KR100321300B1 - Signal Encoding Device, Signal Decoding Device, Signal Encoding Method, Signal Decoding Method - Google Patents

Abstract

It is an object to facilitate error correction in shuffling signals of a signal encoding apparatus, a signal decoding apparatus, a signal encoding method, a signal decoding method, and a signal recording medium.

The shuffling circuits 22 and 25 of the encoding apparatus perform shuffling of input data, and enable data interpolation that is not understood due to a burst error or the like. The frame counters 21 and 24 allow the continuity of the frame to be held in advance as a consecutive number on the frame header. The decoding apparatus detects a missing data due to a burst error or the like in the continuity of the frame number on the frame header and corrects an error from the data after deshuffling.

Description

Signal encoding apparatus, signal decoding apparatus, signal encoding method, signal decoding method

The present invention relates to a coding scheme, a decoding scheme, an encoding apparatus, a decoding apparatus, a recording medium and an apparatus for recording, reproducing and decoding, which are suitable for moving picture and audio data compression.

8 and 9 show examples of the structure of a conventional coding apparatus and decoding apparatus.

In FIG. 8, the audio signal is compressed and simultaneously encoded by the audio encoder 1 and input to the data multiplexing circuit 3. The video signal is compressed by the video encoder 2 and simultaneously encoded and input to the data multiplexing circuit 3. The data multiplexing circuit 3 adds an audio packet header to the encoded audio signal to form an audio packet, and adds a video packet header to the encoded video signal to form a video packet and multiplexes it.

Here, the MPEG (Moving Picture Experts Group) method is widely known as a compression encoding method of a video signal and an audio signal. Regarding the video signal, the MPEG method first reduces the redundancy in the time axis direction by taking a difference between picture frames of the video signal, and then uses an orthogonal transform method such as discrete cosine transform (DCT) to spatially It reduces the redundancy in the axial direction. In this way, the video signal can be efficiently encoded and recorded on a predetermined recording medium.

In the case of reproducing a recording medium on which a high-efficiency encoded video signal is recorded in this manner, the video signal can be reproduced by efficiently decoding the reproduced signal by inverse orthogonal transformation or the like.

Data multiplexed with audio and video by the data multiplexing circuit 3 is input to a DSM (Data Storage Memory) 4 and stored once. The above operation is continued until the input of the video signal and the audio signal is completed. The multiplexed data, once stored in the DSM 4, is input to an ECC (error correction code) encoder 5 together with additional information necessary for decoding, and after a predetermined amount of redundant data (parity) is added, the modulation circuit 6 Is entered. Data modulated by the modulation circuit 6 is recorded by the recording device 7 on a recording medium 8 such as an optical disk.

The multiplexed data recorded on the recording medium 9 in FIG. 9 is reproduced by the reading device 10 and input to the demodulation circuit 11. The multiplexed data demodulated by the demodulation circuit 11 is input to the ECC decoder 12 to perform error detection and error correction of the data. The error corrected multiplexed data is once input to the ring buffer device 13, stored, and read again, and then input to the data separation circuit 14. The input multiplexed data is separated into audio data and video data by the data separation circuit 14, and the audio data is input to the audio decoder 15, decoded, and an audio signal is output. The video data is input to the video decoder 16 and decoded to output an image signal.

However, there is a need to multiplex high-quality, high-quality data that is not compressed in the multiplexed bit stream, that is, uncompressed data, in the MPEG method. At this time, in multiplexing uncompressed data, it is necessary to efficiently perform error correction when a multiplexed uncompressed data frame is dropped by an error in consideration of the characteristics of uncompressed data and multiplexing. The conventional coding apparatus and decoding apparatus have a problem that the above request cannot be realized.

The signal encoding apparatus of the present invention includes encoding means for encoding an input signal into an encoded signal in units of frames, counting means for counting the number of frames of the encoded signal, and shuffling means for shuffling the encoded signals in frame units. The encoding means writes the number of frames in each frame header of the frame unit.

In addition, the signal encoding apparatus of the present invention includes encoding means for encoding an input signal into encoded signals in units of frames, and shuffling means for shuffling encoded signals in units of frames, wherein the encoding means is used for the shuffling. The shuffling parameter is written in each frame header of the frame unit.

Further, the shuffling means of the present invention performs the shuffling based on fixed shuffling parameters, respectively, according to the type of application, and the shuffling parameters include information indicating the size of one sample data of the coded signal and a delay size. And at least one of information indicating whether or not data of a sample point to be processed first is delayed, and information indicating the number of divisions when dividing the coded signal data into a plurality of frames.

In addition, the shuffling means of the present invention performs the shuffling when the encoding means performs uncompressed encoding or interpolation in an adjacent sample with respect to the input signal, and the frame header has a predetermined word size. It has a sync word formed by combining a negative maximum number and a positive maximum number.

