DE19855501A1 - Video coding / decoding method in a fault-tolerant mode and associated device - Google Patents

Description

BACKGROUND OF THE INVENTION FIELD OF THE INVENTION

The present invention relates to a video knockout the decoding (CODEC) method and an associated device tion and in particular to a CODEC process in a tolerant operating mode and an associated device tung. The present invention is also based on the preliminary figen US Patent Application No. 60 / 067,013 entitled "Video CODEC Method in Error Resilient Mode "of the previous application otherwise.

DESCRIPTION OF THE PRIOR ART

In general, in a case where there is an error in one Channel is present, a CODEC this error of the channel cannot cope if it encodes all frames into bitstreams. This is because the bitstreams are not a decision element for determining whether to restore the fault area put or discard. So it is difficult to ent divide whether a certain part of a frame has a bit loss due to an error or whether the entire frame Has errors so one problem is that a start point of the bit stream in the next frame te by ignoring the encoded bit stream of the entire frame is riert. In a conventional CODEC, an additional Liche device used as a decision element to a to solve such problem. As an example, according to CODEC, which is defined in the H.263 standard, an image star code (PSC) used and if an error after this field information is recognized, the area that follows the PSC, ignored and the next PSC is then searched for. Next  towards a group of block start code (GBSC) that the Start of group of block (GOB) indicates used, and if an error is recognized based on this information, then only the corresponding GBSC is ignored, and the next GOB then searched to increase the amount of ignored area Reduce.

Fig. 1 shows an example of a video data packet that was generated by a conventional CODEC in a fault-tolerant mode. The video data packet of Fig. 1 is an example of the video data packet that was generated in the MPEG-4 CODEC. If one refers to this package, the conventional fault-tolerant CODEC data is divided into movement data and texture data for coding. The motion data consists of a macro block identification (COD) bit indicating whether coding is being performed or not, a macro block pattern chrominance (MCBPC) bit indicating the chrominance of each macro block and a motion vector, and the texture data consists of encoded block pattern luminance (CBPY) data, data quantization (DQUANT) data and discrete cosine transform (DCT) data. Such motion data and texture data are divided by a motion marker (MM). The DCT data are also encoded by reversible variable length coding (RVLC).

The decoding process of the bit stream, which is described in the above Art has been encoded is described. First is a re-sync Chronismarking RM sought in the bit stream. If an RM ge is found, the information is stored until the next RM viewed a package. A motion vector data part (MVDP) that consists of motion vectors, only shows predicted moves tion vectors, so that it is only possible to get an actual Lichen motion vector to use if the previously deco dated motion vector exists. Thus, if an error the corresponding one is recognized from the MVDP of the bit stream whole packet is ignored and there will be one RM that the next  Package corresponds to searched for decoding, which is therefore a Loss of a large amount of information.

SUMMARY OF THE INVENTION

To solve the above problems, there is a first task of the present invention, a video encoding-decoding (CODEC) procedure in a fault-tolerant operating mode deliver.

A second object of the present invention is to deliver a computer-readable medium on which a Com Computer program for the video CODEC process in one error tolerant operating mode.

A third object of the present invention is a video CODEC device in a fault tolerant loading drive mode to provide the video CODEC process len.

To solve the first task, a video CODEC supplied, which comprises the following steps: (a) dividing egg header part (HDP) bit area, a motion vector part (MVDP) bit area and a bit area of the data part of a discrete cosine transform (DDP) from each Macroblock of video data in a fault-tolerant mode kind; (b) performing variable length encoding on divided bit areas; (c) performing a reversible va Variable length coding of the bit areas, which are derived from the according to the variable length coding encoded bit areas according to a predetermined priority selected for recovery were; and (d) inserting markings into the according to the variable length coding or the reversible variable length gene coding coded bit areas.

To solve the second task, one is from a computer readable medium on which a computer program for the execution  video CODEC process is delivered, the method comprising the steps of: (a) splitting a header part (HDP) bit area, a motion vector data part (MVDP) bit area and a bit area of a Da part of a discrete cosine transformation (DDP) of each the macroblock of the video data in a fault-tolerant loading mode of operation; (b) performing variable length coding the divided bit areas; (c) performing a reversi variable length coding of the bit areas resulting from the bit areas subject to variable length coding according to a predetermined priority for recovery have been selected; and (d) inserting marks into the bit areas of the variable length coding or the rever sible variable length coding.

