JP3383561B2 - Decoding quality estimation device - Google Patents

Description

Detailed Description of the Invention

[0001]

The present invention relates to relates to the decoding quality estimation equipment suitable for digital mobile communications.

[0002]

2. Description of the Related Art CDMA (Code division multiple acc)
In digital mobile communication such as ess), an error may occur in data received by the receiving device side due to deterioration of a signal transmitted in space or mixing of noise. When communication is performed via a transmission line in which a signal deteriorates during propagation, such a problem occurs regardless of the magnitude. In order to correct such a signal error, a process of including redundant bits such as convolutional coding is performed when the signal is coded on the transmission side. Using these redundant bits, the received data can be repaired to restore error-free data. However, if the signal is severely degraded, it may not be possible to restore the signal. In such a case, the data is discarded in frame units, for example. In order to perform such processing, it is necessary to perform processing for estimating the decoding quality, which is whether or not there is an error in the decoded data.

As such a decoding quality estimation method, for example, a technique described in USP 513400 is known. This technique decodes the received data and then encodes it again. Separately, the received data is binarized separately. The decoded data has been subjected to error correction, and if this error correction is complete, it should have the same content as the transmitted data.

On the other hand, the received data is a transmitted signal including a transmission error and the like. Therefore, when the comparison is performed in the binarized state, the bits that do not match are the portions where the transmission error occurs. If there are many unmatched bits,
It is considered that there are many transmission errors and the transmission status is poor. Moreover, in such a state, it can be determined that an error exists in the decoded data. In this way, the decoding quality is estimated.

[0005]

By the way, the above conventional techniques have the following problems to be solved.
Each bit of the received data is called a reception symbol. Upon reception, each received symbol is processed as a multi-valued digital value. This process is called soft decision. In the soft decision signal, each bit is binarized in the subsequent decoding. This process is called a hard decision. In estimating the decoding quality, the re-encoded data and the hard-decided received symbol are compared. However, the reliability of received data differs depending on whether the soft-decision received symbol value is sufficiently large or conversely small. In the traditional way,
There is a problem in that the contents of the soft decision received symbol are not reflected in the estimation of the decoding quality.

On the other hand, there are cases where the received symbol is replaced with "0" on the receiver side for various purposes such as power control. The symbol which has been replaced with “0” is called a 0 symbol. The 0 symbol is a value that is neither 1 nor 0 even in the case of binary data. Even in such a case, the conventional method ends up estimating the decoding quality including this 0 symbol. However, since the 0 symbol is noise that has no influence on the decoding result, it is preferable to perform an appropriate exclusion process.

[0007]

The present invention adopts the following constitution in order to solve the above points. <Structure 1> A decoding unit that accepts and decodes the encoded reception symbol, an encoding unit that re-encodes the decoded data output by this decoding unit, and an output of this encoding unit for each bit of the reception symbol Controlling the sign of the received symbol, the degree of disagreement between the received symbol and the coded output, assuming that the multi-valued received symbol is binarized as it is to obtain the coded output. A decoding quality estimation apparatus comprising: a dissimilarity calculation unit that digitizes by using the above; and an accumulation unit that cumulatively adds the outputs of the dissimilarity calculation unit to obtain decoding quality estimation data.

[0008]

[0009]

[0010]

< Structure 2 > A zero-replacement unit that accepts encoded reception symbols and replaces the reception symbols at preset positions with zero symbols, and a decoding unit that accepts and decodes the output of this zero-replacement unit. , An encoding unit for re-encoding the decoded data output by this decoding unit, a multiplication unit for multiplying the output of this encoding unit with the received symbol, and an output of this multiplication unit are cumulatively added and decoded. A decoding quality estimation device, comprising: an accumulator for obtaining quality data.

