JP2526949B2 - Erasing method in data recorder - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION [Industrial field of application] The present invention is applicable, for example, when a rotary head type digital audio tape recorder (DAT) is used as a data recorder for recording and reproducing data sent from a computer or the like. The present invention relates to a recording method in a suitable data recorder.

[Conventional technology]

2. Description of the Related Art In a computer, in order to protect (back up) data written on a hard disk or the like, such data is transferred to a data streamer (data recorder) once a day and recorded.

As a conventional data recorder, a normal analog audio tape recorder is often used. However, in such an analog tape recorder, the consumption of the magnetic tape becomes extremely large, and since the data rate at the time of recording is low, it takes a very long time to transfer and record. Cannot be easily performed.

Therefore, it has been considered to use a digital audio tape recorder, a so-called DAT, by a helical scan method using a rotary head that has been commercialized in recent years as a data recorder. When the DAT is used as a data recorder, the data from the host computer is converted into the DAT format and recorded. In the DAT format, two tilted tracks formed when two heads having different azimuths make one rotation are regarded as one frame, and 16-bit PCM audio data is interleaved and recorded with this one frame as a unit. , It is designed to record auxiliary sub data. In that case,
As will be described later, a main area in which PCM data is recorded and a sub area in which sub data is recorded are formed on each track.

Also, in DAT, when another signal is newly recorded on the recorded tape, generally, an erasing head is not used,
By recording a new signal by so-called double writing, the previous signal is erased.

[Problems to be solved by the invention]

As mentioned above, the DAT does not have an erasing head,
If some or all of the recorded data is no longer needed, it must be erased by some method, and at the time of erasing, it is necessary to erase while maintaining the DAT format.

[Means for solving problems]

In the present invention, there is provided an erasing method of erasing data having a second recording length shorter than the first recording length on the past data having the first recording length and erasing the rest of the data as invalid data. A first erasing mode in which the end identification data of a predetermined length is recorded subsequently to the data having a length and the past data after the end identification data is invalidated;
A second erasing code for erasing all the past data after the terminating identification data is recorded while recording the terminating identification data of a predetermined length following the data having the second recording length, based on a command signal. The first and second erase modes are selectively enabled.

[Action]

Data is substantially erased without breaking a predetermined data format such as the DAT format.

〔Example〕

FIG. 1 shows the configuration when a DAT is used as a data recorder, where 1 is DAT, 2 is an interface bus,
Reference numeral 3 is a host computer, and 4 and 5 are inner buses. The DAT 1 includes a recording / reproducing unit 6, a recording amplifier 7, a reproducing amplifier 8, a signal processing circuit 9, a RAM 10, a data controller 11,
The interface board 12, the system controller 13, the servo and motor drive circuit 14, etc. are mainly configured.

The system controller 13, the signal processing circuit 9, and the data controller 13 have an absolute frame number AFN described later.
Predetermined signals such as O, check data, mode instruction, logical frame number LFNO, determination result by check data, and data transfer instruction are transmitted and received.

The recording / reproducing unit 6 is provided with a rotary head drum (not shown), and the magnetic tape is wound around a range of about 90 degrees around the drum and is transported by a capstan. This drum has different azimuth angles A,
B2 heads are provided so that one rotation of the drum records and reproduces two inclined tracks on the tape.

Digital data supplied from the host computer 3 via the buses 5, 2 and 4 is the interface board 12
To the data controller 11, the RAM 10, and the signal processing circuit 9 according to the instruction from the system controller 13.
By performing predetermined signal processing in the above, the conversion to the DAT format described above is performed. The formatted signal is supplied to the recording / reproducing unit 6 through the recording amplifier 7 and recorded on the magnetic tape by the heads A and B.

Further, the signals recorded on the magnetic tape are reproduced by the heads A and B, and the reproduced signals are supplied to the signal processing circuit 9 through the reproducing amplifier 8, and the digital data which is inversely converted and taken out here is transferred to the data controller 11 and the interface. It is supplied to the host computer 3 through the board 12 and the buses 4, 2 and 5.

In the above apparatus, the format of the DAT recorded on the magnetic tape is as shown in FIG.

In FIG. 2, when the heads A and B rotate once on the tape 15, two inclined tracks T _A and T _B are formed from the lower side as shown by an arrow a in the figure. One frame is composed of these two tracks T _A and T _B. One track T _A (or T _B ) is composed of 196 blocks, and one block is composed of 288 bits. Of these, 34 blocks at both ends are sub-areas, and 128 blocks in the center are main areas.