To solve the third task, a video CODEC-Vor direction provided, the following comprises a division unit for dividing a header part (HDP) bit area, a motion vector data (MVDP) bit area and bit areas of a data part of a discrete Kosi nustransformation (DDP) of each macroblock of the video data in a fault-tolerant operating mode; a unit of the varia blen length coding for the variable length coding of the divided bit areas, a coding unit of the reversible variable length coding for the implementation of the reversi variable length coding of the bit areas resulting from the bit areas subjected to the variable length coding a predetermined priority for recovery were selected, and a marker insertion unit for the Insertion of marks in the bit areas of the variable Length coding or the reversible variable length code tion.

BRIEF DESCRIPTION OF THE DRAWINGS

The above objects and advantages of the present invention be made clearer by a detailed description before  embodiments with reference to the attached Drawings:

Fig. 1 shows an example of a video data packet which has been generated during a conventional coding-decoding (CODEC) in a fault tolerant mode of operation;

Fig. 2 is a flow chart of the present showing the main steps of a vi deo CODEC method in a fault tolerant mode of operation according to a preferred embodiment of the invention

Fig. 3 is a block diagram of a video CODEC device in a fault-tolerant mode of operation according to a preferred embodiment of the present invention; and

Fig. 4 shows an example of a video data packet, which was generated by the fault-tolerant video CODEC method according to the vorlie invention.

DESCRIPTION OF THE PREFERRED EMBODIMENTS

Consider Fig. 2, in which the main steps of a video coding decoding (CODEC) process in a fault tolerant mode of a preferred embodiment of the invention are shown before lying in accordance with, the video CODEC method a division step 20, a variable Length coding step 22 , a reversible variable Längenko coding step 24 , which is carried out for each of the selected bit regions, and a marker insertion step 26 .

In step 20 , a header data part (HDP) bit area is formed by dividing the header data. In a similar manner, a motion vector data part (MVDP) bit area and a bit area of a data part (DDP) of a discrete cosine transformation (DCT) are formed by dividing the motion vector data and the DCT data, respectively.

The header data comprise information about the coding status of the current macroblocks. That is, the header data indicate whether the current macroblock is an inside macroblock in which the content of the current frame is encoded as it is, or an intermediate macroblock in which the difference is ge compared to a previous macroblock. So would when an error occurs in the header data, a very result in critical loss of information.

Therefore, due to the importance of the header data, the header data to be recovered first. Because of this, a Macroblock Identification (COD) bit indicating whether a Ko is performed or not, and a macroblock must sterchrominance (MCBPC) bit in an information bit (COD + MDBPC) can be combined for the division. This Kombina tion for the split is due for the following reasons moves. In the H.263 standard, 1 bit is used for the COD bit, and a variable length code is used for the MCBPC bit Header data used. However, since the COD bit is both 0 as well as 1, it is difficult to judge whether the COD bit has an error. Even in the case where that MCBPC bit using a variable length code ei If there is an error, it is very likely that the fault lerable bits in a variable length coding table exi bulls, so that the detection of the error is difficult. Consequently in this embodiment, the COD bit and the MCBPC bit combined for a split, and the combined bit A reversible variable length coding is used in order to easily recognize the change in the bit. For example, in the case of data 0110, an error such as 0101 or 1001 can be detected, making the probability the use of incorrect information is reduced. Wei thereafter, because the reversible variable length coding backward recovery is performed as well  possible. However, the HDP bit area can be divided can be built without a combination of the bits.

Step 20 is also performed in macroblock units, so that preferably the step of inserting a first macroblock index (FMBI) bit area indicating the serial number of the first macroblock in each packet referred to for decoding is included . Preferably, step 20 also includes inserting a last absolute motion vector (LAAV) bit, which includes the absolute motion vector information, rather than a predicted motion vector, according to the MVDP bitge. Preferably, step 20 further includes the step of inserting a packet number (PN) bit area as additional information indicating a serial number of the packet.

In the variable length coding step 22 , the variable length coding is carried out on the HDP bit area, the MVDP bit area and the DDP bit area. The PN and LAMV bit areas are also variable length coded.

