JPS6251541B2 - - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention relates to an adaptive differential PCM (ADPCM) method that is applied to, for example, band compression transmission of voice band signals and is equipped with an adaptive quantizer and an adaptive predictor, and an apparatus used therefor. It is.

As a method for encoding and transmitting an audio signal waveform with a low transmission bit capacity, it utilizes the fact that there is a correlation between the sample values when the audio signal waveform is sampled at the Nyquist rate. In other words, the signal at the current time is predicted using a signal sequence transmitted at least one sample time before the current time, and the residue obtained by subtracting the predicted value from the signal at the current time, that is, the prediction residual, is quantized and encoded. Differential PCM method for transmission, so-called DPCM (Differential Pulse Code)
Modulation method is known.

Furthermore, as a method to reduce the transmission bit capacity, an adaptive quantizer is used in which the quantization step size in the quantizer is adaptively changed according to the input signal of the quantizer, and a prediction coefficient value in the predictor is changed in a predicted residual manner. ADPCM with an adaptive predictor that changes adaptively so that the difference is always minimized
(Adaptive DPCM) method is known. this
The adaptive predictor in the ADPCM method can be divided into two types depending on the method of modifying the prediction coefficients.

One of them is called a forward adaptive predictor, which calculates an autocorrelation function using an input signal of a certain time length, and uses this autocorrelation function to minimize the prediction residual at a certain time interval. It calculates the predicted coefficient value. For details, see the American magazine Bell Systems Technical by BSAtal et al.
Journal, October 1970 issue (BSTJ,
“Adaptive Predictive Coding of Speech Signals” (October 1970, PP, 1973-1986)
Predictive Coding of Speech, Signals.''

The other type is called a backward adaptive predictor, which uses the quantized prediction residual and the input value of the predictor to sequentially modify the prediction coefficient values to minimize the prediction residual. It is something to do.
For details, see the American magazine “IEE Transactions on Communications September 1975” (IEEE
Transactions on Communications;
September 1975, pp.935-941) “The Residential Encoder
Speech Digestification” (“The
Residual Encoder−An Improved ADPCM
It is described in document 2 entitled "System for Speech Digitization").

Adaptive quantizers are also divided into two types depending on the method of modifying the quantization step size. One is called a forward adaptive quantizer,
The quantization step size is calculated at fixed time intervals based on an input signal of a fixed time length. For details, see the American magazine "Bell Systems Technical Journal November 1975 issue" by P. Noll (BSTJ,
“A Comparative Study of Barely Quantization Schemes” published in November 1975, pp. 1597-1614)
Speech encoding (“A
Comparative Study of Various Quantization
This is described in document 3 entitled "Schemes for Speech Encoding."

The other type is called a backward adaptive quantizer, which determines the quantization step size at the next sample time based on the quantization step size at the current time and the adaptive quantizer output level. It is. For details, please refer to NS Giant (NS
Bell Systems Technical Journal September 1973 issue (BST
J., September 1973, pp. 1119-1144), “Adaptive Quantization with One Word Memory”
Quantization With a One-Word Memory.”)
This is described in the document 4 entitled, etc.

Of the ADPCM methods described above, the present invention particularly relates to an ADPCM method using a backward adaptive predictor and a backward adaptive quantizer, and therefore the present invention relates to the basic configuration of this method and the conventionally implemented ADPCM method. This aspect will be explained in more detail using the drawings.

In the following description, in order to make the explanation more detailed and easier to understand, the backward type adaptive predictor is divided into an adaptive predictor and a backward prediction coefficient value correction circuit, and the backward type adaptive quantizer is divided into an adaptive predictor and a backward prediction coefficient value correction circuit. The adaptive quantizer and the packed word quantization step size correction circuit will be explained separately.