In addition, the shuffling means of the present invention is that the dispersion of the data of the coded signal The shuffling is performed so as to have a structure, and the shuffling is performed so that the distribution of data for a plurality of frames of the coded signal is completed in the plurality of frames.

Further, the shuffling means of the present invention performs the shuffling so that the head of the access unit and the head of the completed plurality of frames coincide.

In addition, the signal decoding apparatus of the present invention includes decoding means for decoding a coded signal in units of frames to generate a decoded signal, deshuffling means for deshuffling the decoded signal, and frame header for each coded signal in units of frames. Detection means for detecting an error frame based on the frame number included in the frame and the number of frames decoded by the decoding means, and interpolation means for interpolating the error frame in the deshuffled decoded signal.

In addition, the signal decoding apparatus of the present invention decodes the decoded signal based on decoding means for decoding the coded signal on a frame basis to generate a decoded signal, and on the shuffling parameter included in each frame header of the coded signal on a frame basis. Deshuffling means for shuffling, and interpolation means for interpolating an error frame in the deshuffled decoded signal.

The deshuffling means may perform the deshuffling based on a fixed shuffling parameter according to the type of application, wherein the shuffling parameter is information indicating a size of one sample data of the coded signal. At least one of information indicating a delay size, information indicating whether data of a sample point to be processed first is delayed, and information indicating the number of divisions when dividing the data of the coded signal into a plurality of frames. Error correction means for performing error correction on the coded signal, wherein the interpolation means performs the interpolation on data in which an uncorrectable error has occurred in the error correction means.

Further, the signal decoding apparatus of the present invention includes detection means for detecting an error frame based on the frame number included in each frame header of the frame-coded signal and the number of frames decoded by the decoding means. Interpolation means performs the interpolation on the error frame detected by the detection means.

Further, in the signal decoding apparatus of the present invention, the decoding means detects each of the frame headers by a sync word configured by combining a maximum number of negatives and a positive maximum number in a predetermined word size.

Further, the signal encoding method of the present invention encodes an input signal into an encoded signal in units of frames, counts the number of frames of the encoded signal, shuffles the encoded signal in units of frames, and sets the number of frames in units of frames. Write in each frame header.

In addition, the signal encoding method of the present invention encodes an input signal into a coded signal in units of frames, shuffles the coded signals in units of frames, and sets a shuffling parameter used for the shuffling to each frame header of the frames. Fill in.

The signal encoding method of the present invention is characterized in that the shuffling is set according to the type of application and is performed based on fixed shuffling parameters, respectively, wherein the shuffling parameter is the size of one sample data of the coded signal. At least one of information indicating a delay, information indicating a delay size, information indicating whether data of a sample point to be processed first is delayed, and information indicating a number of divisions when data of the encoded signal is divided into a plurality of frames. The shuffling is performed when uncompressed encoding or compression encoding that is interpolated in adjacent samples is performed on the input signal.

In the signal encoding method of the present invention, the frame header has a sync word formed by combining a maximum number of negative parts and a positive maximum number at a predetermined word size, and the shuffling is performed so that the distribution of data of the coded signal becomes an embedded structure. And the shuffling is performed so that distribution of data for a plurality of frames of the coded signal is completed in the plurality of frames.

Further, the signal encoding method of the present invention performs the shuffling so that the head of the access unit and the head of the completed plurality of frames coincide.

The signal decoding method of the present invention generates a decoded signal by decoding a coded signal on a frame basis, deshuffles the decoded signal, and includes a frame number included in each frame header of the coded signal on a frame basis, and An error frame is detected based on the number of frames decoded by the decoding means, and the error frame is interpolated in the deshuffled decoded signal.

In addition, the signal decoding method of the present invention decodes a coded signal on a frame basis to generate a decoded signal, and deshuffles the decoded signal on the basis of shuffling parameters included in each frame header of the coded signal on a frame basis. And interpolate an error frame in the deshuffled decoded signal.

Further, the signal decoding method of the present invention performs the deshuffling based on fixed shuffling parameters, respectively, according to the type of application, wherein the shuffling parameters are information indicating a size of one sample data of the coded signal and a delay size. At least one of information indicating whether the data of the sample point to be processed first is delayed, and information indicating the number of divisions when dividing the data of the coded signal into a plurality of frames. Error correction is performed, and the interpolation is performed on data in which an uncorrectable error has occurred in the error correction.

Further, the signal decoding method of the present invention detects an error frame based on the frame number included in each frame header of the coded signal in the frame unit and the number of frames decoded by the decoding means. The interpolation is performed on the error frame.

Further, the signal decoding method of the present invention detects each of the frame headers by a sync word formed by combining a negative maximum number and a positive maximum number at a predetermined word size.