In step 24 of the reversible variable length coding, a reversible variable length coding is carried out on the bit areas of variable length coding which have been selected according to a predetermined priority for the recovery. That is, if there is a loss of information during decoding, the reversible variable length encoding is performed with the important information of the entire packet or macroblock that may be lost. The reason for the priority is described in detail. First, as described above, the header data indicates whether the current macroblock is an intermediate macroblock in which the content of the current frame is encoded as it is, or an inside macroblock in which the difference from a previous macroblock is encoded. Almost full macroblocks are built from intermediate macroblocks to increase the effectiveness of the compression. If an error occurs in the HDP of the intermediate macroblock, the MVDP and the DDP cannot be used. Even if an error occurs in the MVDP, it is not possible to use the DDP. Thus, each piece of data is important in the sequence of the HDP, MVDP and DDP. That is, the priority exists within the macroblock according to such importance, and recovery according to the priority is required. Here, with a selectively reversible variable length coding according to the priority due to a limited channel capacity, not all data are considered, while the reversible variable length coding increases the number of bits compared to the variable length coding. In the video de CODEC method according to the present invention, which is not shown in FIG. 2, the information of the channel properties, such as capacity, error status and degree of congestion, is received by a reverse channel to identify the channel properties . In the identification step of the channel property, if a suitable channel status is present, a predetermined bit area, which has a low priority, for example a DDP bit area, is reversibly variable-length coded, otherwise the bit area is only variably length-coded, so that the channel properties as also to meet the increasing fault tolerance. Alternatively, in the channel property identification step, if an appropriate channel status is not present, additional information may be inserted into a bit area having a lower priority, such as a DDT bit area, to refer to for decoding. The use of additional information is described in detail below. First of all, it is possible to push the information from the HDP into the DDP. This is the case if an error occurs in the HDP with priority, not in the MVDP and the DDP that can be realized. By placing the HDP behind the DDP, the current data packet can be retrieved without being discarded. Secondly, it is possible to transfer a CRC (abbreviation for cyclical redundancy check) as additional information into the DDP. For example, in the DDP bit area, an 8-bit fixed length code is used as a direct current (DC) value, and a variable length code is used as an alternating current (AC) value. Relatively speaking, it is difficult to identify an error from an 8-bit code of fixed length. To aid in the detection of an error, the CRC is added to the 8-bit fixed length code, thus improving the recovery properties. Another fixed-length code, for example a quantizing part (QP) that indicates a quantizing value, and the 8-bit fixed-length code are also used to obtain the CRC. That is, by adding the CRC to a fixed length code of the DDP, the error recovery properties can be improved during decoding. That is, the additional information of the above two cases is used to help recover the defective area of the current package, or to identify the area that is different to detect errors therein. In this embodiment of the video CODEC method in a fault-tolerant operating mode, different RVLC tables are also used for the RVLC according to the channel properties in the channel property identification step, so that the error detection becomes more effective.

In the marker insertion step 26 , a resynchronization marker (RM) for the identification of packets and a header marker (HM) are inserted as an information bit for the identification of an HDP and an MVDP. A motion marker (MM) is also inserted for identifying the MVDP and the DDP. The HM is preferably constructed from code words which are not used for coding the HDP, so that they can be distinguished during decoding.

Fig. 3 shows a video CODEC device in a fehlerto leranten mode in accordance with a preferred embodiment of the present invention for achieving the above Vi deo CODEC method. Looking at Fig. 3, the video CODEC device includes a partitioning unit 30 , a variable length coding unit 32 , a reversible variable length coding unit 34 for reversible variable length coding of the variable length coded bit areas selected according to a predetermined priority for retrieval , a channel identification unit 36 for identifying the channel properties by receiving information about the channel properties such as capacity, error status and degree of congestion, for example by a reverse channel (not shown), and a marker inserting unit 38 for that Inserting a marker into the variable length-coded or the reversibly variable length-coded bit area.

The splitting unit 30 receives video data and splits of an HDP bit area, an MVDP bit area and a DDP bit area from each macroblock of the video data in a fault-tolerant mode as described with reference to FIG. 2. The division unit 30 also preferably forms the HDP bit area by combining a predetermined number of bit areas which have a higher priority for the recovery, for example a COD bit and an MCBPC bit, into an information bit (COD + MCBPC) during the Splitting.

The variable length coding unit 32 receives the subdivided bit areas and performs variable length coding, and the reversible variable length coding unit 34 performs the reversible variable length coding on the bit areas selected from the variable length-coded bit areas according to a predetermined priority for the recovery , for example according to the probability of loss in the whole packet or the macroblock during decoding, which leads to good error recovery properties. However, the reversible variable length coding increases the number of bits compared to the variable length coding, so that it is preferable to perform the reversible variable length coding only on the bit areas selected according to the priority for the recovery, taking into account the limited channel capacity becomes.