< Structure 3 > In the apparatus according to Structure 2, the number of zero symbols included in the received symbols is counted, and the influence of the number of zero symbols included in the received symbols on the output of the accumulator is removed. A decoding quality estimation apparatus comprising a zero symbol number adjuster for performing a calculation process for

[0013]

[0014]

BEST MODE FOR CARRYING OUT THE INVENTION Embodiments of the present invention will be described below with reference to specific examples. <First Specific Example> FIG. 1 shows a block diagram of a signal receiving apparatus according to a first specific example. In the transmitter 10, only the encoding unit 11 and the modulation unit 12 are illustrated. The structure of the part of the transmitter 10 is the same as the conventional one. In the encoding unit 11,
The input signal is subjected to processing such as convolutional coding. Then, a redundant bit that enables error correction is added to this input signal and sent to the modulation unit 12. The modulator 12 modulates a binary input signal and transmits it. At this time, for example, PS
It is transmitted after being subjected to digital modulation such as K and FSK. This signal is received by the receiver 20.

The signal received through the antenna 21 is
Processing is performed by the receiver 20 including the demodulation unit 22, the decoding unit 23, the encoding unit 24, the mismatch degree calculation unit 25, the accumulation unit 26, the determination unit 27, and the reproduction unit 28. The demodulation unit 22 is a unit that demodulates a signal received by the antenna 21 and extracts a received symbol. The decoding unit 23 is, for example, a unit that decodes the convolutionally coded received symbols into bits before coding with the Viterbi algorithm. Even if the sign of the received symbol is inverted due to noise or the like, the redundant bits for error correction can be used for decoding without error. The encoding unit 24 is a unit that encodes the output of the decoding unit 23 thus obtained again. The content of this encoding is the same as that done at the transmitter.

The dissimilarity calculation unit 25 compares the output of the encoding unit 24 with the received symbol input to the decoding unit 23, and how much the content of the received symbol matches the content of the re-encoded symbol. To determine if For example, when the output of the encoding unit is "+1" and the received symbol is between +1 and +8, it means that the symbol is correctly received. Further, when the output of the encoding unit is "-1" and the received symbol is between -1 and -8, it means that the symbol is correctly received. In the case of this example, when the signal is correctly received, the non-coincidence degree computing unit 25 sets the sign of the received symbol to-and outputs it. When a signal is erroneously received, it has a function of setting the sign of the received symbol to + and outputting it.

Therefore, when the received symbol is, for example, "-2", "-2" is output to the accumulator 26. If the received symbol is "+2" even though the output of the encoding unit is "-1", the value "+2" is output to the accumulation unit 26 because the content is incorrect.

The accumulating section 26 is a section for accumulating the input signals as they are and outputting the result to the judging section 27. This is the data for decoding quality estimation. The determination unit 27 is a unit that compares an appropriate threshold value prepared in advance with the output of the accumulation unit 26 to determine the degree of error in the received signal. The determination result is output to the reproducing unit 28.
The reproducing unit 28 is a unit that reproduces the output of the decoding unit 23, and outputs the sound for each frame to a circuit for output processing in the case of a mobile phone, for example. The accumulator 26 is
The accumulated result is output to the determination unit 27 for each received frame,
It is initialized before processing the next frame.

The output of the accumulator 26 has a negative sign and a larger absolute value as the ratio of the output of the encoder 24 and the content of the received symbol is higher. If there are many disagreements, the values will be negative or positive with a small absolute value. Signals are actually transmitted and received under various environments, an appropriate threshold value is set, and when the output of the accumulator 26 exceeds the threshold value, the frame is discarded. That is, by making the part silent, reproduction of an erroneous signal is prevented.

In the present invention, the non-coincidence degree computing unit 25 accepts multi-valued received symbols and changes the codes depending on whether they agree with the result of encoding, and then accumulates the multi-valued values as they are. I do. Therefore, the feature is that the soft decision result of the received signal is reflected in the accumulation result, and the decoding quality of the signal can be estimated more accurately.