Figure 1 shows a conventional method using a backward adaptive predictor and a backward adaptive quantizer.
This shows an example of the ADPCM method. In FIG. 1, on the transmitting side 11, a subtracter 12
Then, the predicted value x^ _j , which is the output of the adaptive predictor 14, is subtracted from the input signal xj from the input terminal 13 to obtain _the predicted residual _ej . The prediction residual e _j is quantized by the adaptive quantizer 15 to obtain a quantization output level l _j . The encoder 16 outputs its quantized output level l
_j is converted into a code q _j and transmitted to the transmission line 17. Further, the quantized output level l _j is dequantized by an adaptive dequantizer 18 to obtain a signal e^ _j . The signal e^ _j is added to the output predicted value x^ _j of the adaptive predictor 14 by an adder 19 to obtain a locally decoded signal _x〓j . The local decoded signal x〓 _j is input to the adaptive predictor 14 to obtain the predictor output value, that is, the predicted value x^
get _j .

Here, a backward type quantization step size correction circuit 21 is provided, which is described in, for example, the previous document 4.
The correction method shown in is applied. In that case, the quantization output level l _j of the adaptive quantizer 15 is used to modify the quantization step size according to the following equation.

△ _j+1 = △ _j・M(l _j )+C (1) In equation (1), △ _j corresponds to the quantization step size at time j, and M(l _j ) corresponds to the quantization output level l _j . C represents a constant that determines the lower limit of the quantization step size.

A backward prediction coefficient value modification circuit 22 is provided, and in this example, a learning identification method is used as a method for modifying the prediction coefficient values by this circuit 22. For details, see, for example, J. Nagumo (J.
IEEE Transactions Automation Control June 1967 issue (IEEE
Transactions Automatic Control, June1976,
“A Learning Method for System Identification” published in “PP.282-287”)
In other words, the prediction coefficient value is corrected according to the following equation.

In equation (2), a ^j _i is (i=
1, 2, ..., N)-th prediction coefficient value,
N represents the prediction order in the adaptive predictor 14. Also, g is selected to be a minute amount of about 0.1. However, in general, for the purpose of temporally attenuating the influence of transmission code errors caused by noise occurring in a line, the following equation is used instead of the equation (2), as shown in the above-mentioned document 2, for example.

Here, δ is a minute amount less than 1, and serves to attenuate the influence of transmission code errors over time with a time constant determined by 1/δ. Other symbols are the same as in formula (2) above.

On the receiving side 20, the code q received from the transmission path 17
_j is input to the decoder 23 to obtain a signal l _j . signal l _j
is input to the adaptive inverse quantizer 24 to obtain the signal e^ _j . The signal e^ _j is added to the output value x^ _j of the adaptive predictor 26 by the adder 25, and a reproduced signal _x〓j is obtained and supplied to the output terminal 27. Here, the backward prediction coefficient value modification circuit 28 and the backward quantization step size modification circuit 29 on the receiving side are the same as the backward prediction coefficient value modification circuit 22 and the backward quantization step size modification circuit 21 on the transmission side, respectively. The former modifies the prediction coefficient value according to the above equation (3), and the latter modifies the quantization step size according to the above equation (1), for example.

It has the conventional backward adaptive quantizer and backward adaptive predictor described above.
The ADPCM method (hereinafter referred to as the backward type ADPCM method) has a simpler device configuration than the ADPCM method (hereinafter referred to as the forward type ADPCM method) which has a forward type adaptive quantizer and a backward type adaptive predictor. There is an advantage that the amount of transmitted bits does not increase because there is no need to transmit the quantization step size and the predictor prediction coefficient values. However, as will be explained below, it has the disadvantage of being sensitive to transmission code errors.

In other words, in a backward adaptive quantizer, when a transmission code error occurs, the quantization step sizes on the transmitting and receiving sides become inconsistent, and the quantization step size becomes the predetermined upper limit of the step size or the above (1). The disadvantage is that the quantization step sizes on the transmitting and receiving sides do not match until the lower limit of the step size shown in equation (2) is reached, and it takes time to recover from the effects of transmission code errors.

In addition, in the backward type adaptive predictor, when a transmission code error occurs, the prediction coefficient values on the transmitting and receiving sides become inconsistent, and the influence of the transmission code error changes over time with a time constant determined by 1/δ. However, the disadvantage is that it takes a considerable amount of time until the predicted coefficient values completely match.

