RU2229174C2 - Mpeg format data signal recording on data medium and data reproduction - Google Patents

Abstract

FIELD: channel signal recording devices; time-critical data transfer means using asynchronous channel for their transfer. SUBSTANCE: device has means affording time-critical data in the form of stream of sequential time-critical transmission blocks. In addition it has means for discriminating stream recovery information out of time-critical data. This information includes at least one time information item for transmission blocks. Timecritical data stream can be recovered from mentioned stream recovery information. Device also incorporates means for tagging at least some of transmission blocks with tagged time information items over asynchronous channel. The latter has digital recorder. Time-critical data can be recorded in Moving Pictures Expert Group (MPEG) format where transmission blocks are given in the form of transport bursts. EFFECT: provision for recording data signals of various types in desired format. 14 cl, 27 dwg

Description

This application is a continuation of application No. 08 / 225.193 dated April 8, 1994 for "Recording information signals in MPEG format on a storage medium and reproducing them" by W.J. Van Gestel, R.W.J.J. Saeijs u I.A. Shah.

BACKGROUND OF THE INVENTION

The present invention relates to a recording device for recording an information signal on tracks of a storage medium containing

- an input device for receiving an information signal,

- a channel encoder for encoding an information signal in a channel to obtain a channel signal suitable for recording on a track of said information medium,

- a recording device for recording a channel signal onto a track, wherein the channel signal includes one subsequent or sequence of signal blocks, each signal block contains a first part of a block containing a synchronization signal, and a second part of a block containing a certain number of channel bytes for recording onto a medium using the device recording and for reproducing the information signal from the storage medium using the playback device.

The aforementioned recording device is known from EP-A 492704 - document [1] in the attached list of references.

This known device is a screw scanning recorder, such as that used in a digital video cassette recorder (CCM), designed to record an information signal including a digital audio signal and a digital video signal in the audio recording sectors and video recording sectors, respectively, in sequential tracks where, when recording a track the video recording sector is the first, followed by the audio recording sector. However, the sectors on the track can also be in the reverse order. In addition, other sectors can be recorded on the track, such as the synchronization area located at the beginning of the track so as to synchronize the internal system timer with the signals read from the track, as well as areas, preambles and conclusions that are placed between different sectors and function as a separator.

References here can be made to previously filed European applications No. 93202950 and No. 93201263 - respectively [2] and [3] in the attached list of references.

These well-known sources refer to proposals for implementing a new standard for digital recording on video tapes, which makes it possible to record and play digital video and audio signals onto / from magnetic media with a longitudinal method of recording information, such as magnetic tape. This new standard for digital video recording will lead to new digital recording / playback devices of the so-called DVC (digital video tape recording) type.

SUMMARY OF THE INVENTION

An object of the present invention is to provide a recording device that is capable of recording other types of information signals in a format of a known type defined in the preamble. A recording device in accordance with this invention is characterized in that the information signal is a signal of MPEG standard (moving image compression standard); this information signal of the MPEG standard contains one subsequent or sequence of transmission packets (transport packets), the channel encoding means are adapted to store each time information included in x MPEG transmission packets of the information signal in the second part of the group block of the channel signal signal blocks, this second part block at least the first signal block includes a third part of the block for storing identification information identifying a temporary signal when there is a vy signal block of the group of y signal blocks, and that x and y are integers, and wherein x≥1 y≥1. More specifically, the recording device of the above type according to the invention is characterized in that this information signal is an information signal in MPEG format, this MPEG information signal contains serial transmission packets (transport packets), channel encoding means are adapted to store each time information included in x MPEG transmission packets of the information signal in the second part of the group of blocks of the channel signal blocks, these second parts of the signal blocks include the third part of the block eye for storing information about the sequence number of the signal blocks, and x and y are integers, x≥1 and y≥1.

The invention is based on the following premises. The Grand Alliance HDTV System Specification project of February 22, 1994, document [4] in the attached list of references, in particular sections V and VI, includes a description of a transmission system designed to transmit MPEG information signal, which includes data that is compressed digital video signal, and corresponding data that is compressed digital audio for transmission over a cable network. MPEG information signal is presented in the form of transmission packets (transport packets) having either the same or variable length in time. In both cases, however, the transmission packet includes 199 bytes of information, whose first byte is a synchronization byte.

The transmission of such an MPEG information signal in a form suitable for recording and reproducing from an information medium, such as a magnetic information medium, such as magnetic tape, requires special means to implement this method of transmission using a known magnetic tape format. More specifically, the invention relates to the storage of transmission packets in signal units of a known magnetic tape format.

In the general case, it can be said that when storing information in x MPEG transmission packets of the information signal in the signal blocks, some unallocated space in the y signal blocks remains available for storing additional information; this additional information relates to special cases of recording and playing MPEG of an information signal to / from an information carrier. In a specific example of DVC format, second block parts are available for five signal blocks. Then 11 bytes (= 5 × 77-2 × 187) remain free in five signal blocks. These 11 bytes can be divided into second parts of blocks of five signal blocks in various ways so as to obtain third parts of blocks. One such way is that the first two bytes of all the second parts of the blocks are provided as the third parts of the blocks, and the last byte provided can be considered as the third part of the block to indicate the boundary between the information of the two transmission packets, which is stored in the five signal blocks.

In the above example, identification information identifying the signal unit as the first signal unit from the group of y signal units can be stored in the third part of the first signal unit in the group of y signal units. Or, a sequence number relating to a sequence of signal blocks may be stored in the third parts of the blocks. This sequence number can also be identified as a continuity counter. This embodiment provides several advantages.

The advantage of using identification information identifying the signal block as the first signal block in the group of y blocks is that the beginning of the group can be determined, which simplifies reading the data during playback.

One of the advantages of sequence numbering is that by reproducing the signal blocks, it is possible to determine from the sequence numbers whether the signal block was skipped due to a playback error or not, so error correction or error concealment is necessary.

Another advantage is that information in the signal blocks can be moved during recording. When extracting sequence numbers, you can carry out their corresponding reverse movement to obtain the original data stream.

In addition, having the sequence numbers included in the third parts of the signal blocks, it is possible to repeat the signal blocks if the MPEG data transmission packet stored in these signal blocks requires a high degree of protection against errors that may appear during recording and subsequently reproduction process.

The recording device of the above type may also differ in that the information signal is an MPEG information signal in accordance with the MPEG format, this MPEG information signal contains serial transmission packets, channel encoding means are used to store each time information included in x MPEG information signal transmission packets in the second parts of the blocks of the first group of the first signal blocks of the above signal blocks of the channel signal, so as to make it possible normal playback mode using the video information stored in the first group from the first signal blocks during the normal playback mode, channel encoding means are also designed to extract the video signal of the complex encoding mode from the MPEG information signal and to memorize the specified video signal of complex or special mode in the second parts of the blocks of the second group of z second signal blocks of the above signal blocks of the channel signal so as to make possible a complex playback mode using video information stored in said second signal blocks, the second signal blocks of at least one signal block in the first and second group of the first and second signal blocks respectively comprise a third part of the block for storing identification information indicative whether the group is the first signal blocks or the second signal blocks, where x, y and z are integers, x≥1, y> 1 and z> 1.

More specifically, the information signal is MPEG information signal in MPEG format, MPEG information signal contains serial transmission packets, channel encoding means is used to store each time information included in x MPEG information signal transmission packets, in the second parts of blocks in the group of signal blocks channel signal, the second parts of at least those signal blocks in the group of signal blocks that contain the initial part of the transmission packet contain the third hour block for storing sequence number information related to the sequence number of the transmission packet having its initial part stored in the second part of the signal block, where x and y are integers, x≥1, y> 1. This enables a normal playback mode in the playback device using the first signal blocks and playback in the complex playback mode using the second signal blocks in response to the detection of information indicating groups containing the first signal blocks or second signal blocks, respectively.

A recording device of the above type may also be characterized in that the second parts of all signal blocks in the first and second groups of the first and second signal blocks respectively comprise a third part of blocks for storing identification information indicating whether the group contains the first signal blocks or second signal blocks. More specifically, the second parts of the blocks in the group of signal blocks each contain a third part of blocks for storing the sequence number of the transmission packet corresponding to the transmission packet, information about which is stored in the indicated signal block.

