JP2000181807A - Method and device for inspecting data in recording medium - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a data inspecting method based on CRC for a recording medium such as a memory card and a flash memory, by which a processing time is shortened with a small calculation load. SOLUTION: When data are changed in the data area 2 of the recording medium 1, data of a data change area A is replaced with differential data with data before change without re-calculating CRC in a whole from the top to the end of the data area 2, CRC re-calculation is executed concerning a part after that, the calculation result is added to a CRC calculation result before change to obtain the CRC calculation result after change.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a method and an apparatus for checking data recorded on a recording medium for errors.

[0002]

2. Description of the Related Art Parity check (a method of adding 1-bit parity to each byte to match an odd or even number in byte units) is used for error checking of data in a recording medium such as a memory card or a flash memory. A sum check (a method of adding 2 bytes which is the addition result of each byte over the entire medium) is generally used. These methods are simple error checking methods that require less computational load and test storage capacity. However, these methods do not have high inspection accuracy and often fail to perform accurate error inspection.

As an error checking method having a higher checking accuracy than such checking methods, a CRC (Cyclic Redundancy Ch.
eck: cyclic redundancy check) is known. The CRC is often used for inspection when data reliability is insufficient, and is mainly used for applications such as serial communication and access to a hard disk.

Hereinafter, the concept of the CRC will be briefly described. CRC uses a cyclic code (a code in which a vector obtained by cyclically shifting an arbitrary code word of a certain code by an arbitrary bit is also a code word) that is a code effective for error checking, and uses a predetermined code as a code. Divide by the number and use the remainder for inspection. This cyclic code data block is composed of k information bits and (nk) check bits.

[0005] Then, consider the data as a polynomial of a variable x,
Treat mathematically. For example, for data "101011",
The "1" of the leftmost x ^5, 2 th "1" from the left end x ^3, also x ¹ the third "1", if the "1" rightmost were considered constant term, the data Is expressed as a polynomial x
⁵ + x ³ + x + 1.

For such a polynomial expression, the following modulo 2 addition of a polynomial is defined and an operation based on this is performed. In general, mod N addition indicates addition within a range of a number N, and portions exceeding N are discarded. Therefore, in mod 2 addition, 0 + 1 = 1, 1 + 1 = 0, which means “binary addition without carry”, which is an exclusive OR operation of both bits in addition of binary values. (EX-O
R). Specifically, it is as follows. 1x ⁿ + a ^{1x n = 0x n 1x n +} 0x n = 0x n + 1x n = 1x n 0x n + 0x n = 0x n -1x n = 1x n above, in the mod 2 adder, associativity, distributivity, exchange It can be seen that the rule holds. Further, the mod 2 adder, because it is -1x ⁿ = 1x ^n, subtraction is the same as additive.

A cyclic code can be uniquely defined by a (n−k) -th order residue generating polynomial G (x) that determines a check bit of data. This residue generating polynomial G
Generation of a cyclic code by (x) is performed as follows.

Message polynomial M indicating information bits
In order to encode (x), x ( ^nk ) is multiplied by M (x) to obtain x ( ^nk ) M (x), and the multiplication formula is divided by the remainder generator polynomial G (x). The quotient and remainder of this division are Q
Assuming that (x) and R (x), the following equation (1) holds. x ^nk · M (x) = Q (x) · G (x) + R (x) (1)

For example, transmission data U (x) in communication
Is generated by dividing the remainder R (x) by x ^nk・ Addition to M (x)
Calculate. Addition and subtraction are the same in mod 2 addition
Then, U (x) is expressed by the following equation (2). U (x) = x^nkM (x) + R (x) = Q (x) G (x) (2) Therefore, in the transmission data, the information bits of the upper k bits are
x^nk-Inspection of lower (nk) bits, representing M (x)
The bits represent R (x). From the above equation (2),
Expression (3) is derived. U (x) ÷ G (x) = (x^nkM (x) + R (x)) ÷ G (x) = Q (x) (3) As can be seen from equation (3), the data including the check bits
Dividing the entire data U (x) by the remainder generator polynomial G (x)
The remainder is 0. Therefore, the remainder of this division is 0 on the receiving side.
Whether the transmission data is correct or not.
Can be inspected. Here, the division is mod 2 addition, that is, 2
EX so that carry and borrow do not occur.
-OR. Therefore, in this case, the remainder is always
One bit less than the divisor, so that the remainder is
And the divisor is a remainder generator polynomial.