According to the encoding method of the present invention, an error can be corrected by detecting the dropping of a frame due to a burst error and interpolating the detected dropping frame by the deshuffling and adjacent sample data. In addition, according to the encoding method of the present invention, the shuffling method itself can be transmitted.

In addition, according to the encoding method of the present invention, it is possible to vary the minimum unit of relocation during shuffling. Further, according to the encoding method of the present invention, the distance between the data to be rearranged at the time of shuffling can be varied.

Embodiments according to the present invention will be described with reference to the drawings. In the present invention, the basic compression encoding method is the same as the MPEG encoding method.

First, the first embodiment will be described. First, the encoder will be described, and then the decoder will be described. Finally, bit strip syntax, shuffling, and deshuffling operations are described.

1 shows a first embodiment of a signal encoding apparatus according to the present invention. Since the same reference numerals can be used for the same reference numerals as in FIG. 8 shown in the example of Fig. 1, detailed descriptions are omitted. In FIG. 1 an audio signal is input to an audio encoder 28. At the same time, a signal for selecting to transmit the audio signal by either compression or uncompression increase, and a parameter at the time of shuffling are input to the shuffling control circuit 20. The shuffling control circuit 20 inputs a control signal to the audio encoder 28 and the shuffling circuit 22 according to the input parameter. In this embodiment, the shuffling circuit 22 performs a shuffling operation when uncompressed encoding is selected.

The audio inloader 28 compresses or decompresses the input audio signal according to the control input from the shuffling control circuit 20 and outputs the encoded data to the shuffling circuit 22. The audio encoder 28 simultaneously writes normal header information specified in MPEG into the frame header when compression encoding is selected, and audio frame count 21 whenever a predetermined audio frame is encoded when uncompressed encoding is selected. Increments and writes the count value and the shuffling parameter into the frame header. The shuffling circuit 22 shuffles the bit stream from the audio encoder 28 based on the shuffling parameter from the shuffling control circuit 20 when uncompressed encoding is selected, and when compressed encoding is selected. Outputs the strip as it is.

In FIG. 1 the video signal is input to video encoder 29. At the same time, a signal for selecting whether to transmit the video signal either compressed or uncompressed, and a parameter for shuffling are input to the shuffling control circuit 23. The shuffling control circuit 23 inputs a control signal to the video encoder 29 and the shuffling circuit 25 according to the input parameter. In the present embodiment, when uncompressed encoding is selected, the shuffling circuit 25 is to perform a shuffling operation.

The video encoder 29 inputs, to the shuffling circuit 25, data that is compressed or uncompressed and encoded by the video signal input in accordance with the control input from the shuffling control circuit 23. The video encoder 29 simultaneously writes normal header information specified in MPEG into the frame header when compression encoding is selected, and the video frame counter 24 whenever a predetermined video frame is encoded when uncompression encoding is selected. Increments and writes the count value and the shuffling parameter into the frame header. The shuffling circuit 25 shuffles the bit stream from the video encoder 29 based on the shuffling parameter from the shuffling control circuit 23 when uncompressed encoding is selected, and when compressed encoding is selected. Outputs the bit stream as it is.

Audio data shuffled or unshuffled by the shuffling circuit 22 and video data shuffled or unshuffled by the shuffling circuit 25 are input to the data multiplexing circuit 3.

The data multiplexed by the data multiplexing circuit 3 on time-multiplexed audio and video is input to the DSM 4 and stored once. The above operation is continued until the input of the video signal and the audio signal is completed. The multiplexed data once stored in the DSM 4 is input to the ECC encoder 5 together with additional information necessary for decoding, and is then input to the modulation circuit 6 after a predetermined amount of redundant data (parity) is added. Data modulated by the modulation circuit 6 is recorded on the recording medium 8 by the recording apparatus 7.

2 shows a first embodiment of the signal decoding apparatus according to the present invention. In Fig. 2, the same reference numerals as those in Fig. 11 shown in the prior art can be used, and detailed description thereof will be omitted.

In FIG. 2, the multiplexed data recorded on the recording medium 9 is reproduced by the reading apparatus 10 and input to the demodulation circuit 11. The multiplexed data demodulated by the demodulation circuit 11 is input to the ECC decoder 12 to perform error detection and error correction of the data. The error corrected multiplexed data is once inputted to the ring buffer device 13, stored therein, then read again, and inputted to the data separation circuit 14. The input multiplexed data is separated into audio data and video data by the data separating circuit 14, and the audio data is input to the audio decoder 40 and decoded. The video data is input to the video decoder 41 and decoded. The audio decoder 40 determines whether the target audio frame is compressed or uncompressed data in the frame header of the input bit stream, and if it is uncompressed data, writes the deshuffling control circuit 42 on the audio frame. The completed shuffling parameters. In addition, the audio decoder 40 loads the frame number read from the audio frame header of the first frame into the audio frame counter 44, and increments the audio frame counter 44 whenever the next audio frame is decoded. I do it. The count value of the audio frame counter 44 and the frame number read out from the audio frame header by the audio decoder 40 are input to the comparison circuit 46.