Storing the packet sequence number has its advantages if the received MPEG data stream has a constant bit rate and contains several different video programs that are embedded interleaved into the MPEG data stream. In general, such data streams have too high a bit rate to write the entire data stream to the storage medium. A typical bit rate in MPEG format is 45 Mb / s (megabits / second), while the bit rate of writing to a storage medium is typically 25 Mb / s. The recording device now includes a program selector for selecting one video program and the corresponding audio signal from the MPEG standard data stream to obtain an MPEG information signal for recording. Since the information corresponding to only one video program is included in the MPEG transmission packet (transport packet), such a program selector selects only those transmission packets from the MPEG data stream that include information corresponding to only one specified video program. This means that some packets of the resulting source MPEG data stream are deleted. When playing, however, MPEG video signal in accordance with the MPEG standard, although it includes only one video program, must be restored or recreated. Such a regenerated data stream should contain transmission packets that were selected during recording and located in the same order, i.e. empty packets corresponding to packets removed during recording, in one way or another, should be inserted into the regenerated data stream. During recording, a sequence number is added to each received transmission packet, including for packets to be deleted. The sequence numbers of the packets that are selected and stored are stored in the third part of the signal blocks in which the transmission packets are stored. During playback, sequence numbers are restored, and it is not necessary that subsequent numbers are next higher numbers. In this situation, one or more empty transmission packets must be inserted in order to restore the original MPEG data stream.

The recording device, as defined above, can also be characterized in that the information signal is an MPEG information signal in accordance with the MPEG format, the MPEG information signal contains sequential transmission packets, the recording device contains a detecting device for determining when transmission packets are received and for generating time information for each received transmission packet, while the temporary information for the transmission packet corresponds to the aforementioned moment of reception of the above transmission packet, channel encoding means is intended for storing information included in x MPEG transmission packets of the information signal in the second parts of the group of blocks of the signal blocks of the channel signal, while the second parts of the blocks of at least those signal blocks in the group of the signal blocks that contain the initial part the transmission packet, include the third part of the block for storing time information for the specified transmission packet having an initial region stored in the second part of the signal block, and x and y are tsya whole, x≥1, y> 1. More specifically, the second parts of the blocks of the group of signal blocks each comprise a third part of blocks for storing temporal information corresponding to a transmission packet that has information stored in a second part of said signal block. Remembering the temporary information corresponding to the transmission packet requires that the recording device be equipped with a detection device for determining the time of reception of the transmission packet. This tool has its advantages if an MPEG data stream is received, which has a variable bit rate and contains a set of different video programs included interleaved in the MPEG data stream. As mentioned above, in the general case, such a data stream has a too high bit rate for writing the entire data stream to a storage medium. The recorder in this case contains a program selector for selecting one video program with its corresponding audio signal from the MPEG data stream to obtain an MPEG information signal for recording. If the information corresponding to only one video program is included in the MPEG transmission packet, such a program selector selects only those transmission packets from the MPEG data stream that contain information corresponding to the specified one video program. By determining and storing the temporal information corresponding to the transmission packet, the recording device should be capable of reproducing temporal information and recovering the MPEG information signal using this temporal information.

It should be noted that the means discussed above can be used in the recording device separately or in combination with each other. As a result, an information carrier having signal blocks stored on tracks of the information carrier will be obtained, the signal blocks having a first part of a block that contains a synchronization signal and a second part of a block that contains many bytes of channel data, with x MPEG information signal transmission packets stored in the second parts of the blocks in the group of the signal blocks of the channel signal. In addition, in accordance with the present invention

- the second part of the block of at least the first signal block in the group of y signal blocks contains a third part of the block for storing identification information identifying the signal block as the first signal block in the group of y signal blocks, or

- the specified identification information is information about the sequence number, and the second parts of the blocks in the group of the signal blocks all contain a third part of the blocks for storing information about the sequence number related to the sequence numbers of the signal blocks, or

- the second parts of the signal blocks each contain a third part of the blocks for storing identification information indicating whether the signal block contains data for a normal playback mode or data of a "complex" mode, or

- the second parts of the blocks of at least those signal blocks in the group of signal blocks that contain the initial part of the transmission packet, contain the third part of the blocks for storing identification information related to the sequence number of the transmission packet corresponding to the transmission packet having its initial part stored in the second part of the signal unit, or

- the second parts of the blocks of at least those signal blocks in the group of signal blocks that contain the initial part of the transmission packet, contain the third part of the blocks for storing temporary information for the specified transmission packet having its initial part stored in the second part of the signal block, or

- the third parts of the blocks containing information that is a combination of one or more of the above.

It should be obvious that the playback device must be adapted to each specific implementation of the recording device, so that you can restore MPEG information signal recorded on the storage medium. Such a reproducing apparatus is characterized in the claims relating to the reproducing apparatus.

The present invention also includes a method for transmitting time critical data via an asynchronous channel without changing any time critical information; in other words, allows you to make the asynchronous channel transparent for the transmission of time-critical data. This is achieved, generally speaking, by tagging (i.e., accompanying the transmitted data with tags (tags)) of one or more units of the transmitted data with temporary information before sending them to the channel, and then using the tagged information to recreate the correct data at the other end of the channel. By way of example, such data may be an MPEG data signal, and a channel may be a DVCR. But the present invention is not limited only to this application and can be applied to asynchronous channels, such as a computer or telephone network or digital interface.

Brief Description of the Drawings

These and other aspects of the invention will be apparent from the following description of exemplary embodiments with reference to the drawings, in which the following is presented:

1 is a track format of a DVC-type information carrier.

Figure 2 is a schematic representation of the contents recorded on the track in accordance with Figure 1 of the video sector.

Figure 3 is a schematic representation of a serial MPEG data stream and the format of transmission packets included in this serial MPEG data stream.

4 is an example of storing two transmission packets in five signal blocks.

Figure 5 - track contents when recording MPEG information on it.

6 is an example implementation of a recording device.

7 is an example implementation of a playback device.

Fig. 8a is an example of an initial MPEG data stream having a constant bit and packet rate, Fig. 8b shows a recorded MPEG data stream, and Fig. 8c shows a restored original MPEG data stream signal.

Fig.9 is an example implementation of a processing device for the "normal playback" mode in the recording device of Fig.6.

Figure 10 is an example of a sequence of three groups of five signal blocks in each.

11 is another example of a sequence of three groups of five signal blocks in each.

12 is an example embodiment of a processing device for the “normal playback” mode in the playback device of FIG. 7.

Fig. 13a is an example of an initial MPEG data stream having a variable bit and packet rate, Fig. 13b shows a recorded MPEG data stream, and Fig. 13c shows a regenerated initial MPEG data stream.

Fig. 14 is another example of a processing device for the "normal playback" mode in the recorder of Fig. 6.

FIG. 15 is another example of a processing device for “normal playback” mode in the playback device of FIG. 7.

Fig. 16 illustrates a storage medium and a read head scanning a storage medium in a "complex" playback mode.

17 is a sequence of signal blocks on a track forming an area of “complex” playback on a track.

Fig. 18 is another exemplary embodiment of the invention combining recording and reproduction systems.

Fig. 19 is an example of input and output data streams for the device of Fig. 18.

FIG. 20 is another example of input and output data streams for a device similar to that shown in FIG.

Fig - a possible form of a data block in accordance with the invention for transmission through an asynchronous channel.

Fig. 22 is a more detailed representation of the body of the data block shown in Fig. 21.

Fig - possible tagging in accordance with the invention.

Detailed Description of Preferred Embodiments

Figure 1 shows the format of the signals when they are recorded on a track of a magnetic information carrier by means of a DVC type screw-type video recorder. The left edge of track 1 in FIG. 1 is the beginning of the track, and the right edge is the end of the track. A track contains several parts of a track. The portion of the track designated G1 is the initial portion of the track. An example of the initial part G1 of the track is described in detail in [1].

After part G1 of the track, the tuning part TP1 of the track is indicated, designated 1T1 (insert temporary information) and contains the track setting, synchronizing information and identification (or temporary) information. A further explanation of the contents of the 1T1 part of the track can be found in [3].

After part TP1 of the track, a separator G2 follows. The splitter G2 is followed by the portion TP2 of the track, which is the audio recording sector and contains digital audio information. The splitter G3 is followed by a portion of the TP3 track, which is a video recording sector and contains digital video information. The separator G4 is followed by the part TP4 of the track, marked INDEX and which contains, among other information, additional codes, for example absolute and / or relative temporary information and a table of contents (TOC). The track ends with part of the G5 track. It must be said that the order in which parts TP1, TP2 and TP3 follow on the track may be different.

The content of the video signal recorded in sector TP3 is shown in FIG. 2. Figure 2 schematically depicts a set of 149 horizontal lines, denoted j = 1 to j = 149 and storing bytes of information. 149 lines are actually 149 signal blocks (or synchronization blocks) that are stored sequentially in the recording sector TP3 of the video signal. Each signal block stores 90 bytes of information, numbered from i = 1 to i = 90.