For this purpose, the channel property identification unit 36 controls the variable length coding unit 32 and the reversible variable length coding unit 34 so that when a suitable channel state exists (does not exist), a bit area having a low priority, such as a DDP bit, is reversibly variable length encoded, and otherwise the bit area is variable length encoded. Alternatively, an additional information insertion unit (not shown) may also be included to insert additional information into a low priority bit area, such as the DDP bit area, if the channel property identification unit 36 determines that an appropriate channel condition exists (not exists). Preferably, leads when the channel characteristics identification unit 36 determines that a suitable channel condition exists (does not exist), the reversible Variablenlängenkodiereinheit 34, the reversible variable length coding using different Ko decoding tables through to the same effect as that described with reference to FIG. 2, to to reach.

The CODEC device according to the present invention may further include an FBMI bit area inserting unit (not shown) for inserting an FMBI indicating a serial number of the first macroblock of each packet so as to achieve the same effect as with reference to FIG Fig. 2 be written, and the reversible variable length coding unit 34 can perform the reversible variable length coding on a LAMV bit area. Preferably, the video de CODEC device includes a LAMV bit area inserting unit (not shown) which includes absolute motion vector information behind the MVDP bit area.

The marker insert unit 38 also pushes a resynchronization marker (RM) for separation into packets, a header marker (HM) consisting of code words that are not used to encode the HDP bit area, as an information bit for dividing the HDP and the MVDP, and a motion marker (MM) for dividing the MVDP and the DCT. Here the HM consists of code words that are not used to encode the HDP so that they can be distinguished during decoding.

Fig. 4 shows an example of a video data packet obtained by the video CODEC method according to the present invention. Referring to FIG. 4 so comprises video data packet, the method video CODEC generated by the a RM-bit area, a PN-bit area, a FMBI-bit area, an HDP-bit area, a HM-bit area, a MVDP Bit area, a LAMV bit area, an MM bit area and a DDP bit area. In the video data packet, which includes such bit areas, the RM bit denotes the start of the video data packet, so that the start point of a packet for decoding can be found, as is described in the prior art. The PN denotes a serial number of each packet and can be used as additional information, and the FMBI bit denotes a serial number of the first macro block of each packet. In this embodiment, the HDP bit is converted into an information bit (COD + MCBPC) by the combination of a COD bit and an MCBPC bit defined in the MPEG-4 or H-236 standard and then performing the reversible variable Received length coding on the combined bit. The HM bit as an information bit for dividing the HDP and the MVDP also consists of code words which are not used to code the HDP bit. The MVDP bit is obtained by performing the reversible variable length coding on the information generated by the motion vector prediction. A coded block pattern luminance (COBY) bit, the data quantizer (DQUANT) bit, and a DCT coefficient are encoded in the DDP bit area containing discrete cosine transform (DCT) information. It is also preferred to insert this information, for error identification and recovery during decoding, and the information used in the current packet into the DDP bit. Such insertion of information can be applied depending on the channel properties selected. The MM bit is used to split the MVDP bit and the DDP bit.

The decoding method for the video data packet, as above has been encoded is described. First, when an RM is found in the received bit stream, the bit stream between the found RM and the next RM as a package hen as described in the prior art. When a Failure in the HDP bit occurs, the HDP bit can reverse as well can be decoded as well as forward, since the HDP is reversible length-coded. For example, the HDP bit by using the PN and FMBI of the next package be won again. That is, since the macro block number of the current package corresponds to the value that can be obtained from the This can decrease the FMBI of the next packet received by 1 HP bit to the previous area of the fake macroblock area can be recovered by reverse coding. The MVDP bit is also decoded until MM is found to thus decoding the motion vector by prediction. If an error occurs during such a decoding process occurs, decoding in the reverse direction is also possible, since the MVDP is a reversible variable length coding was thrown. In comparison there exists in a decoration dierverfahren for a video data packet that through the conven tional CODEC method was coded, only the predicted one Motion vector in the MVDP bit so that it is possible to get one use the actual motion vector only if the previously decoded motion vector exists. In the video CODEC The method according to the present invention is the last one motion vector LAMV as an absolute motion vector not reversibly variable as a predicted motion vector length coded to be sent. Thus, in the case of reversible variable length decoding using the LAMV det, this is independent of the previous motion vector. Also in the event that an error occurs in the DDP bit area, if the DDP bit area has been variable length encoded, the Package discarded. However, if the DDP bit area is reversed has been variably length-coded, the decoding is in  a reverse direction possible, which is an increased tolerance against channel errors. Even in the case where additional information necessary for recovery dig is encoded in the DDP bit area, such can be too additional information can be used for decoding.

