JPS6052965A - Error correcting method - Google Patents

Abstract

PURPOSE:To correct exactly an error without overlooking an existence of an error by executing a parity check of a PCM reproducing signal, and thereafter, correcting an error. CONSTITUTION:A voice data word D of a reproduced PCM sound signal, and a parity word of P and Q sequences are written in a memory, a parity check of the Q sequence X, etc. in the diagonal direction is executed, and unless an error pointer based on an error detection by a CRC is set in data (a)-(j) in which an error is detected, it is decided that an error detection of the CRC has been overlooked. In that case, the error pointer is set in all the data (a)-(j), and the error correction is executed by a parity of the P sequence. Accordingly, impossibility of an error correction when an error of the CRC is overlooked is prevented, and the error correction is executed exactly.

Description

DETAILED DESCRIPTION OF THE INVENTION Field of the Invention The present invention relates to an error correction method when recording, reproducing, transmitting and receiving PCM signals.

Conventional Structure and Problems There is a method for improving the sound quality of recorded and reproduced audio by temporarily converting the audio signal into a digital signal.Hereinafter, a device in which this method is applied to a video tape recorder (VTR) will be described.

Converting an audio signal to a digital signal increases the frequency band, so the audio digital signal can be recorded and played back using the same video head as the video signal. Therefore, the first
As shown in the figure, the tape 2 is wound around the cylinder 1 at an angle of 1000 degrees or more, and the audio digital signal and video signal are connected to the same video head (for example, 3
Recording and playback are performed continuously in A). Figure 2 shows the recording pattern on the tape, and Figure 3 shows the timing during recording.
The video signal V is recorded alternately by the heads 3A and 3B for each field. On the other hand, the audio digital signal S is 1
Each field is compressed, and the winding is at the part where the angle is 180° or more. Recorded by head 3A.

Here, the audio digital signal has a frame structure so that error correction can be performed, and parity and CRC (Cyclic
c Redundancy Check Code) is added and recorded as a PCM audio signal. 1 in Figure 4
O8 indicating the data structure of the frame is a frame synchronization signal;
An is frame number %DnlDa-1-n"2a+n
'D3a+o...D7a+1 is the audio data word, Pn.

Qn is a parity word for error correction, and CRCn is a CRC word for error detection.

CRC is a method used to detect code errors. First, data is expressed as a polynomial, and the calculation is performed using a predetermined generator polynomial. The remainder is added to the data as a CRC. During playback, the data including the CRC is converted to the same generator polynomial. The presence or absence of an error is detected by performing division by 75 = if there is no error in the data, it is divisible, and if there is an error in the data, it cannot be completely divided and a remainder is generated.

Next, we will explain the conventional error correction method. First, the audio data word D and parity words Q and P of the reproduced PCM audio signal are written into the data memory, and at the same time, a CRC is calculated for each frame to detect errors. conduct.

If there is an error, a pointer is placed in the error pointer memory corresponding to all audio data words and parity words in that frame. Next, data is read from the data memory in the order of the parity series, and at the same time, the error pointer corresponding to the data is read from the error pointer memory, and error correction is performed using the erasure method.

In the error detection method using 80RC, which performs error correction using the above method, the presence or absence of an error is detected from the CRC calculation result as described above, so even if there is an error, it may be possible to divide the error. Sometimes we judge that something that exists is correct.

Therefore, even if an error is detected in the parity, the error pointer is not set, so there is a drawback that error correction cannot be performed in some cases.

Purpose of the Invention The present invention solves the above problems, and makes it possible to correctly correct errors without overlooking the existence of errors. This is an error correction method that performs a parity check before performing error correction processing on a PCM signal to prevent CRC error detection from being missed.

DESCRIPTION OF THE EMBODIMENTS FIG. 6 is a diagrammatic representation of the contents of the data memory, where the vertical direction represents one frame, the numbers at the top are frame numbers, and the parity series is taken in the diagonal direction.

The reproduced PCM audio signal is written to the data memory and at the same time error detection is performed by CRC calculation.If there is an error, a pointer is set in the error pointer memory.0 Next, in the order of the Q parity sequence Read data from data memory and perform parity check. If there is an error in the data d, the error is detected by the parity check of the Q sequence X. At this time, an error has been detected in the CRC of frame 5, and the error pointer of data d is standing. However, if frame 60 CRC misses error detection, no error pointer is placed on data d, and error correction cannot be performed. Therefore, all data a, b, c, d of the Q sequence.

"+'+q+hti+] error pointer is set. In this way, even though the parity check of the Q sequence is performed and an error is detected, if the error pointer is not set, it means that a CRC error detection was missed. judge,
An error pointer is placed on all Q-sequence data in which an error has been detected. After this operation, normal error and correction processing is performed. In other words, even if error pointers are set for all of the Q sequences, error correction can be performed using the parity of the P sequence, so error pointers will also be set for correct data, but if the CRC error detection is missed. This can prevent the inability to make corrections due to

Effects of the Invention According to the present invention, missed CRC error detection can be prevented by performing a parity check before performing error correction processing, and subsequent error correction processing can be reliably performed using only the normal erasure method. It is possible to provide a simple and excellent error correction method.

[Brief explanation of drawings]

Figure 1 is a block diagram of the head section of a VTR that records and plays PCM audio signals, Figure 2 is a diagram showing the recording pattern on the tape, Figure 3 is a timing diagram during recording, and Figure 4 is a diagram of the PCM audio signal. FIG. 5, which is a diagram showing the frame structure, is a state diagram illustrating a temporary data memory for explaining the present invention. Name of agent: Patent attorney Toshio Nakao and 1 other person No. 1
Figure A Figure 3-4IC. Figure 5×

Claims (1)

[Claims] An error correction method characterized by performing a parity check before performing error correction processing on a received or reproduced PCM signal.