The first two bytes (i = 1 and i = 2) of each signal block form a synchronization code 2 bytes long. The next three bytes in each signal block form an identification code ID that contains, among other information, one that indicates the sequence number of the signal block in the recording part TP3 of the video signal. The last eight bytes in the signal blocks constitute information for horizontal parity. Information for vertical parity is stored in an array with numbers i = 6 to i = 82 inclusive in the last 11 signal blocks.

The bytes of the video signal information are stored in an array located in the signal blocks with numbers i = 6 through i = 82 inclusive, having sequential numbers c j = 3 through j = 137 inclusive. Bytes of additional data are stored in an array located at numbers i = 6 through i = 82 inclusive in the signal blocks having sequential numbers j = 1, 2 and j = 138. The signal blocks are stored sequentially in the TP3 part of the video signal path starting with the signal block indicated by j = 1, the next signal block indicated by j = 2, and so on until the signal block indicated by j = 149.

Additional data to be stored in the signal blocks indicated by j = 1, 2, and 138 may be teletext or control data.

It should be noted here that it can be determined that additional data should be recorded in a different place from the specified. In this regard, it is recommended to refer to the document [1], Fig.13, where additional data is located in the part of the memory designated 111.

Figure 3 schematically shows the MPEG data stream arriving at the recording device in accordance with the present invention. The MPEG data stream contains serial transmission packets designated ..., P _k-1 , P _k , P _{k + 1} , .... Each such packet contains a 4-byte PH header and a 184-byte packet body. Transmission packets may be transmitted in a data stream having a constant bit rate. This means that packets are of equal length, visible in time and received at a fixed packet rate. Transmission packets may also be transmitted in a data stream having a variable bit rate. In this situation, it is not necessary that the packets are of the same length, are visible in time, and they can be received with a variable packet rate. The first byte in such a packet, entitled PH, is a sync byte. The synchronization byte is identical for all transmission packets (transport packets). The remaining three bytes in the header contain ID information, such as a packet identifier. For further consideration, it is recommended to refer to [4] in the list of references, and more precisely, to part V, paragraph 5.1 on p.27.

The body of each transmission packet contains 184 bytes for storing video and audio information, which must be transmitted in accordance with the MPEG format. The body of one transmission packet may store either audio information corresponding to a certain video signal or a video signal. In addition, in the event that several programs are transmitted in the MPE & data stream, the body of the transmission packet stores a video signal corresponding to one of such transmitted video programs.

One aspect of the invention now consists in recording a video signal and its corresponding audio signal, corresponding to one of these video programs transmitted in an MPEG data stream, to an information medium having a track format shown in FIGS. 1 and 2. The information stored in the transmission packets should be stored in signal blocks, for example, 135 signal blocks, designated from j = 3 to j = 137 in the portion of track TP3 intended for recording a video signal. Two synchronization bytes, marked i = 1 and i = 2, identification (ID) information in the form of three ID bytes, marked i = 3, 4 and 5, as well as 8 bytes of horizontal parity, marked i = 83 to 90 in These signal blocks are required for proper recording and playback. As a result, only 77 bytes, designated i = 6 to 82, in the signal blocks designated j = 3 to 137, are provided for storing transmission packets with MPEG information. The part of the signal blocks formed by 77 bytes i = 6 to 82 is defined as the second parts of the signal blocks.

Since synchronization during recording and playback is carried out by means of synchronization words in each signal unit, there is no need to transmit synchronization bytes of the transmission packet by the information carrier. Thus, before storing the information contained in the transmission packets in the second parts of the blocks designated from j = 3 to j = 135, the synchronization bytes of all transmission packets are discarded. As a result, only 187 bytes of information should be recorded in the signal blocks for each transmission packet.

Simple calculations show that two transmission packets can be stored in five signal blocks and 11 bytes remain free to record other information. Figure 4 shows how two transmission packets can be stored in the second parts of the blocks in a group of 5 signal blocks, denoted SB1 ... SB5 in Figure 4. Figure 4 shows only the contents of the second parts of blocks of 77 bytes in length included in the signal blocks. As can be seen from Figure 4, 11 bytes are allocated in a group of five signal blocks so that every second block part contains the third part of the block from TB3.1 to TB3.5 two bytes long at the beginning of the second block sections of five signal blocks from SB1 to SB5, respectively, and the third part of the signal block FB is in the third signal block SB3. 187 bytes of the first transmission packet are stored in signal blocks SB1, SB2 and SB3, with the three ID bytes of the header packet of the first transmission packet, indicated by TH1, being stored first in the signal block SB1, immediately after the third part of block TB3.1, and the subsequent 72 first body bytes the first transmission packet is stored in the second part of the signal block SB1. The next 75 bytes of the body of the first packet are stored in the second part of the signal block SB2, after the third part of the block TB3.2, and the last 37 bytes of the body of the first packet of transmission are stored in the second part of the signal block SB3, after the third part of the block TB3.3.

Then follows the FB byte, which indicates the boundary between the information of the first and second transmission packets stored in a group of five signal blocks. 187 bytes of the second transmission packet are stored in the signal blocks SB3, SB4 and SB5, with the three ID bytes of the header packet packet of the second transmission packet, marked TH2, being stored first in the signal block SB3, immediately after the byte of BB. The next 34 first bytes of the body of the second transmission packet are stored further in the second part of the signal block SB3. The next 75 bytes of the body of the second transmission packet are stored in the second part of the signal block SB4, after the third part of the block TB3.4 and the last 75 bytes of the body of the second transmission packet are stored in the second part of the signal block SB5, after the third part of the block TB3.5.

It should be noted that another distribution of 11 bytes across five signal blocks is also possible. For example, 11 bytes can be located in two third parts of blocks; the first third part of the block has, for example, 6 bytes and is located at the beginning of the first signal block SB1, and the other third part of the block of 5 bytes in length is located in the third signal block and defines the boundary between two transmission packets stored in five signal blocks. Another example may be one third of the block located at the beginning of the signal blocks SB1 and SB3, and another third part of the block in the third signal block SB3, indicating the boundary between two transmission packets stored in five signal blocks, and the third part of the signal block SB1 can have, for example, 4 bytes, the third part of the signal block SB3, for example, 3 bytes, and the third part of the signal block SB3, indicating the mentioned boundary, for example, 4 bytes.

The third parts of blocks TV3.1 to TV3.5 can be used to store additional information. First example: the third block part TB3.1 may include an indication identifying the signal block SB1, which is the first signal block in a group of five signal blocks. This can be realized by storing a special position bit in the third block part of TB3.1 - a bit having the value "0" or "1". In the same bit, located in the third block parts of TB3.2 to TB3.5, its opposite value is stored. In another example, information about the sequence number, for example, sequence numbers from 1 to 5, can be stored in the third block parts of TB3.1 to TB3.5, respectively, in a group of five signal blocks, where the third block part of TB3.1 contains it sequential number "1", and the third block part of TV3.5 - respectively, filled in her serial number "5". To memorize consecutive numbers in the third block parts of TB3.1 to TB3.5, three position bits are required. Sequential numbers can, however, cross the boundaries of groups so as to identify large sequences of signal blocks, for example within a single track or even more than one track.

In another example, a special position bit in the third block parts of TB3.1 to TB3.5 in a group of five signal blocks can be set to “0” or “1” and used to indicate that the video data included in the signal block has so-called data of the "normal" playback mode, and the opposite value of this bit can be used to indicate that the video data included in the signal unit has so-called data of the "complex" playback mode. The use of these modes of "normal" playback and "complex" playback will be explained below.

In another example, consecutive numbers are generated according to received transmission packets in the MPEG data stream. As previously explained, such an MPEG data stream may include more than one video program. Since the bit rate of the MPEG data stream is usually higher than the bit rate of the signal that can be recorded, only one video program can be selected from the sequential MPEG data stream for recording. The selection of one video program means the selection of transmission packets from the MPEG data stream, which contain information related to the specified video program, and the removal of the remaining packets. Therefore, the sequential array of transmission packets to be recorded has sequence numbers, which are optionally next in order, since the sequence numbers of the remote transmission packets are absent. If the sequence numbers in the third block parts are stored, these numbers can be restored during playback. By checking successively received sequence numbers, it can be determined whether the original MPEG data stream received by the recorder contained remote transmission packets between two reproducible transmission packets. If so, then the original MPEG data stream can be restored by inserting one or more empty packets between two playable packets.

In one example, temporal information is stored in the third block parts for the same purposes as above, namely, to restore the original MPEG data stream, in the case when this data stream is a data stream having a variable bit rate.

It should be clear that also a combination of the additional information described above can be included in 11 bytes available for storing such information in a group of five signal blocks.