FIG. 2 shows an operation explanatory diagram of the apparatus of the first specific example. The operation of the apparatus shown in FIG. 1 and the contents of the output signal in each section will be described with reference to this figure. First, the encoding unit 11 of the transmitter 10 shown in FIG. 1 encodes a signal to be transmitted. At this time, a predetermined error correction bit is included in this signal as in the convolutional coding process. The content of the output signal is shown in T1 of FIG. In the following description of signal contents, a binary signal is "+1".
And the value "-1". Of course, the same applies when the binary signal is represented by "1" and "0". This is for distinguishing from the zero symbol value “0”. A zero symbol is a zero symbol inserted by depuncturing or a zero symbol replaced by a symbol used for other purposes. In addition, the mark * is placed in the part where there is no signal in the figure.

For example, when the content of the data before encoding is "+ 1 + 1-1-1 + 1 + 1 + 1-1 + 1-1" as shown in T1, the content becomes as shown in T2 after encoding. This is an example of performing a convolutional coding process with a rate of 1/2. Thus, the 9-bit information becomes 18 bits, and the redundancy (property) increases. Next, the modulator 12 digitally modulates this signal. Although the modulation method is arbitrary, for example, the well-known PSK or FSK method is used. This signal is transmitted from the transmitting antenna 13 and received by the receiving antenna 21. When a radio wave propagates in the air, the level of the signal is lowered and noise is mixed. This affects the decoding quality at the receiver 20. The same thing happens not only when propagating in the air but also when transmitted through a transmission line such as a cable. Therefore, the present invention can be used for estimating the decoding quality of signals transmitted and received through various signal transmission paths.

In the receiver 20, the demodulation section 22 digitally demodulates the received signal. The contents of the received symbols after demodulation are shown in R1 of FIG. Each received symbol is soft-decided and is represented by, for example, 4-bit multivalued data. A large absolute value of the received symbol means that the level of the received signal is high. The point where the sign is inverted from that at the time of transmission is where a signal error occurs.

Next, the decoding unit 23 performs a decoding process for each received symbol. As shown by R2 in FIG. 2, the decoded signal is restored by the error correction information, and the same content as the original signal T1 is faithfully reproduced. Here, the encoding is performed again in the encoding unit 24. The result of this encoding is shown in R3 of FIG. This content is the same as the output of the encoding unit 11 of the transmitter 10. The example shown in this figure is an example in which error correction is successful. if,
When the error correction is not successful, the decoding result is different from that of the transmitting side, and the encoding result is also different from that of the transmitting side.

However, the evaluation cannot be made only by looking at the decoding result. Therefore, in the dissimilarity calculation unit 25, the received symbol output from the demodulation unit 22 is compared with the encoding result output from the encoding unit 24. At this time, the received symbols are input to the disagreement degree computing unit 25 as they are in multi-value. For example, the 0th bit of the encoded signal R3 is “+1”. The corresponding received symbol is “+1” as indicated by R1. If the received symbol is in the range of +1 to +8, the contents are in agreement. The sign of is set to −, and the absolute value of the received symbol is left unchanged, and “−1” is output.

The first bit of the encoded signal R3 is "+1". The corresponding received symbol is “+3” as indicated by R1. Since the received symbol is in the range of +1 to +8, the contents thereof match. At this time, the dissimilarity calculation unit 25 determines the code of the received symbol. -, And the absolute value of the received symbol is left unchanged and "-3" is output.
On the other hand, the third bit of the encoded signal R3 is "+
1 ". In contrast, the corresponding received symbol is R
As shown in 1, it is "-3". Received symbols are +1 to
Since it is not within the range of +8, it can be seen that the content is incorrect. At this time, the non-coincidence degree calculation unit 25 outputs “+3” with the sign of the received symbol as + and the absolute value of the received symbol as it is.

