, address identifying each sync block within the group of five sync blocks and a signal indicating whether norrital replay data or special replay data is contained are recorded in ID 30.5. and the remainder is recorded in the header 310 dispose(] after the ID. for each sync block.

SB1, S132, SB3 arid SB4 record an ID and a header, like SBO. In this embodiment, the size of the sync block is 82 bytes (excluding the Cl area), the size of each sync is 2 bytes. the size of each ID is 3 bytes, arid the size of each header is one byte. The size of the transport packet is 188 bytes. Accordingly, two transport packets (188 x 2 = 376 bytes) can be recorded in the data regions of five sync 99 blocks (76 x 5 = 300 bytes). The remaining four byte may be allocated for dummy areas 319a and 319b, shown in Fig. 57. two bytes each, and a predefined values may be recorded there. In this way, two transport packets can be recorded in five syric blocks.

Fig. 58 shows a modification of the data packet format of Fig. 57. It is similar to that of Fig. 57. But in place of the two dummy areas 319a and 319b, a single dummy area 319c is provided, and four bytes of predefined values may be recorded in the dummy area 319c.

In the above-described embodiment, the size of the header is one byte. By removing, at the time of recording. the byte indicating the synchronization within the transport header. the size of the transport packet can be reduced, and the area spared may be added to form a larger header. Necessary signals other than those described in this embodiment may be recorded in the area spared in that way.

In this way, an identical format may be used in the main areas and the special replay areas, and reproduction can be made in the form of transport packets. therefore unnecessary to newly form transport time of replay. Embodiment 1.3 Embodiment 13 relates to an arrangement with which a password can be recorded together with a video program, and the recorded video program can be replayed only upon input of a password identical to the password recorded with the program. By the use of a password, the program can be protected from unauthorized replay. The password can be recorded in the area which is used as the dummy area in Embodiment 12.

Fig. 59 is a circuit for signal processing after the error correction decoding in the replay system according to Embodiment 13. In the drawing, reference numerals 390 to It is packets at the 396 denote members identical to those in Fig. 55. Reference numeral 420 denotes an SB/TP converter for converting the syric blocks into transport packets, and separatirig the password from the replay data. Reference numeral 422 denotes a password input terminal for input of a password by a user, zinc] 424 denotes a password check circuit for comparing the password input by the user wit the password from the replay data. Reference numeral 426 denotes a message signal generator for generating a video signal for displayijig a message to the user (viewer) iii(licating that the recorded program is accompanied with a password, and cannot be replayed unless a correct password is input. nie message signal is selected and output when the recorded program being reproduced from the tape is accompanied with a password, arid tio password is input. by the user (viewer) at the time. of replay, or the password Diput by the user (vi ewe r) at the time of repay does not match the recorded password. Reforerice numeral 430 deuotes a replay SB output terminal.

Fig. GOA and Fig. 60B show the corifiguration of the password area according to Embodiment 13. In Fig. 60A, reference numeral 440 denotes a dummy area 319a or dummy area:31.9b in Fig. 57. Subareas 441, 442, 443 and 444, each havirig four bits, are formed by dividing the dummy area 440 into four, and are called password subareas A, B, C and D.

In Fig. GOB, reference numeral 450 denotes a dummy area 319c in Fig. 58. Subareas 451, 452, 453 an(] 454, each having eight bits, are formed by dividing the dummy area 440 into four, and are called password subareas E, F, G and Ii.

Since the password subareas 441 to 444 each have four bits, each password subarea can express a number of 0 to 9, so that password of four digits can be recorded. Since the password subareas 451 to 454 each has one byte, each subarea can record aii English alphabetic letter, or a riumber, so a 101 password of four digits, each digit b-elilg either a iiumber or aii Eiiglish alphabetic letter, cail be recorded. The password caii be set by the user at the time of recordiilg a program, aiid recorded. Wheri the user does iiot set the password, a predefiiied value, e.g., of "1" for all the bits. may be recorded to iiidicate that iio password has beeii set.

Now the deseriptioii is made of replay operatioii. The replay syilc blocks are iiiput to tile SB/TP coriverter 420, where five syiic blocks are syiittiesizecl, aiicl two traiisport packets are extracted from the five syiic blocks. The data (of four digits) recorded at the password area is extracted, <-iiid while the supplied to the password check circuit 424, trailsport packet is supplied to the selector 428. The data front the password area is checked by the password cileck circuit 424. If the data is of a predef iried valtie. i. e., if the data coiisists of bits which are all -i" iri the example uiAer c.oiisi(leratioii, theii the program is treated as beiiig iiot protected by a password. If tile data front the password,area is tiot of the predefiried value, atid if a password is iriput by the user (viewer), which is supplied via the password iiiput terniiiial 422 to the password check circuit 424, the iiiput password is compared with the recorded password. If they niatch, the pro(.essiiig will be the same as iii the ease where the password is iiot recorded, iii(l the selector 428 is niade to select the transport packets forniiiig the replay data. If the passwords do iiot match, or if iio password is iiiput by the user (aild if the recorded progrant is accomparlied with a password) the selector 428 is made to select the sigilal front tile message sigrial geiierator 426, alid a niessage is displayed, indicatirig that the progrant is protected by a program aiicl c.,iiiriot be replayed iiiiless a correct password is iriput. It may alternatively be so arraiiged that wlieii rio password is iiipxjt while the recorded program is protected by a password a message promptiiig the 102 user to input a password is displayed, and when a wrong password is input a message indicating the input password is wrong, and a correct password should be input is displayed.

When the program is protected by a password. and a correct password is not input, display of the program is inhibited. This is achieved by the operation of the selector 428, which does not select the transport packets forming the replay data. Additionally (or alternatively), the tape transport and head scanning may also be stopped, unlesq or until a correct password is input.

With the configuration and operation described above, it is possible to protect the program from being seen by an unauthorized user. Embodiment 14 In Embodiment 10, replay is performed at a speed set by Embodimept.5. In Embodiment 14, replay is performed from special replay areas for a specific speed, at a speed lower than the specific speed.

