JP2003283468A - Error correction processing method and transmission apparatus - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a transmission apparatus for transmitting a digital signal, to which transmission line coding processing such as error correction coding is applied, that enhances a successful rate of error correction so as to realize transmission with higher reliability. <P>SOLUTION: The transmission apparatus transmits the digital signal to which transmission line coding processing such as error correction coding is applied. A receiver side applies preliminary correction processing to error information on a signal form of a received signal in the case of applying decoding processing to the signal and then applying prescribed error correction processing thereto. Thus, even a TS signal including many errors and having been so far disabled of correction can be corrected and the frequency of occurrence of stopping the image demodulation processing can be decreased. <P>COPYRIGHT: (C)2004,JPO

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to improving the reliability of digital data output from a digital transmission device.

[0002]

2. Description of the Related Art Recently, a technique is widely used in which a video or audio signal is converted into a digital signal, data is compressed by MPEG processing, and then transmitted by a modulator or a demodulator capable of transmitting digital data. An example of the configuration of the transmission device is shown in FIG. 5 and will be described below. On the transmitting side, the video signal is MPE
The compressed data is input to the G encoder 1T. The structure of this compressed data is 188 with 8 bits as 1 word.
The form is called a transport stream (hereinafter referred to as TS) signal composed of a large number of packets in units of words. The TS signal starts with a 47h code, and the following 187 words mainly contain compression information, but there are also packets having other information shown below. 1) PID information consisting of 13 bits corresponding to a name tag when a plurality of streams are mixed. 2) A 4-bit cyclic counter value indicating the order of TS signals. 3) P consisting of 42 bits indicating the time timing of decoding
CR (Program Clock Reference) signal. This TS signal is DVB-ENC (Digital Video Broadband).
It is input to casting-Encoder 2 and DVB (Digital Vide)
o Broadcasting) processing is performed to obtain a DVBe signal. This DVBe signal is, for example, OFDM-modulated by the modulator (MOD) 3T and becomes an IFt signal which is an intermediate frequency in the 130 MHz band. The 180 MHz band IF signal is sent to the transmission high frequency section (Th) 4-1 and converted into a microwave band signal and then power amplified. Then, it is radiated as a radio wave from the transmitting antenna 4-2.

The radiated radio wave passes through the transmission line 5 and
It reaches the receiving antenna 6-1. It should be noted that in the state of the transmission path 5, the strength of the radio wave is also attenuated according to the distance, so that noise and the like are mixed. The radio waves collected by the reception antenna 6-1 are input to the reception high frequency section (Rh) 6-2,
After being amplified, the frequency is converted to the intermediate frequency 130 MHz band and output as an IF signal in the 130 MHz band. This 130 MHz IFr signal is used for the demodulation unit (D
EM) Input to 3R, O corresponding to the same processing as modulation
FDM demodulation is performed and output as a DVBr signal.
This DVBr signal is input to the DVB-DEC (decoder) 8 to become a TSr signal and is returned to the original transport signal. The MPEG decoder 2R expands the TSr signal and outputs the original video signal.

FIG. 6 shows an example of a concrete configuration of the DVB-ENC2, which will be briefly described. In the TS signal, the head word 47h is replaced with B8h in the B8h replacing section 2-1, the energy spreading processing is performed in the energy spreading section 2-2, and the error correction parity information is set in the Reed Solomon (RS) coding section 2-3. 16 words are added. By this processing, the TS signal has a 204-word structure. Then, the interleaving unit 2-4 performs interleaving processing which is data order rearrangement for effectively performing error correction. Since the above-mentioned energy diffusion processing has a period of 204 words × 8, the head word 47h of the TS signal is converted to B8h as a reference. In Figure 7, D
An example of a specific configuration of the VB-DEC 8 will be shown and briefly described. The input DVBr signal is detected by the 47h detection unit 8-
The first word 47h is detected at 1, and with this 47h as a reference, the reverse interleaving unit 8-2 performs the reverse interleaving process to restore the original data order rearrangement, and the RS correction unit 8-3 performs the Reed-Solomon error correction to reverse the energy. Diffuser 8-4
Then, the energy despreading process is performed, and the original TS signal is obtained. In the error correction performed here, if the error is up to 8 words, which is a half of the number of additional parities, the original correct word can be corrected. However, when the number of error words is 9 or more, the error becomes uncorrectable and the error remains. Further, when the number of errors is large, it cannot be recognized that the error cannot be corrected, and the error is corrected. That is, not only the word having an error is left as it is, but also a word having a correct value having no error is erroneously corrected to an erroneous value. As a result, the number of errors is increased compared to the original state.