By the above processing, the dissimilarity calculation unit 25
Outputs a signal having the contents shown in R4 of FIG. Next, the accumulator 26 cumulatively adds the signals thus obtained. The cumulative addition result is shown in (b) R5. Here, it is "-41". This is the decoding quality estimation value. When the above processing is performed, the more negative the sign of the output R5 of the accumulator 26 and the larger its absolute value, the higher the decoding quality, and conversely,
Even if the sign is negative, it can be determined that the decoding quality is poor when the absolute value is small or the sign is positive.

The judging section 27 judges the output result of the accumulating section 26. The output R5 shown in (b) is "-41". If the threshold serving as a criterion for determination is set to, for example, "-10", both are compared and it is determined that the decoding quality is good. As a result, the output of the decoding unit 23 is directly output to the reproducing unit 28 and processed as a received signal. In the case of a mobile phone, the received frame is converted into a voice signal and output from the speaker.

On the other hand, as shown in FIG. 2 (c), when the output R5 of the accumulator 26 has a value such as "-5", the decoding quality is below the threshold, so the frame is discarded. To do. The above process is thus executed for each frame of the received signal. Therefore, reproduction or discard processing is performed for each frame of the received signal. In the case of a voice call, if there are few frames with poor decoding quality, substantially good noise-free communication is continued. On the other hand, if there are many discarded frames, the call will be partially interrupted.

<Effect of Concrete Example 1> As described above,
The absolute value of the multi-valued reception symbol is the reception level of the signal. In an environment in which the transmission state is good, the absolute value of the signal R1 shown in FIG.
The code of 1 is the same. On the other hand, when the transmission state is poor, the absolute value of the signal R1 shown in FIG. 2 becomes small. If noise or the like is further mixed, the sign of the signal R3 and the sign of the signal R1 become opposite. According to this specific example, by reflecting the absolute value of the multi-level received symbol in the cumulative addition result, information including the signal level thereof is added, so that the reliability of the decoding quality estimation is improved.

As described above, the non-coincidence degree computing unit 25 determines whether or not the result of encoding and the content of the multi-level received symbol match, and outputs data for reflecting this in the cumulative addition result. It only has to be the part that does. Therefore, the configuration is arbitrary, and the decoding quality may be high when the cumulative result is negative.

<Second Specific Example> FIG. 3 shows a block diagram of a signal receiving apparatus according to a second specific example. This device is a specific example 1.
When compared with the case of
The output side of is different. The output of the mismatch calculation unit 25 is input to the hard decision unit 29, and the output of the hard decision unit 29 is input to the accumulation unit 26. That is, the difference is that a new hard decision unit 29 is inserted between the dissimilarity calculation unit 25 and the accumulation unit 26. The hard decision unit 29 monitors the output of the dissimilarity calculation unit 25, and if the result of the calculation processing is positive, "-1",
If it is negative, it has the function of outputting "+1". This circuit operation will be described by comparing the contents of signals as in the first specific example.

FIG. 4 shows an explanatory diagram of the operation of the apparatus of the second specific example. The output signals T1 and T2 in the transmitter 10 in the figure and the output contents of the decoding unit 23, the encoding unit 24, and the inconsistency degree calculation unit 25 in the receiver 20 are the same as those described in the first specific example. Here, the output R5 of the hard decision unit 29 will be described.

In the output R4 of the dissimilarity calculating section 25 shown in FIG. 4A, the 0th output is "-1". The sign is-. Therefore, the output of the hard decision unit 29 is "-
Further, the second signal is "-4" and its sign is-. Therefore, the output of the hard decision unit 29 is "-".
Further, the third signal is "+3". Since the sign is +, the output of the hard decision unit 29 is "+1".

As a result, the hard decision unit 29 shown in FIG.
Output R5 of the third, fourth and twelfth signals is "+1" and the other signals are "-1". When this data is cumulatively added in the accumulation unit 26, the result R6
Is "+3". Since the outputs of the hard decision unit 29 are only "-1" and "+1", the outputs of the accumulation unit 26 are all positive. It can be said that the larger the absolute value is, the lower the decoding quality is.