Fig. 61 shows head scanning traces of the rotary head during six-time speed replay of eight-time speed replay data in a recording format of Fig. 29, using a 1 cli x 2 head system, according to Embodiment 14. The arrows indicate the head scanning traces. The special replay data for the sixtime speed replay is obtained by reproducing the eight-tiMe speed replay data, four units of which are repeatedly recorded in the eight- time speed replay areas. When reproducing at a six-tinte speed from the eight-time speed replay areas, tile servo is locked at the eight-time speed replay areas. By this method, identical special replay data may be reproduce(] twice. In that case, one of them is discarded, to achieve replay at a six-time speed.

Fig. 62 shows Ilead scanning traces of the rotary head during six-time speed replay of eight-time speed replay data in a recording format of Fig. 29. using a 2 ell x 1 [lead 103 1 ', - system, according to Embodiment 14. The arrows Indicate the Ilead scanning traces. The special replay data Cor the sixtime speed replay is obtained by reproducing the tile eighttime speed replay data, four units of wfi,ich are repeatedly recorded in the eight-time speed replay areas. When reproducing at a six-time speed from the eight-time speed replay areas, the servo is locked at the eight-1.1me speed replay areas. By this method, identical special replay data may be reproduced twice. In that case, one of them is discarded, to actiieve replay at a six-time speed.

Fig. 6.3 shows head scanning traces of the rotary head during six-time speed replay of eight-time speed replay data in a recording format of Fig. 29, using a 2 ch x 2 head system, according to Embodiment 14. The arrows indicate the head scatining traces. The special replay dat.1 for the sixtime speed replay is obtained by reproducing the the eighttime speed replay data, four units of which are repeatedly recorded in the eighttime speed replay areas. When reproducing at a six-time speed from the eight-time speed replay areas, the servo is locked at the eight-time speed replay areas. By this method, identical special replay data may be reproduced twice. In that case, one of them is discarded, to achieve replay at a six-time speed.

In EmbodiRierit 14, description is made of the cases where the replay from the eight-time speed areas is conducted at a six-time speed. But the inventive concept described above cari be applied to situations where replay from special replay areas for a set replay speed is conducted (it a replay speed lower than the set speed. Embodiment 15

In Embodiment 15, replay is made frorn special replay areas for a specific replay-speed. at a replay speed Iiigher than the specific replay speed. Description is made for the case in which replay is effected from the area-, for four-

104 time speed in Embodiment 5, at 12-time speed.

Fig. 64 shows head scanning traces of the rotary head during twelve-time speed replay of four-time speed replay data in a recording format of Fig. 29, using a 1 ch x 2 head system, according to Embodiment 15. The arrows indicate the head scanniiig traces. The special replay data for the twelve-time speed replay is obtained by reproducing the four-time speed replay data, two units of whicli are repeatedly recorded in the four-time speed replay areas. Illien reproducing at a twelve-time speed from the folir-time speed replay areas, the servo is locked at the four-time speed replay areas.

Fig. 6.5 shows head scanning traces of the rotary head during twelve-time speed replay of four-time speed replay data iti a recordirig format or Fig. 29, using a 2 ch x 1. Itead system, accordiiig to Embodiment 15. 'I'lie arrows indicate the head scanning traces. The special replay data for the twelve-time speed replay is obtained by reproducing the four-time speed replay data, two units of which are repeatedly recorded in the fourtime speed replay areas. When reproducing at a twelve-time speed from the fOIJr-tiITIe speed replay areas, the servo is locked at tlie four-time speed replay areas.

Fig. 66 shows head scanning traces of the rotary head during tivelve-time speed replay of four-time speed replay data in a recording format of Fig. 29, using a 2 ch x 2 head system, according to Embodiment 15. The arrows indicate the head scanning traces. The special replay data for the twelvetime speed replay is obtained by reproducing the four-time speed replay data, two units of which are repeatedly recorded in the four-time speed replay areas. When reprodticing at a twelve-time speed from the four-time speed replay areas, the servo is locked at the four-time speed replay areas.

Fig. 67A and Fig. 67B are used to explain the fast replay according to Embodiment 15. Fig. 67A shows the configiiration of the recording areas of the four-time speecl replay data. Fig. 67B shows the positions on the screen. In each of the cases shown in Fig. 64 to Fig. 66, it is necessary to record the data in the four-time spee(] special replay areas in the form shown in Fig. 67A and Fig. 67B. III the drawing, reference numeral 242 denotes a special replay area for four-time speed, recorded by an Achannel head, 244 denotes a special replay area for four-time speed, recorded by a B-CIM11nel head, 460 denotes a one intra-frame or one intrnfield screen as a whole, and 462 denotes I central part of' the one intraframe or one intra-field screEl.n.

Of the data recorded in the special replay area 242 for four-time speed, recorded by the A-channel head, t1le Central part (in the embodiment under consideration, the ser-,,o is assumed to be locked at the central part of each special replay area) is used to record the data of the central part 462 of the screen of one intra-fralfle or intra-field pictilre. This data is part of the four-time speed data, and no additional four-time speed areas are used. It is slifficient if the four- time speed special replay areas 242, recorded by the A-channel head, is recorded at an interval of a predefined number of tracks. In this embodiment, since twelve-time speed replay is effected, the interval consists of six units, each unit consisting of four tracks. Of' the the special replay areas 242 recorded by the A-channel head, the areas other than the areas where the data of' the central part 462 of' the screen (whole picture) of one intra-frame or intra-field picture, and the four- time special replay areas 244 recorded by the B-channel head are used to record the data other than the data of' the central part 462 of the screen of one intra-frame or intra-field picture, that is the data of the screen 460 of one intra-frame or intra-field

106 picture minus the data of the central part 462 of the screen of the one intra-frame or intra-field picture. By replaying the signal for the central part of the screen, the special replay, with a high picture quality and with frequent refreshing can be obtained.

In Embodiment 15, description is made of the cases where the replay from the four-time speed areas is conducted a t a twelve-time speed. But the inventive concept described above can be applied to situations where replay from special replay areas for a set replay speed, in a format in which the special areas for the set replay speed is collectively disposed, Is cov)ducted at a replay speed higher than the set speed.