Here, the concept of RS correction will be briefly described. In the RS correction, when 16 words of parity are given, eight error positions and eight values thereof can be calculated, so that eight unknown errors can be corrected. That is, with respect to the error of half of the added parity number, the position and the wrong value can be calculated. Since equations can be created by the number of parities, it is possible to obtain as many unknowns as there are equations. Since we can make equations for the number of parity 16,
16 unknowns are found. However, since the formula used for the verification is eliminated, when many errors are mixed in, the frequency at which erroneous correction cannot be detected increases. If the number of corrections is reduced, some formulas can be used for verification and the frequency of erroneous corrections is reduced. The RS correction can reduce the frequency of erroneous recognition of the number of errors unless the correction capability is fully utilized.

[0006]

In the conventional configuration described above, when the number of errors is large, the number of errors is further increased due to incorrect correction processing, and the MPEG decoding processing is further deteriorated. This deterioration sometimes gives a big error to the MPEG decoder, such as waiting for data that should not arrive, and also lengthens the number of seconds in which an abnormal image occurs. The present invention aims to eliminate these drawbacks, utilize the regularity of the TS signal, reduce the number of errors, and increase the success rate of error correction.

[0007]

In order to achieve the above object, the present invention provides a decoding process on the receiving side in a transmission device for transmitting a digital signal which has been subjected to a transmission line coding process such as error correction coding. At this time, a predetermined error correction process is performed after the pre-correction process of the error information in the signal format of the received signal. In addition, at least a video signal is digitally compression-encoded, a transmission path encoding process such as error correction encoding is performed, and a transmission unit that digitally modulates and then converts into a transmission signal, and a digital demodulation process that receives the transmission signal,
In a transmission device comprising a transmission line decoding process such as error correction and a digital compression information decompression process reception unit, the transmission line decoding unit of the reception unit pre-corrects error information in the signal format in the received signal. After that, a means for performing a predetermined error correction process is provided. In addition, the pre-correction process,
1) If the word value of the first word of the compressed and encoded transport stream packet of the received signal is not 47h, 47h
2) When the word value of the 1632nd word is not B8h,
B8h, 3) If the 13-bit PID value is a value that is not used in the target compressed data, replace it with a value that is used, and 4) add it just before the adaptation field of the transport stream. If the 4-bit cyclic counter value differs from the previous cyclic counter value of the transport stream having the same PID by +1, +1
5) If the upper bits of the SCR code and PCR code are different from the previous SCR code and PCR code, at least one of the previous upper bits is replaced. It was done. That is, the word structure of a compression-encoded transport stream such as MPEG is not completely random but a certain rule (signal format).
There are parts described according to. Therefore, in the present invention, the word portion that is contrary to the above-mentioned signal format is pre-corrected to a correct word in advance, the number of errors is reduced to a predetermined number (8 words) or less, and thereafter, error correction is performed at full power to eliminate residual errors. There is no or nothing.

[0008]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS FIG. 1 shows an embodiment of the configuration of a transmission apparatus of the present invention, which will be described in detail below. This is the DVB-DEC8 shown in FIG.
This is a configuration in which B-DEC7 is substituted. An example of a specific configuration of the DVB-DEC 7 with pre-correction is shown in FIG. 2 and will be described. The 47h detection unit 8-1 detects the leading word 47h of the DVBr signal, and the DVBr signal is input to the deinterleaving unit 8-2 with reference to this 47h and becomes a dii signal. Further, the 47h detection unit 8-1 outputs the ST1 signal indicating the 47h detection. This dii signal is a pre-RS correction unit 7-1.
To be a dfrs signal. The dfrs signal is input to the energy despreader 8-4 and becomes a die signal. Then, the die signal is input to the pre-correction unit 7-2 and becomes the dpc signal. The ST1 signal is input to the deinterleaving unit 8-2, the pre-RS correction unit 7-1, the energy despreading unit 8-4, and the pre-correction unit 7-2 as a timing start signal. The pre-correction unit 7-2 receives the input ST1.
The ST2 signal delayed by the pre-correction processing time is output. The dpc signal is input to the energy diffusion unit 2-2 and becomes a dpe signal. The dpe signal is applied to the RS correction unit 8
-3 and becomes a dprs signal. The dprs signal is
It is input to the energy despreading unit 8-4 and becomes a TS signal.
The ST2 signal includes an energy diffusion unit 2-2 and an RS correction unit 8
-3, and is input to the energy despreading unit 8-4.

