JPH08162963A - Data encoder and decoder - Google Patents

Abstract

PURPOSE: To minimize the propagation of errors to encoded data of the errors generated in allocation data in the case of expanding and decoding data compression-encoded by a sub-band encoding system. CONSTITUTION: This data encoder 10 sub-band encodes input data, generates ALLOC data, SCF data, the encoded data and instruction data and generates the bit streams. An ECC encoder 20 adds ECC codes to the bit streams and a recording system 22 records them. A reproduction system 26 reproduces the bit streams and an ECC decoder 28 performs the error detection processing or the like of the bit streams. The data decoder 30 separates the respective data from the reproduced bit streams and expands and decodes them to original base band data.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a data coding apparatus for compressing and coding baseband data such as audio data by a subband coding method or the like by utilizing the correlation between preceding and following data, and a compression method as described above. The present invention relates to a data decoding device that expands and decodes encoded data, and a data encoding / decoding device, a data recording device, and a data reproducing device using these.

[0002]

2. Description of the Related Art An encoding technique such as a sub-band encoding method for providing a high quality audio signal is MP
It has been advanced by EG. In the sub-band encoding method, one block of 320-sample baseband audio data is frequency-divided into 32 sub-bands, signal strength is calculated for each of these sub-bands, bits are reallocated, and compression encoding processing is performed. I do. The encoded data (DATA) of each subband obtained by the compression encoding process is the allocation (ALLOCATION; ALLOC) data indicating how many bits are allocated to that subband and the maximum value of the signal strength in that subband. With scale factor (SCALE) data showing

The encoded data and these data are 1
Each block is recorded or transmitted as a bit stream multiplexed in the data format shown in FIG. 8 on a recording medium such as a magneto-optical disk or a magnetic tape. Of these data, allocation data is data that is allocated to all subbands by 4 bits each. Further, the scale factor data is 6 only when the allocation data is not 0, that is, when the bit is assigned to the corresponding subband.
It is data that is assigned bit by bit. The encoded data is data that is allocated by the number of bits indicated by the corresponding allocation data only when the allocation data is not 0, that is, only when the bit is allocated to the corresponding subband. In this way, the compression efficiency is increased by including only the coded data and the scale factor data of the subband to which the bits are allocated in the bitstream.

[0004]

Among the bit streams obtained by the above-mentioned sub-band coding method, etc., FIG.
As shown by arrows a and b in (A), p,
q-th subband (subbands p and q, where 1 ≦
When an error occurs in the coded data (coded data p, q) of p, q ≦ 32 (the same applies hereinafter), only that coded data p, q is invalidated.

Further, as shown by an arrow a in FIG.
When an error occurs in the scale factor data (scale factor data p) of the subband p, FIG.
As shown by the arrow b in (B), the data after the encoded data p becomes invalid.

Further, as shown by an arrow a in FIG.
If an error occurs in the allocation data (allocation data 32) of the subband 32, it is impossible to determine whether or not a bit is allocated to the subband 32 and the head position of the encoded data cannot be identified. All data after the scale factor data 32 becomes invalid. Similarly, as shown by an arrow a in FIG. 8D, when an error occurs in the allocation data p, it is impossible to determine whether or not a bit is assigned to each subband after the subband p, and the encoding is performed. Since the head position of the data cannot be identified, all data after the scale factor data p shown by the arrow b becomes invalid.

As described above, even if an error occurs in any of the allocation data, all the encoded data are invalidated, so that the encoded data of the entire block cannot be decoded. That is, FIG.
As shown in (A), (B), and (C), the error in the allocation data generated in the arrow a invalidates all the data after the arrow b, and therefore the error is propagated to at least all encoded data. There is a problem that it will end up.

In the data formats shown in FIGS. 8 and 9, allocation data, scale factor data, and encoded data are arranged earlier in the subband data having a lower frequency. In the decoding process of audio data, it is known that lower frequency sub-band data is more important auditorily. This means that even if some of the higher frequency sub-band data located later in the data format is invalid,
If most of the data in the subband with the lower frequency can be decoded, the listener will not noticeably deteriorate the sound quality.

The present invention has been made in view of the above-mentioned problems of the prior art. Baseband data such as an audio signal is divided into subbands and compression-encoded, and the compression-encoded data is obtained. Even if an error occurs in the allocation data when decompressing and decoding the data, the propagation of the error to the encoded data can be limited to a part of the range, and the deterioration of the data after decoding can be minimized. An object is to provide an encoding device and a data decoding device. It is another object of the present invention to provide a data encoding / decoding device, a data recording device, and a data reproducing device that use the data decoding device and the data encoding device to reduce data deterioration.

