JPS63185125A - Multi-stage decoding method - Google Patents

Abstract

PURPOSE:To attain the decoding of a simple data while omitting an optional decoding by constituting a coding method so as to attain two or more stages of multi-stage coding and giving a discriminating signal representing whether or not each coding is applied, thereby preventing mis-decoding in a direction not applying any coding. CONSTITUTION:In processing the digital information of NXq-digit (N, q are integers) with the unit of q-digit, the data N are arranged to 2-dimension or above and the n-stage of coding from a 1st direction to the n-th direction (n is an integer, n>=2) from the 1st direction without duplication in an optional direction, with a code Ci on the Galois field GF(2<q>) in the i-th coding (i is an integer, 1<=i<=n) by the code length Ki (Ki is an integer) is applied. Thus, is decoding a data with the discriminating signal representing whether or not each coding is applied, whether or not decoding is applied is selected in response to the discriminating signal. Thus, since decoding is applied to the coding processing only in this way, mis-decoding is prevented.

Description

Detailed Description of the Invention [Field of Industrial Application] This invention uses Reed-Solomon codes ( This invention relates to a decoding method when using an error correction code such as R5 code (hereinafter referred to as R5 code).

[Conventional technology]

Generally, when data is recorded on and reproduced from a recording medium such as a magnetic tape or a magnetic disk, data errors may occur depending on the condition of the recording medium. Data error traps include pernut errors due to dropouts and random errors due to S/N deterioration. In order to correct these errors, a two-stage coded error correction code is often used. -For example, q
Consider the decoding of a two-stage code using an R8 code on GF(28) when =8. The sixth encoder is a two-stage encoder.
There is one shown in the figure. ill is a C1 encoder, - is an interleaving circuit, and - is a C1 encoder. First, C encoding is performed on the original data, and after interleaving, C1 encoding is performed and output. . 2. The post-decoding device shown in Figure 7 is shown in Figure 7. 0 In Figure 11 is a C1 decoder;
C) is C! The decoder performs dinterleaving after C1 decoding, and then performs C1 decoding. Here, as a conventional two-stage encoding method, N
When handling data arranged in two dimensions of 1 x 8 digits, N digits in the Y direction, and dividing them into 8 digit units in the X direction, the first check code of K, -N, is added. There is a two-stage encoding method in which a second check code of Kt - Kt is added after that. Ct(K,,N
, ) R3 code and C1 as (Kt = Kt ) RS code. Note that (α, β) R8 code has information length β and code length α
represents the R5 code of

Next, specific decoding will be explained with reference to FIG.
For example, when recording digital audio data on a magnetic tape using a helical scan video tape recorder C (hereinafter referred to as VTR), the audio data is recorded in two channels with a quantization pit count of 16 bits and a sampling frequency of 48 KHz. In the case of NTSC television standard signals, 800.8 samples of data are included in one field period. If this data is arranged 6 sample data each in the X direction, 184 samples will be required in the Y direction. If each sample data is treated in units of 8 bits, there will be 4 data for 2×2 channels, and 4×6=24 data will be arranged in the X direction. Therefore, with N, = 24 and N2 = 184, C! with a code length of 30 in the second direction. If you perform encoding and perform C1 encoding with a code length of 84 in the first direction,
K! = 80, K, = 84, and Ct is (80, 2
4) R8 code, C1 becomes (84, 80) R8 code.

In addition, in the data array, the 1st in the X direction and the mth in the Y direction (1*
m integer* Kt≧7≧1, N,≧rn≧1)
Let the O data be expressed as al + m.

First, calculate the syndrome by performing C1 decoding on the data in each row 84 (-) arranged in the first direction,
The location and size of the error are determined and output as a flag. Shift this by 1 in the Y direction and repeat in the same way as C1
In the primary direction, which performs decoding, in the second direction, e.g. all+
ay+5+ 33+11+""""'+ 336+ 1
Perform VcC decoding on the 80 (-Kt) data of 21 based on the C1 decoding flag, calculate the syndrome from the reproduced data, find the position and size of the error, and perform correction. , flags are output for those that cannot be corrected. This is shifted by 1 in the Y direction and repeated in the same manner to perform C1 decoding.