The data decoded by the audio decoder 40 is input to the deshuffling circuit 48. The deshuffling circuit 48 receives the control signal from the deshuffling control circuit 42 to perform the deshuffling operation when the decoded data is the shuffled data, and the rearranged data is sent to the interpolation circuit 52. Is entered. The comparison circuit 46 compares the read frame number with the audio frame counter value. When the value is different, the comparison circuit 46 determines that the target audio frame is dropped due to an error, and inputs the dropped frame detection signal to the interpolation circuit 52.

When the interpolation circuit 52 receives the dropped frame detection signal, the interpolation circuit 52 performs interpolation processing from a sample adjacent to the sample dropped from the signal input from the deshuffling circuit 48 to output an audio signal with error correction.

The video decoder 41 determines in the frame header of the input bit stream whether the target video frame is compressed or uncompressed data, and if it is uncompressed data, writes the deshuffling control circuit 43 on the video frame. The completed shuffling parameters. In addition, the video decoder 41 loads the frame number read from the video frame header of the first frame into the video frame counter 45 and increments the video frame counter 45 each time the next video frame is decoded. do.

The count value of the video frame counter 45 and the frame number read out from the video frame header by the video decoder 41 are input to the comparison circuit 47. Data decoded by the video decoder 41 is input to the deshuffling circuit 49. The deshuffling circuit 49 receives the control signal from the deshuffling control circuit 43 and performs the deshuffling operation when the decoded data is the shuffled data. The rearranged data is input to the interpolation circuit 53.

The comparison circuit 47 compares the read frame number with the video frame counter value t and determines that the target video frame has dropped due to an error when the value is different, and inputs a drop frame detection signal to the interpolation circuit 53. When the interpolation circuit 53 receives the drop frame detection signal, the interpolation circuit 53 performs an interpolation process based on, for example, an intermediate value of adjacent samples from a sample adjacent to the dropped sample from the signal input from the deshuffling circuit 49 to correct an error. The output video signal is output. In the above description, the dropped frame detection signal is obtained by the comparison circuits 46 and 47. However, the dropped frame detection signal may be output when the uncorrectable data is generated by the ECC decoder.

Although two types of shuffling circuits of an audio signal and a video signal are shown in this embodiment, both are substantially the same in terms of algorithm and hardware configuration, and differ only in the shuffling parameters to be given. Therefore, the following describes the method of shuffling audio data. That is, the following description can be applied to an audio signal and a video signal or a general signal.

3 illustrates an example of the shuffling method in the present invention. In FIG. 3, when the original data sample string (60 (for example, one word is one word)) is set to 10 to 49, the sample string 61 is caused by the shuffling circuit 22 at the time of encoding. Relocated shuffling data is shown. Sample points of odd number, such as input data 11, 13, 15, 17, etc., are delayed 10 samples by passing through the delay circuit 26 in the shuffling circuit 22. The other data, that is, the sample number of the even number, is directly output from the shuffling circuit 22 without passing through the delay circuit 26. Here, the shuffling parameters such as the delay amount and the odd sample data or the flag for delaying even sample data and the word size of one sample are specified at the time of encoding and recorded on the frame header. In addition, the frame header is directly output without passing through the delay circuit 26.

Next, an example of the deshuffling method at the time of decoding is shown. In FIG. 3, the sample string 62 represents shuffling data, and is the same as the shuffling data 61 shuffled at the time of encoding. However, assuming that a frame with sample data numbers 15, 26, 17, 28, and 19 was dropped due to a burst error, error correction will be described.

The deshuffled data 63 is a signal whose input data 62 has been rearranged by the deshuffling circuitry 48. The deshuffling parameter is a signal recorded in the frame header at the time of encoding, but at the time of shuffling, the data of the odd number is delayed and repositioned, whereas the even-numbered data, which is inverse parity, is a delay circuit in the deshuffling circuit 48. 10 samples are delayed after passing through 50, and rearrangement is performed in order of sample data before encoding.

The interpolation data 64 includes a signal interpolated from adjacent data of data dropped by the burst error. Since the sample points of errors are dispersed by shuffling and deshuffling, interpolation from two adjacent sample points becomes possible.

In this embodiment, two types of shuffling circuits of an audio signal and a video signal are shown. However, as described above, since both are substantially the same in terms of algorithm and hardware configuration, since only the shuffling parameters given are different, the uncompressed frame is summed up. I will explain in the form of. That is, the following description can be adapted not only to audio signals or video signals but also to signals such as measurement data and text data.