[0010]

In order to achieve the above-mentioned object, a data encoding apparatus according to the present invention divides baseband data into a predetermined number of subbands and compression-encodes the subbands. Fixed-length allocation data indicating bit allocation, scale factor data indicating the maximum value of encoded data only for the subband to which bits are allocated, and encoded data only for subband to which bits are allocated. Encoding means for generating, and data multiplexing for generating a bitstream in which the allocation data for each data block, the scale factor data, and the encoded data are associated with each other and multiplexed in accordance with a predetermined data format. Means and the position of the encoded data in the data format And a instruction data generating means for generating instructs data. Suitably, the said instruction | indication data production | generation means produces | generates the data which shows substantially the number of the sub-band to which the bit was allocated as said instruction | indication data.

A data decoding device according to the present invention is a data decoding device for decoding the baseband data from a bit stream generated by any of the above data encoding devices and the instruction data, Based on the separation means for separating the encoded data from the bit stream, and from the beginning, using the allocation data and the corresponding scale factor data until the allocation data is first invalid. It has a decompression / decoding means for decompressing / decoding the separated encoded data. Preferably, the separating means detects the start position of the encoded data in the data format based on the instruction data, and separates the encoded data from the bit stream.

A data encoding / decoding device according to the present invention has any one of the above data encoding devices and one of the above data decoding devices. A data recording device according to the present invention includes:
Recording means for recording the allocation data, the scale factor data, the encoded data and the instruction data on a predetermined recording medium. Further, a data reproducing apparatus according to the present invention includes the data decoding apparatus according to any one of the above, and the allocation data, the scale factor data, the encoded data, and the instruction data recorded on a predetermined recording medium, on the recording medium. And a reproducing means for reading and reproducing.

[0013]

In the data encoding device according to the present invention, the encoding means converts the baseband data such as an audio signal,
For example, compression coding is performed by a subband coding method, and coded data, fixed-length allocation data indicating the bit allocation of each subband, and the maximum value of the coded data corresponding to the subband to which the bits are allocated are respectively shown. Generate scale factor data. The data multiplexing means multiplexes the generated allocation data, scale factor data, and encoded data in association with each other for each data block in accordance with a predetermined data format to generate a bitstream. The instruction data generating means generates, as the instruction data indicating the position of the encoded data in the data format, data indicating the number of subbands to which bits are assigned, for example.

Further, in the data decoding device according to the present invention, the separating means separates the encoded data from the bit stream based on the instruction data. That is, in the above data format, since the allocation data of fixed length (32 × 4 bits) is followed by the scale factor data of 4 bits for each of the number of subbands to which bits are allocated, the separating means is the data format. From the nth bit represented by the following equation from the beginning,
The data up to the allocation data of the next block is separated as encoded data.

[0015]

## EQU00001 ## n = 128 + 4k + 1 (1) where k is the value of the instruction data indicating the number of subbands to which bits are assigned.

The decompression / decoding means decompresses / decodes all the encoded data when no error occurs in the allocation data, and when there is an error in the allocation data, the allocation data at the beginning is sequentially read. , Decompression / decoding processing is performed using allocation data before the allocation data in which the error first occurs, and corresponding scale factor data and encoded data.

[0017]

Embodiments of the present invention will be described below. Figure 1
It is a figure which shows the structure of the data recording / reproducing apparatus 1 using the data encoding device 10 and the data decoding device 30 which concern on this invention. As shown in FIG. 1, a data recording / reproducing apparatus 1
Is composed of a data encoding device 10, an ECC encoding device 20, a recording system 22, a reproduction system 26, an ECC decoding device 28, and a data decoding device 30, and, for example, converts baseband digital audio data (input data). Each block is subband-encoded, recorded on a recording medium 24 such as a magnetic tape, and the data recorded on the recording medium 24 is reproduced.

As shown in FIG. 2A, the data encoding apparatus 10 processes 320 samples of input data as one block, frequency-divides each sub-block into 32 sub-bands, and outputs the signal strength of each sub-band. Value is calculated, bits are reallocated for each sub-band, compression coding processing is performed, and as shown in FIG.
(ALLOC), scale factor data (SCF), encoded data (DATA) and instruction data (SCALE FACTOR NUMBER; SCFN)
Generate The data encoding device 10 multiplexes allocation data, scale factor data, and encoded data into blocks according to the data format shown in FIG. Output to. The configuration and operation of the data encoding device 10 will be described later with reference to FIGS. 3 and 4.