The reproduction of data is 24 (-N,
) is sent out while checking the C2 decoding flag, and interpolation is performed for portions where errors cannot be corrected. If such two-stage coding is performed, C1 code will be used for burst error correction, and C1 code will be used for random error correction.
C. code is involved, and error correction can be performed efficiently.

[Problem that the invention seeks to solve]

Since the conventional decoding method is configured as above,
Since decoding is performed regardless of whether or not encoding has actually been performed, unencoded data may be erroneously decoded, and since all decoding is performed, processing is complex and time-consuming. There was a problem.

This invention was made to solve the above-mentioned problems, and since it decodes only what has already been encoded, it is possible to prevent erroneous decoding, and it also prevents some decoding from occurring. Even if omitted, it is an object of the present invention to provide a multi-stage decoding method that can perform simple decoding without having to change any other decoding at all.

[Means for solving the problem] The multi-stage decoding method of the present invention determines whether or not to perform decoding based on an identification signal added during encoding indicating whether or not encoding has been performed. Furthermore, some of the decoding operations are omitted to enable simple decoding.

[Effect]

According to this invention, it is possible to prevent erroneous decoding in which unencoded data is decoded, and instead of decoding all encoded data, -
By omitting erroneous decoding, it is possible to shorten the decoding time.

[Embodiments of the invention]

An embodiment of the present invention will be described below with reference to the drawings. n=
8, and 8-stage encoding will be described. FIG. 2 is a block diagram showing the 8-stage encoding device of this embodiment, where e]) is a C8 encoder, d) is an interleave circuit, and (e) is a C8 encoder;
(to) is a C1 encoder, (2) is an interleave circuit, and (a) is a C1 encoder. First, C3 encoding is performed on the original data, and then interleaved and C2
Perform encoding. After further interleaving, C
Figure 8 is a block diagram showing the 8 post-coding device. The interleaving circuit (to) is a C8 decoder. After the reproduced data is C1 decoded, it is subjected to dinterleaving, and then C1 decoded based on the flag information. After further dinterleaving, C3
Decryption is then performed. FIG. 1 is a data array diagram showing the 8-stage encoding method of this embodiment. Here again, let us consider the case where two channels of digital audio data with a quantization bit count of 16 bits, a sampling frequency of 48 KHz, and two channels are recorded on a helical scan VTR that handles NTSC television standard signals, as in the conventional example. As with the conventional example, each sample data is treated in 8-bit units, and the X direction K 24
(-Ns), Y direction K 184 (=Nt)
O2'eX Arrange the data originally, perform C8 encoding with a code length of 8 degrees in the eighth direction, perform C1 encoding with a code length of 82 in the second direction, and perform C1 encoding with a code length of 82 degrees in the first direction. 84
If we perform C3 encoding with = so , K, = 8
21 K, = 84 and C1 is (go, 24) R5 code,
C! is (82, 80) R5 code, C1 is (84, 82)
It becomes R3 code. In addition, in the data array, the first in the X direction,
Let m-th data in the Y direction be expressed as aplm.

First, if it is identified that 01 encoding has been performed on a portion of the header for the data in each row 84 (-) arranged in the first direction, C1 encoding is performed. Calculate the syndrome, determine the location and size of the error, and make corrections.
Flags are output for those that cannot be corrected. This is shifted by 1 in the Y direction and repeated in the same manner to perform C1 decoding.

Then a second direction, e.g. al+1182m 41 a
Ill? .......

aS! If it is identified that C1 encoding has been performed in a part of the header for the data of +, 8, 82 (=,), C2 decoding is performed.The syndrome is calculated, and the position and size of the error are determined. is calculated, corrections are made, and flags are output for those that cannot be corrected. C1 decoding is performed by repeating this process in the Y direction by -r by 1. Furthermore, an eighth direction, e.g. alB + a2+51 a
3+9+ """+ asol+tt's 80 (-)
If it is identified that C3 encoding has been performed on a portion of the header of the data, C3 decoding is performed. It calculates the syndrome, determines the position and size of the error, performs correction, and outputs a flag if it cannot be corrected. Shift this by 1 in the Y direction and repeat in the same way as C
3 decoding is performed, and the data is reproduced by looking at the flag output of 24 (-N,) data arranged in the first direction.
The data is sent out, and portions where errors cannot be corrected are interpolated.