Next, such a specific example will be described. For example,
Residue generating polynomial G (x) = x^Five+ X ^Four+ X^TwoUsing +1
Consider a cyclic code where n = 15, k = 10, and nk = 5. Me
Message polynomial M (x) = x⁹+ X^Five+ X^TwoCorresponds to +1
To encode the message “1000100101”
x^Five-Divide M (x) by G (x) to get the remainder R (x)
Ask like.

[0011]

(Equation 1)

Therefore, x ⁵ · M (x) is expressed by the following equation (4), and transmission data U (x) is expressed by x ⁵ · M (x)
And the remainder R (x) = x + 1, so that the following equation (5) is obtained. ^{x 5 · M (x) =} x 14 + x 10 + x 7 + x 5 = (x 5 + x 4 + x 2 +1) × (x 9 + x 8 + x 7 + x 3 + x 2 + x + 1) + (x + 1) ... (4) U (X) = (x ¹⁴ + x ¹⁰ + x ⁷ + x ⁵ ) + (x + 1) (5)

The transmission data U (x) generated by the (nk) -order remainder generation polynomial G (x) in this way can detect any burst error of length (nk) or less. Proven.

[0014]

In order to improve the accuracy of error checking of data in a recording medium such as a memory card or a flash memory, a CRC having such a high checking accuracy is used.
It is conceivable to use. However, the large computational load and the necessity of recalculating the entire message even when a part of the message is changed, require that the CR to these recording media be used.
This is a major barrier to C application.

The present invention has been made in view of such circumstances, and has a small calculation load and a short processing time, and can be applied to a recording medium such as a memory card or a flash memory. And an apparatus.

[0016]

According to a first aspect of the present invention, there is provided a data inspecting method for a recording medium, which inspects whether or not there is a data error in a recording medium in which a part of recording data has been changed based on a CRC calculation result. In this method, the data in the changed portion is replaced with the difference data between the data before the change and the changed data, the CRC calculation is re-executed from that portion onward, and the CRC recalculation result is stored in the CRC before the data change.
It is characterized in that the presence or absence of an error in the data of the recording medium is checked based on the addition result added to the calculation result.

According to a second aspect of the present invention, there is provided a data inspection apparatus for a recording medium for inspecting whether or not there is a data error in a recording medium in which a part of the recording data has been changed. Replace with the difference data between the data before change and the changed data, and
Calculating means for re-executing the C calculation;
Adding means for adding the C recalculation result to the CRC calculation result before data change, and checking whether there is an error in the data of the recording medium based on the addition result by the adding means. And

According to a third aspect of the present invention, there is provided a method for inspecting data recorded in a recording area of a recording medium, based on a result of a CRC calculation. The recording area is divided into a plurality of small areas, and the CRC calculation result is recorded for each small area.
When data content is changed in the recording area, the CRC calculation is re-executed for the small area in which the data content has been changed.

Unlike serial communication in which the entire message sequence is different each time, in the inspection of the entire database such as a recording medium such as a memory card or flash memory, the entire data does not change every time, and the access to the database is not changed. Only part of it changes each time.

In consideration of such a characteristic of data change of the recording medium, the first invention replaces the data of the changed portion with
Replace with the difference data between the data before the change and the changed data, re-execute the CRC calculation from that part onward, add the CRC recalculation result to the CRC calculation result before the change, and change the addition result after the change. The result is the CRC calculation. Therefore, when there is a data change, it is sufficient to perform the CRC recalculation from the changed part to the subsequent part, and it is not necessary to perform the whole CRC recalculation from the beginning, so that the calculation load can be reduced and the processing time can be reduced. Can be shortened.