Therefore, for example, the determination unit 27 determines the threshold value as "-12" and determines the output R6 of the accumulation unit 26. In the example shown in FIG. 4B, the output is "-15", so it is determined that the decoding quality is good. And
The determination unit 27 outputs to the reproduction unit 28 a control signal R7 indicating that normal reception should be performed. On the other hand, for example, the accumulator 26
Output R6 is "-10" as shown in FIG. 4 (c). In this case, the value is equal to or larger than the threshold value set by the determination unit 27. From this, it is determined that the decoding quality is poor, and the control signal R7 indicating that the frame should be discarded is output to the reproducing unit 28. These controls are performed for each frame as in the case of the first specific example.

In the above process, it is sufficient to send only the sign bit included in the calculation result of the mismatch calculation section 25 to the hard decision section 29. This is one line for code control.
It can be realized only by making a main connection. That is, the dissimilarity calculation unit 2
It is not necessary to output the output of 5 to the hard decision unit 29 in a multivalued or collectively stored state in the memory. Further, since the hard decision unit 29 outputs only "-1" or "+1", the accumulation result of the accumulating unit 26 has a smaller numerical value than that of the output of the inconsistency degree calculating unit 25 as it is. Expressed. If this is stored in the memory, the word length of the accumulator can be made sufficiently small.

For example, in the above example, when the outputs of the hard decision section 29 are all "+1", the accumulation result of the accumulation section 26 becomes the maximum value, which is "18". Therefore, the result can be expressed by a 4-bit digital signal. on the other hand,
If the output of the dissimilarity calculator 25 is cumulatively added as it is,
The maximum value is 8 × 18 or 144. This must be represented by an 8-bit digital signal. As a result, in the case of the specific example 2, the information for decoding quality estimation can be obtained with a word length of ½.

<Effect of Concrete Example 2> As described above, according to the concrete example 2, the register for holding the result of the mismatch degree calculation processing is not required, the operation in the hard decision section is simplified, and the accumulation is further improved. It is possible to shorten the word length of data in the arithmetic unit.

<Specific Example 3> FIG. 5 is a block diagram of a signal receiving apparatus according to a specific example 3. In the case of this specific example, the comparison unit 30 is inserted between the mismatch degree calculation unit 25 and the accumulation unit 26 of the receiver 20. The comparison unit 30 compares the output of the dissimilarity calculation unit 25 with a preset threshold value, and “+1” when the output of the dissimilarity calculation unit 25 is larger than the threshold value, and “−1” when the output is less than or equal to the threshold value. It has a function to output the result "". For example, here, the description will be given by setting the threshold value to “−1”.

FIG. 6 is an operation explanatory diagram of the apparatus of the third specific example. 6A, T1, T2, R1, R2, R3, R
The signal of No. 4 is the same as that of the specific example 2. The comparison unit 30 illustrated in FIG. 5 compares each signal of the output R4 of FIG. 6A with the threshold “−1”. As a result, the 0th signal "-
Since 1 ”is a value equal to or larger than the threshold, the comparison unit 30 sets“ +1 ”.
Is output. Since the second signal "-3" has a value less than the threshold value, the output of the comparison unit 30 is "-1". When such processing is performed, 0, 3, 4, 5, 8, 12, 14
The output of the comparison unit 30 for each th signal is "+1", and the output for the other signals is "-1". The accumulator 26 thus adds the outputs of the comparator 30. As a result, R6 becomes "-11".

The judging section 27 sets a threshold value of "-8", for example. The larger the value of the output R6 of the accumulator 26, the worse the decoding quality. Therefore, when the output less than the threshold value "-8" is obtained from the accumulation unit 26, the determination unit determines that the reception is normal, and outputs the control signal R7 to that effect. On the other hand, as shown in FIG. 6C, when the output R5 of the accumulating unit 26 reaches a value of "-6", for example, it exceeds the threshold value, and therefore a control signal R6 of frame discard is output. Become.