The sub-area is further divided into several sections.
That is, from the bottom end of the track, the margin part, the preamble part for the PLL of the subcode, the first subcode part, the postamble part, the gap part of the block part, the tracking (ATF) signal part, the gap part between blocks, Data PLL
Preamble part, inter-block gap part, ATF signal part, inter-block gap part, preamble part for subcode PLL, second subcode part, postamble part,
A margin part is provided. Of these, the first and second sub-code parts are each composed of 8 blocks, and the other parts are each composed of a predetermined number of blocks. still,
The length measure in the figure is not accurate.

The main area is composed of 128 data blocks, and one block is composed of 8 bits each of a synchronization signal, W of PCM-ID, a part, a block address, and a parity from the head of the block, as shown in FIG. The main data is arranged in the next 256-bit period. When handling an audio signal, this main data is 16-bit PCM data for each of the L channel and the R channel. These 16-bit main data consist of two tracks T _A , T
_It is interleaved in the main area of _B (1 frame) and placed with parity. In this case, approximately 5760 bytes of data are recorded in the main area of one frame. When used as a data recorder, the data sent from the host computer 3 is converted into 16-bit data and treated as the above PCM data, and these data are formatted as shown in FIG. It is recorded in the main area of the frame.

That is, in FIG. 4, the above-mentioned 5760 bytes are formed into words (0 to 1439) of 4 bytes (32 bits), and each of these words has 16 bits (2 bytes) in accordance with the audio signal. It is divided into an L channel and an R channel. And in this format,
First, a sync part is provided in 3 words (12 bytes), the first 1 byte of this sync part is all "0", the following 10 bytes are all "1", and the last 1 byte is all "0". "
To be.

Next, an 8-word (32-byte) header section is provided.
The MSB side 4 bits of the first half byte of the word 3L channel in this header section are used as a format ID indicating the format of the data decoder, and the latter 4 bits are undecided. The remaining 3 bytes of this word 3 are used as a logical frame number (LFNO) area, which will be described later. With this LFNO consisting of 8 bits, for example, 23 frames are set to 1
As a unit, a binary value indicating the frame serial numbers 1 to 23 is provided.

Then, in the subsequent words 7 to 1438, a data portion of 5728 bytes in total is provided, and the data signal from the host computer 3 is sequentially provided to each word by 4 bytes.

Further, the word 1439 is used as an error detection code (EDC) part, and an error detection code (EDC) is generated for a data string in which each data bit is viewed vertically for each data signal of the header part and the data part described above. Is provided.

Here, in the DAT format, the data of the L channel and the data of the R channel are arranged so as to be alternately recorded into two tracks by 2 bytes and recorded, and for example, tracks are formed on both sides of the drawing. The azimuth (±) of the heads A and B is shown.
Therefore, as described above, the EDC can be generated for the two tracks forming one frame by generating the EDC for the data string in which each data bit is viewed in the vertical direction.

Therefore, according to this format, it is possible to determine the unerased portion of the main area by performing EDC calculation. That is, in the above-mentioned format, if one track is left unerased, every other data for generating the EDC is different, and the EDC operation cannot be performed correctly. Therefore, by detecting this value, it is possible to determine the unerased portion.

Next, regarding the above logical frame number LFNO,
It will be described with reference to the drawings.

As described above, the LFNO has, for example, 23 frames as one unit, and each frame is given a serial number of 1 to 23. That is, 1 to 23, 1 to 23 every 23 frames.
Will be repeated. In FIG. 5, the entire LFNO is composed of 8 bits, of which
The MSB must be the last frame of the unit,
When the unit is 23 frames, the last frame ID (LASTF-ID) is the 23rd frame.
The next 1 bit of the MSB is an ECC frame ID (ECCF-ID) indicating that it is an ECC frame for error correction, and the next 6 bits represent 1 to 23 LFNO. Therefore,
LFNO is selected from the range of 1 to 64 by
You can choose any length other than 23 frames. In that case, it can be known from the LASTF-ID that the frame is the last frame of the unit. Also ECCF
The ID may be multiplexed not only in one frame but also in a plurality of frames.

By using a unit with such an LFNO,
The division of a predetermined amount of data becomes clear, which facilitates signal processing. Also, by changing the maximum value of LFNO and selecting an appropriate length for the unit, signal processing can be performed more easily.

Next, the data configuration of the first and second subcode portions in the subarea will be described.

Each of the first and second sub-code parts is composed of eight sub-code blocks, and each can record 2048-bit data.

FIG. 6 shows the structure of the subcode block. After the 8-bit synchronizing signal, W ₁ , W ₂ and the parity are arranged, 256-bit subcode data including the parity is arranged. This subcode data is composed of four packs, and one pack is composed of 8 × 8 bits (8 symbols).