Further, in consideration of such a characteristic of data change of the recording medium, in the second invention, the recording area of the recording medium is divided into a plurality of small areas.
The C calculation result is recorded, and when the data is changed, the C calculation is performed only for the small area where the data change occurs.
Perform RC recalculation. Therefore, only the CRC recalculation of the changed small area needs to be performed, and the entire CRC is calculated from the beginning.
Since there is no need to perform recalculation, the calculation load is reduced and the processing time can be reduced.

Therefore, in the present invention, the calculation load is reduced and the processing speed can be reduced. As a result, the CRC can be applied to a recording medium such as a memory card or a flash memory.

[0023]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS The present invention will be specifically described below with reference to the drawings showing the embodiments. FIG. 1 is a functional block diagram of an inspection device 10 for inspecting a data error of a recording medium by CRC. The inspection apparatus 10 includes a reading unit 11 for reading recording data from a recording medium 1 to be inspected, for example, a memory card, a writing unit 12 for writing a CRC inspection result to the recording medium 1, and a ROM 13 for storing a CRC inspection program. And a calculation unit 14 for performing various calculations of the CRC check
And a RAM 15 for storing intermediate values calculated at the time of CRC inspection, etc.
And a CPU 16 for controlling these components.

FIG. 2 is a schematic diagram showing a recording area of the recording medium 1. The recording area of the recording medium 1 includes a data area 2 for recording data and an inspection result area 3 for recording a CRC inspection result. Having.

When a memory card is used as a database,
The entire data does not change with each access, only the data in a part of the area changes. FIG. 2 shows an example of a data change area A in which such data change has been made, with hatching.

When such data change occurs, it is necessary to perform a CRC check. Conventionally, the entire CRC recalculation from the beginning to the end of the data area 2 has been performed. Therefore, even with small data changes, the calculation load has increased.

(First Embodiment) In the first embodiment,
Only for the data change area A and thereafter, the CRC is recalculated to obtain a new CRC calculation result.
In other words, the data of the changed portion is replaced with the difference data between the data before the change and the changed data, the CRC is recalculated from that portion onward, and the CRC recalculation result is added to the CRC calculation result before the change. Then, the result of the addition is set as the changed CRC calculation result. By doing so, only the CRC recalculation for the part after the change has to be performed, and it is not necessary to recalculate the entire CRC from the beginning, so that the calculation load can be reduced and the processing time can be reduced. it can.

Hereinafter, an example of a specific method of the first embodiment will be described using a 16-bit CRC as an example. 16
The bit CRC, a n-k = 16, the remainder generator polynomial is ^{G (x) = x 16 +} x 15 + x 2 +1, burst error detection can be performed up to 16 bits long. It has also been proven that if the burst of errors is larger than 16 bits, the error can be detected with a 99% probability.

Consider recalculation of a 16-bit CRC when 8-bit data D '(x) at the s-th byte from the end of the data string is changed to another data D (x). In this case, the calculation may be performed again from the beginning to the end of the message after the change, but a slightly simpler recalculation method will be described here.

Assuming that the data string before the change is M '(x),
Since addition and subtraction are the same in mod 2 addition, the data string M (x) after the change is represented by the following equation (6). M (x) = M '(x) ^-x8s * D' (x) + ^x8s * D (x) = M '(x) + ^x8s * D' (x) + ^x8s * D (x) (6) )

If the quotient and remainder of the data before the change are Q '(x) and R' (x), respectively, the following equation (7) holds for the data before the change, and the relation of the equation (7) is obtained. The following equation (8) is established using ^{x 16 · M '(x)} = Q' (x) · G (x) + R '(x) ... (7) x 16 · M (x) = x 16 · {M' (x) + x 8s · D ' ^{(x) + x 8s · D} (x)} = x 16 · M '(x) + x 16 + 8s · D' (x) + x 16 + 8s · D (x) = Q '(x) · G (x) + R ′ (x) + x ^{16 + 8s} · {D ′ (x) + D (x)} (8)