The invention can ver in a digital universal computer become a reality of a program from a computer usable medium expires, which does not rule out in the way a storage medium, such as magnetic storage medium (e.g. ROMs, floppy disks, hard drives, etc.), an optically readable medium (e.g. CD-ROMs, DVDs, etc.) and carrier waves (e.g. transmissions over the Internet). Thus the present invention can run as a medium that can be used by the computer, the one computer readable program code unit, the trained on it for a video CODEC, including the computer readable program code device in the Com medium usable on the computer includes: one from the computer Readable program code device for dividing a HDP bit area, an MVDP bit area and a DDP bitge offers each macroblock of video data in one error learning type; a program code readable from the computer device for variable length coding of the subdivided Bit domains; a computer readable program code device for reversible variable length coding of the bit areas, the from the variable length-coded bit areas according to a predetermined priority selected for recovery were; and a computer readable program code device for inserting markers in the variable length dated or reversibly variable length-coded bit areas. A functional program, a code and code segments that can be used to implement the present invention animals can be done by an experienced computer programmer derived from the description of the invention contained herein become.  

As described above, the CODEC method can be performed according to tolerance of the present invention Deliver channel errors, so that the transmissions under Bedin in which mistakes are a serious problem, how for example in a wireless communication channel, weni are affected by errors.

Although specific terms in the above embodiment ba based on the MPEG-4 visual and H.263 standards were like RM, PN, FMBI, HDP, HM, MVDP, MM and DDP, so who which they only use in a general and descriptive sense applies and should not constitute a restriction. So it can Video CODEC method of the present invention on all Vi deo-CODECs are used that use other bit areas, which are the same as the bit areas defined by the above speci Expressions are described.

Claims (29)

1. A video encoding-decoding (CODEC) method for encoding video data to generate a video data packet and for decoding the packet, the method comprising the steps of:

• (a) subdividing a header data part (HDP) bit area, a motion vector data part (MVDP) bit area and a data part (DDP) bit area of a discrete cosine transformation of each macroblock of the video data in an error tolerant manner;
• (b) variable length coding of the subdivided bit regions;
• (c) reversible variable length encoding of the bit areas selected from the variable length encoded bit areas according to a predetermined priority for recovery; and
• (d) inserting markers into the variable length-coded or reversibly variable-length-coded bit areas.

2. The method according to claim 1, wherein in step (c) The HDP bit area and the MVDP bit area are reversibly variable in length be encoded. 3. The method of claim 1, wherein the HDP bit area is through combining a variety of bit areas that are strongly related to Contribute recovery, formed into an information bit becomes. 4. The method of claim 3, wherein the predetermined Bitge offer a macro block identification (COD) bit and a macro block pattern chrominance (MCBPC) bits are important to that Recovery, and the COD bit and the MCBPC bit in a new syntax (COD-MCBPC) can be combined. 5. The method of claim 1, wherein step (a) one Step of inserting a first macroblock index (FMBI) -  Bit area, which is the serial number of the first macro block denoted each package. 6. The method of claim 2, wherein step (a) one Step of inserting a last absolute movement vector (LAMV) bits, which is the absolute motion vector information tion includes after the MVDP bit area. 7. The method of claim 4, wherein the LAMV bit area rever is variably length-coded. 8. The method of claim 1, further comprising a step inserting a packet number (PN) bit area as to additional information for displaying a serial number of the packet after the RM bit area. 9. The method of claim 1, wherein step (d) comprises the following steps

• (d1) inserting a resynchronization marker (RM) for dividing each packet;
• (d2) inserting a head marker (MM) as an information bit for dividing the HDP and MVDP bit areas, which consists of code words that are not used to code the HDP bit area; and
• (d3) Insert a motion marker (MM) for the subdivision of the MVDP and DDP bit areas.