4 shows an example of a bit stream syntax for recording a shuffling parameter and a continuous frame number for detecting frame drop in the frame header in the present invention.

In FIG. 4, (a) is a bit stream syntax of an uncompressed data frame, that is, "non_compressed_data_frame". In the figure, "syncword" indicates either Ox80007fff (hexadecimal; "1000 0000 0000 0000 0111 1111 1111 1111" in binary) or Ox807f807f (hexadecimal; "1000 0000 0111 1111 1000 0000 0111 1111" in binary). When the detection was detected, it was confirmed that it was a frame header. The two sync words are each a combination of the maximum number (minimum) and the maximum number of negatives in 16 bit units, and the maximum number (minimum) and the maximum number of negatives in 8-bit units. It is unlikely to be mistaken as a sync word by accident.

"Flame_length" represents the length of the target uncompressed frame. "Continuity_counter" indicates a continuous frame for detecting dropped frames. "Data_descriptor_flag" is a flag indicating the presence or absence of a data descriptor. "Access_unlt_start_flag" is a flag indicating that the head of the access unit and the head of the frame match.

"Interleave_parameter_flag" indicates the presence or absence of a shuffling parameter. "Data_descriptor" is a parameter required for decoding an uncompressed frame and varies from application to application. "Interleave_unit_wordsize" is the minimum word size for shuffling. "Interleave_delay" is a delay amount necessary for performing a predetermined shuffling. "Delay_flag" is a flag indicating whether the leading data of the frame is delayed or transmitted without delay. "Data_bytes" is an uncompressed data string to be shuffled.

"L-PCM_descriptor" shown in (b) of FIG. 4 is an example of a bitstream syntax that shows the details of "data_descriptor" of FIG. 4 (a) when linear PCM audio is uncompressed. In terms of contents, the contents of "sampling_frequency" (sampling frequency) 2 bits, "copyright" (right) 1 bit, "original / home" 1 bit, "emphasis" (emphasis) 2 bits, and "reserved" 2 bits are 8 bits in total.

As a modification of the first embodiment, the shuffling parameter is fixed in advance for a specific application for each application, and a predetermined shuffling process can be realized without transmitting "interleave_parameter_flag". In this case, deshuffling is performed with a fixed shuffling parameter according to the type of application.

Next, another shuffling method will be described with reference to FIG. 5 as a second embodiment of the present invention. In the second embodiment, the shuffling circuit 30 shown in Fig. 6A is used instead of the shuffling circuit 22 of the first embodiment shown in Fig. 1 as the shuffling circuit. As the deshuffling circuit, a deshuffling circuit having the same configuration as that of the shuffling circuit 30 shown in FIG. 6A is used instead of the deshuffling circuit 48 of the first embodiment shown in FIG. In FIG. 5, when the original sample sequence 70 (for example, one sample is 1 word) is set from 10 to 49, the sample sequence 71 is rearranged by the shuffling circuit 30 at the time of encoding. The shuffled data is shown. Sample points of even numbers, such as input data 10, 12, 14, 16, etc., are delayed 10 samples by passing through the delay circuit 32 in the shuffling circuit 30. In addition, the sample points of the input data 11, 13, 15, 17, and 19, that is, the sample points of the odd number of the first frame and the second frame, pass through the delay circuit 31 in the shuffling circuit 30 to delay 20 samples. do. In addition, the sample points of the input data 21, 23, 25, 27, 29, that is, the sample points of the odd number of the third frame and the fourth frame, are assigned to the delay circuit 31 and the delay circuit 32 by the shuffling circuit 30. Direct output without passing through In decoding, on the other hand, the sample string 72 of FIG. 5 is rearranged like the sample string 73 by the deshuffling circuit. The relocation method is the same as in encoding, and the sample number of the even number is delayed by 10 samples, the sample point of the odd number is delayed by 20 samples for the first frame and the second frame, and the third frame and The fourth frame is output without delay. When there is a burst error, interpolation processing is performed on the deshuffled sample string 73 as shown in FIG. 3 to obtain the sample string 74. FIG.

Here, comparing the first and second embodiments, according to the shuffling method (FIG. 3) described in the first embodiment, the sample data distributed in multiple frames by shuffling has a so-called embedded structure, It does not have a structure completed in the frame. Therefore, when a predetermined access unit is accessed by random access or the like and decoded in the middle frame, the required sample point cannot be obtained in the first frame, so interpolation processing is necessary. For example, when the decoding starts in a frame including the sample points 30, 21, 32, 23, and 34 of the sample sequence 62 in FIG. 3, one of the sample points 21, 23 and one in the original sample sequence 60 is obtained. Since the sample points 20, 22, and 24 that constitute the frame have no data, interpolation processing is required.