The contents of W ₁ and W ₂ are different between the block having the block address EVEN and the block having the ODD, and the pack numbers (1 to 7) are attached to the EVEN block and the ODD block as shown in the figure. . The eighth pack is provided with a C ₁ code for error detection.

W ₁ are each 4-bit area ID of the EVEN block, data
It consists ID, W ₂ is "1" in the upper one bit, consists pack ID, 4 block address bits of 3 bits. W ₁ of the ODD block is composed of a 4-bit undetermined portion and a format ID, and W ₂ is composed of upper 1 bit “1”, 3-bit all-zero code, and 4-bit block address. Also when 4-bit area ID in W ₁ of the EVEN block is "0100", EOD-I indicating the end of data
D (end of data ID).

Each of the packs 1 to 7 is divided into 8 words of 8 bits, and each word includes, for example, a code indicating the lead-in area of the recording start portion on the tape, a code indicating the lead-out area of the recording end portion, Various kinds of codes such as a code indicating a recording date and time, an absolute frame number, a logical frame number, etc. are provided together with the parity.

FIG. 7 shows the structure of the pack 3 as an example of the seven packs.

This pack 3 consists of 8 words of 8 bits each, PC1-8
Consists of. The upper 4 bits of PC1 are provided with a pack number (here, “0011” indicating pack 3), and the lower 4 bits are provided with a format ID. PC2 is an undecided part. An area ID indicating the lead-in area or the lead-out area is provided in the upper 4 bits of PC3. An absolute frame number (AFNO) is represented by the lower 4 bits of PC3 and a total of 20 bits of PC4 and PC5. This AFNO is a serial number attached to all frames recorded on one roll of tape. Next, a total of 16 with PC6 and PC7
Check data CD relating to the present invention is provided by bits, and PC8 is provided with parity for PC1 to PC7.

The 16-bit check data CD is the exclusive OR of all data (data from the computer) existing in the main area provided in one track (T _A or T _B ) in which the pack 3 is provided. It is a thing. Alternatively, as the check data CD, an error detection signal such as CRC for all the data existing in the main area may be used.

Further, the check data can be provided in the undetermined parts of the other packs PC1 and PC2 to 7 to perform multiple writing, thereby improving the reliability. In that case, one track is provided with the first and second sub-code parts of 8 blocks respectively, and two blocks have 7 packs, so that one track records up to 56 same check data CDs. be able to.

Such check data CD is used as follows.

At the time of reproduction, the check data CD read from one track is compared with the exclusive OR of the data read from the main area of the track. As a result, if they do not match, it is determined that either the entire main area has been erased or the entire sub area has been erased (the previous check data CD was also erased). can do. Further, when the both match, it can be determined that both the entire main area and the entire sub area are correct or both are incorrect. In order to know which of the main area and the sub area is correct or incorrect, it is possible to make a determination by using LFNO and AFNO by the method described below.

FIG. 8 shows the order of judgment, and each step S1 to S4
It is possible to obtain the judgment criteria shown in FIG. 9 based on the judgment according to, and to know the contents of the judgment shown in FIG. 10 based on the judgment criteria.

In FIG. 8, first, in step S1, it is checked whether or not the data in the main area (hereinafter referred to as main data) can be corrected by the C ₁ and C ₂ codes. If the correction cannot be made, it is determined that there is a dropout in the main data of the track as C in FIG. 10, and a message to that effect is sent. Next step when correction is possible
Use S2 to check the continuity of LFNO in the main area. If there is no continuity, it is determined to be B in FIG. 10 and it is determined that a drop-in has occurred on the track, and a message to that effect is sent. If the LFNO has continuity, the check of the main area is terminated, and then the sub area is checked in step S3. The reproducibility of the pack data in the sub area pack is checked by C ₁ code correction, pack parity, multiple matching, and the like. If this check cannot be made, it is determined to be (5) in FIG. 10, and it is determined that there is a drop-in in the sub-area data (sub-data) and the main data is correct. When the reproducibility check is completed, the continuity of the AFNO and the main data are checked by the check data CD in step S4. Based on the results of each check, the judgment criteria shown in FIG. 9 are obtained. The contents of the judgment in FIG. 10 are obtained based on the judgment criteria.

 The contents of each judgment are: A- (1) ............ Main data is correct.

A- (2) ............ There were drop-ins in both the main data and sub-leader.

A- (3) ………… There was a drop-in in the main data. AFNO has been continuous.

A- (4) ............ There was a drop-in in the sub data. In this case, AFNO should be self-propelled and the main data should be correct.

A- (5) ………… There was a dropout in the sub data. In this case, AFNO should be self-propelled and the main data should be correct.

