JP2003111018A - Recording and reproducing device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To revive and reproduce data recorded on a recording medium by tracing the data back to a time when a defect takes place even when the data having been not recognized because the defect takes place during recording resulting in impossible recording of management data. SOLUTION: The recording and reproducing device adopts a constant data management unit (cluster) used for writing/reading data, applies compression processing to data divided into data having the management unit or below, and records data resulting from attaching particular pattern information to the data after the compression processing, references information at a particular position in the management unit of the read data and applies revising processing to the data when the information is the particular pattern (MPEG header part).

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a recording / reproducing apparatus, and more particularly to a recording / reproducing apparatus capable of reproducing recorded data even when a problem occurs during recording.

[0002]

2. Description of the Related Art As an apparatus for dividing an image signal or an audio signal into predetermined units and performing processing such as compression, MPEG (M
Video recording and recording devices using video compression and audio compression of the Otion Picture Experts Group) are known. In these devices, the data management of the recording medium used is DO
The data management method is standardized by ISO / IEC-9293 or the like. Further, a technical magazine or the like describes the data management method in DOS. (See CQ Publisher, Tora Gi Computer 1991.1, etc.) Here, an outline of a data management method in DOS will be described. An IPL (Initial Program) is set in the first area of the recording medium.
Loader) area, in which information related to the recording medium (maker name, version information, information necessary to access the recording medium (number of heads, number of logical sectors, capacity per sector, etc.), Information such as how the management information to be recorded is recorded. IPL
Next to the area, there is a FAT (File Allocation Table) area, and in the FAT area, which area of the recording medium the data recorded on the recording medium uses is recorded.

Further, there is a directory area following the FAT area, and information about a file (data) recorded on a recording medium (file name, file creation date, type, size, and FAT area used by the file). (Start address, etc.) is recorded. Further, there is a data area following the directory area, in which actual data is recorded. In the recording medium, the unit of writing and reading of data (cluster) is determined, and the size of the cluster is generally 2048 Bytes or 4096 Bytes.

When recording data on a recording medium, the size of one data is usually larger than the size of a cluster.
Therefore, the data is recorded as one file across a plurality of clusters. To read the data recorded across multiple clusters, access the directory area, search the area where the information about any file is recorded, and read the top address of the FAT area used by that file. , The corresponding address in the FAT area is accessed. At this address
In the recording medium, the cluster numbers used by the file are recorded in order, and one file can be read by sequentially accessing the corresponding cluster numbers on the recording medium. Further, by reading the information regarding all the files recorded in the directory area and searching the FAT area from the information, the used cluster can be identified.

Further, since the number of logical sectors of the recording medium is recorded in the IPL area and the number of clusters on the recording medium is known, if the used cluster is known, the unused cluster can be known.

Next, in the actual recording / reproducing apparatus, recording / recording is performed.
The outline of the processing until recording is performed and the recorded data can be recognized as one file will be described.

First, according to the procedure described above, an unused cluster on the recording medium is searched in advance before recording is started, and data is written to the searched unused cluster. At this time, the cluster to which the data was written is recorded in the work memory. Also, when recording / stop recording, FAT
By recording the number and order of the cluster that wrote the data in the free area in the area, and recording the arbitrary file name and management data related to the file in the free area in the directory area, it becomes possible to recognize as one file. .

By the way, during recording by the recording / reproducing apparatus, power failure (battery exhaustion) or some trouble occurs,
When recording is finished without writing the management data such as the cluster number used in the FAT area or the directory area (hereinafter referred to as the management area) and the file name,
Since the data recorded from the start of recording to the time when the trouble occurs cannot be recognized as a file, it cannot be reproduced.

[0009]

As a method of preventing such a recording failure, there is a method of backing up a work memory for temporarily recording data to be recorded in the management area with a battery or a large capacity capacitor. In this method, a flag is set in a specific area of the work memory during recording, and the flag is turned off when the management data can be recorded normally. Therefore, even if the management data could not be written to the management area due to a problem during recording, the flag is referenced when the device is restarted, and if the flag is set, the management data is normally written to the management area. It is possible to judge that the processing has ended without doing so, and add the management data backed up in the work memory to the management area. However, in this method, it is essential that the management data is backed up normally, for example, when the microcomputer for system control is latched up and the management data is not recorded normally in the work memory, or when the backup battery is used. In case of a break, management data cannot be restored. Further, if the medium is replaced before the device is restarted, there is a risk that unintended management data may be written on the replaced medium.