In Embodiment 15, description is made of the cases sylicre the central part of' UIC screell of' '111 or intra-field image is recorded in part of the special replay area recorded by an A-chamiel head. The invention is not limited to this particular arrangement. The central part of the screen of an iritraframe or intra- field image may be recorded in such part of the special replay area for a set replay speed fron) which data can be reproduced at a speed higher than the set replay speed. in the recording format in which the special replay data is recorded at one. location where the special replay area for- the set replay speeds are concentrated as shown in Fig. 29. Embodiment 1.6

In the following Embodiments 16 to 19, description is made of' various de, ,,-jces for removirig the effects of fluctuation in the head position to ensure reproduction of replay data at a Iligh speed.

As an example, it is assumed, according to the basic specification of' the prototype consumer digital WIT, each track on the tape corresponds to 186 sync blocks (SBs), the difference between the starting positions of the adjacent

107 1 An n 1 track in the track lorig.itti(Iiiiil dIrection is (1 sync blocks (d = 0. 35 SB), and the track width wid the head width are ideritical. Eiiil)o(ILrFictil-l 1.6 is describe(] on the il)o,,:.e. ass ump tion.

Fig. 68 is a block diagram showing a recording system of a digital VTR according to Embodiment 16. In the drawing, reference numeral 470 denotes in input terminal for an ATV signal bit stream, 472 denotes a variable-length decoder, 474 denotes a coutiter, 476 denotes a data extractor, 478 denotes an EOB (end of block) appending circult, and 480 denotes a sync signal generator. Reference numeral 482 denotes a syne block generator, which appends the bytes to the bit stream, oii the basis of the sync signal from the sync sigmil generator 480, to form sync bl-ocks to be recorded 1n the main areas oii the tracks, and forms Cast. replay syne blocks on the basis of the, fast replay signal from the EOB appending circuit 478, to thereby form a sigrial to be recorded in the predefined sym. blocks.

Reference riumeral 484 denotes a recording signal processor for performing recording signal processing such as recording modulation and recording amplification, 70 denotes heads of two different azimuths, and 10 denotes a magnetic tape.

The recording operation by the above recording system will next be described Iri detail. MPEG2 bit stream is input. via the input termiiial 470, and supplied to the sync block generator 482, where sync bytes are appended, on the basis of the sync signal from the the syne signal generator 480, to form sync blocks. The bit streant receise(l at the input terminal 470 is also supplied to the variable-lerigth decoder 472, syliere the syntax of the MPEG2 bit stream is analyzed, and intrainiages are extracted, and timing signals are generated by the counter 474. and the low-frequeney components of all the blocks of the intra-images are extracted at the data extractor, and E0Bs are appended at 108 the EOB apperid! rig c. 1 r.cti J t 478. to f orni f ast replay data, which is otll,.1)tit to I-Ihe sym., bloc]( generator 482. On the basi.s of' the sync signal. Crom the syne signal generator 480, the syric. block generator 482 appends sync bytes to the fast replay signal front the EOB appending circuit 478, to forin the syne blocks for fast replay, zinc] forms a recording signal to be recorded in tile predefined sync blocks.

The recording signal formed of the respective sync blocks front the sync block generator 482 is supplied to recording signal processor 484, where various recording signal 1)ro(.,.,ssirig, c;iic.li as digital recording modulation, recording amplification. are applied, and then supplied the heads 70 of two different azimuths, and recorded oil the and to the magnetic tape 10.

Next. description 1.s made of Hie dispos.11--ton oil tile tracks for recording fast replay syric blocks which are fast replay data.

Fig. 69 shows a scanning trace of a rotary head on the tracks during fast replay. The drawing shows the case where tile the replay speed is fivetime speed. i.e., tile speed multiplier m is five, and the length of the tracks in terms of the number of the sync blocks sync block is 186 SBs, and the difference d between tile starting positions of the adjacent tracks A and B, in the track longitudinal direction is 0.35 SB. The relationship between the difference D between crossing positions in the track longitudinal direction, and the length Te of tile areas of the track front which reproduction is possible is illustrated. If the tape speed in is in integer-multiple speed, and the phase. lock is controlled, the headscanning is in synchronism with tile tracks of the identical azimuth, arid the positions of the data which is reprodticed are fixed.

Referring to Fig. 69, if it is assurned that such part of the replay signal whose Output level is -6dB or greater 109 is reproduced, the head A (.an reproduce data from the hatched regions. If the track width and the hend width are identical, the different D between the crossing positions of the head A in the track longitudinal direction is D = Te + Tu, where Te = Tu, and the total length of the regions from which reproduction is possible is Te = S - (m - 1) x d}/(ni - 1) Fig. 70 shosys a scanning trace of a rotary head during replay at 56-time speed. Fig. 71A to Fig. 71C are for explaining the position fluctuation of the rotary head scanning trace. Fig. 71A shows the scanning trace by which three sync blocks can be reproduced, while Fig. 71B and Fig.

71C show the scanning traces shifted forward and backward.

The reg.ioiis rrom MiLch I-Ihe or the. is ensured during 56time speed replay is hatched regions.

Each of the regions from which reproduction is possible, as determined by the above-recited equation, amounts to:

Te = (S - 55 x C/55 = 3.0 SB The ni;l.xtiiixiTii ntimber n (ii being an fixteger) of consecutive syne blocks which can always be reproduced from the above region Te = 3.0 SB, in other words, minirmini number ri (n being an integer) of consecutive sync blocks ivitliiii the above region Te = 3.0 SB from which reproduction of data is ensured, is n = 2 SB. This is because the limits of the region f rom which reproduction is possible do not necessar ily coincide with the boundaries of the sync blocks, aS ShOWII iTl Fig. 71A to Fig. 71C. For instance, the sync block J2 is read in the case of Fig. 71A, but not in the case of Fig. 71B. The sync block j4 is read in the case of Fig. 71A, but not in the case of Fig. 71C. Accordingly, the maxintum number of the consecutive sync blocks, within the region from which reproduction is possible, from which reproduction is possible without fail is 2 SB if Te = 3 SBs.