Next, the operation of each section will be described. Four
The 7h detection unit 8-1 detects, for example, 204 bytes in the DVBr signal.
ST1 which is a pulse indicating a temporal position of 47h (0100 0111) words existing in the word cycle is searched.
Output a signal. The deinterleaving unit 8-2 is ST1.
Rearrange the order of data according to the signal. The purpose of this is to disperse the data with the error correction code in advance and transmit the continuous error mixed in the transmission line by returning the data sequence to the original order, and the continuous error will be dispersed to 8 words or less. This is because the frequency of corrections increases. According to the ST1 signal, the pre-RS correction unit 7-1 adds 16th to 189th words to 204th words of 204 words.
Error correction processing is performed according to the word parity. Note that the parity is 16 words, but correction is not performed up to 8 words, and is limited to, for example, about 5 words. The energy despreading unit 8-4 multiplies the previously promised spreading pattern according to the ST1 signal so that the data does not have a DC bias.

The pre-correction unit 7-2 corrects a word that violates the law of compressed data. The processing performed here will be described below. (1) 47 if the word value of the first word is not 47h
Replace with h. (2) If the word value of the 1632nd word is not B8h, it is replaced with B8h. (3) If the 13-bit PID value is a value that is not used in the target compressed data, it is replaced with the value that was used. (4) If the 4-bit cyclic counter value added just before the adaptation field of the TS signal is different from the previous cyclic counter value of the transport packet having the same PID, the value is incremented by +1. Replace. (5) SCR (System Clock Reference) which is time information
In the code and PCR code, the upper bits that should not change in a short time are the previous SCR code and P
If it is different from the CR code, it is corrected to the value of the previous high-order bit.

The energy diffusing unit 2-2 includes a pre-correcting unit 7
In order to convert the output dpc from -2 into a form in which the RS correction code is added, the previously promised spreading pattern is again multiplied according to the ST2 signal. The RS correction unit 8-3 is ST
The error correction is performed by setting the correction capability to the maximum with reference to the timing of the start word of 204 words according to the two signals. Since the words that are mixed in the transmission path and violate the law of compressed data are corrected by the pre-correction unit 7-2, the frequency of errors is reduced. That is, the number of errors contained in more than 8 words in 204 words is corrected to 8 words or less. Therefore, the frequency with which all errors in the 204 words can be corrected also increases. The energy despreading unit 8-4 multiplies a previously promised spreading pattern according to the ST2 signal again so as to prevent DC bias in the data, and restores the TS signal.

FIG. 3 shows an example of a specific configuration of the pre-correction unit 7-2, which will be described. Die signal is 47h correction unit 7
-2-1 is input. The output of the 47h correction unit 7-2-1 is input to the B8h correction unit 7-2-2. The output of the B8h correction unit 7-2-2 is input to the PID correction unit 7-2-3. The output of the PID correction unit 7-2-3 is input to the cyclic counter correction unit 7-2-4. The output of the cyclic counter correction unit 7-2-4 is the SCR & PCR upper correction unit 7-2.
Input to -5. SCR & PCR upper correction unit 7-2
The output of -5 is output as the dpc signal. The ST1 signal is 47h correction unit 7-2-1 and B8h correction unit 7-2-
2, PID correction unit 7-2-3, cyclic counter correction unit 7-
2-4, the SCR & PCR upper correction unit 7-2-5, and the delay 7-2-6. delay7-2
The output of -6 is output as the ST2 signal.

Next, the operation will be described. Each part indicates the position of the first word of the TS signal according to the ST1 signal for each 204 words. 47h correction unit 7-2-1
When the first word of the input TS signal, ie, the die signal, has a value other than 47h, only the first word is 47h.
Replace with. Specifically, a total of 8 bits from the 1st bit to the 8th bit are replaced with 47h. B8h correction unit 7-2-
2 is 1632 of the die signal which is the input TS signal
If the word value of the word is a value other than B8h, only this word is replaced with B8h. Specifically, a total of 8 bits from the 1632nd bit to the 1639th bit are replaced with B8h. The PID correction unit 7-2-3, from the 12th bit to the 24th bit, until the fixed time when the TS signal is input.
Various PID values existing in many packets are stored for a total of 13 bits up to the bitth bit, and the PI with the highest appearance frequency is stored.
Obtain the D value. Then, after receiving the transmitted TS data for a certain period of time, when a new PID value that has not been stored in the memory up to now, that is, has not appeared until then, is detected, this PID value is regarded as an error and the frequency of occurrence is the highest. High PID
Replace with the value. Alternatively, the PID value is estimated and replaced from the increase status of the cyclic counter value described below.