Here, x ^{16 + 8s} · in equation (8)
Term {D '(x) + D (x)} , because the order is greater than the ^16th order x can be divided by the remainder generating polynomial G (x), the quotient of the time, the remainder of each Q (x), R If (x) is used, it can be expressed by the following equation (9). x ^{16 + 8s} · {D ′ (x) + D (x)} = Q (x) · G (x) + R (x) (9)

When the equation (18) is further rewritten using the relation of the equation (9), the following equation (10) is obtained. ^{x 16 · M (x) =} Q '(x) · G (x) + R' (x) + Q (x) · G (x) + R (x) = {Q '(x) + Q (x)} · G (X) + R '(x) + R (x) (10)

Since the first term in equation (10) is divisible by G (x), the remainder obtained by dividing x ¹⁶ · M (x) by G (x) is R ′ (x) + R (x). You can see that
Therefore, the remainder of the changed data is R ′ (x) + R
(X). Note that R '(x) is the remainder of the data before the change, and R (x) is ^x8s ｛{D' (x) + D (x)}.
Is the remainder of

Therefore, to summarize the above, by performing the following steps, a new data D '(x) in the s-th byte from the end of the data string is changed to data D (x). The remainder is obtained. (Step 1): Data D ′ (x) and data D (x) are added by mod 2. (Step 2): A remainder R (x) in a 16-bit CRC with the dividend being the product of the addition result and ^x8s is determined. (Step 3): mod 2 is added to the obtained remainder R (x) and the remainder R ′ (x) of the data before the change, and the addition result (R
Let (x) + R ′ (x)) be the new remainder.

Next, a specific example of the CRC recalculation at the time of such data change will be described. Two continuous data "0101
0001 "and" 10010001 "are converted to a 16-bit CRC (G (x) = x
Remainder ¹⁶ + x ¹⁵ + when divided by x ² +1) is "0110010101100011" as follows.

[0037]

(Equation 2)

Here, for example, the first data is "010100
01 ”to“ 00100010. ”In this case,
When the 16-bit CRC calculation is performed from the beginning, the remainder “0100111101100101” is obtained as follows.

[0039]

(Equation 3)

When the processing for obtaining the remainder is performed by the method of steps 1 to 3 as described above, the following is obtained. First,
The data before change (corresponding to D ′ (x)) “01010001” and the data after change (corresponding to D (x)) “00100010” are added by mod 2 to obtain “01110011”.

[0041]

(Equation 4)

The changed part is the first byte (s =
1), x ^8s · {D ′ (x) + D (x)} is
"0111001100000000", which is subjected to a 16-bit CRC calculation as follows, and the remainder (corresponding to R (x)) "001010
1000000110 ”.

[0043]

(Equation 5)

The remainder “0010101000000110” is the remainder of the data before change (corresponding to R ′ (x)) “0110010101100011”
Is added by mod 2 to obtain the final remainder “0100111101100101”. This remainder result matches the remainder result calculated from the beginning.

The superiority of the method of the first embodiment as described above over the conventional method of recalculating the entire data area 2 from the beginning to the end will be described. FIG. 3 is a diagram showing the position of the data change area A in which data has been changed. According to the method of the first embodiment, when the data change area A occurs near the head of the data area 2 (see FIG.
(A)) Although the advantage over the conventional method is not very high, when data change occurs near the end of the data area 2 (FIG. 3 (b)), the calculation load is greatly reduced, The advantage over the method becomes significant.

In the first embodiment, the method of steps 1 to 3 is used as a method for performing recalculation from the area after the data change occurs. However, this is merely an example. But of course it is good.

(Second Embodiment) FIG. 4 is a schematic diagram showing a recording area of a recording medium 1 according to a second embodiment.
In the recording medium 1 of the second embodiment, the entire data area 2 is divided into a plurality of data small areas, and an inspection result area is provided independently for each data small area.

In the example shown in FIG. 4A, the data area 2 is divided into the first, second, third, and fourth data small areas 2a, 2b, 2c, and 2d.
It is divided into a first data small area 2a, a first inspection result area 3a for recording a CRC inspection result corresponding to the first data small area 2a, a second data small area 2b, and a second data small area 2b. A second inspection result area 3b for recording the corresponding CRC inspection result, a third data small area 2c, and the third data small area 2c
Inspection result area for recording the CRC inspection result corresponding to
3c, a fourth data small area 2d, and a fourth inspection result area 3d for recording a CRC inspection result corresponding to the fourth data small area 2d are provided in this order, and the entire data area 2 is physically divided into four parts. Configuration.