If Te is not ail integer, such maximum number is ii = t - 1 SB, where t is a maximum integer which does not exceeds Te.

it is seen from the above that, in tile case of 56-time speed replay, fast replay sync blocks should be recorded in the areas 1 to 3 in Fig. 70.

When, however, fast replay is conducted using a rotary drum, the position at which the head crosses the respective tracks may be shifted because of the fluctuation in the head scanning trace due to the tape speed fluctuation, the drum rotational speed fluctuation, arid like. In such a case, it is necessary to read the data of 2 SBs for fast replay, without fail. If the maximum value of' shift, from the reference position, of the actual position at is,llic.li the llead crosses a specific. track during fast replay at a certain speed is w syme blocks (having rounded ill) to the next. integer,. i. e., the actual shif t being not more than w sync. blocks, but more than (sy-1) sync blocks), the range of shift is +w SB in the track longitudinal directiori from the reference position which is attained when the phase is locked.

Fig. 72 shosys disposition of the fast data according to Embodiment 16. It is assumed that the shift at 56-time.speed is w = 4 SB. The region within which the syric blocks may be seanned because of' the shift extend (ii + 2 x w) = 10 SB. Accordingly, if' the data for 2 S13 is designated by D1 and D2. the data D] and data D2 are repeatedly and cyclically recorded for the range of (ii + 2 x w) syric bloc.ks. Fig. 73 shows ail example of disposition of the fast replay data oil a track according to Embodiment 16. Highspeed replay data is sequeritially (in the ascending order of suffix i to D) and repeatedly (or cyclically) recorded over 10 sync blocks, numbered X, X+l,... X+9, centered on the reference position of the region where replay is possible by the head crossing a specific. track.

ill 1. - Witil this arrangement, it is possible to ensure reproduction of the recorded sync block data D1 and D2 of 2 SB for fast replay during fast replay at a certain speed, even when the position at which the head crosses the specific track is shifted.

Fig. 74 is a block diagram showing a replay system of a digital VT1R of Embodiment 16. In the drawing, reference numerals 70 and 10 denote members identical to those in the recording system shown in Fig. 68. Reference numeral 490 denotes a replay sigrial processor for performing replay signal 1) rocess i rigs, such <is waveform equalization, signal detection and recording demodulation, 492 clenotes a replay data separat. or for separating tile nornial replay data and tile fast replay data in the replay signal. 494 (leriotes a sele(.t-.or. 496 deriotes a replay mode signal gerierator for generating a signal indicating the replay mode. and 498 denotes an output terminal.

During replay, the replay signal replayecl by the head 70 front the magnetic tape 10 is supplied to the replay signal processor 490, where replay signal processings, such as waveform equalization, signal detection, and recording demodulation, are applied. The replay signal is theri supplied to the replay signal separator 492. where the signal. replaye(l front the tracks is separated irito the bit stream (g) for normal replay data, and the sync block data for fast replay, which are then supplied to the selector 494. Oil the basis of the signal indicating tile replay mode, from the replay rnode signal generator 496. the selector 494 selects the normal replay data (g) during norinal replay, and the fast replay data during fast replay, and the selected data is output via the output terminal 498, and sent to an MPEG2 decoder provided outside of the digital VTR.

In the manner described above, by recording ii pieces of 112 data 1)i ( i = 1 ' 2,... ' n) each of' which can be recorded in one sync block sequeritially and repeatedly in (n + 2 x w) consecutive sync blocks from which data is reproduced <it mtime speed, it is ensured to read fast replay data even syllen the position of the head scanning trace fluctuates, because of the tape transport speed fluctuation, or the drum rotary speed fluctuation, and fast replay pictures with a good quality can be obtained, and much of the data for fast replay can be recorded and replayed. Embodiment 17 In Embodiment 16, sync block data for fast replay is recorded in predefined positions on predefined tracks which are scarmed during ni- time speed replay.

It is also possible to repeatedly record the fast replay data so that the fast replay dala cim be can be read regardless of' the. azimuth track at which (al. whose end) the rotary head begins scanning. In that case, the pull-in of the servo system is quick and the fast replay image can be obtained instantly.

Fig. 75 shows the positiorial relationship between the scanning traces and the fast replay data according to Embodiment 17. Identical. sync bloc]( positions on various icleritl(.,il-,i7,Iiiitit.li tracks arp scaimed. from respective starting poilits. To enable m-time speed replay. ldentical sym. block data for. -rast: replay is repentedly rpeorded over (ii + 2 x w) consecutive sym. blocks at identical position on ench or at least m consecutive identical-azimuth tracks, as shown, by way of example, in Fig. 76. In this way, regardless of the track (of the identical iZi[Tlllt,tl) (It WhiCh the fast replay is started, the replay data can be obtained.

Iii the manner described above, by repeatedly recording mtime speed signal is recorded in (n + 2 x si.) consecutive. syne blocks at identical positions on ru consecutive ideritical-azimuth tracks, reading of the fast replay data is ensured in the event or fl.tictuation in the hend scanning 113 traces due to tal)e sl)eeci fluctuation and drimi rotary, speed f luctuation, and reproduction of good citiali I,.)- pi c. 1Aires is ensured, and much fast replay data can be recorded and replayed. Embodiment 18 Embodiment 18 relates to a bit stream recording and replay device capable of fast replay, Ivith a different example of disposition, on tracks, of fast replay syric block forming fast. replay data.

Fig. 77 to Fig. 79 show rotary liend scanning traces during 56-time speed replay according to Embodiment 18. Fig. 77 to Fig. 79 show examples of. 536-time speed replay, with different phase control. positions and different llead traces. Replay signals are picked up from the hatched 1) o r t i o ns. For instance, in Fig. 77, tfie f'oiii-l-.11 1-.o sixt..II sync blocks i.e., from the beginning of' the fourth syllc block to the end of sixth sync block, or 4.0-th to 7.0-th sync blocks are read. Similarly, in Fig. 78. the 4.7-th to 7.7-th syne blocks are read, and in Fig. 79. the.7).7-1-.)1 to 8.7-th sync blocks are read.