A flag can be set when a drop-in is detected by the check data CD, and error correction can be performed by the error correction code ECC based on this flag. Further, in order to prevent erroneous reading of the check data CD and the data in the main area, the detection by the check data CD may be performed plural times.

Next, the data erasing method according to the present invention in the data recorder using the DAT format described above will be described with reference to FIG.

FIG. 11A shows a recording pattern of the magnetic tape 15 before erasing data. Data 1 to 4 are recorded by one backup of the data record, and the EOD-ID (the first (See Fig. 6) is recorded over 300 frames, and before that, there is old data recorded before.

On the tape 15 of FIG. 11A, for example, data 1,
A case will be described in which the data 2 and the data below the data 3 are deleted while leaving 2. In this case, the contents of erasure include (1) erasure indicating that all the data of data 3 and below are unnecessary (hereinafter referred to as unnecessary erasure).

(2) All data of data 3 and below are erased (hereinafter referred to as complete erasure).

There are two types. A first command signal is sent from the host computer when the unnecessary erasure is requested, and a second command signal is sent from the host computer when the complete erasure is requested.

(1) When unnecessary erasing is performed, when the first command signal is received, first, as shown in FIG. 11B, an amble portion is formed at the beginning of the data 3 for one or two frames or several frames. The amble is formed by dummy data such as all zeros. Next, the EOD-ID is recorded in the sub area over 300 frames.

According to the above, at the time of reproducing the tape 15 of FIG. 9B, the amble portion is first detected, and then the EOD-ID is detected, so that a part of the data 3 and the data below it are all unnecessary. I understand.

(2) In case of complete erasure When receiving the second command signal, first, as shown in FIG. 6C, the EOD-ID is recorded in the beginning part of the data 3 for 300 frames, and then the amble part remains in the entire section. It is formed. The amble portion may be formed in several frames before the EOD-ID of the data 3.

According to the above, when the tape 15 of FIG.
-Because the ID is detected and then the amble part is formed,
It can be seen that all data after that has been erased.

The amble part is formed by setting the above-mentioned LFNO in the main area to all zeros, or setting the above-mentioned AFNO and check data CD in the sub-areas to all zeros, or both of them.

According to the data erasing method described above, the erasing contents at the time of erasing are distinguished by the first and second command signals, and the erasing contents can be known by detecting the recording pattern at the time of reproduction. Further, according to this erasing method, the data can be substantially invalidated without using the erasing head without destroying the DAT format.

〔The invention's effect〕

According to the present invention, it is possible to erase data without using an erase head while maintaining a predetermined data format such as a DAT format. Also, two types of erasing modes can be selected.

[Brief description of drawings]

FIG. 1 is a block diagram showing an embodiment of the present invention, and FIG.
FIG. 3 is a diagram showing the format of T, FIG. 3 is a data configuration diagram of a main data block, FIG. 4 is a data configuration diagram of a main area of one frame, FIG. 5 is a data configuration diagram of a logical frame number, and FIG. FIG. 7 is a data configuration diagram of a code block, FIG. 7 is a data configuration diagram of a pack of sub code blocks, and FIG. 8 is a flowchart showing an error determination method.
FIG. 9 is a diagram showing an error determination standard, FIG. 10 is a diagram showing the contents of the determination, and FIG. 11 is a recording pattern diagram of a magnetic tape showing a data erasing method according to the present invention. In still code used in the drawings, a T _A T _B ...... track 1 ...... DAT 2 ...... host computer 13 ...... system controller EOD-ID ...... data end ID.

 ─────────────────────────────────────────────────── ─── Continuation of the front page (72) Kentaro Odaka 6-735 Kita-Shinagawa, Shinagawa-ku, Tokyo Sony Corporation (72) Masaki Yamada 6-35 Kita-Shinagawa, Shinagawa-ku, Tokyo Within Sony Corporation (72) Inventor Katsumi Inazawa 6-735 Kita-Shinagawa, Shinagawa-ku, Tokyo Within Sony Corporation (56) Reference JP-A-1-72361 (JP, A)

Claims (1)

(57) [Claims] 1. An erasing method for leaving data having a second recording length shorter than the first recording length on past data having a first recording length and erasing the rest of the data as invalid data is the second recording. A first erasing mode in which the end identification data of a predetermined length is recorded subsequently to the data having the length to invalidate the past data after the end identification data, and the data having the second recording length is subsequently supplied. A second erasing mode for recording the terminating identification data of a predetermined length and erasing all the past data after the terminating identification data is provided, and the first and second erasing modes are selectively selected based on a command signal. An erasing method for a data recorder, which is characterized by the above.