In recent years, systems have appeared that improve the above-mentioned drawbacks and have a certain degree of robustness against transmission code errors.
An example of a backward adaptive quantizer is the one published by DJ GOODMAN and others in the American magazine “IEE Transactions on Comics.
November 1975 issue” (IEEE Transactions on
Communications November1975 PP.1362−
1365) “A Robust Adaptive Quantizer”
In addition, an example of a backward type adaptive predictor is JP-A-53-123255 (Reference 7). A backward ADPCM method using a backward adaptive quantizer and a backward adaptive predictor will be explained with reference to the drawings.

FIG. 2 is an example of a block diagram of a backward ADPCM system that uses a backward adaptive quantizer according to Document 6 and a backward adaptive predictor according to Document 7 and is less susceptible to transmission code errors. In the figure, the backward adaptive quantizer is divided into the adaptive quantizer 15 and the backward quantization step size correction circuit 21, and the backward adaptive predictor is divided into the adaptive predictor 14 and the backward quantization step size correction circuit 21. It is described separately from the numerical value correction circuit 22. Also, in the figure, a subtracter 12 and an encoder 16
The decoder 23 and adder 25 perform the same operations as the components with the same numbers in FIG.

In Figure 2, the backward quantization step size correction circuit is on the transmitting side (component number 2).
1), the receiving side (component number 29) performs the same operation. As a method for correcting this quantization step size, the algorithm described in the above-mentioned document 6 is used. The algorithm is shown in the following equation.

△ _j+1 = (M(l _j )・△ _j ) (4) In equation (4), △ _j is the quantization step size at time j, and M (l _j ) is the quantization output level l _j
The coefficient β determined corresponding to is a positive constant smaller than 1. The fact that this algorithm is not easily affected by transmission code errors is described in detail in the above-mentioned document 6, so the explanation will be omitted here.

Regarding the adaptive predictor, compared to the adaptive predictor in FIG. 1 which configures a recursive filter on the receiving side, _the adaptive predictor in FIG. Since the output of the adaptive inverse quantizer 24 is supplied to the predictor 26 without being fed back to the predictor 26 to form a nonrecursive filter, it is less susceptible to transmission code errors. In addition, the backward prediction coefficient value correction circuit in FIG. 2 performs the same operation on both the transmitting side (component number 22) and the receiving side (component number 28), and belongs to the same type as the formula described in Document 7. Modify the prediction coefficient value according to the following equation.

In the above equation, a _i ^j represents the i-th predictor coefficient value at time j. N is the number of prediction taps of the adaptive predictor. δ and g are the same as in equation (3). According to equation (6), even if a transmission code error occurs at time j, the output e^ _j of the adaptive inverse quantizer on the transmitting and receiving sides is different, and the prediction coefficient values are different, the number of predicted taps is According to equation (6), if the same time has elapsed, the influence of the transmission code error on the second term on the right side of equation (6) becomes zero.

The adaptive predictor operates on both the transmitter side 124 and the receiver side 26 as shown in the following equation.

In the above formula, x^ _j is the adaptive predictor output value, and N
represents the predicted number of taps.

Therefore, compared to the ADPCM system shown in Figure 1, the backward ADPCM system with the configuration shown in Figure 2 has the advantage of greatly improved resilience against transmission code errors and extremely short recovery time. .

However, the backward type described in Figure 2
The use of the ADMPM method has the drawback of degrading the symbol error rate for data modem signals. This is the prediction coefficient value correction algorithm (formula (6) above) in the backward prediction coefficient value correction circuit.
This is thought to be due to. Generally speaking, short-term deterioration of the input signal waveform due to encoding and decoding hardly causes quality deterioration in voice communication, which is communication between humans. However, data communication, which is machine-to-machine communication, is extremely sensitive to the short-term deterioration and is easily susceptible to quality deterioration. Generally, the symbol error rate is used as a measure to evaluate the degree to which a data signal deteriorates when encoding and decoding is performed.If this value is very bad, the receiving side will It becomes impossible to demodulate.

The purpose of this invention is to use a backward type system that can transmit high-quality voice signals with the same equipment configuration as the conventional ADPCM system, and that can obtain a symbol error rate extremely close to that of data modem signals.
The purpose of the present invention is to provide an ADPCM method and a device used therefor.