To ensure reading of replay data at fast replay, at whichever phase the head traces is achieved, it is so arrange(] that, even sylien the fast replay signal is not obtained from one track during fast replay, reading of the signal from the next identical-azimuth track is ensured. That is, even when the head trace position is shifted due to phase fluctuation, the recorded sync block data for fast replay, can be obtained from the total of one track and a next identical-azimuth track.

Fig. 80 shows the positional relationship between the scanning trace and the fast replay data according to Embodiment 18. It shows the positions of the regions Te in two identical-azimuth tracks A1 and A2 from which reproduction is possible during fast replay. In Fig. 80, if 114 the reproductioll is possible from the porti.ons wliere the level of the output replay signal is greater tlian -6(113, the signals are reproduced from the hatched regions in tile. tracks A1 and A2. If the track width and the head width are identical, the length Tu which is the difference between the upper end an(] lower ends of the regions on the tracks AI and A2 is given by:

Tu = 3S - (ill - 1) x C/(m - 1) The position of the sync block in track A2 is 2d sync blocks higher than the position in track AI.

Fig. 81A and Fig. 81B show the fluctuation in the position of the rotary head scanning trace according to Embodiment 18. (A) shows the scanning trace by which 3 sync blocks can be reproduced, and (B) shows the scanning trace followed wlien the position is varied. To ensure reproduction of fast replay sync,. block data from the two identical- azimuth tracks AI and A2 during fast replay, even when head scanning phase is changed in the two identicalazimuth tracks AI and A2, sync block data of a length of not less than (Te + Tu) sync blocks is repeatedly recorded oil track AI, from the starting point of the region syliere the fast replay sync block is recorded, toward the tail end of the track.

For instance, when fast replay is performed at 56-time speed, the tiiaxiiiitirij number of sync blocks wbich (.,in always be consecutively reproduced from the track region on the tape is ii = 2. and the length of the region from which the replay signal (.,in be obtained is Te = 3 S13. If fast replay sync block data Dl and D2 is repeatedly record over 6 SB in the direction of from sync. block 1 to sync block 6, in Fig. 81A and Fig. 8113, the fast replay data earl be read, even if tile phase is shifted in the track longitudinal direction, toward the tail end of the track. In track A2 also, if fast replay data is repeatedly recorded over (Tu + Te + 2(1) front the tail eiid of the regioil front which the reproductioit. is possible toward the head eii(l oC the track, as showm iii Fig. 80, the fast replay data DI, D2 caii be read front the track A2, everi if the phase is shifted iii the track loiig.Lttidiiizil directioii, toward the head eii(l of the track.

Let us riow coiisider the case where the sytic block data D1, D2 is to be obtairied front the track A2 oii.l.y, or the ease where the sy,iic block D1 is obtaiiied from track Al, arid D2 is obtaiiied front track A2. Iii the case where the syric block data DI, D2 is to be obtaiiied from track A2, the syiic block data should be recorded up to such a 1)ositioii that. sytic. block data D1 eari be read front track AI arid syric block data DI, D2 cari be read front the track A2. Fig. 82 is a schematic diagratn slioiv.liig the positiori at which syric block data DI cail be read from track AI atid,,-yfic block data D1, D2 earl be read front the track A2. The fast replay sigrial D1, D2 stiould be disposed iii the sytic blocks iii the liatelied regiori of front the (Tu + 2d + 1)-th sytic block to (D + 2d + 1) mi the track A2.

Iii the case where syiic block data D1 is obtairied front the track AI, arid the syiic block data D2 is obtairied front the track A2, the syiic block data should be recorded to stich a positimi that the syric block data D1, D2 cart be read front tile track AI arid the syilc block data D2 cari be read front the track A2. Fig. 83 shows the schematic diagram showing the position at which the syilc block data D1, D2 caii be read front the track AI arid tile syiic block data D2 cart be read from the track A2. The fast replay sigiial D2 should be disposed iii the syiic blocks iii the hatched regiori of front (Tu + 2d + 2)-th syiic block. to (D + 2d + 2) syiic block oil the track A2 iii the drawiiig.

Front tile above it is seeri. that, ivlieii fast replay is performed at 56-tirne speed for iristmice, tile maximum Tiumber of syiic blocks which cart, always be reproduced coiisecutively 116 is ii = 2 SB, the length of the region from whiell the replay signal call be obtained is Te = 3 SB, and where the sync block data D1 is read from the track Al, and the sync block data D1, D2 is read from the track A2, the fast replay signal D1, D2 are disposed ill the sixth and seventh sync blocks ill the region oil the track A2 from which reproduction is possible, as shown ill Fig. 84. When the syne block data D1, D2 is read from the track A1 and the sync block data D2 is read from the track A2, the fast replay signal D2 is disposed ill the seventh sync block as shown ill Fig. 8.5. Ill this case, the fast replay speed signal is repeatedly recorded ill the respective identical-azinitith tracks, and, ill doing so, the two pieces of sync block data D1, D2 are sequentially (ill the ascending order of the suffix t 1-1c, D) repeatedly recorded ill seven consecutive sync blocks at identical position oil the respective tracks, and the data are so disposed that the seventh data of the track identical to the second data (D2 ill the example of Fig. 8.5) of the seven pieces of fast replay data recorded ill the identical sync block position oil the immediately preceding identicalazimuth track. and the disposition ill the fast replay regions oil the tracks is as shosyn ill Fig. 86.

Fig. 87 shows the length of the.sync 1)1ocks for the fast repLay data where the fast replay is performed at mtime speed, the maximum number of the sync blocks which call almlys be reproduced consecutively from the region oil the track of the tape is ii, and the length of region from which the reproduction signal call be obtained is Te. the difference between the head crossing positions ill the track longitudinal direction is D = Te + Tu. and n pieces of sync block data D1, D2,... Dn are consecutively recorded. When the minimum integer which is not smaller than (Tu + 2(1) corresponds to L (here, Tu = D - Te), n sync block data are sequentially (ill the ascending order of the suffix i to D) 117 repeatedly recorded in (L + ii + 1) consecutive sync blocks at identical positions oil the tracks and the data are so disposed that the (L + ii + 1)-th data in the track is identical to the n-th data (Dn in the example shown in Fig. 87) of the fast replay data recorded in the identical sync block position oil the immediately preceding identicalazimuth tr,,1c.k, aild recorded oil at least ni ideritical-azirlititli track. With such an arrangement, the reading of the fast replay signal is ensured even if the phase is varied.