On the other hand, according to the shuffling method according to the second embodiment (FIG. 5), since the original sample sequence 70 is shuffled as shown in the sample sequence 71, the shuffling is necessarily completed with 20 samples (4 frames). It is. Therefore, if shuffling is performed so that the head of the access unit and the head of the 20 samples coincide with each other during random access, the data is output without interpolation from the head of the access unit even when decoding starts in the middle frame due to random access or the like. can do. For example, when decoding is started in a frame including sample points of 30, 41, 32, 43, and 34 of the sample string 72 of FIG. 5, the original sample string is reproduced when data for 4 frames including the frame is reproduced. Since the continuous data (No. 30 to No. 49) for four frames (70) can be obtained, data can be output without performing interpolation processing. All other configurations can be the same as the first embodiment.

Next, Fig. 6B shows the configuration of the shuffling circuit and the deshuffling circuit in accordance with the third embodiment of the present invention. Here, the shuffling circuit and the deshuffling circuit have the same configuration. According to the first embodiment, when shuffling, the sample data string of one frame is rearranged and distributed into two frames depending on the presence or absence of a delay. In the third embodiment, as shown by reference numerals 91, 93, and 95 in FIG. 6B, three delay circuits are provided, and one frame of sample data can be distributed into four frames. The output buffers 92, 94, 96, and 99 are buffers for adjusting the phase of the output data. If the number of divisions is added to the frame header within 4 divisions as a parameter of shuffling, shuffling of any number of divisions can be realized. The delay channel can also be increased like the delay circuit 97 and the output buffer 98 shown by the dotted lines in FIG. 6B. In this case, arbitrary division number shuffling and deshuffling can be performed.

The shuffling method of the third embodiment will be described according to FIG. The original data sample 80 is shuffled every four samples and distributed over four frames (sample sequence 81). Here, the representation of the data will be changed as shown by reference numeral 82. That is, the third embodiment is equivalent to scanning the raster format data in the direction perpendicular to the raster direction and changing the shuffling order. Thus, it is possible to adapt to shuffling of two-dimensional data. All other configurations are the same as those in the first embodiment and are omitted. In the above-described embodiment, the shuffling is performed only on the uncompressed data. However, when the compression method capable of interpolating between adjacent samples is used, the shuffling may also be performed on the compressed data.

According to the encoding method of the present invention, data that has been replaced by so-called shuffling of sample data over a plurality of frames with respect to uncompressed data is transmitted, and when an error occurs, the influence of a part dropped by a burst error is distributed, In the encoding method of performing interpolation processing on the basis of deshuffled data in adjacent frames of the apparatus, header information for transmitting a continuous value for each frame for each packet type is prepared, and a counter for counting the number of transmission frames is prepared for the continuous value. Prepare a comparator that compares the header information with the counter value and checks the discontinuity of the header information which should be a continuous value, detects the dropping of the frame due to a burst error, and detects the dropped frame from the deshuffling and By interpolation, the error can be corrected.

In addition, according to the encoding method of the present invention, the detection of dropped frames in the error correction by the shuffling can be controlled according to a control signal from an ECC decoder regardless of the detection of the discontinuity of the frames.

Further, according to the encoding method of the present invention, the shuffling method itself can be transmitted by recording the parameters necessary for the shuffling operation in the frame header at the time of encoding.

Further, according to the encoding method of the present invention, the minimum unit of relocation during shuffling can be changed by transmitting the size of one sample data as a parameter required for the shuffling operation.

Further, according to the encoding method of the present invention, a delay size is transmitted as a parameter necessary for the shuffling operation to vary the distance between the data to be rearranged during shuffling.

Further, according to the encoding method of the present invention, a sampling data string subjected to a shuffling process performed by switching between an output after performing a delay with respect to an input sample string in the shuffling operation and an output without delay is first performed. A flag for setting whether or not the sample point data to be processed is output after performing a delay or outputting without delay is transmitted to enable correct deshuffling.

Further, according to the encoding method of the present invention, in all uncompressed frames that are completed shuffling in a plurality of frames, a random access is performed by encoding so that the head of the data in the frame always coincides with the head of the processing unit so-called access unit suitable for handling the input data. Even if the data is not interpolated, data can be output.

In addition, according to the encoding method of the present invention, the transmission of the shuffling method by the above parameters can be limited for specific application purposes, thereby fixing the parameters to improve efficiency.

In addition, according to the encoding method of the present invention, the establishment of emulation is reduced by using a combination of values that are unlikely to occur at a predetermined word size as a sync word for confirming that the data string to be processed in the bit stream is a frame header. In this case, the frame header can be reliably detected at the time of decoding, and correct deshuffling can be performed.