The threshold value set in the comparison unit 30 may be determined empirically, and the standard is as follows. That is,
When the signal attenuates on the transmission path from the transmitter to the receiver,
The absolute value of the received symbol received in multi-value becomes small.
In some cases, it may be determined that the symbol having an absolute value near “+1” has low reliability. Therefore, as shown in the above example, the threshold is set to "-1", and the received symbol when the absolute value is "+1" is processed not to be included in the decoding quality judgment.

<Effect of Concrete Example 3> According to the above apparatus,
Since an appropriate threshold value is set for the output of the accumulating unit and the results are accumulated, more reliable decoding quality evaluation can be performed.

<Fourth Embodiment> FIG. 7 shows a block diagram of a signal receiving apparatus according to a fourth embodiment. This device has a demodulation unit 22 and a decoding unit 2 when compared with the example of the specific example 1 shown in FIG.
The difference is that the zero-replacement portion 32 is provided between the three. Further, a multiplication unit 33 is arranged here instead of the dissimilarity calculation unit 25. Further, the hard decision unit 29 is arranged on the output side of the multiplication unit 33, and the output thereof is cumulatively added in the accumulation unit 26.

The zero-substitution unit 32 performs zero-substitution processing on a portion preset as a zero symbol in the signal output from the demodulation unit 22. Further, the multiplication unit 33 receives the output of the encoding unit 24 and the output of the zero substitution unit 32 and outputs the multiplication result of both. The hard decision unit 29 binarizes the output of the multiplication unit 33. The device configured as described above operates as follows.

FIG. 8 shows an operation explanatory diagram of the apparatus of the fourth specific example. The output R1 from the demodulation unit 22 of the receiver 20 is exactly the same as the previous examples. Here, the zero substitution unit 32 receives the output of the demodulation unit 22 and obtains the output as shown by R2 in FIG. That is, the portion corresponding to the zero symbol included in the output of the demodulation unit 22 is replaced with “0”. In this example, the portion corresponding to the zero symbol is the 2nd, 4th, 8th, 10th, and 16th signals. It is clear in advance which position the zero symbol is included in one frame on both the transmitting side and the receiving side. The signal in this portion is forcibly replaced by "0" in the zero replacement unit 32.

The coding section 24 decodes the output of the zero substitution section 32, and the coding section 24 codes the output of the decoding section 23. Its function is exactly the same as that of the specific examples so far. And
The outputs shown at R3 and R4 in FIG. 8 are obtained.

Next, the multiplication unit 33 receives the output of the encoding unit 24 and the output of the zero substitution unit 32, and multiplies both of them. For example, the output R4 and the output R2 shown in FIG.
The second signals are "+1" and "+1", respectively. Therefore, the multiplication result of both is "+1". The second signals are "+1" and "+3", respectively. Therefore, the multiplication result is "+3". On the other hand, the R3 of the third signal is “+1” on the output side of the encoding unit, but the output R2 of the zero substitution unit 32 is “0”. This is because the part corresponding to the zero symbol of the received symbol is forcibly "0".
It was replaced by. Therefore, the multiplication result of both is "0".

As described above, when the zero symbol portion is replaced with "0" in advance, the multiplication result of the corresponding portion is always "0". That is, regardless of the content of the portion corresponding to the zero symbol in the received symbol, the multiplication unit 33
The output R5 of the above becomes "0". The result thus obtained is
A value that is input to the hard decision unit 29 and that exceeds “0”, that is, a value that is “1” or more, is set to “+1”, and a value that is less than “0”, that is, “0, −
1, -2, ... "Are set to" -1. "Thus, the output R6
Is obtained. When this output R6 is cumulatively added, the accumulator 2
The output R6 of 6 becomes "+1". This is the information for estimating the decoding quality.