Disposing the data such that the (L + ii + 1)-th data oil the track to be identical to the ii-th data of the fast replay data recorded at the same sync block position oil the immediately preceding identical-azimuth track mearis recording the data Di to satisfy the relationship e2 = mod [ {el + n - mod (ii + L + 1. n), ri 1 where mod (a. b) expresses the remaInder Of IMITieral a divided by numeral 1), and the suffixes of D recorded first on tracks A1 an(] A2 are el and e2 (integers not less than 1 and not rnore than n).

When the fast replay signal for ni-time speed is recorded in the above manner, n pieces of data Di (i = 1. 2,... ii) each of which call be recorded in one sync block are sequentially (in the ascending order of the suffix i to D) and repeatedly recorded in (L + ri - 1) consecutive sync blocks, and the data are so disposed that the (L + ii + 1)-ti data oil the track is identical to tile ii-th data of the fast replay data recorded at the same sync block position on the immediately preceding identical-azimuth track. Accordingly reading of fast replay data is ensured even when the head trace phase is varied clue to variation in the head scanning traces, and fast replay images with a good quality earl be obtained. and much fast replay data can be reproduced. Embodiment 19 In Embodiment 18. fast replay data of the maximum 118 - ' - - -T number n of sync blocks whic.11 can always be reproduced consecutively from the region of the track oil the tape, during ni-time speed replay of fast replay data, is repeatedly recorded in a necessary number of sync blocks, (L + n + 1). The number p of the fast replay data may be less than n (1) being a natural number), and the number of the regions for the fast replay may be more than (L + n + 1).

Fig. 88 is a schematic view showing the data oil the respective tracks in the case where the data of the maximum number ri of sync blocks which can always be reproduced consecutively from the region of the track oil the tape during the fast replay at 30-time speed is recorded <is the fast replay data. At 30-time speed, the maximum number of syne blocks which rail always be reproduced con.secutively from the regi.ori or the track ori the Lape is five. and tile length of the region from which the replay signal can be obtained is Te = 6 SB, and the length of the sync blocks of the fast replay data where the five sync block data D1, D2,... D5 are consecutively recorded is (L + n + 1) = 13 sync blocks which c-ire consecutive at identical positions oil the tracks. (Here, L is again a minimurn integer not smaller than (Tu + 2(1).) The data are sequentially (in the ascending order of the suffix i to D) and repeatedly recorded, and the data are so disposed that tile 13-th data oil tile track is identical to the fifth data of the fast replay data recorded at the identical sync block position oil the immediately preceding identical- azimuth track. In this way, reading of the fast replay signal is ensured even if the phase fluctuates.

Fig. 89 shows disposition of the data iii the fast replay region in the case where tile fast replay data is for the fast replay at 30-time speed and is formed of p = 2 sync blocks. For conducting 30-time speed replay, the length of the sync blocks of tile fast replay data used for recording 119 the two sync block data D1, D2 consecutively.1s (L + p + 1) = 10 sync blocks an(] these 10 syric blocks are consecutive at the same position oil the track. (Here, L is again a minimum integer riot smaller than (Tu + 2d).) The data are sequentially (in the ascending order of the suffix 1 to D) arid repeatedly recorded. arid the data are so disposed that the lotll data of the track is identical to the 1) = 2nd data of the fast replay data recorded at the same sync block position oil the immediately preceding i dent i cal -azimuth track. In this way, even when the phase fluctuates the reading of the fast replay signal is ensured. Fig. 90 shows an example of disposition of fast replay data. Specifically, it shows tile disposition of data in the fast replay region for the case where the data for the 30-tirite speed replay is formed of 1) = 2 sync blocks.

Since the length of the region for the fast replay data is 10 sync blocks, the 56-time speed replay according to Embodiment 18, and the fast replay at a speed with which the maximuril number of sync blocks which can always be reproduced consecutively is not less than 2 and not more than 6 may be performed, arid yet the reading of the fast replay signal is ensured even if the phase fluctuates. Fig. 91 shows scanning traces in 56- time speed replay. In this ease, the length of sync blocks necessary for always reading two data is 7 as was explained in connection with Embodiment 18, arid with the arrangement of Fig. 90, reading is ensured regardless of the pliase. Fig. 92 shows disposition of the fast replay data and head traces during 44-time speed replay. The maximum number of sync blocks which can always be reproduced consecutively is 3, and Te = Tu = 4.0 SB, so (L + p + 1) is 8 SB. With the disposition of Fig. 90, too. reading is ensured at 44-time speed regardless of the phase. Accordingly, the example of Fig. 90 enables fast replay from 30-tinte speed to 56- tinie speed.

1 -;;i -, Disposing the data such that the (L + p + 1)-th data oil the track is identical to the p-th data of the fast replay data recorded at the same sync block position on the immediately preceding identical-azimuth track means recording data Di in such a mariner as to satisfy the relationship:

e2 = mod [ { el + p - mod (p + L + 1, p), p 1 where mod (a,))) expresses the remainder of a divided by 1); and el, e2 (integers not less than 1 and not more thail il) are suffixes to data D which are recorded first oil the tracks A1 and A2, respectively.

In the manner described above, in recording the rri-time speed fast replay signal orl tile tracks, p pieces of data Di ( i = 1, 2.... p) each of which can be recorded in one sync block are sequentially (in the ascending order of the suffix i to D) and repeatedly recorded in the (L + 1) + 1) consecutive sync blocks at tile same position oil the identical- azimuth tracks, and the data are so disposed that the (L + p + 1)-th data oil tile track is identical to the pth data of the fast replay data recorded at the identical syne. block position oil the immediately preceding identicalazirituth track, and the data is recorded oil at least m identical-azimuth tracks. With such tin arrangement, even when the head scanning traces fluctuates or the head trace phase is shifted, reading of the fast replay data is ensured, and fast replay image of a good quality is obtained, and much fast replay data can be recorded and replayed.