Of the data generated by the data encoding device 10, allocation data is assigned to each sub-band by 4 bits, so that it has a fixed length of 128 bits, and scale factor data is used when the allocation data is not 0. Since only 6 bits are allocated for each block, the number is different for each block, and the encoded data has a variable length because it is allocated by the number of bits indicated by the corresponding allocation data only when the allocation data is not 0. Data recording / reproducing apparatus 1 described above
The following table shows the operation and allocation data, scale factor data, and encoded data contents.

[0020]

[Table 1] Number of block samples; 320 samples Number of sub-band divisions; 32 bands Allocation data (ALLOC); 32 pieces × 4 bits Scale factor data (SFC); 32 pieces or less × 6 bits Encoded data (DATA); 32 bands Below x 10 bits or less x 16 bits or less However, the above numerical values are per one block. It should be noted that scale factor data for subbands whose allocation data value is 0 is not generated. Further, the number of bits of the encoded data is equal to the number of bits indicated by the corresponding allocation data, and the encoded data of the subband whose allocation data value is 0 is not generated.

The ECC encoder 20 has an error in the bit stream generated by the data encoder 10.
A correction code (ECC code) is added and output to the recording system 22. The recording system 22 records the bitstream to which the ECC code is added on the recording medium 24.

The reproduction system 26 reads out and reproduces the bit stream recorded on the recording medium 24, and outputs the reproduced bit stream to the ECC decoding device 28. E
The CC decoding device 28 performs error detection processing and error correction processing of the bit stream using the ECC code included in the reproduced bit stream, and outputs the processed bit stream to the data decoding device 30. The data decoding device 30 separates the encoded data by using the instruction data included in the bit stream that has been subjected to the error detection processing and the error correction processing, and invalidates the head allocation data and the corresponding scale factor data due to an error. The allocation data before the allocated allocation data and the corresponding scale factor data are sequentially used, and the corresponding encoded data is expanded and decoded and output as baseband digital audio data (output data). The configuration and operation of the data decoding device 30 will be described later with reference to FIGS. 5 and 6.

The configuration and operation of the data encoding device 10 will be described below. FIG. 3 is a diagram showing the configuration of the data encoding device 10 shown in FIG. FIG. 4 is a diagram showing allocation data generated by the data encoding device 10 shown in FIG. 3, in which (A) shows a data format,
(B) shows the format of allocation data.
As shown in FIG. 3, the data encoding device 10 includes a timing generation circuit (TG) 100 and a subband encoding circuit (S).
BC) 102, data packing circuit (DP) 104,
It is composed of an allocation data counting circuit (ALCC) 106.

The timing generation circuit 100 uses a frame pulse signal (FP) indicating a block division of input data.
The operation timings of the subband coding circuit 102, the data packing circuit 104, and the allocation data counting circuit 106 are controlled based on the above. The subband coding circuit 102 subband codes the input data to generate allocation data, scale factor data and coded data, and outputs the data to the data packing circuit 104 and the allocation data counting circuit 106. As described above, the allocation data is 32
It is a fixed length data of 4 x 4 bits, as shown in FIG.
The allocation data corresponding to each subband indicates the number of bits of the corresponding encoded data. When the signal strength of the corresponding subband is 0, the value of the allocation data is 0.

The allocation data counting circuit 106 is
Of the allocation data generated by the subband coding circuit 102, for example, the number of data whose value is not 0 is counted and output to the data packing circuit 104 as instruction data (SCFN). As the instruction data, any data can be used as long as it can indicate the position of the encoded data in the data format shown in FIG. 4, but the value is 0.
The number of non-allocation data (the number of subbands to which bits are assigned) is used as the instruction data.
This is preferable because the data length becomes the shortest. The data packing circuit 104 multiplexes each data output from the subband coding circuit 102 according to the data format shown in FIG. 4A, further adds instruction data generated by the allocation data counting circuit 106, and adds the bit to the bit data. Generate and output a stream.

The configuration and operation of the data decoding device 30 will be described below. FIG. 5 shows the data decoding device 3 shown in FIG.
It is a figure which shows the structure of 0. FIG. 6 is a diagram for explaining the operation of the timing generation circuit 300 of the data decoding device 30 shown in FIG. As shown in FIG. 5, the data decoding device 30
Is a timing generation circuit (TG) 300, a data unpacking circuit (DUP) 302, a concealment circuit (CN).
L) 304 and a subband decoding circuit 306.