According to the present invention, in the ADPCM method, a predictor gain in an adaptive predictor is calculated, the predictor gain is compared with a predetermined value, and the time when the predictor gain exceeds the predetermined value is determined. In this case, the speed at which the prediction coefficient values are modified in the adaptive predictor is reduced or stopped. The backward type ADPCM method according to the configuration of this invention is as follows:
Compared to the conventional ADPCM system having the configuration shown in FIG. 2, this method has the advantage that an extremely low symbol error rate can be obtained for data modem signals when there is no transmission code error. Furthermore, the backward type ADPCM method according to the configuration of the present invention has the advantage that even if a transmission code error occurs, the recovery time from the influence of the error is extremely short.

FIG. 3 shows an embodiment of the ADPCM method according to the present invention. Components in FIG. 3 having the same numbers as in FIG. 2 perform the same operations as in FIG. 2. The backward type quantization step size correction circuit performs the same operation on both the transmitter side 21 and the receiver side 29, and corrects the quantization step size according to equation (4) above. The backward prediction coefficient value correction circuit performs the same operation on both the transmitting side 22 and the receiving side 28, and the prediction coefficient value correction method is based on the above equation (6).

Next, in this embodiment, components related to the present invention added to the components shown in FIG. 2 will be explained. On the transmitting side 11, an adder 31 is used to add the input value e^ _j and the output value x^ _j of the adaptive predictor 14 to obtain an adder output _x〓j . The product-sum circuit 32 receives this addition output x〓 _j and outputs a squared leakage integral value X. The product-sum circuit 32 is configured as shown in FIG. 4, for example. That is, the input, here (x〓
_j-i ) ² is input and output through the adder circuit 33, and its output is passed through the 1-sample time delay circuit 34, further multiplied by the leakage amount related to the leakage integral in the coefficient circuit 35, and fed back to the adder circuit 33. Further, the operation of the product-sum circuit 32 is equivalent to the following equation.

In this equation, ρ ₁ is a leakage amount related to leakage integration, and is a non-negative constant that is sufficiently smaller than 1.

On the other hand, in FIG. 3, the output value e^ _j of the adaptive inverse quantizer 18 is input to the product-sum circuit 36, which outputs the squared leak integral value Y. The product-sum circuit 36 is constructed in the same manner as shown in FIG. The operation of this product-sum circuit 36 is equivalent to the following equation.

In the above formula, ρ ₂ is the leakage amount related to the leakage integral, which is 1
is a non-negative constant that is sufficiently smaller than .

The discrimination circuit 37 takes the ratio of input X and input Y,
Find the value corresponding to the predicted gain, and calculate the value of the ratio,
The discrimination circuit 37 compares a predetermined value equal to the average prediction gain that the adaptive predictor can take at every sample time, and at a time when the ratio value exceeds the predetermined value, the discrimination circuit 37 A determination signal Z is output to the adaptive prediction coefficient value correction circuit 22. At the sample time when the discrimination signal Z is input, the prediction coefficient value correction circuit 22 changes the value of g in the prediction coefficient value correction formula (6) to the value of g used one sample time earlier than the sample time. The prediction coefficient value is corrected by setting it to a predetermined value smaller than or to 0. On the other hand, at the sample time when the discrimination signal Z was not input,
The prediction coefficient value is corrected according to equation (6) above.

On the receiving side 20, the product-sum circuit 38 performs the same operation as the product-sum circuit 32 on the transmitting side, and the product-sum circuit 39 and the discrimination circuit 41 operate the same as the product-sum circuit 36 and the discrimination circuit 37 on the transmitting side, respectively. Do this.
Therefore, the discrimination circuit 41 receives the signal X and the signal Y and outputs the discrimination signal Z. The backward prediction coefficient value correction circuit 28 performs the same operation as the backward prediction coefficient value correction circuit 22 on the transmitting side 11.

According to the configuration of this embodiment, when a prediction coefficient value suitable for the input signal is found in the backward prediction coefficient value correction circuit, the prediction gain in the adaptive predictor is equal to or less than the predetermined average prediction gain. By determining this time using the determination circuit, it is possible to reduce or stop the correction speed in the prediction coefficient value correction circuit. Therefore, once a prediction coefficient value suitable for the input signal is determined, no further correction is performed.