The invention may be embodied in other specific forms and manners without departure from its scope.

121

Claims

1. A dicital VTR for recording recording digital video and audio signals in respective desi-conated areas of oblique tracks in a predetermined track format, using a rotary drum on which head of two different azimuths are mounted, comprising: data separating means for extracting a fast replay signal from the normal recording sig g gnal; recording means for recording the fast replay signal in one region C-5 in one track per one scanning of the head, of the regions covered by, the head traces and in the tracks of identical azimuth; identification signal recording means for recording an identification sigmal for identifying the track; and replay means for replaying the identification signal.

2. The digital VTR as set forth in claim 1, wherein a first recording region is provided in one track of one azimuth in which said fast replay signal is recorded. and a second recording region for recording the fast replay sigmal is also provided in the track of the other azimuth, and succeeding said one track; the length of the second recording region is about half the length of the first recording region, and the center of the second recording region within the track is at about the same position as the center of the first recording region within the track.

3. The digital VTR as set forth in claim 2, wherein, in the upper and lower end parts of the first recording region which extend out of the region corresponding to the second recording region of the adjacent 122 track of a different azimuth, the signal identical to those in said second recording region is recorded.

4. The digital VTR as set forth in any of claims 1 to 3, wherein said recording, means forms the fast replay signal dedicated for th g e particular fast replay signal for each of the fast replay speeds, and records the fast replay signal at different positions on the magnetic recording tape.

5. The digital VTR as set forth in claim 4, wherein said recording means repeatedly records the fast replay signal for (M x i)-time speed replay (i = 1, 2,... n) at predetermined positions in predetermined tracks of consecutive M (M being a natural number) tracks, and repeatedly records the fast replay signal for (M x i)-time speed replay, 2 x i times, taking the M tracks as one unit for each speed.

6. The digital VTR as set forth in claim 5, wherein said recording means repeatedly records the fast replay signal for 4i-time speed replay (i = 1, 2,... n) at predetermined positions in predetermined tracks of consecutive four (M being a natural number) tracks, and said identification signal recording means records three types of frequency signals as pilot signal for tracking control on these four tracks, being in superimposition with the digital data.

C

7. The digital VTR as set forth in claim 1, further comprising error correction code appending means for appending the error correction code formed of a predetermined number of sync bits inserted at a predetermined period in the signal sequence recorded in the magnetic 123 recording tape, a predetermined number of ID bits succeed:-iúy said sync bits, a predetermined number of first parity bits generated from the ID bits, second parity bits generated from a predetermined number of digital data succeeding the first parity bits, third parity bits generated from a plurality of digital data extending over said sync bits, and fourth parity bits generated from the digital data and positioned at the back of said digital data; erroneous correction detection means for comparing the fourth parity bits with the first parity bits reproduced by said replay means, and detecting erroneous correction on the basis of the result of comparison.

8. The digital VTR as set forth in claim 7, wherein said error correction code appending means appends the fourth parity bits only to the fast replay signal.

9. A digital VTR for recording digital video and audio signals, in Z designated areas on oblique tracks of a magnetic recording tape, in a predefined format, using a rotary drum on which heads of two different azimuths are mourited, and replaying from the areas, comprising: data separating means for extracting digital video signal (hereinafter referred to as fast replay signal) used for fast replay, from a normal recording signal; recording means for recording the fast replay signals for the respective fast replay speeds, in predefined consecutive regions in a predefined track of a group of four consecutive tracks; identification signal recording means for recording identification sigan , al for identifying the tracks; 124 replay means for replaying the recording signal for normal replay, or fast replay signals for +2-time speed replay, or +4N-time speed replay or (-4N+2)-time speed replay (N being a positive integer); and tracking control means for performing tracking control so that said head scans the redefined regions in the predefined track of the four p Z:' tracks in accordance with the identification signal.

10. The digital VTR as set forth in claim 9, wherein said identification si2:,,nal recording means comprises: recording means for recording, as said identification signal, pilot Sigmals of two different frequencies alternately, every other tracks; and said tracking control means includes comparison means for comparing the levels of the identification signals of the two different frequencies contained in the replay signal, while the head is scanning the position corresponding to the center of the area where the fast replay signal for the particular fast replay speed is recorded.

11. The digitaf VTR as set forth in claim 9, wherein said identification signal recording means comprises: recording means for recording, as said identification signal, pilot signals of two different frequencies alternately, every other tracks; and said recording means records sync block numbers together with the fast replay signal; said tracking control means compares the levels of the identification signals of the two different frequencies contained in the replay signal, when the sync block number of the predefined sync block in the area where the fast replay signal for the particular fast replay speed is recorded, to achieve trackina g control.

12. A digital VTR for recording digital video and audio signals, in In I-D desio-pated areas on oblique tracks of a magnetic recording tape, in a predefined format, using a rotary drum on which heads of two different azimuths are mounted, and replaying from the areas, comprising: data separating means for extracting digital video signal C lt (hereinafter referred to as fast replay signal) used for fast replay, from a normal recording signal; appending means for appending sync byte, ID byte, header byte to the fast replay signal, in the same sync block configuration as said recording signal; recording means for recording the fast replay signal in areas on tracks, such that during fast replay, only one location on one track of an azimuth identical to the head is covered by the head scanning trace; identification signal recording means for recording identification signal for identifying the tracks; and replay means for replaying the identification signal.

13. The digital VTR as set forth in claim 12, further comprising: input means for inputting a password from outside; recording means for recording the password together with the C1 digital video signal; replay means for replaying the password at the time of replay of the digital video sig gnal; and replay inhibiting means for inhibiting display of the digital video signal unless a correct password is input at the time of replay.

126

14. A digital VTR for recording digital video and audio signals. in -no tape, in a designated areas on oblique tracks of a magnetic recordi CID predefined format, using a rotary drum on which heads of two different azimuths are mounted, and replaying from the areas, comprising: data separating means for extracting digital video signal 0 Zn (hereinafter referred to as fast replay signal) used for fast replay, from a normal recording signal; recording means for disposing a fast replay signal for an (M x i)time speed replay (i = 1, 2,..., n), at predefined positions on predefined tracks of consecutive M tracks (M being a natural number), and I 1:1 repeatedly recording the fast replay signal for (M x i)-time speed replay, (2 x i) times; identification signal recording means for recording identification signal for identifying the tracks on which the fast replay signal is recorded; and replay means for performing replay at an arbitrary fast replay speed which is an even-number multiple of the normal speed and is lower than the (M k n)-time speed, using the fast replay signal recorded for (M x n)-time speed replay.