As a method of preventing the recording failure, there is a method of recording the management data into a plurality of files by recording the management data on the medium at regular intervals. With this method, it is necessary to create the management data and record it on the medium so that the written data can be recognized as a file by utilizing the interval between the data compression and the data writing process after the data compression. Alternatively, the data writing process after compression may not be in time.

Further, in order to process the divided files as one file, it is necessary to have management information uniquely, and an extra area for data management is required. In addition, there is a problem that overhead is increased because a plurality of files must be handled when reproducing. Further, even if the data is divided into a plurality of files and recorded, the data immediately before the trouble occurs and before the management data is written cannot be reproduced.

The present invention has been made in view of these problems, and a problem occurs during recording, management data cannot be recorded, and even data that cannot be recognized as data is recorded on a recording medium by the time the problem occurs. There is provided a recording / reproducing apparatus capable of resuming and reproducing the data recorded in (1).

[0013]

The present invention adopts the following means in order to solve the above problems.

Data management units (clusters) used when writing and reading data are constant, and the data divided into the data management units or less are subjected to compression processing, and specific pattern information is added to the data after the compression processing. A recording / reproducing apparatus for recording the added data (frame), the recording / reproducing apparatus refers to information of a specific position in a management unit of the read data, and the information is the specific pattern (MPEG header section). When the data is restored.

[0015]

BEST MODE FOR CARRYING OUT THE INVENTION There are various image or audio compression systems, and various media are used as recording media. MPE, which is an international standard for audio compression systems, is used.
Using G (Motion Picture Experts Group) audio,
PCMCIA-ATA (Personal Computer M
emoryCard International Association-AT Attachment)
A case where a flash memory card is used will be described as an example.

The PCMCIA-ATA card has an integrated controller (IDE) integrated by incorporating a dedicated controller.
It can control the same as a hard disk of evice E Electronics. PCMCIA and PCMCIA-A
The details of TA are published by PCMCIA and Japan Electronic Industry Development Association (JEIDA). PC
For details of CARD STANDARD and 1DE (ATA), please refer to ATA- issued by ANSI.
1 to ATA-4.

[0017] MPEG audio utilizes the psychoacoustic characteristics and reduces the amount of information by omitting detailed information with low auditory sensitivity, but the outline is the latest MPEG textbook (first edition in August 1994, ) ASCII Publishing), and details on MPEG audio standards (ISO / IE
Cll172-3 and ISO / IEC13818-
3). Needless to say, the image and sound compression methods and the recording medium may be other ones.

Next, the compression and expansion processing of the audio signal will be described. FIG. 1 is a diagram for explaining compression and expansion processing of an audio signal. FIG. 1 (a) is a schematic diagram when compression processing is performed with MPEG1 layer 2 audio, and FIG. 1 (b) is the same expansion processing. FIG. Further, FIG. 2 shows a format of a frame after compression processing with MPEG1 layer1 audio.

In FIG. 1A, the input audio sample data is input to the band division unit 101, and 32
It is converted into sample data for each sub-band (SB) divided into a number of frequency bands (step ST1). At the same time, the input audio sample data is also input to the masking threshold value calculation unit 102, and the FFT (Fast Fourie
r Transform) processing is performed to calculate the power level for each subband. By calculating the power level for each subband, the masking threshold value calculation unit 102 outputs the masking threshold value of each subband due to the masking effect (step ST2). In the bit allocation / scale factor calculation unit 103, the masking threshold value for each subband is compared with the level of the sample data in each subband which is the output of the band division unit 101, and the predetermined total number of assignable bits is obtained. The amount of information (the number of bits) to be assigned to each subband is determined so as to fit (step ST3). Then, the requantization unit 104 requantizes the sample data assigned to each subband according to the number of bits assigned to each subband,
The sample data is compressed (step ST3).
Further, the formatting unit 105 arranges and outputs data such as a scale factor (SF) and sample data so that the signal has a frame format determined by the MPEG1 audio standard (step ST4).