The timing generation circuit 300 controls the operation timing of each component of the data decoding device 30 based on the frame pulse signal (FP) indicating the block delimiter in the bit stream, and also has the data format shown in FIG. The area from the rear of the allocation data to the m-th data represented by the following equation is determined to be scale factor data indicated by arrow a in FIG. 6, and m scale indicated by arrow b in FIG.
A timing signal indicating the period from the + 1st data to the last encoded data is generated and output to the data unpacking circuit 302 to separate these data.

[0028]

## EQU00002 ## m = 128 + 4k (2) where k is a value of instruction data indicating the number of subbands to which bits are assigned.

The data unpacking circuit 302 separates allocation data, scale factor data, encoded data, and instruction data from the bit stream and conceal circuit 304 and timing generation circuit 300.
Output to

The conceal circuit 304 has a range from the encoded data corresponding to the allocation data invalidated first due to an error or the like to the last encoded data, the encoded data corresponding to the invalid scale factor data, and , The encoded data that has not been error-corrected is invalidated (concealed) so as not to be decoded, and is output to the subband decoding circuit 306. The subband decoding circuit 306 expands and decodes the concealed encoded data using the corresponding allocation data and scale factor data, and outputs it as output data.

Hereinafter, referring to FIG. 7, data recording /
The operation of the reproducing apparatus 1 will be described. FIG. 7 is a diagram showing a range of encoded data that can be decoded by the data decoding device 30 of the data recording / reproducing device 1 shown in FIG. 1 when an error occurs in allocation data. (B),
(C) shows the case where an error occurs in the allocation data of the 32nd, pth, and 1st subbands, respectively. The data encoding device 10 frequency-divides the input data into 32 subbands, compression-encodes them, generates allocation data, scale factor data, encoded data, and instruction data, and further, these data are given a predetermined value. A bitstream is generated by multiplexing according to the data format. The ECC encoding device 20 adds an ECC code to the bitstream. The recording system 22 is an EC
The bitstream to which the C code is added is recorded on the recording medium 24.

The reproduction system 26 reads the bit stream from the recording medium 24 and reproduces it. ECC decoding device 28
Performs error detection processing and error correction processing of the bitstream using the ECC code included in the reproduced bitstream. The data decoding device 30 separates each of the above-mentioned data from the bit stream that has been subjected to error detection processing and the like, decompresses and decodes the original baseband data to generate output data.

In the data encoding device 10, the instruction data is further added to the allocation data, the scale factor data and the encoded data, and the data decoding device 3
At 0, since decompression decoding is performed using this instruction data, if the allocation data 32 becomes invalid due to an error as shown by the arrow a in FIG.
All the other coded data 1, ..., 31 can be decoded simply by invalidating the previously coded data. Also, FIG.
As indicated by arrow a in (B), allocation data p
, 32 are invalid, the other encoded data 1, ..., P-1 are all decodable. FIG. 7 (C)
As indicated by the arrow a, all the encoded data after the arrow b becomes invalid only when the allocation data 1 becomes invalid. In the data recording / reproducing apparatus 1, if an error occurs in the encoded data,
As described above with reference to (A), only that encoded data is invalidated. Further, in the data recording / reproducing apparatus 1, when an error occurs in the scale factor data, as described above with reference to FIG.
Only the corresponding encoded data is invalid.

FIG. 7 shows a case of performing compression encoding and decompression decoding using instruction data, and FIG. 9 shows a case of performing compression encoding and decompression decoding without adding instruction data. By comparison, in the latter case, if any of the allocation data becomes invalid due to an error, all encoded data becomes invalid and decoding becomes impossible, whereas in the former case, the allocation data at the beginning becomes invalid. The encoded data corresponding to the allocation data before the allocated allocation data can be decoded. Therefore, the use of the instruction data is very effective because the propagation of an error that has occurred in the allocation data can be greatly suppressed. Moreover, although such an effect can be obtained, the amount of data increased by adding the instruction data is very small.

In many cases, the coded data at the top is not invalidated. The coded data at the beginning is the coded data of the subband with a low frequency, and as described above, these coded data are more important to the auditory sense than the coded data of the subband with a high frequency. . Therefore, even if an error occurs in the allocation data, the quality of hearing is not deteriorated.