As an example, we explain the above that a 3-bit word is necessary in the third block parts to indicate the sequence numbers of signal blocks in a group of five signal blocks. In particular, the 3-bit words "000", "001", "010", "011" and "100" can be used to identify the sequence. This means that the 3-bit words "101", "110" and "111" remain free for future use. For example, the 3-bit words “101”, “110” can be used to identify “normal” playback data or “complex” playback data.

Figure 5 shows the format of the track, if MPEG information is stored in the second part of the signal block part of the track TP3 (Figure 1), indicated here TP3. 5 shows the first two signal blocks (j = 1, 2) in the portion of track TP3, which also includes auxiliary data, followed by 135 signal blocks (j = 3 to j = 137), now containing MPEG information and additional information, described above. The next signal block (j = 138), also containing auxiliary data, is followed by 11 signal blocks containing parity information. Storing MPEG information and additional information in 135 signal blocks may require additional coding operations from errors that must be performed on the specified information, leading to the appearance of additional information about the parity, which should also be stored on the track. When MPEG information, which includes video information and corresponding audio information, is stored in 135 signal blocks in a portion of track TP3 ’, it is not necessary to store audio information in a portion of track TP2 shown in FIG. This part, designated TP2 ’in FIG. 5, can be used to store the parity information obtained from additional error coding operations.

Figure 6 schematically shows one possible embodiment of a recording device. This recorder comprises an input 11 for receiving an MPEG serial data stream for recording transmission packets included in the data stream in the signal blocks on parts of TP3 ’tracks. Input 11 is connected to input 12 of processing device 14 for “normal” playback mode. In addition, a processing device 16 for the "complex" playback mode may be used, having an input 17 also connected to the input 11. The outputs 19 and 20, respectively, of the processing device 14 for the "normal" playback mode and the processing device 16 for the "difficult" playback mode (if it is present) connected to the corresponding inputs of the multiplexer 22. It is clear that in the absence of the processing device 16 for the mode of "complex" playback, the multiplexer 22 should also be absent.

The auxiliary generator 24 is designed to supply a signal of auxiliary information for storing in the signal blocks indicated by j = 1, 2 and 138 (Figure 2). The outputs of the multiplexer 22 and the generator 24 are connected to the corresponding inputs of the error correction device 26. This error correction device 26 provides the first coding operation from the error indicated by ECC3 and the second coding operation from the error indicated by ECC2. The next, third error coding operation, designated ECC1, is performed by the error correction coding device 28.

The playback device includes a generator 30 for adding ID information to bytes i = 3, 4 and 5 of the signal blocks (FIG. 2) for adding index information for storing in part of the track TP4 (FIG. 5) and separating information for forming separators G1 ... G5 (Figure 5). After combining these signals in device 32, the combined signal is sent to device 34, where encoding is performed, in which 24-bit words of the incoming bitstream are converted into 25-bit words, and a synchronization word is added so as to obtain the first two bytes (i = 1, 2) in the signal blocks, and ITI information is added for storing in part of the track TP1.

The conversion of 24-bit words to 25-bit words by encoding device 34 is well known in the art. Information on this is contained, for example, in the source [5] from the attached list of references, which also describes the addition of synchronizing words to the data stream.

The output of the encoding device 34 is connected to the input of the recording device 36, in which the data stream coming from the encoding device 34 is recorded on the oblique tracks of the information medium by means of at least one recording head 42.

The first error coding operation, designated ECC3, is required in order to implement additional error protection of MPEG information, which must be recorded on the storage medium, and obtain parity information that must be stored in part of track TP2, as explained earlier. The second error coding operation, designated ECC2, results in vertical parity information that must be stored in 11 signal blocks (j = 139 ... 149) of the TP3 ’part of the track (FIGS. 2 and 5). The third error coding operation, designated ECC1, results in horizontal parity information that should be stored in the last 8 bytes of the signal blocks on a portion of the TP3 ’track (FIGS. 2 and 5).

Before further describing the processing device 14 for the normal playback mode and the processing device 16 for the complex playback mode in the recording device of FIG. 6, a description will first be made of the reproducing device. This has the advantage that, with a further description of the means used in the processing devices 14 and 16, it is possible to establish a direct connection with the advantages of these means for reproduction.

FIG. 7 schematically shows an exemplary embodiment of a reproducing apparatus for reproducing from an information medium 40 information obtained by the recording apparatus shown in FIG. 6. The reproducing device comprises a reader 50 having at least one read head 52 for reading information from the oblique tracks of the information carrier 40. The output of the reader 50 is connected to the input of the decoding device 54, which performs decoding of the “25 into 24” read signal, i.e. . converts 25-bit words of the input data stream to 24-bit words. Further, after deleting in the device 56 all the information that is not required to restore the original MPEG data stream, error correction is performed in the device 58. It is clear that error correction is performed in three steps. The first correction operation is based on the ECC1 operation using horizontal parity information (Figure 2), the second correction operation is based on the ECC2 operation using vertical parity information, and the third correction operation is based on the ECC3 operation using parity information stored on part of track TP2 '(Figure 5).

The output of the error correction device 58 is connected to the input of the processing device for the "normal" playback mode 60. Additionally, a processing device 62 for the "complex" playback mode may also be present, having an input also connected to the output of the error correction device 58. The outputs 64 and 65 of the processing device for the “normal” playback mode 60 and the processing device 62 for the “complex” playback mode (if present) are connected to the corresponding inputs a and b by a switch 66, the output of which is connected the exit 68. It is clear that in the absence of the switch 66 processing apparatus 62 will be absent. If the playback device switches to the "normal" playback mode, this means that the storage medium moves at a nominal speed. The “normal” playback processing device 60 is activated and the switch 66 is set to ac. If the playback device is switched to the mode of "complex" (or "special") playback, this means that the storage medium moves at a different speed than the nominal speed, the processing device 62 for the mode of "complex" playback is activated and the switch 66 is set to b-c.

Further discussion will be given for devices 14 and 16 (FIG. 6) in conjunction with processing devices 60 and 62 (FIG. 7).

It is assumed that the recording device is able to select one video program and its corresponding audio signal from the serial MPEG data stream, which is input 11, in accordance with the selection signal received by the user from the device. As mentioned earlier, only those transmission packets in the serial MPEG data stream should be selected that include information related to the selected video program. Fig. 8a shows a serial MPEG data stream as a function of time containing transmission packets indicated by P _k . It should be noted that the MPEG transmission packets of the data stream do not contain a packet number. The packet number indicated in FIG. 8a is therefore the number to be generated by the packet number generator 86 in FIG. 9, which will be discussed later.

The selection of only those packages from the packages P _k in Fig. 8a that include information related to the selected video program means, for example, that the packages P _k-4 , P _k-1 , P _k , P _{k + 2} , P _{k + 4} , P _{k + 8} must be selected, and that intermediate packets must be removed. The resulting data stream obtained in the recording device for recording on the storage medium is shown in Fig. 8b, which shows the data stream as a function of time. No conclusion should be drawn either about the location of the time information in the time scale in Figs. 8a and 8b, nor from the relative position of the packets in Figs. 8a and 8b. This is because, as mentioned above, the bit rate of the initial MPEG data stream (Fig. 8a) differs (is greater) from the bit rate at which the selected transmission packets must be recorded on the storage medium.

An exemplary embodiment of the processing device 14 for the “normal” playback mode for recording the data stream shown in FIG. 8b is shown schematically in FIG. 9. The device 14 ’in FIG. 9 comprises a selector (selector) 76 having an input connected to the input 11 of the device 14’. The selection device 76 has another input 78 for receiving a selected signal supplied by the user. The output 79 of the selection device 78 is connected to the input of the synchronization data deletion device 80, whose output is connected to the signal combining device 82. Next, the output 81 of the selector 76 is connected to the control input 83 of the combining device 82, for supplying a control signal to the combining device 82.

The input 11 is also connected to the input of the packet detector 84, which has an output connected to the input of the packet number generator 86. The output of the packet number generator 86 is connected to the second input 83 of the combining device 82.

The selector 76 selects transmission packets from P _k-4 , P _k-1 , P _k , P _{k + 2} , P _{k + 4} , P _{k + 8} from the sequential MPEG data stream supplied to input 11 in accordance with the selection signal received at input 78. The selected packets are sent to the synchronization data deletion device 80, in which the first synchronization byte in the packet, called PH (Figure 3), is removed from the packets in accordance with the above description. A packet detector 84 detects the reception of each packet in the original MPEG data stream supplied to input 11 and generates a time pulse for each detected packet. The generator 86 contains a counter, which increases its value under the influence of pulses from the timer supplied to the generator 86. As a result, the next large counter value is supplied to the output for each received clock pulse. At the output 87 of the generator 86, thus, the numbers k-4, k-3, k-2, k-1, k, k + 1, k + 2, k + 3, k + 8, ... appear. Under the influence of the control signal supplied to the control input 83 of the combining device 82, the control device combines the packages P _k-4 , P _k-1 , P _k , P _{k + 2} , P _{k + 4} , P _{k + 8} , selected by the selector 76, so as well as the values of the counter k-4, k-1, k, k + 2, k + 4, k + 8 from the stream of counter numbers supplied by the packet number generator 86 for storing in the signal blocks.