The cyclic counter correction unit 7-2-4 extracts the cyclic counter value existing just before the adaptation field. Specifically, the 4th bit from the 29th bit to the 32nd bit is the target. Originally the same P as last time
Since it is a value obtained by adding +1 to the cyclic counter value of the packet having the ID value, if the previous value is not +1 it is replaced with the previous value +1. The SCR & PCR upper correction unit 7-2-5 stores the SCR and PCR values which are included in the adaptation field and whose presence is determined by the flag. This SCR value is a 42-bit value obtained by measuring the passage of time at 27 MHz intervals. The existing cycle is about 100 ms at the maximum, and there is little change in the lower 16 bits, so if the upper 26 bits differ from the previous SCR and PCR values,
Replace with the previous value. FIG. 4 shows the configuration of a transmission line decoding unit that does not perform pre-RS correction. This configuration is shown in Figure 2.
The pre-RS correction unit 7-1 is removed from the decoding unit shown in FIG. As an effect, since the correction processing limited to about 5 words is not performed, the performance of the comparison judgment with the rule performed in the pre-correction is slightly lowered, but an equivalent effect is obtained.

[0015]

As described above, according to the present invention, even in the case of a TS signal which has a large number of errors and cannot be corrected by the conventional method, a case can be corrected, and the frequency of stopping the image demodulation processing is reduced. be able to.

[Brief description of drawings]

FIG. 1 is a block diagram showing an embodiment of the overall configuration of a transmission device of the present invention.

FIG. 2 is a block diagram showing an embodiment of the configuration of a DVB-DEC with pre-correction according to the present invention.

FIG. 3 is a block diagram showing an embodiment of a configuration of a pre-correction unit of the present invention.

FIG. 4 is a block diagram showing an embodiment of the configuration of a DVB-DEC with pre-correction according to the present invention.

FIG. 5 is a block diagram showing an example of the overall configuration of a conventional transmission device.

FIG. 6 is a block diagram showing an example of a configuration of a conventional DVB-ENC.

FIG. 7 is a block diagram showing an example of a configuration of a conventional DVB-DEC.

[Explanation of symbols]

1T: MPEG encoder, 2: DVB-ENC, 2
R: MPEG decoder, 3T: Modulator (MOD), 3R:
Demodulation unit (DEM), 4-1: Transmission high frequency unit (Th), 4-
2: Transmitting antenna, 5: Transmission line, 6-1: Receiving antenna, 6-2: Receiving high frequency part (Rh), 7: DV with pre-correction
B-DEC, 2-2: Energy diffusion unit, 7-1: Pre-RS correction unit, 7-2: Pre-correction unit, 8-1: 47h detection unit, 8-2: Deinterleave unit, 8-3: RS Correction Department,
8-4: Energy despreading unit.

Claims (3)

[Claims] 1. In a transmission device for transmitting a digital signal which has been subjected to a transmission path coding process such as error correction coding, in a decoding process on a receiving side, a pre-correction process is performed on error information in a signal format of a received signal. After that, a predetermined error correction processing is performed, and an error correction processing method characterized by the above-mentioned. 2. A transmitting unit for digitally compressing and coding at least a video signal, performing transmission path coding processing such as error correction coding, and digitally modulating and then converting it into a transmission signal, and a digital demodulation processing for receiving the transmission signal. In a transmission device including a receiving unit that performs a transmission line decoding process such as error correction and a decompression process of digital compressed information, the transmission line decoding unit of the receiving unit is configured to display error information in a signal format in a received signal. A transmission device comprising means for performing a predetermined error correction process after the correction process. 3. The transmission device according to claim 2, wherein the pre-correction process is performed by: 47h if 1) the word value of the first word of the transport stream packet of the received signal compressed and encoded is not 47h.
2) When the word value of the 1632nd word is not B8h,
B8h, 3) If the 13-bit PID value is a value that is not used in the target compressed data, replace it with a value that is used, and 4) add it just before the adaptation field of the transport stream. If the 4-bit cyclic counter value differs from the previous cyclic counter value of the transport stream having the same PID by +1, +1
5) If the upper bits of the SCR code and the PCR code are different from the previous SCR code and PCR code, at least one of the processing of replacing with the previous upper bit is performed. Characteristic transmission device.