15. A digital VTR for recording digital video and audio signals, in designated areas on oblique tracks of a magnetic recording tape, in a predefined format, using a rotary drum on which heads of two different azimuths are mounted, and replaying from the areas, comprising: sync block forming means for forming sync blocks by appending sync bytes to digital signal recorded in the magnetic recording tape at a predetermined interval; 127 data separating means for extracting a fast replay signal from the normal recording signal; recording means for sequentially and repeatedly recording n pieces of data Di (i = 1, 2,... n, n being a natural number) each of which can be recorded in one sync block, over (n + 2 x w) consecutive sync blocks Sj 0 = 1, 2,... (n + 2 x w)) at identical positions on predefined tracks; wherein n is a maximum number of sync blocks which can always be reproduced from the track regions overlapping with the head scanning traces during m-time speed replay, w is a minimum natural number which is not smaller than the maximum shift from the reference position at which the head crosses a specific track, during mtime speed replay.

I

16. The digital VTR as set forth in claim 15, wherein said recording 1 means repeatedly records the fast replay signal in (n + 2 x w) consecutive sync blocks Sj at an identical sync block position on each track, on at least m consecutive identical-azimuth tracks.

17. A digital VTR for recording digital video and audio signals, in designated areas on oblique tracks of a magnetic recording tape, in a predefined format, using a rotary drum on which heads of two different azimuths are mounted, and replaying from the areas, comprising: sync block forming means for forming sync blocks by appending sync bytes to digital signal recorded in the magnetic recording tape at a predetermined interval; data separating means for extracting a fast replay signal from the normal recording signal; =1 recording means for sequentially and repeatedly recording p pieces of data Di (i = 1, 2,... p p being a natural number not more than n) each of which can be recorded in one sync block, in (p + L + 1) consecutive sync blocks Sj 0 = 1, 2,... (p + L + 1)) at the same position in each track, in at least m tracks of consecutive identical-azimuth tracks in such a manner as to satisfy ek+l =mod [{ek+p-mod(p+L+ 1, p)),pl where ek and ek+ 1 (integers not less than 1 and not more than p) are the suffixes i to the data D first recorded, where n is the maximum number of sync blocks which can always be reproduced consecutively from the region of the track on the tape overlapping with the head scanning trace during m-time speed replay, L is the number of sync blocks which is a minimum integer not smaller than (D - B + C) where C is the difference between the starting positions of the tracks Tk and Tk+1 in the track longitudinal direction, D is the difference between the positions, in the track longitudinal direction, at which the head crosses with the respective tracks, B is the length of the region from which the reproduction from one track is possible consecutively, during m-time speed replay, and mod [a, b] expresses the remainder of a divided by b.

18. A digital VTR for recording digital video and audio signals, in designated areas on oblique tracks of a magnetic recording tape, in a predefined format, using a rotary drum on which heads of two different azimuths are mounted, and replaying from the areas, comprising:

I 129 data separating means for extracting intra-frame encoded image data, from an input bit stream; recording means for forming fast replay signals for a plurality of fast replay speeds from the image data, and recording the nl-time fast replay signal in an area therefor, at positions designated according to the corresponding position on the screen of the signals, with the signals corresponding to the edges of the screen being positioned at the ends of I ZD the recording region on the oblique track, and with the signals corresponding to the position toward the center of the screen being positioned toward the center of the recording region on the oblique track; and replay means for performing fast replay at an n2 time speed (n2 > nl) by replaying the nl-time speed fast replay signal.

19. A video recording apparatus adapted to record in a plurality of C tracks on a data medium auxiliary data specifically for replay at other than nonnal speed, wherein a portion of the auxiliary data is recorded in a location of the auxiliary data recording area on the track according to the part of the picture that the portion represents.

20. A video playback apparatus comprising an error correction means for error correction of replayed data, wherein the error correction means operates in respective modes corresponding to replay at normal speed and replay at other than normal speed, the operational parameters of the error correction means being automatically modified according to the mode.

21. A digital VTR substantially as hereinbefore described with reference to Figures 24to 29 of the accompanying drawings.

22. A digital VTR substantially as hereinbefore described with CI reference to Figure 30 of the accompanying drawings.

23. A digital VTR substantially as hereinbefore described with 1 reference to Figure 31 and Figure 32 of the accompanying drawings.

W

24. A digital VTR substantially as hereinbefore described with reference to Figures 33 to 42. 44 and 45 of the accompanying drawings.

25. A digital VTR substantially as hereinbefore described with reference to Figures 43 to 45 of the accompanying drawings.

1 1 --

26. A digital VTR substantially as hereinbefore described with reference to Figures 46 to 55 of the accompanying drawings.

27. A digital VTR substantially as hereinbefore described with reference to Figure 56 of the accompanying drawings.

28. A digital VTR substantially as hereinbefore described with reference to Figure 57 and Figure 58 of the accompanying drawings.

29. A digital VTR substantially as hereinbefore described with reference to FieUre 59 and FicUre 60 of the accompanying drawings.

C1 13 1

30. A digital VTR substantially as hereinbefore described with reference to Figures 61 to 6-33 of the accompanying drawings.

31. A digital VTR substantially as hereinbefore described with reference to Figures 64 to 67 of the accompanying drawings.

32. A digital VTR substantially as hereinbefore described with reference to Figures 68 to 74 of the accompanying drawings.

33. A digital VTR substantially as hereinbefore described with reference to Figure 75 and Figure 76 of the accompany ing drawings.

34. A digital VTR substantially as hereinbefore described with reference to Figures 77 to 87 of the accompanying drawings.

35. A digital VTR substantially as hereinbefore described with reference to Fiegures 88 to 92 of the accompanying drawings.

132