Figure 10 shows an example of how transmission packets and corresponding packet numbers can be stored in a group of five signal blocks. Figure 10 shows three consecutive groups of five signal blocks, designated G1, G2 and G3, in which information is stored. In the third block part TB3.1 of the first signal block SB1 in group G1, a packet number k-4 is stored, and the information contained in packet P _k-4 is then stored in the signal blocks SB1, SB2 and SB3 in group G1. In the third block part FB of the third signal block SB3 in the group G1, the packet number k-1 is stored, and the information contained in the packet P _k-1 is then stored in the signal blocks SB3, SB4 and SB5 in the group G1. In the third block part TB3.1 of the first signal block SB1 in the group G2, the packet number k is stored, and the information contained in the packet P _k is then stored in the signal blocks SB1, SB2 and SB3 in the group G2. In the third block part FB of the third signal block SB3 in the group G2, the packet number k + 2 is stored, and the information contained in the packet P _{k + 2} is then stored in the signal blocks SB3, SB4 and SB5 in the group G2. In the third block part TB3.1 of the first signal unit SB1 in the group G3, the packet number k + 4 is stored, and the information contained in the packet P _{k + 4} is then stored in the signal blocks SB1, SB2 and SB3 in the group G3. In the third block part FB of the third signal block SB3 in the group G3, the packet number k + 8 is stored, and the information contained in the packet P _{k + 8} is then stored in the signal blocks SB3, SB4 and SB5 in the group G3. As long as the number of bits in the packet number is less than or equal to 8, this number will fit into the third block part of FВ, whose length is 1 byte.

Another example of storing the packet number in the third block part is shown in Fig.11. In the third block parts TB3.1, TB3.2 and TB3.3 of the signal blocks SB1, SB2 and SB3, respectively, the packet number k-4 is stored in the group G1, and the information contained in the packet P _k-4 is stored in the signal blocks SB1, SB2 and SB3 in group G1, as explained previously in accordance with FIG. 4. In the third block part FB of the third signal block SB3 in group 61, as well as in the third block parts TB3.4 and TB3.5 of signal blocks SB4 and SB5 in group G1, the packet number k-1 is stored, and the information contained in the packet P _k-1 is then stored in the signal blocks SB3, SB4 and SB5 in the group G1, as explained previously in accordance with FIG. 4.

In the third block parts TB3.1, TB3.2 and TB3.3 of the signal block SB1, SB2 and SB3, respectively, the packet number _k is stored in the group G2, and the information contained in the packet P _k is stored in the signal blocks SB1, SB2 and SB3 in group G2. In the third block part FB of the third signal block SB3 in group G2, as well as in the third block parts TB3.4 and TB3.5, the signals of blocks SB4 and SB5, respectively, in group G2, the packet number k + 2 is stored, and the information contained in the packet P _{k + 2} is stored in the signal blocks SB3, SB4 and SB5 in group G2. In the third block parts TB3.1, TB3.2 and TB3.3 of the signal blocks SB1, SB2 and SB3 in the group G3, the packet number k + 4 is stored, and the information contained in the packet P _{k + 4} is stored in the signal blocks SB1, SB2 and SB3 in group G3. In the third block part FB of the third signal block SB3 in the group G3, as well as in the third block parts TB3.4 and TB3.5 of the signal blocks SB4 and SB5 in the group G3, the packet number k + 8 is stored, and the information contained in the packet P _{k + 8} is stored in the signal blocks SB3, SB4 and SB5 in group G3.

Instead of storing the packet number k-4 in the third block portion TB3.3 of the signal block SB3 in group G1, the packet number k-1 in the indicated third block portion can be stored. Instead of storing the packet number k in the third block part TB3.3 of the signal block SB3 in group G2, the packet number k + 2 in the indicated third block part can be stored. Instead of storing the packet number k + 4 in the third block portion TB3.3 of the signal block SB3 in group G3, the packet number k + 8 in the indicated third block portion can be stored.

Fig. 12 shows an exemplary embodiment of a processing device 60 for the "normal" playback mode of the reproducing apparatus of Fig. 7 for recovering the original MPEG data stream of Fig. 8a from the data stream, as shown in Fig. 8b, using packet number information, also stored in the signal blocks as described above. The recovered MPEG data stream is shown in FIG. The processing device for the “normal” playback mode 60 '(Fig. 12) comprises a demultiplexer 90 having an input connected to an input 59 of the processing device 60' for receiving a sequence of groups of signal blocks, such as groups G1, G2 and G3 shown in FIG. 10 and 11 to restore packets that are sent to output 91 and to restore an array of packet numbers ... k-4, k-1, k, k + 2, k + 4, k + 8 ... from the third block parts in signal blocks, and to supply the specified array of packet numbers to output 92. Received packets are sent to the summarizer synchronization data 94, in which the synchronization signal of a packet of one byte length is reinserted as the first byte in all packets. The packets thus obtained are fed to the input 95 of the combining device 96. The output 92 of the demultiplexer 90 is connected to the inputs 97 and 98 of the combining device 96 and the generator 100 of empty packets, respectively. The output 102 of the empty packet generator 100 is connected to the input 103 of the combining device 96. The output 105 of the combining device 96 is connected to the output 64 of the processing device for the normal playback mode 60 '.

Suppose that packet P _k-4 and packet number k-4 are recovered from the first group G1, consisting of five signal blocks, and are supplied to combining device 96 and empty packet generator 100. This affects the packet P _k-4 output 105 by combining device 96. Next, packet P _k-1 and packet number k-1 are restored from group G1 and fed to combining device 96 and empty packet generator 100. Using a comparator and / or subtractor (not shown), it is established that packet number k -1 is not the next large number n compared to k-4 adopted before, and therefore two packets are missing. Therefore, the empty packet generator 100 twice produces empty packets of the same length, and combining device 96 inserts these two empty packets into the serial data stream immediately after packet P _k-4 (Fig. 8c). Then, combiner 96 inserts packet P _k-1 into the serial data stream.

It should be noted here that the generator 100 is not required to be the generator of empty packets. It is acceptable that the generator 100 be a generator of information of some length about an empty packet, which would be equal to the length of the packet or equal to several packet lengths.

The rocket R _k is the next reproducible demultiplexer 90 packet, and this packet is supplied after adding the synchronization byte to the input 95 of the combining device 96. The packet number k is supplied to the inputs 97 and 98 of the combining device 96 and the empty packet generator 100. Since the packet number k is in order by number with respect to packet k-1, then empty packets are not generated and packet P _k is output 105.

The next reproducible package is P _{k + 2} . After comparing the packet number k + 2 with the previous reproduced number k, it turns out that one empty packet must be inserted into the serial data stream. Next, the packet P _{k + 2 is} added to the data stream (Figs). This process continues for other packets to obtain the restored MPEG data stream (FIG. 8c). When comparing FIGS. 8a and 8c, it can be seen that FIG. 8c shows an MPEG data stream having the same bit and packet rate as the MPEG data stream in FIG. 8a. This data stream can now be fed to a standard MPEG decoder, which is capable of decoding a single video program selected by the recorder during recording from the MPEG data stream in FIG. 8c.

13a shows a sequential MPEG data stream as a function of time; the data stream contains packets P _k having a variable length, the bit rate in the data stream is also variable. It should be noted that the MPEG transmission packets of the data stream do not contain packet numbers. The packet number k related to the packet (FIG. 13 a) is added in this description for identification purposes only. FIG. 14 shows an example implementation of a processing device for “normal” playback 14 for recording a single video program that includes a serial data stream, as shown in FIG. 13 a. The device 14 ″ of FIG. 14 is very similar to the device of FIG. 9. The device 14 ″ differs from the device in FIG. 9 in that instead of the packet number generator 86, a time detector is present having an input connected to the output of the detector 84 and an output 111 to the input 112 of the combining device 82.