In addition, according to the encoding method of the present invention, in the shuffling method, the number of divisions when shuffling can be varied by transmitting the number of divisions when distributing consecutive input sample data into a plurality of frames in a frame header.

1 is a block diagram showing a configuration example of an encoder device according to the present invention.

2 is a block diagram showing a configuration example of a decoder device according to the present invention.

3 is a diagram showing a shuffling method of a first embodiment in the present invention.

4 is a diagram showing an example of a bit stream syntax in the present invention.

5 is a diagram showing a shuffling method of a second embodiment in the present invention.

FIG. 6 shows a shuffling circuit of a third embodiment in the present invention. FIG.

7 is a diagram showing a shuffling method of a third embodiment in the present invention;

8 is a diagram showing the configuration of a conventional signal encoding apparatus.

9 is a diagram showing a configuration of a conventional signal decoding apparatus.

Explanation of symbols for the main parts of the drawings

3: data multiplexing circuit 4: DSH

5: ECC encoder 6: modulation circuit

7: recording device 8: recording medium

20, 23: shuffling control circuit 21: audio frame counter

22, 25: shuffling circuit 26, 27: delay circuit

28: Audio Encoder 29: Video Encoder

Claims (42)

In the signal encoding apparatus for encoding an input signal, Shuffling control means for inputting a compression / uncompression selection signal and a shuffling parameter; Encoding means for encoding an input signal according to a control input from the shuffling control means into an encoded signal in units of frames; Counting means for counting the number of frames of the coded signal; And A shuffling means for shuffling the coded signal in units of frames based on the shuffling parameter, And the encoding means writes the number of frames in each frame header of the frame unit. In the signal encoding apparatus for encoding an input signal, Suffling control means for inputting a compression / uncompression selection signal and shuffling parameters; Encoding means for encoding an input signal according to a control input from the shuffling control means into an encoded signal in units of frames, and A shuffling means for shuffling the coded signal in units of frames based on the shuffling parameter, And the encoding means writes a shuffling parameter used for the shuffling into each frame header of the frame unit. The signal encoding apparatus according to claim 1, wherein the shuffling means performs the shuffling based on fixed shuffling parameters according to the type of application. The method according to claim 2, wherein the shuffling parameter includes information indicating a size of one sample data of the coded signal, information indicating a delay size, information indicating whether or not data of a sample point to be processed first is delayed, And at least one piece of information indicating the number of divisions when dividing the data into a plurality of frames. The signal encoding apparatus according to claim 1, wherein the shuffling means performs the shuffling when the encoding means performs uncompressed encoding or compression encoding which can be interpolated from adjacent samples with respect to the input signal. 2. The signal encoding apparatus of claim 1, wherein the frame header has a sync word configured by combining a maximum number of negatives and a maximum positive number at a predetermined word size. The signal encoding apparatus according to claim 1, wherein the shuffling means performs the shuffling so that a distribution of data of the coded signal is embedded. The signal encoding apparatus according to claim 1, wherein the shuffling means performs the shuffling so that variance of plural frame data of the coded signal is completed in the plural frames. 9. The signal encoding apparatus according to claim 8, wherein the shuffling means performs the shuffling so that a head of an access unit and a head of the completed plurality of frames coincide. In the signal decoding apparatus for decoding a received signal, Decoding means for decoding a coded signal on a frame basis to generate a decoded signal; Deshuffling means for deshuffling the signal decoded by the decoding means; Detection means for detecting an error frame based on a frame number included in each frame header of the coded signal in frame units and the number of frames decoded by the decoding means; And And interpolation means for interpolating the error frame from the deshuffled decoded signal. In the signal decoding apparatus for decoding a received signal, Decoding means for decoding a coded signal on a frame basis to generate a decoded signal; Deshuffling means for deshuffling the signal decoded by the decoding means based on a shuffling parameter included in each frame header of the coded signal in units of frames; And And interpolation means for interpolating the error frame from the deshuffled decoded signal. 11. The signal decoding apparatus according to claim 10, wherein the deshuffling means performs the deshuffling based on fixed shuffling parameters, respectively, according to the type of application. The method according to claim 11, wherein the shuffling parameter includes information indicating a size of one sample data of the coded signal, information indicating a delay size, information indicating whether or not data of a sample point to be processed first is delayed, And at least one piece of information indicating the number of divisions when dividing data into a plurality of frames. 12. The signal decoding according to claim 11, further comprising error correction means for performing error correction on said coded signal, wherein said interpolation means performs said interpolation on data in which an uncorrectable error has occurred in said error correction means. Device. The signal decoding apparatus according to claim 11, wherein the interpolation means performs the interpolation on the error frame detected by the detection means. 