The size of the frame after compression differs depending on the compression ratio. For example, when a stereo sound is compressed to 1/6 by the MPEG1 layer 2 audio system with a sampling frequency of 48 kHz, a quantization bit number of 16 bits, it is 768.
It becomes Byte. (24 msec unit) As will be described later, the frame (768 Byte
e) and the cluster (20
Taking the common multiple of 48 Bytes or 4096 Bytes, 12,288 Bytes (6 clusters or 3
Cluster, 16 frames or 32 frames). Therefore, every time 6 or 3 clusters are read, the beginning of the cluster and the beginning of the frame match (see FIG. 5).
(See (a)).

FIG. 2 is a diagram showing an MPEG audio frame. In the figure, the header portion is the first 4 bytes in the MPEG audio frame, and a specific pattern is to be written in this area. The first 12 bits of the header part is a sync SYNC word, and all 12 bits are "1". SYNC WOR
1 bit next to D indicates an ID, which is "1" in the case of MPEG audio. The next 2 bits of the ID indicate the layer of MPEG audio, and in the case of the above-mentioned layer 2, "1".
0 ". The next 1 bit of the layer indicates the presence or absence of error correction, and it is" 1 "when there is no error correction. The next 4 bits of the presence or absence of error correction indicate the transmission bit rate,
Under the above conditions (sampling frequency 48 kHz, quantization bit number 16 bits, stereo audio is compressed to 1/6), the transmission bit rate is 256 kbit / se.
c, which is “1100”. The next 2 bits of the transmission bit rate indicate the sampling frequency, which is "01" in the case of 48 kHz. After that, data indicating various modes continues. However, since these data do not change unless the recording is stopped and the recording mode is changed, the data becomes information of a specific pattern as long as the recording is continued.

Therefore, even in an unused cluster obtained from the management area of the memory card, a specific part of the cluster matches the MPEG header information, and further M
If the previous information is read by the size of the PEG frame (768 bytes) and this matches the MPEG header information, this part can be regarded as being compressed by MPEG audio.

As for the data regarded as MPEG audio, the management data can be recognized as a file by writing the management data in the management area. Therefore, the management data of the data recorded on the medium due to a problem during recording is written. Even if it is not included, it is possible to restore the part recorded before the trouble occurs as described above.

FIG. 3 is a block diagram showing a recording / reproducing apparatus according to this embodiment. The audio signal input in FIG. 3 is input to the buffer 301 and the low-pass filter (LP
F) 302 is converted into digital audio data by an A / D (Analog / Digital) converter 303.
Further, the digital filter 304 becomes audio data from which frequency components outside the audible band are removed. Next, after the audio data is taken into the buffer memory 305,
Encoder / decoder unit 307 in microcomputer 306
Are stored in the buffer memory 305 and are temporarily stored in the buffer memory 305. When a certain amount of compressed audio data is accumulated in the buffer memory 305, title data (text data) is put into the auxiliary data area of this compressed audio data (frame) and recorded in the memory card 308.

Next, when reproducing the audio data, the reproduction button on the operation unit 309 is operated. When the play button is pressed, the compressed audio data is read into the buffer memory 305 for each reading unit of the storage medium (memory card 308), the compressed audio data in the buffer memory 305 is fetched into the encoder / decoder unit 307 in the microcomputer 306, and the voice is expanded. And returns it to the buffer memory 305. The decompressed data returned to the buffer memory 305 has a frequency component outside the audible band removed by a digital filter 311, is analog-converted by a D / A (Digital / Analog) converter 312, and has a low-pass filter (L).
It is reproduced as an analog audio signal via the PF) 313 and the output amplifier 314.