Since the multi-stage decoding method according to the present invention is configured as described above, for example, the number of quantization bits of audio data can be reduced to 1.
When increasing from 6 bits to 20 bits, the number N of data arrays arranged in the
If the shaded area in FIG. 1 is also used as a data area, 80 pieces of data can be arranged in the X direction.

Moreover, the overall data array format and capacity are the same, and the decoding method is also C! Decoding There is no need to change C1 decoding at all.Furthermore, C8 encoding is not performed, and C! By using an identification signal indicating that encoding or C-encoding is being performed, it is possible to prevent erroneous decoding in which decoding is performed in a direction that has not been encoded.

In this case, the encoding device in FIG. 2 first performs C1 encoding (82, 80) to obtain an R8 code using a C1 encoder (E), and then performs interleaving using an interleaving circuit (2), followed by a K Cs encoder (2). 01 encoding is performed at (84°82) to obtain an R3 code. The encoding device is shown in FIG.
First, the C1 decoder ell performs C1 decoding, and after performing dinterleaving in the dinterleave circuit (2), the C1 decoder ell1 performs C1 decoding. Perform decryption. Further, in the above example, the number of quantization bits of the audio data is increased, but other data may be included until the number of digitized bits reaches t.

Furthermore, in the above embodiment, C8 encoding and decoding were not performed due to the increase in the amount of information, but even if encoding was performed, the error correction rate and speed of the system would be affected, and arbitrary decoding, e.g. C1°C! without C8 decoding. Simple decoding may be performed by performing only decoding.

In the above embodiment, the data array is a two-dimensional array, but it may be an eight-dimensional array as shown in FIG.
In a code configured to perform n-stage decoding,
The number of stages for which decoding is not performed can be arbitrarily selected below n- stages,
Also, it is possible to arbitrarily choose which one to use.

Furthermore, even if the data group to be decoded is a complete type that treats the data in the data library as one unified unit,
Either type may be used, such as an incomplete type in which data is written and encoded sequentially.

In addition, although we considered recording digital audio data on a helical scan VTR, the data content may be any digital data, and the recording medium may be tape, tape, etc.
It can be anything, such as a disk. Also, the applicable device is VTR.
Other recording devices such as digital audio tape recorders and floppy disk devices may also be used. moreover,
The identification signal indicating whether or not encoding has been performed may be added to a part of the header at the end of the data, within the data, or in a completely different part.

〔Effect of the invention〕

As described above, according to the present invention, the encoding method is configured to enable multi-stage encoding of two or more stages, and since there is an identification signal indicating whether or not each encoding has been performed, encoding is not performed. A highly flexible decoding method that can prevent erroneous decoding in the direction, omit arbitrary decoding and easily decrypt data, and does not require any changes for other decoding. There is an effect that can be obtained.

[Brief explanation of the drawing]

FIG. 1 is a data array diagram showing an 8-stage encoding method according to an embodiment of the present invention, FIG. 2 is a block diagram showing an 8-stage encoding device according to a partial implementation of the present invention, and FIG. Example 8
FIG. 4 is a block diagram showing a posthumous encoding device; FIG. 4 is a data array diagram showing an eight-stage encoding method according to another practical example of the present invention; FIG. 5 is a data array diagram showing a conventional two-stage encoding method; FIG. 6 is a block diagram showing a conventional two-stage encoding device, and FIG. 7 is a block diagram showing a conventional two-stage decoding device. In the figure, q is the number of digits per data, N is the number of data, Kl e Kt +..., Kn is the code length, and n is the number of encoding stages. Note that the same reference numerals in the figures indicate the same or equivalent parts.

Claims (2)

[Claims] (1) Handle digital information of N x q digits (N, q: integer) in units of q digits, arrange N data in two or more dimensions, and arrange the data from the first direction to the first one without duplication in any direction. Up to n direction (n: integer, n≧2), the i-th encoding (i: integer, 1≦i≦n) has code length K in the i-th direction.
i (Ki: integer) constitutes the code Ci on the Galois field GF (2^q) and is configured so that n-stage encoding can be performed, and an identification indicating whether each encoding has been performed or not. 1. A multi-stage decoding method characterized in that when decoding data to which a signal is added, it is possible to select whether or not to perform decoding according to the identification signal. (2) Multi-stage decoding according to claim 1, characterized in that when decoding the data, decoding of some of the n stages is omitted to simplify the decoding. method.