11. The signal decoding apparatus according to claim 10, wherein the decoding means detects each of the frame headers by a sync word configured by combining a maximum number of negatives and a positive maximum number of predetermined word sizes. In the signal encoding method for encoding an input signal, Encoding the input signal into a coded signal in units of frames, Counting the number of frames of the encoded signal; Shuffling the encoded signal in units of frames, and And writing the number of frames in each frame header of the frame unit. In the signal encoding method for encoding an input signal, Encoding the input signal into a coded signal in units of frames, Shuffling the encoded signal in units of frames, and And writing a shuffling parameter used for the shuffling into each frame header of the frame unit. 18. The signal encoding method of claim 17, wherein the shuffling is set according to a type of application and is performed based on fixed shuffling parameters, respectively. 19. The method of claim 18, wherein the shuffling parameter includes information indicating a size of one sample data of the coded signal, information indicating a delay size, information indicating whether or not data of a sample point to be processed first is delayed, And at least one piece of information indicating the number of divisions when dividing the data into a plurality of frames. 18. The signal encoding method according to claim 17, wherein the shuffling is performed when performing uncompressed encoding or compression encoding which can be interpolated from adjacent samples with respect to the input signal. 18. The method of claim 17, wherein the frame header has a sync word configured by combining a maximum number of negatives and a maximum number of positives at a predetermined word size. 18. The signal encoding method according to claim 17, wherein the shuffling is performed so that the distribution of data of the encoded signal becomes an embedded structure. 18. The signal encoding method according to claim 17, wherein the shuffling is performed so that distribution of data for a plurality of frames of the coded signal is completed in the plurality of frames. The signal encoding method according to claim 24, wherein the shuffling is performed such that a head of an access unit and a head of the plurality of completed frames coincide with each other. In the signal decoding method for decoding a received signal, Generating a decoded signal by decoding a coded signal on a frame basis; Deshuffling the decoded signal; Detecting an error frame based on a frame number included in each frame header of the encoded signal in the frame unit and the number of frames decoded by the decoding means, and Interpolating the error frame from the deshuffled decoded signal. In the signal decoding method for decoding a received signal, Generating a decoded signal by decoding a coded signal on a frame basis; Deshuffling the decoded signal based on a shuffling parameter included in each frame header of the coded signal in units of frames; and Interpolating an error frame from the deshuffled decoded signal. 27. The signal decoding method according to claim 26, wherein said deshuffling is performed based on fixed shuffling parameters, respectively, according to the type of application. 29. The method of claim 27, wherein the shuffling parameter includes information indicating a size of one sample data of the coded signal, information indicating a delay size, information indicating whether or not data of a sample point to be processed first is delayed, And at least one piece of information indicating the number of divisions when dividing data into a plurality of frames. The method of claim 27, wherein error correction is performed to the coded signal, And the interpolation is performed on data in which an uncorrectable error has occurred in the error correction. The method according to claim 27, wherein an error frame is detected based on a frame number included in each frame header of the coded signal in the frame unit and the number of frames decoded by the decoding means, And the interpolation is performed on the detected error frame in the detection step. 27. The signal decoding method according to claim 26, wherein each of said frame headers is detected by a sync word formed by combining a negative maximum number and a positive maximum number at a predetermined word size. 3. The signal encoding apparatus according to claim 2, wherein the shuffling means performs the shuffling when the encoding means performs uncompressed encoding or compression encoding that can be interpolated from adjacent samples with respect to the input signal. The signal encoding apparatus according to claim 2, wherein the frame header has a sync word configured by combining a maximum number of negatives and a maximum positive number at a predetermined word size. 3. The signal encoding apparatus according to claim 2, wherein the shuffling means performs the shuffling so that a distribution of data of the coded signal is embedded. 3. The signal encoding apparatus according to claim 2, wherein the shuffling means performs the shuffling so that variance of data for a plurality of frames of the coded signal is completed in the plurality of frames. 12. The signal decoding apparatus according to claim 11, wherein the decoding means detects each of the frame headers by a sync word configured by combining a maximum number of negatives and a positive maximum number of predetermined word sizes. 19. The signal encoding method according to claim 18, wherein the shuffling is performed when performing uncompressed encoding or compression encoding which can be interpolated from adjacent samples. 19. The signal encoding method of claim 18, wherein the frame header has a sync word configured by combining a negative maximum number and a positive maximum number at a predetermined word size. 19. The signal encoding method according to claim 18, wherein the shuffling is performed so that the distribution of data of the encoded signal becomes an embedded structure. 19. The signal encoding method according to claim 18, wherein the shuffling is performed so that distribution of data for a plurality of frames of the coded signal is completed in the plurality of frames. 28. The signal decoding method according to claim 27, wherein each frame header is detected by a sync word formed by combining a negative maximum number and a positive maximum number at a predetermined word size.