Here, the outline of the expansion will be described with reference to FIG. First, the signal separation unit 201
At, the header of the compressed audio data is detected, and compression mode information, bit allocation information, scale factor, and audio data of each subband are cut out from the compressed audio data (step ST1). The inverse quantization unit 202 inversely quantizes the audio data of each subband by the number of bits according to the bit allocation information separated by the signal separation unit 201 (step ST
2). Further, in the inverse scaling unit 203, the signal separation unit 201
The signal inversely quantized by the scale factor information separated by is converted into a signal having an amplitude before normalization (step S
T3). By synthesizing these signals by the band synthesizing unit 204 (step ST4), it is possible to restore as an uncompressed audio signal.

FIG. 4 is a flow chart for explaining the restoration process of the recorded data. Hereinafter, it is possible to restore the data by using the MPEG header information without using the management information of the data recorded on the recording medium (that is, when the management information is not recorded). explain.

First, after the occurrence of the above trouble, the apparatus is restarted and the data restoration button provided on the operation unit 309 shown in FIG. 3 is pressed (step ST1). The microcomputer 306 reads the management area on the memory card 308,
The used and unused areas are mapped (step ST2). Read the data of the cluster in the area that can be judged not to be used from the management data (step ST
3). It is determined whether or not the head of the cluster is a specific pattern (a pattern peculiar to the MPEG header), and if it is a specific pattern, the process proceeds to step ST5, and if not, the process proceeds to step 4b (step ST4a). Next, a specific pattern (M
PEG header specific pattern),
If it is a specific pattern, the process proceeds to step ST5, and if not, the process proceeds to step 4c (step ST4b). Next, it is determined whether 512 bytes or more from the beginning of the cluster is a specific pattern (a pattern peculiar to the MPEG header). If the specific pattern, the process proceeds to step ST5, and if not, the process proceeds to step ST10 (step ST4c). ). (That is, step ST4a
In step ST4c, a portion where the beginning of the cluster and the beginning of the frame coincide is searched, and the same compressed data (data in which information other than the bit rate in the frame header portion coincides) is recorded in successive clusters. To search for the frames that are open. Although an example of a fixed bit rate is shown here for the sake of easy understanding, in the case of a variable bit rate, the frame size can be obtained by reading the bit rate (“BITRATE” shown in FIG. 2) of an arbitrary frame. Can be calculated, the head position of the next frame can be detected based on this size. That is, FIG. 5 is a diagram showing the relationship between the management unit (cluster) of the recording medium and the recorded compressed data. FIG. 5A shows the case where the compressed data is recorded continuously in the cluster. b) shows a case in which compressed data is recorded without being continuous to the cluster (with other compressed data sandwiched in the middle).

First, as shown in FIG. 5A, when compressed data is continuously recorded in a cluster, 12.288B, which is the least common multiple of the cluster (4096 bytes) and the MPEG frame (768 bytes).
The beginning of the cluster and the beginning of the MPEG frame match for each yte. That is, although clusters 1 and 4 coincide with the beginning of the frame in FIG. 5A,
In clusters 2, 3, 5, and 6, the start of the frame does not match. For clusters that do not match the beginning of these frames,
256 bytes or 512 bytes from the beginning of the cluster
The beginning of the frame comes to the y-th eye part.

That is, if the degree of shift from the frame head of consecutive clusters is sequentially grasped, as shown in FIG. 5B, it is known whether or not there is another compressed data in the data of consecutive clusters. be able to. 5B shows an example in which other compressed data (frames F (M-1) to F (M + 10)) are recorded in the middle of certain compressed data (frames Fl to F22). First, when recording compressed data, the medium is recorded. In the state where the empty clusters are scattered (the clusters N + 1 to N + 2 have already been used and the empty cluster next to the empty cluster N is the cluster N + 3 or later), the certain compressed data (frame Fl to F
22), the frames Fl to F5 and the leading 256 bytes (F6a) of the frame F6 are recorded in the cluster N, and then the remaining 512 bytes of the frame F6 are recorded.
e, frames F7 to F21, and frame F2
The first 256 bytes of 2 are recorded in clusters N + 3 to N + 5.