In this example, the larger the accumulation result of the accumulator 26, the higher the decoding quality. Therefore, for example, the threshold value in the determination unit 27 is set to "-2".
As a result, the output result R6 of the accumulator 26 is estimated to be in a normal state. In this way, from the determination unit 27 to the reproduction unit 2
A control signal R7 for instructing normal reception is output to No. 8.
Note that, for example, if the output of the accumulating unit 26 is "-5", the value is less than or equal to the threshold value, and therefore the determining unit 27 outputs the control signal R6 for instructing frame discard to the reproducing unit 28.

In this specific example, the portion corresponding to the zero symbol included in the output of the demodulation section is forcibly replaced with the signal of the value "0" by the zero replacement section. This serves to uniformly process the signal of the part corresponding to the zero symbol from the output of the encoding part in the multiplication part. The signal corresponding to the zero symbol in the output of the encoding unit is not substantially related to the quality of the voice call itself, for example. That is, this is a signal for communication control, and the content of the signal is processed separately. Therefore, it is not necessary to include it in the criteria for determining whether or not to discard frames.

However, in the above-described specific examples, all of them are uniformly used as the information for estimating the decoding quality. In this specific example 4, as a result of being processed by the multiplication unit, all outputs of the portion corresponding to the zero symbol are “0”, which are constant values. Therefore, the decoding result of the portion corresponding to the zero symbol is a constant value even after the hard decision processing and the accumulation processing are performed thereafter. It has no effect. In the above example, the portion corresponding to the zero symbol as a result of the hard decision is "-1", so the total "-6", which is the processing result for the 2nd, 4th, 8th, 10th and 16th signals, is ignored. By doing so, it is possible to estimate the decoding quality only for the signal for which frame discard control should be originally performed.

In the above example, the result of the multiplication section is processed by the hard decision section before cumulative addition is performed. However, the output of the multiplication unit may be directly output to the accumulation unit 7 to obtain the accumulation result using the same multivalued absolute value as that of the first specific example. Even in this case, the zero symbol does not affect the output result of the accumulating unit 26 due to the action of the zero replacing unit 32.

<Effect of Concrete Example 4> As described above, the zero symbol does not affect the accumulation result, so that the reliability of the decoding quality estimation information can be further improved.

<Fifth Embodiment> FIG. 9 shows a block diagram of a signal receiving apparatus according to a fifth embodiment. Receiver 20 of this specific example
In this example, a zero counting unit 34 and a 1/2 dividing unit 35 are provided on the output side of the multiplying unit 33 of the device according to the fourth specific example. further,
A subtraction unit 36 is provided that takes the difference between the output of the accumulation unit 26 and the output of the 1/2 division unit 31. The zero count unit 34 is the multiplication unit 33.
This is the part that counts the number of zero symbols contained in it and outputs the numerical value. Further, the ½ division unit 35 is a unit that divides the output of the zero count unit 34 into halves. In addition,
When the position and number of zero symbols are clearly known from the beginning, the zero count unit 34 can output a constant numerical value without using the output of the multiplication unit 33.

As will be described later, the ½ division unit 35 is a unit that adjusts the number of zero symbols to a constant value in consideration of the influence of zero symbols on the output result of the decoding quality. The subtraction unit 36 is a unit that subtracts the output of the 1/2 division unit 35 from the output of the accumulation unit 26 and uses the result as information for decoding quality estimation.

FIG. 10 shows an operation explanatory diagram of the specific example 5.
The number of zero symbols is not always constant.
In some cases, the location and the number are unknown on the receiving side. In such a case, the zero counting unit 34 counts the data whose value included in the output of the multiplication unit 33 is "0". In addition,
For example, in the example shown in FIG. 8, the 14th signal happens to have a value of "0" even though it is not a zero symbol. Although this is included as noise, it is unlikely that the number will affect the whole, so it can be ignored and counted. The zero symbol number adjuster 40 including the zero counting unit 34, the 1/2 dividing unit 35, and the subtracting unit 36.
I will call it.