Selecting only those packets P _k from the serial data stream of FIG. 13a that contain information related to one video program to be selected, again means, for example, that packets P _k-4 , P _k-1 , P _k , P _{k + 2} , P _{k + 4} , P _{k + 8} should be selected, and intermediate packets should be removed. Fig.13b shows the data stream of the selected packets, which should be stored in a group of signal blocks, as has already been explained with reference to Fig.10 and 11. It should be noted that there is no relationship between the time axes in Fig.13a and 13b. It should be further noted that although the packets in the data stream of FIG. 13a have unequal lengths, they all contain 188 bytes of information. Therefore, the packages selected and shown in FIG. 13b are shown as having equal lengths.

The device shown in Fig. 14 receives the data stream in Fig. 13a and selects packets P _k-4 , P _k-1 , P _k , P _{k + 2} , P _{k + 4} , P _{k + 8 from it} . A packet detector 84 detects the reception of each packet in the original sequential MPEG data stream supplied to input 11 and generates time pulses for each detected packet. In accordance with each received time pulse, the time detector 110 determines a point in time t _k (FIG. 13 a) in the case of event P _k . Thus, at the output 111 of the detector 110, times t _k-4 , t _k-3 , t _k-2 , t _k-1 , t _k , ... etc. are formed. Moreover, the time detector 110 determines the duration of time intervals dt _k between two consecutive times, where dt _k is equal to the time interval t _{k + 1} -t _k . These values of the time intervals dt _k also go to the output 111. Under the influence of the control signal supplied to the control input 83 of the combining device 82 ', this device combines the packets P _k-4 , P _k-1 , P _k , P _{k + 2} , P _{k + 4} , P _{k + 8} selected by the selector 76, as well as time instants and corresponding time intervals t _k-4 , dt _k-4 , t _k-1 , dt _k-1 , t _k , dt _k , t _{k + 2} , dt _{k + 2} , t _{k + 4} , dt _{k + 4} , t _{k + 8} , dt _{k + 8} , ... from the information stream supplied by the temporary detector 110 for storing in signal blocks.

The storage of transmission packets in the signal blocks will be performed in the same manner as discussed above for FIGS. 10 and 11. The storage of temporary information in the third block parts may be as follows.

In the third block part TB3.1 of the first signal block SB1 of group G1 in FIG. 10, temporary information t _k-4 and dt _{k-4 are} stored. In the third block part FB of the third signal block SB3 of group G1, time information t _k-1 and dt _{k-1 are} stored. In the third block portion TB3.1 of the first signal block SB1 of the group G2, time information t _k and dt _{k are} stored. In the third block part FB of the third signal block SB3 of the group G2, time information t _{k + 2} and dt _{k + 2 are} stored. In the third block part TB3.1 of the first signal block SB1 of group G3, time information t _{k + 4} and dt _{k + 4 are} stored. In the third block section FB of the third signal block SB3 of the group G3, time information t _{k + 8} and dt _{k + 8 are} stored.

It may be that the third block part TB3.1 of the first signal block SB1 in groups and / or the third block part FB of the third signal block SB3 in groups will be too small to store temporary information. In this case, temporary information can be stored elsewhere or partially recorded in the third block part of TB3.1 and FВ, and partially elsewhere, see below.

According to the example of FIG. 11, time information t _k-4 and dt _k-4 is stored in the third block portions TB3.1, TB3.2 and TB3.3 of signal blocks SB1, SB2 and SB3, respectively, in group G1. The storage of temporary information can be performed once in a common memory array of the third block parts TB3.1, TB3.2 and TB3.3 or can be repeated at least once. For example, temporary information t _k-4 and dt _k-4 is stored in each of the third block parts TB3.1, TB3.2 and TB3.3. Temporary information t _k-1 and dt _k-1 can be stored in the third block parts FB, TB3.4 and TB3.5 of the signal blocks SB3, SB4 and SB5, respectively, in group G1. The storage of temporary information can be performed once in a common memory array of the third block parts FB, TB3.4 and TB3.5, or may be repeated at least once. For example, temporary information t _k-4 and dt _k-4 is stored in each of the third block parts TB3.4 and TB3.5. It is possible to store temporary information for the packet P _k-1 in the third block part of TB3.5. Further, it is possible to store temporary information in the third block parts of TB3.4 and TB3.5, and not in the block part of FВ.

Temporary information for packet P _k can be stored in the third block parts of group G2 in the same way that temporary information for packet P _k-4 was stored in third block parts of group G1. Temporary information for packet P _{k + 2} can be stored in the third block parts of group G2 in the same way as temporary information for packet P _k-1 was stored in the third block parts of group G1.

Temporary information for packet P _{k + 4} can be stored in the third block parts of group G3 in the same way as temporary information for packet P _k-4 was stored in the third block parts of group G1. Temporary information for packet P _{k + 8} can be stored in the third block parts of group G3 in the same way that temporary information for packet P _k-1 was stored in the third block parts of group G1.

On Fig shows an example implementation of the processing device 60 for the mode of "normal" playback of the playback device shown in Fig.7. The device 60 is designed to receive the restored MPEG signal of the data stream shown in FIG. 13 a from the data stream shown in FIG. 13 b using time information also stored in the signal blocks in the manner described above. The reconstructed MPEG signal of the data stream is shown in FIG. 13c. The device 60 '' of FIG. 15 is very similar to the processing device of FIG. 12. The demultiplexer 90 'is designed to recover packets from consecutive groups of signal blocks and provides packets to the output 91. The demultiplexer 90' is also used to recover time information t _k and dt _k from the third block parts of the signal blocks and to supply the specified time information to the output 92. Synchronizing bytes are added to each packet in the clock adding device 94. The received packets are fed to the input 95 of the combiner 96 '. The output 92 of the demultiplexer 90 is connected to the inputs 97 and 98 of the combining device 96 'and the empty information generator 100', respectively, for supplying temporary information to the combining device 96 'and the generator 100'. Assume that the packet P _k-4 and the corresponding time information are recovered from the first group G1, consisting of five signal blocks, and are supplied to a combiner 96 'and an empty packet generator 100'. This acts on the packet P _k-4 supplied to the output 105 by the combining device 96 '. The length of the packet P _k-4 should be equal to dt _k-4 , and the packet must be submitted to the output 105 at a time corresponding to t _k-4 . Next, the packet P _k-1 and time information corresponding to the packet P _k-1 are restored from the group G1 and supplied to the combiner 96 'and the empty packet generator 100'. By means of a comparator and / or subtraction device (not shown), it is established that the time t _{k-1 is} not equal to t _k-4 + dt _k-4 . Therefore, at least one packet following packet P _k-4 was skipped during recording. Therefore, the empty packet generator 100 'produces empty information so as to fill the gap between the end of the packet P _k-4 at time t _k-4 + dt _k-4 and time t _k-1 (Fig.13c).

Then combiner 96 'inserts a packet P _{k-1 of} length dt _k-4 into the serial data stream.

The packet P _k is the next reproducible demultiplexer 90 packet, and this packet is supplied after adding the synchronization byte to the input 95 of the combiner 96 '. Temporary information corresponding to the packet P _k is supplied to the inputs 97 and 98 of the combining device 96 and an empty information generator for packets 100 '. Since t _k is equal to t _k-1 + dt _k-1 , it is not necessary to generate empty information, and the packet P _k is output 105.

The next is the package P _{k + 2} . After comparing t _{k + 2} with t _k + dt _k, it turns out that one empty information must be inserted by the generator 100 '. Next, a packet P _{k + 2} having a length dt _{k + 2} is added to the data stream (Fig. 13c). This process continues for other packets to obtain the restored MPEG data stream shown in FIG. 13c. When comparing FIG. 13 a and FIG. 13 c, it can be seen that FIG. 13 c shows an MPEG data stream having the same (variable) bit and packet rate as the MPEG data stream in FIG. 13 a. This data stream can now be fed to a standard MPEG decoder, which is capable of decoding a single video program selected by the recorder during recording from the MPEG data stream of FIG. 13c having a variable bit and packet rate.

Next, other information will be described, which should be inserted into the free space in the groups of signal blocks separately or together with the packet numbers and / or temporary information described above.

Examples of such other information are information identifying a signal block in a group of y (y = 5) signal blocks, which should be the first signal block in a group of signal blocks. Such information can be stored in the third block parts TB3.1 of the signal blocks SB1 in groups G1, G2 and G3 in FIGS. 10 and 11.

Another example of such other information is the inclusion of the signal block number in the third block parts of the signal blocks, as shown in FIG. 11. The numbering of signal blocks can be performed within a group, as shown in Fig. 11, numbers 1 through 5 are stored in the third block parts TB3.1 through TB3.5, respectively, in signal blocks SB1 through SB5 in each group. The numbering of signal blocks can be implemented for a large number of signal blocks belonging to more than one group of signal blocks. It can be imagined that all signal blocks on the same track have a unique signal block number stored in the third block parts of the signal blocks. Now all signal blocks on the track can be identified by their unique number.