When data is restored, the frame F6a in the cluster N has only the first 256 bytes recorded, so if a frame (F7) having the same header information as F6a exists at a position 512 bytes from the beginning of the cluster, the cluster Frame written on (F7)
May be a frame following F6a, so that it can be a target for restoration (cluster N + 3 in FIG. 5B corresponds to this). In other words, the capacity that could not be recorded in the last frame recorded in any cluster is set as the offset amount from the cluster head, and in another cluster, there is a frame having the same header information as the last frame at the position considering the offset If this is done, it can be targeted for revival.

Next, as shown in FIG. 4, the restoration frame list in the work area of the microcomputer 306 is updated by the above processing (step ST5), that is, the head of the MPEG header can be predicted (frame head, head). +256 Byte, beginning +512 Byte
If it is in the eye), it will be listed in the revival frame list as being able to revive. If the data of consecutive clusters is other compressed data, it can be restored as another file when the data is restored.

Next, the next frame to be read is incremented within the same cluster (step ST6). Next, it is determined whether or not the end of the cluster has been reached (step ST7). If the end of the cluster has been reached, the process proceeds to step ST10, and if not, the process proceeds to step ST8. Next, the incremented frame is read, and it is judged whether or not the head of the read frame matches the MPEG header (step ST8). If they do not match, step S
Proceed to T10, and if they match, microcomputer 3
Update the revival frame list in the work area of 06,
After allowing the read frame to be restored, the process proceeds to step ST6 (step ST9).

Next, the read cluster is incremented in the unused area (step ST10), and if it is not the end of the unused area, the process proceeds to step ST3, and if it is the end, the process proceeds to step ST12. When the end of the unused area is reached, management data is created from the list of frames to be restored, and the management data is written to the management area on the memory card 308 (step ST12). By the above processing, the data in which the management information is not recorded can be restored.

As described above, according to the present embodiment, for example, a predetermined common multiple is set between the management (read / write) unit of the recording medium and the size of the data to be recorded, and further recording is performed. Since the frame continuity is detected by detecting the header information (specific pattern) of the data, the data can be restored even if the data management information could not be written during the recording on the recording medium.

[0036]

As described above, according to the present invention, even if the management data cannot be recorded due to a problem occurring during recording and the data cannot be recognized as data, the data cannot be recorded on the recording medium by the time the problem occurs. The recorded data can be restored and played back.

[Brief description of drawings]

FIG. 1 is a diagram illustrating compression and expansion processing of an audio signal.

FIG. 2 is a diagram showing an MPEG audio frame.

FIG. 3 is a diagram showing a recording / reproducing apparatus according to an embodiment of the present invention.

FIG. 4 is a flowchart illustrating a restoration process of recorded data.

FIG. 5 is a diagram showing a relationship between a management unit of a recording medium and recorded compressed data.

[Explanation of symbols]

101 band division unit 102 masking threshold calculation unit 103 bit allocation / scale factor calculation unit 104 requantization unit 105 formatting unit 201 signal separation unit 202 inverse quantization unit 203 inverse scaling unit 204 band synthesis unit 301 buffer 302 low-pass filter 303 A / D converter 304 Digital filter 305 Buffer 306 Microcomputer 307 Encoder / decoder 308 Memory card 309 Operation unit 310 Display unit 311 Digital filter 312 D / A converter 313 Low-pass filter 314 Output amplifier

Claims (3)

[Claims] 1. Data in which a management unit of data used for writing and reading data is constant, and compression processing is applied to data divided into smaller than or equal to the management unit, and specific pattern information is added to the data after the compression processing. A recording / reproducing apparatus for recording data, wherein the recording / reproducing apparatus refers to information on a specific position in a management unit of the read data, and when the information is the specific pattern, restores the data. Recording / playback device. 2. Data in which a management unit of data used when writing and reading data is constant, and compression processing is applied to data divided into the management unit or less and specific pattern information is added to the compressed data. And
A recording / reproducing apparatus for recording the size of the processed data such that a common multiple of the management unit can be set, wherein the recording / reproducing apparatus is a specific position in the management unit of the read data. The information recording / reproducing apparatus is characterized in that when the information is the specific pattern, the data is restored. 3. The recording / reproducing apparatus according to claim 1, wherein the specific pattern is an MPEG header section.