The horizontal axis of FIG. 10 shows the decoding quality, and the vertical axis shows the number of frames. That is, when the receiver receives a large number of frames, the horizontal axis is the decoding quality value obtained as a result of arithmetic processing, and the number of frames with the same decoding quality value is plotted and graphed. Like For example, when only frames that do not contain zero symbols are received and the decoding quality is calculated, it is considered that the number of corresponding frames is approximately evenly distributed around the X1 in the figure. However, when zero symbols are included, the decoding quality value shifts by the number of zero symbols,
The center becomes X2 and shows the same distribution. When a constant zero symbol is always included, the determination threshold of the determination unit 27 may be set to be constant.

However, if the zero symbol is included or not included, or if the number of zero symbols fluctuates, if the threshold value of the decision unit is kept constant, the decision of the decoding quality is erroneous. Cause Therefore, the apparatus shown in FIG. 9 counts the number of zero symbols and removes the influence of the zero symbols when the result of cumulative addition, that is, the output of the accumulator 26 changes. Zero symbol number adjuster 40
Is a circuit having such a function. Therefore, the zero symbol number adjuster 40 is not limited to the above example, and for example, accepts the output of the zero replacement unit 32 as it is, counts the number of zero symbols, or multiplies the number of zero symbols by a value other than 1/2. Or addition / subtraction can be performed to make adjustment so that the decoding quality is not significantly affected by the number of zero symbols.

<Effect of Concrete Example 5> According to this example, the decoding quality can be judged by the judging section using a constant threshold value without affecting the fluctuation of the number of zero symbols. Therefore, the information for decoding quality estimation can be made more reliable.

[Brief description of drawings]

FIG. 1 is a block diagram of a signal receiving device according to a specific example 1;

FIG. 2 is a diagram for explaining the operation of the device of the first specific example.

FIG. 3 is a block diagram of a signal receiving device according to a second specific example.

FIG. 4 is a diagram for explaining the operation of the device of the second specific example.

FIG. 5 is a block diagram of a signal receiving device according to a specific example 3;

FIG. 6 is a diagram illustrating the operation of the device of the third specific example.

FIG. 7 is a block diagram of a signal receiving device according to a specific example 4;

FIG. 8 is an explanatory diagram of the operation of the device according to the fourth specific example.

9 is a block diagram of a signal receiving device according to a specific example 5. FIG.

FIG. 10 is an operation explanatory diagram of the fifth specific example.

[Explanation of symbols]

10 transmitter 11 Encoding section 12 Modulator 20 receiver 22 Demodulator 23 Decryptor 24 Encoding unit 25 Disagreement calculator 26 Accumulation part 27 Judgment section 28 Playback section

─────────────────────────────────────────────────── ─── Continuation of the front page (58) Fields surveyed (Int.Cl. ⁷ , DB name) H04L 1/00 H03M 13/12

Claims (3)

(57) [Claims] 1. A decoding unit that accepts and decodes encoded received symbols, an encoding unit that re-encodes the decoded data output by this decoding unit, and an output of this encoding unit for each bit of the reception. Compare and collate with the symbol, and the received symbol expressed in multi-value is directly 2
A disparity degree calculation unit that digitizes the dissimilarity degree between the received symbol and the coded output when it is assumed that the coded output is obtained by the value conversion, and the disparity degree calculation unit A decoding quality estimation device, comprising: an accumulator that cumulatively adds outputs to obtain decoding quality estimation data. 2. A zero-replacement unit that accepts coded received symbols and replaces the received symbols at preset positions with zero symbols, and a decoding unit that accepts and decodes the output of the zero-replacement unit. An encoding unit that re-encodes the decoded data output by the decoding unit, a multiplication unit that multiplies the output of this encoding unit and the received symbol, and the output of this multiplication unit is cumulatively added to obtain the decoded quality data. A decoding quality estimation device comprising: 3. The apparatus according to claim 2, wherein the number of zero symbols included in the received symbols is counted, and the influence of the number of zero symbols included in the received symbols on the output of the accumulator is eliminated. And a decoding quality estimation apparatus including a zero symbol number adjuster for performing a calculation process.