The numbering of the signal blocks described above has several advantages. The numbering of the signal blocks makes it possible to move the signal blocks in a different order than their original order, as is done within a group of signal blocks identified by unique signal block numbers. By determining the signal block number during playback, reverse movement can be performed to obtain the original sequence of signal blocks.

Another means that can be implemented is to repeat the signal blocks, since the recording and subsequent reproduction of information included in the signal blocks require a high degree of protection against transmission errors. Repeated signal blocks have the same signal block numbers, so they can be identified during playback.

Further, when determining the signal block numbers, it can be determined whether the signal blocks were lost during transmission due to errors that occurred during recording and playback. If there is no signal block number in the sequence of numbers, it can be decided that the signal block with this missing number is lost. With this determination, error correction or error concealment can be performed to correct or mask the missing signal block.

Next, operation of the processing devices for the “complex” playback mode 16 and 62 in the recording apparatus of FIG. 6 and the reproducing apparatus of FIG. 7, respectively, will be explained. To perform the mode of "complex" playback (or "special" mode) in the playback device, the storage medium 40 moves at a speed different from the nominal one. On Fig shows the storage medium 40 having the number of the oblique track recorded on it. FIG. 16 also shows the path indicated by 120 along which the read head 52 scans the storage medium in said “complex” mode. In general, information on tracks is recorded by at least two heads having a gap with different azimuth angles, so that even tracks have one azimuth and odd tracks have another. This means that when the storage medium is scanned along path 120, head 52 having one of two azimuth angles is only able to read information from even or odd tracks.

To ensure the playback of video information in the mode of "complex" playback, especially if the video information is recorded in an abbreviated form, it is necessary to add special information of the "complex" mode to special places on the track so that these places are scanned by the head 52 for different speeds of the information carrier, which are possible in the mode of "complex" playback. This information of the “complex” (or “special”) mode is special video information recorded in addition to the information of the “normal” mode, which is recorded on the track in the manner described above. As a result, some of the signal blocks on the track contain this "special" information, which must be scanned and read by the head 52 in the "complex" playback mode.

It should be noted that MPEG data, such as video data in an MPEG data stream, contains abbreviated video information. In order to realize such reduced video information, information corresponding to one picture is additionally encoded so as to obtain so-called I-frames. A high degree of data reduction can be achieved by performing inter-frame coding of at least two consecutive images, resulting in an I-frame for the first and a P-frame for the second image. To restore two images, it is necessary to perform inter-frame decoding, the inverse of inter-frame coding, on the I-frame information in order to restore the first image, and in order to restore the second image, inter-frame decoding, inverse to inter-frame coding, must be performed using both the I-frame information and the P- frame.

In the "complex" playback mode, only I-frame information can be used to reconstruct the video signal, while it is not possible to extract I-frame information and the corresponding P-frame information to implement inter-frame coding. Therefore, to obtain information of the "complex" playback mode, only information stored in I-frames, included in the sequential MPEG data stream and used as data of the "complex" playback mode is extracted.

It should be noted that in a special place on the track, for example, in the shaded area 122 on the track 124 in FIG. 16, several signal blocks are inserted that contain information of the “complex” playback mode. On Fig shows the sequence of signal blocks on the track 124. The shaded area 122 in Fig.16 is formed by a sequence of signal blocks SB _i through SB _j contained in the sequence shown in Fig.17. The third block parts of the TV in the signal blocks contained in the track in the “complex” playback area 122 now contain an indication that the signal blocks contain information of the “complex” playback mode. This information is indicated by the symbol "T" in the third block parts of TB3 in the signal blocks SB _i to SB _j inclusive. The signal blocks stored in the track primarily in the “hard” mode area 122, as well as the signal blocks stored in the track after the area 122, contain information indicating that the information stored in the signal blocks is “normal” mode information reproduction. This information is indicated by the symbol "N" in the third block parts of TB3 in the signal blocks SB _i-2 , SB _i-1 , SB _{j + 1} .

The processing device 16 (Fig. 6) is thus capable of generating information of "complex" playback from the MPEG data stream supplied to its input 17 by storing this information in those signal blocks that are specifically designed to store such information in a special place on the track and for inserting identifying information indicating that the signal unit is a signal unit in which “complex” playback mode information is stored in the third block parts of these signal units.

The processing unit of the "normal" playback mode 14 provides the storage of identifying information indicating that the signal blocks generated by the device 14 contain information of the "normal" playback in the third block parts of these signal blocks.

When the playback device switches to its “special” mode, the complex playback mode processing device 62 should detect the signal blocks that have the identifier “T” stored in their third block parts and restore information from these signal blocks for further processing so to realize the ability to view in the mode of "complex" playback.

The technical implementation described above describes a scheme for receiving MPEG transmission stream, asynchronously recording it to DVCR and restoring the original MPEG transmission stream during playback. This can be broadly understood as a method of transmitting time critical data (MPEG data stream) via an asynchronous channel (DVCR).

In addition to transmitting the MPEG data stream, there are various other applications that may require the transmission of time-critical data via an asynchronous channel. “Asynchronous” means that the physical speed of the data on the channel is different from the data rate, so the time allotted for the transmission of data through the channel is not constant.

In an MPEG data stream as an example of time critical data, the relative time of arrival of the data, which is the temporary information of the transmission stream, i.e. program time reference (PCR) must not change beyond a certain tolerance during transmission without a corresponding change in PCR value. This is because, otherwise, the phase-locked loop in the decoder will not be able to restore the data clock signal, and the buffers may be empty or full.

This problem of transmitting time-critical data via an asynchronous channel without changing any data that must be transferred also exists when the asynchronous channel is a computer or telephone network or a digital interface, for example P1394.

As explained previously in connection with MPEG applications, the main technique is to tag each transmitted unit, for example an MPEG transmission packet, with temporary information before transmission and using this tagging information to restore the correct data at the other end of the channel. Three situations are possible.

1. If the unit size of the transmitted data and the transmission rate are known and constant.

The transmitting side of the channel assigns a serial number to each transmitted unit, which increases by 1 for each transmission unit. Tagging is done, for example, by inserting a fixed number of bits before the beginning of each transmitted unit. The number of bits for tagging can be chosen so that the number of represented sequences is large enough for copying with the most moderate loss of transmitted units due to, for example, channel errors. The receiving side of the channel has a buffer and a counter. The counter is loaded with the serial number of the first received transmission unit at the beginning of each transmission session and is incremented by 1 each time a transmission unit or empty packet is received. If the serial number of the received transmission unit coincides with the counter value, then the received transmission unit is sent further, otherwise the necessary number of empty units is sent until the counter value becomes equal to the serial number. The receiver removes the added tagging information and sends a transmission unit or an empty unit with a known and constant speed.

2. If the unit size of the transmitted data is known and constant, but the transmission rate is unknown.

The transmitting side of the channel assigns to each transmitted unit, for example, two information items that are sufficient to determine the start time and end time of the transmission unit. For definiteness, pairs of start and end time or start time and duration of a transmission unit can be used. The receiving side of the channel should restore the transmission size from the additional information and determine the initial time of each transmission unit. In this case, there is no need to insert empty units.

3. If only the time parameter of the specific data is important.

If there is no need to maintain the characteristics of the entire transmission stream, i.e. only the time of the transmitted data should be supported, then there is no need to restore the size of the transmission, based on additional tagging information produced by the receiver, the size of the transmission of information should be sufficient so that there is no overlapping of the transmitted units. The transmitting side of the channel assigns to each transmitted unit the position of time-critical data and their expected arrival time. If the position is known and constant, the positioning information may be omitted. If in each unit there is more than one time-critical data, then several additional information pairs can be added accordingly. The receiving side of the channel determines the initial time of each transmitted unit from the output transmission size, the position and the expected time of arrival of time-critical data. In this case, there is no need to insert empty units.

On Fig presents a block diagram of another form of the basic system used for MPEG stream, where R represents the transmission rate MPEG data stream divided by a transmission unit in the form of a sequence of transmission packets from the digital interface (D-I / F).

A selector 130 corresponds to a selector 76 in FIG. 9. This implementation corresponds to situation 2 and, thus, contains a reference clock 132 and a counter 133 for determining the temporal information of the incoming data stream and tagging each transmission packet (TP) by means of block 135 with superimposed marks "time of arrival" (TOA) and "duration of arrival "(DOA) according to the local counter. These tag bits are written, via buffer 137, to the magnetic tape along with the transmission packets, using additional bits available from 2 to 5 sync blocks located as described previously.