In the example shown in FIG. 4B, the entire data area 2 is divided into four parts as in FIG. 4A, but the first data small area 2a, the second data small area 2b, and the third data small Region 2c,
The fourth data small area 2d, the first inspection result area 3a, the second inspection result area 3b, the third inspection result area 3c, and the fourth inspection result area 3d are provided in this order, and a configuration considering the continuity of the data area is provided. Has made.

For example, when a memory card is used as a database, the entire data does not change each time it is accessed, but only data in a part of the area.
Therefore, when such data change occurs, the second
In the embodiment, the data sub-areas (the first, second, third, and fourth data sub-areas 2a, 2b, 2c,
C for only one data small area of 2d)
It is sufficient to perform the RC recalculation, and it is not necessary to perform the entire CRC recalculation from the beginning to the end of the data area 2 as in the related art, so that the calculation load can be reduced and the processing time can be reduced.

Although the data area is divided into four,
It goes without saying that the number of divisions of the data area 2 in the embodiment may be any number.

[0052]

As described above, according to the present invention, when data is changed in the recording medium, the data in the changed portion is replaced with the difference data between the data before the change and the changed data, and the subsequent portion is replaced by the difference data. Re-run the CRC calculation for
The CRC recalculation result is added to the CRC calculation result before the change, and the addition result is used as the CRC calculation result after the change.
It is sufficient to perform C recalculation, and it is not necessary to recalculate the entire CRC from the beginning, so that the calculation load can be reduced and the processing time can be reduced.

In the present invention, the data area of the recording medium is divided into a plurality of small data areas, and the CRC calculation result is recorded for each small data area. In this case, it is sufficient to perform the CRC recalculation only for one data small area, and it is not necessary to perform the CRC recalculation for the entire data area, so that the calculation load can be reduced and the processing time can be reduced.

[Brief description of the drawings]

FIG. 1 is a functional block diagram of an inspection device that inspects a data error of a recording medium by using a CRC.

FIG. 2 is a schematic diagram illustrating a recording area of a recording medium.

FIG. 3 is a diagram showing a position of a data change area.

FIG. 4 is a schematic diagram illustrating a recording area of a recording medium according to a second embodiment.

[Explanation of symbols]

 Reference Signs List 1 recording medium 2 data area 2a first data small area 2b second data small area 2c third data small area 2d fourth data small area 3 inspection result area 3a first inspection result area 3b second inspection result area 3c third inspection Result area 3d Fourth inspection result area 10 Inspection device 13 ROM 14 Calculation unit 16 CPU

Claims (3)

[Claims] In a method for checking whether or not there is a data error in a recording medium in which a part of recording data has been changed based on a CRC calculation result, the data in the changed part is compared with the data before the change. The CRC data is re-executed from the part after the difference data with the changed data, and the CRC recalculation result is added to the CRC calculation result before the data change. A method for inspecting data of a recording medium, wherein the method inspects the presence or absence. 2. An apparatus for inspecting whether or not there is a data error in a recording medium in which a part of recording data has been changed, wherein the data in the changed part is a difference between the data before the change and the changed data. Calculation means for re-executing the CRC calculation from the part after replacing the data with data, and addition means for adding the CRC recalculation result by the calculation means to the CRC calculation result before data change, wherein the addition result by the addition means is provided. A data inspection apparatus for a recording medium, wherein the presence or absence of an error in the data of the recording medium is inspected based on the data. 3. A method for checking whether data recorded in a recording area of a recording medium has an error based on a CRC calculation result, wherein the recording area of the recording medium is divided into a plurality of small areas. The CRC calculation result is recorded for each small area, and when the data content is changed in the recording area, the CRC is performed for the small area whose data content has been changed.
A data inspection method for a recording medium, wherein the calculation is performed again.