The instruction data may be located at any position in the data format, for example, multiplexed between allocation data or the like, and may be placed before or after the bit stream. In addition, the data recording / reproducing device 1
The recording system 22 and the reproduction system 26 may be removed from the device and used as a data encoding / decoding device. Further, the instruction data may be, for example, data indicating the start position of the encoded data by the number of bits from the start of the bit stream, or may be data indicating the number of allocation data whose value is 0. In particular, the latter is equivalent to data indicating the number of allocation data whose value is not substantially zero. In addition to the embodiments described above, the data encoding device 10 and the data decoding device 30 according to the present invention can be used in, for example, a data transmission device.

[0037]

As described above, according to the present invention, even if an error occurs in allocation data when compression coding is performed by the subband coding method and the compression-coded data is expanded and decoded. , It is possible to limit the error propagation for encoded data to a part of range. Therefore, according to the present invention, it is possible to minimize the propagation of an error that has occurred in the allocation data. Further, according to the present invention, it is possible to provide a data encoding / decoding device, a data recording device, and a data reproducing device with little data deterioration.

[Brief description of drawings]

FIG. 1 is a diagram showing a configuration of a data recording / reproducing device 1 using a data encoding device and a data decoding device according to the present invention.

FIG. 2 is a diagram illustrating subband coding,
(A) shows a baseband audio signal, (B) shows a bitstream after subband compression encoding,
(C) shows the data format of the bit stream.

FIG. 3 is a diagram showing a configuration of the data encoding device shown in FIG. 1.

4A and 4B are diagrams showing allocation data generated by the data encoding device shown in FIG. 3, in which FIG. 4A shows a data format and FIG. 4B shows a format of allocation data.

5 is a diagram showing a configuration of a data decoding device shown in FIG.

FIG. 6 is a diagram for explaining the operation of the timing generation circuit of the data decoding device shown in FIG.

7 is a diagram showing a range of encoded data that can be decoded by the data decoding device of the data recording / reproducing device shown in FIG. 1 when an error occurs in allocation data;
(A), (B), and (C) show cases where an error occurs in the allocation data of the 32nd, pth, and 1st subbands, respectively.

FIG. 8 is a diagram showing propagation of an error in data compression-coded by conventional subband coding.

FIG. 9 is a diagram showing error propagation when an error occurs in allocation data in data compression-coded by conventional subband coding.

[Explanation of symbols]

1 ... Data recording / reproducing device, 10 ... Data encoding device,
100 ... Timing generation circuit, 102 ... Subband coding circuit, 104 ... Data packing circuit, 106 ... Allocation data counting circuit, 20 ... ECC coding device,
22 ... Recording system, 24 ... Recording medium, 26 ... Reproduction system, 28 ...
ECC decoding device, 30 ... Data decoding device, 300 ... Timing generation circuit, 302 ... Data unpacking circuit,
304 ... Conceal circuit, 306 ... Subband decoding circuit

Claims (7)

[Claims] 1. Baseband data is divided into a predetermined number of subbands, compression-encoded, and fixed-length allocation data indicating bit allocation of each of the subbands,
Scale factor data indicating the maximum value of the encoded data of only the subband to which bits are allocated,
Coding means for generating coded data of only subbands to which bits are allocated, the allocation data for each data block,
Data multiplexing means for generating a bitstream in which the scale factor data and the encoded data are associated with each other and are multiplexed according to a predetermined data format, and an instruction indicating a position of the encoded data in the data format. A data encoding device having instruction data generating means for generating data. 2. The data encoding device according to claim 1, wherein the instruction data generating means generates data, which substantially indicates the number of subbands to which bits are allocated, as the instruction data. 3. A data decoding device for decoding the baseband data from a bit stream generated by the data coding device according to claim 1 or 2 and the instruction data, wherein the data decoding device is configured to perform the operation based on the instruction data. Separation means for separating the encoded data from the bitstream, and the separated encoding using the allocation data and the corresponding scale factor data in order from the beginning until the allocation data becomes invalid first. A data decoding device having a decompression / decoding means for decompressing / decoding data. 4. The data decoding according to claim 3, wherein said separating means detects the start position of the encoded data in said data format based on said instruction data and separates said encoded data from said bitstream. apparatus. 5. A data encoding / decoding device comprising the data encoding device according to claim 1 or 2, and the data decoding device according to claim 3 or 4. 6. The data encoding device according to claim 1 or 2, and recording means for recording the allocation data, the scale factor data, the encoded data and the instruction data on a predetermined recording medium. Data recording device. 7. The data decoding device according to claim 3 or 4, wherein the allocation data, the scale factor data, the encoded data and the instruction data recorded on a predetermined recording medium are read from the recording medium. And a data reproducing device having a reproducing means for reproducing.