During playback, each recorded packet is read at the corresponding time and size, taking into account the “time of arrival” (TOA) and “duration of arrival” (DOA) under the control of the reading control unit 138. The TOA mark indicates the time when this part of the TP should begin reading from buffer 131 and issues it to the DI / F interface. The DOA mark indicates the time for which this transmission packet (TP) should be read, i.e. 188 bytes of this TP should be read during the DOA of this packet, so the size can be easily calculated. The MUX 139 block is used in normal mode to separate tagged packets with a “special” mode and zero packets as desired, and the DE-MUX 140 block is used to delete added packets during playback. This circuit reproduces all the desired packets at the exact time and at the right speed. However, the output stream is not filled in the places corresponding to the missing packets, and therefore has gaps. The main reason for this is that the input stream may have spaces that are not equal to the integer number of transmission packet lengths. For example, the Grand Alliance stream is expected to have spaces, only 20 bytes long. Obviously, they cannot be filled with VCR. In addition, in this scheme, unrecorded PID packets also appear as spaces in the output stream.

19 is an example of a resulting transmission stream. In the upper diagram, for example, the input stream has two program streams: programs A and B. We record only program A. During playback, all packets belonging to program A are restored exactly according to their original time and length, but there is no filling for unrecorded program B. Therefore, the output stream shown in the bottom diagram is more sparse than the input. If the input stream has a real MPEG signal, then the output stream must also be real.

Temporary information in the input data stream R can be determined in various ways that are not essential for the present invention. For example, an input data stream may be provided to a known packet detector 134, which can determine the start and end times for each input transmission packet. The outputs of the packet detector can be used to determine the value of the counter 133, which represents the desired time information. Relative clocks 132 must have the same frequency for recording and reproducing processes, for sequentially interpreting tagging information and the additional steps necessary to initialize the counter 133 in a predetermined manner that is performed during recording and playback. The recorded timer information is extracted from the transmission stream in the read control unit 138 and is used to determine when the stored transmission packet can be issued.

For situation 1, where the transmission rate of the input transmission stream is constant and known, changes to the previous method can be used to restore this stream during playback. The main capabilities of this implementation require only a local counter (instead of the reference clock and the counter in Fig. 18), increasing its value upon the arrival of each transmission packet, and tagging each transmission packet with the “arrival sequence” mark (SOA), using additional bits available from 2 to 5 positions. During playback, each moment of interruption is determined in SOA, it is determined that a transmission packet has been received that has not been recorded. These "missing" packets are replaced by null null packets. All transmission packets are issued with a known and constant length. 20 is a timing chart illustrating input and output streams showing the content of time critical information.

In the flowchart of FIG. 18, arrows indicate only data streams. The specialist will be clear that some blocks are interconnected by commands and control signals that are not shown in the drawing.

As previously shown, the invention is also applicable to other data formats and other time-critical data storage methods. FIG. 21 shows a data unit transmission unit on one channel in which each data block 150 comprises a preamble 151 in the form of a block header, a block body 152 for transmitting a data stream, and a trailing portion 153 in the form of a CRC block. FIG. 22 shows an example of formatting the body of block 152 with respect to MPEG. In this case, the body of block 152 carries 5 188-byte MPEG transmission packets 160-164. There are 6 pointers 166-171 for indicating the byte position of transmission packets. Data 173 ... 178 associated with each pointer 166 ... 171 is temporary information. All pointers and associated data corresponding to time information require more than 24 bytes, as in this example, which must be stored in addition to the five transmission packets. They are stored in the body of the tag or heading 180. Taking for example 10 bits for each byte position indicator x 6 pointers (166 ... 171) = 60 bits; 21 bits for each temporary data pointer x 6 pointers (173 ... 178) = 126 bits; total 186 bits or approximately 24 bytes.

It should be noted that the byte position indicator is not required for all start and end points of a single packet. In this example, which is preferred, only the byte position indicators 167 and 168 indicate the start and end position of the second transmission packet 161. Another byte position indicator indicates a position offset from the beginning. This position offset is preferably the PCR position, usually located in the 12th byte after the start of the transmission packet and which is used sequentially to restore the timer as a clock pulse to ensure synchronization of this device with the encoding source of the data stream. With this choice of the byte position indicator, only 6 pointers 166 ... 171 are required to uniquely identify time information for a sequence of 5 packets 160 ... 164 with 1 packet having 2 pointers and with the remaining 4 packets having only one pointer. The associated data 173 ... 178 in this case are points in time, which are indicated by 6 pointers. Associated data may also include a packet number. When the bit rate is known and constant, the data may contain one packet number. The total length is five packets 160 ... 164 plus a header or tag 180, which contains temporary information for the next 5 transmission packets.

It should be understood that heading 180 has two areas associated with a unit of information. For example, it should contain a position byte pointed to by a pointer 166 and a second field containing, for example, temporary information for this position byte.

To implement the circuit shown in Fig.21 and 22, you can use the same device as shown in Fig.18, together with the cascade format blocks located in front of the buffer cascade 137 to obtain a formatted block, including the title block, calculation and adding a tag packet bodies and combining transmission packets and a CRC block before recording.

Various tagging options possible in accordance with the present invention are presented in FIG. Tagging option 1 includes points in time of the beginning and end of the packet; tagging option 2 includes a point in time of start and duration of the packet; tagging option 3 includes time critical data, such as a PCR whose position is known, i.e. only 1 pointer or position of information, compared to other options that require at least two information data.

It should be understood that the expression “temporary information” is used here in a broad sense and includes not only the information shown in FIG. 23, but also sequential packet numbers and generally all the information necessary for temporary “transparency”, which means that from the point of view The channel does not affect or change the time data of an outside observer.

The invention describes a situation where real-time data (distribution or transmission via an MPEG cable of an information signal) is recorded on non-real-time media (DVCR). Another important example is known as P1394 D-I / F, which includes asynchronous transmission, for which it is important to restore the times of the original data stream.

In the P1394 D-I / F interface protocol, data blocks similar to those shown in FIG. 21 are used to transfer data, for example, between a computer and consumer electronic equipment. If the data is MPEG transmission packets, the description given above with reference to Figs. 21 and 22 may be used. In this case, the channel does not remember the MPEG data stream, but due to the asynchronous nature of the P1394 DI / F interface, the correct synchronization of MPEG packets is lost without using means provided by the invention.

List of references

1. European patent application No. 492704 (PHN 13.546).

2. European patent application No. 93202950 (PHN 14.241).

3. European patent application No. 93201263 (PHN 14.449).

4. Grand Alliance HDTV System Specification, Draft document, February 22, 5 1994.

5. US patent specification No. 5142421 (PHN 13.537).

Claims (14)

1. A device for transmitting time-critical data through an asynchronous channel, comprising means for providing time-critical data in the form of a stream of successive time-critical transmission units, characterized in that it further comprises means for determining stream restoration information from time-critical data, which contains at least one element of temporal information for transmission units, moreover, time-critical flow can be restored from said stream recovery information audio data, means for tagging at least some of the transmission units with said at least one element of temporal information for transmitting said transmission units with the tagged means of temporal information elements for the asynchronous channel, and wherein the asynchronous channel comprises a digital recording device. 2. The device according to claim 1, characterized in that time-critical data is an MPEG data stream (moving picture standard), transmission units are transport packets. 3. The device according to claim 1, characterized in that said channel contains a data network. 4. The device according to claim 1, characterized in that the temporal information for one transmission unit contains only one indicator of the byte position of the transmission unit and the associated temporary information. 5. The device according to claim 1, characterized in that the temporal information for the plurality of transmission units comprises two byte position indicators of some of the plurality of transmission units and one byte position indicator for the remaining of the plurality of transmission units together with associated temporal information. 6. The device according to claim 5, characterized in that the temporary information for a plurality of transmission units is stored in a data unit associated with a plurality of transmission units. 7. The device according to claim 5, characterized in that the two pointers indicate respectively the start and end time for one transmission unit. 8. The device according to claim 7, characterized in that one pointer is an indicator of the byte position in the transmission unit. 9. The device according to claim 8, characterized in that the data associated with one pointer corresponds to a point in time of some of the transmission units. 10. The device according to claim 9, characterized in that the plurality of transmission units are transport packets of the MPEG information signal. 11. The device according to claim 10, characterized in that one pointer indicates a time reference program (PCR). 12. The device according to claim 10, characterized in that the plurality of transmission units contains 5 transmission units. 13. The device according to claim 1, characterized in that the time-critical data is formatted as a data block containing, in order, a block header, a block payload including a plurality of transmission blocks with said time information and an error checking block. 14. The device according to item 13, wherein the block payload contains many transport packets and one tag packet containing temporary information for many transport packets.

Images