10. The method of claim 1, further comprising a step for identifying channel properties, the step of identifying the channels continue a step of the reversible variable length gene coding of a predetermined bit area that a low priority if there is an appropriate channel status, and a variable length coded bit area if a ge own channel status does not exist, includes. 11. The method of claim 10, wherein the bit area is the DDP bit area is.   12. The method of claim 1, further comprising the steps of:
Identify channel properties; and
Inserting additional information into a predetermined bit area which has a low priority if it is determined in the step of identifying the channel properties that a suitable channel state is present. 13. The method of claim 12, wherein the bit area is the DDP bit area is. 14. The method of claim 1, further comprising a step identifying channel properties, where if in the step of identifying the channel properties is determined that a suitable channel state is present, the reversible variable length coding under Ver using various coding tables. 15. A computer readable medium on which a computer program is configured to perform a video encoding-decoding (CODEC) method of encoding video data to generate a video data packet and to decode the packet, which Video CODEC process includes the following steps:

• (a) subdividing a header data part (HDP) bit area, a motion vector data part (MVDP) bit area and a data part (DDP) bit area of a discrete cosine transformation of each macroblock of the video data in an error tolerant manner;
• (b) variable length coding of the subdivided bit regions;
• (c) reversible variable length encoding of the bit areas selected from the variable length encoded bit areas according to a predetermined priority for recovery; and
• (d) inserting markers into the variable length-coded or reversibly variable-length-coded bit areas.

16. A video encoding-decoding (CODEC) device for encoding video data to generate a video data packet and for decoding the packet, comprising:
a subdivision unit for subdividing a header data part (HDP) bit area, a motion vector data part (MVDP) bit area and the data part (DDP) bit areas of a discrete cosine transformation of each macroblock of the video data in an error-tolerant manner;
a variable length coding unit for variable length coding of the subdivided bit areas;
a reversible variable length coding unit for reversible variable length coding of the bit areas selected from the variable length-coded bit areas according to a predetermined priority for the recovery; and
a marker insert unit for inserting markers into the variable length-coded or reversibly variable-length-coded bit areas. 17. The apparatus of claim 16, wherein the reversible varia ble length coding unit the HDP bit area and the MVDP bitge offers a reversible variable length coding. 18. The apparatus of claim 16, wherein the subdivision unit forms the HDP bit area as an information bit by combining a variety of bit areas that contribute strongly to recovery. 19. The apparatus of claim 18, wherein the predetermined Bit areas include a macro block identification (COD) bit and a Macroblock Pattern Chrominance (MCBPC) bits used for replication are important, and the COD bit and the MCBPC bit can be combined into a new syntax (COD + MCBPC). 20. The apparatus of claim 16, further comprising an on unit for inserting a first macroblock index (FMBI) Bit area, which is the serial number of the first macro block each package indicates includes.   21. The apparatus of claim 16, further comprising an on unit for the insertion of a last absolute movement vector (LAMV) bits, which is the absolute motion vector information tion includes after the MVDP bit area. 22. The apparatus of claim 21, wherein the reversible varia ble length coding unit the LAMV bit area of a reversible undergoes variable length coding. 23. The apparatus of claim 16, further comprising an on unit for inserting a packet number (PN) bit area as additional information for displaying a serial Number of the packet included after the RM bit area. 24. The apparatus of claim 16, wherein the marker insert unit a resynchronization marker (RM) for the Un of each packet, a head marker (MM) as one Information bit for the subdivision of the HDP and MVDP bits areas that consist of code words that are not used to encode the HDP bit area and a move marker (MM) for the subdivision of the MVDP and DDP bit areas. 25. The apparatus of claim 16, further comprising a cana property identification unit includes, the channel property identifier unit the right Versible variable length coding unit and the variable length gene coding unit controls so that if a suitable channel status is present, a predetermined bit area that has a low priority, reversible variable length coding, and if there is no appropriate channel status, the bit area is length-coded variable. 26. The apparatus of claim 25, wherein the bit area is DD-P bit area. 27. The apparatus of claim 16, further comprising:
a channel property identification unit for identifying the channel properties; and
an additional information insertion unit for inserting additional information into a predetermined bit area which has a low priority when it is determined in the channel property identification unit that an appropriate channel condition is present. 28. The apparatus of claim 27, wherein the bit area is the DD-P bit area. 29. The apparatus of claim 16, further comprising a channel property identification unit for identifying which includes channel properties, the reversible variable length coding unit a reversible variable length coding using ver of various coding tables when in the channel is identified as a suitable Channel status is present.