In the above explanation, the method of delaying the correction of the prediction coefficient value was to reduce the value of g in the second term of equation (6), or to set it to 0, but the value of δ in the first term of equation (6) was A similar effect can be obtained by increasing the value.

By adopting the configuration of the embodiment shown in FIG. 3, an extremely low symbol error rate can be obtained for data modem signals in the absence of transmission errors, and high quality transmission can be achieved. Furthermore, since the backward adaptive quantizer and the backward adaptive predictor use a system that is less susceptible to the effects of transmission code errors, there is an effect that the recovery time from the effects of transmission code errors is short.

In the embodiment described above, the above (8) and (9)
In the equation, X and Y are the leakage integral values of x〓 _j ² and e^ _j ^2, respectively, but they may also be the leakage integral values of |x〓 _j | and |e^ _j |. By doing so, there is no need to calculate the squares of x〓 _j and e^ _j , so there is an advantage that the amount of calculation can be significantly reduced.

In this embodiment, the adaptive predictor shown in FIG. 2 is used as the backward adaptive predictor, but the adaptive predictor shown in FIG. 1 may also be used. Further, the functions of each part shown in FIG. 3 may be performed by program processing using a microcomputer.

[Brief explanation of the drawing]

Figure 1 is a block diagram showing an example of a conventional backward ADPCM system having a backward adaptive quantizer and a backward adaptive predictor.
The figure is a block diagram showing an example of a backward type ADPCM system that is less susceptible to transmission code errors and has a backward type adaptive quantizer and a backward type adaptive predictor that are less susceptible to transmission code errors. A block diagram showing an embodiment of the ADPCM method according to the present invention. FIG. 4 is a block diagram showing an example of a leak integral value circuit in the first embodiment. 12: Subtractor, 13: Input terminal, 15: Adaptive quantizer, 17: Transmission line, 18, 24: Adaptive inverse quantizer, 25: Adder, 14, 26: Adaptive predictor, 21, 29: Backward type quantization step size correction circuit, 22, 28: Backward type prediction coefficient value correction circuit, 16: Encoder, 23: Decoder, 31: Addition circuit, 32, 36, 38, 39:
Product-sum circuit, 37, 41: Discrimination circuit.

Claims (1)

[Claims] 1. On the transmitting side, adaptive prediction means adaptively predicts an input signal, and adaptive quantization means adaptively quantizes the difference between the input signal and the output value of the adaptive prediction means. quantizes, encodes, and transmits the input signal using an adaptive inverse quantizer that receives the code on the receiving side and has characteristics opposite to that of the adaptive quantizer; In an adaptive differential PCM method for reproducing an original signal using adaptive prediction means having the same characteristics as the prediction means, means for calculating a gain of the prediction means in the adaptive prediction means; Comparison with a predetermined value, and at a time when the gain of the prediction means exceeds the predetermined value, the correction speed of the prediction coefficient value in the adaptive prediction means is reduced or stopped. adaptive differential
PCM method. 2. On the transmitting side, an adaptive prediction means for adaptively predicting an input signal, an adaptive quantization means for quantizing the difference between the input signal and an output value of the adaptive prediction means, and the adaptive quantization means. an adaptive differential PCM transmitting device, comprising: an encoder that converts an output value of the code into a code sequence, and transmits the code sequence at a predetermined sampling time, and calculates a prediction means gain in the adaptive prediction means. a gain calculating means for inputting the calculated value of the gain calculating means and comparing the calculated value with a predetermined value;
inputting the output value of the comparison means to the adaptive prediction means, and reducing the correction speed of the prediction coefficient value in the adaptive prediction means at a sample time when the calculated value in the gain calculation means exceeds the predetermined value; an adaptive differential PCM transmitting device comprising means for causing or stopping the code sequence, on the receiving side, the code sequence is received, the code sequence is decoded using a decoder, and the output value of the code sequence is determined from the output value of the decoder. Adaptive difference that reproduces the original signal using adaptive inverse quantization means having characteristics opposite to those of the adaptive quantization means on the transmission side, and adaptive prediction means having the same characteristics as the adaptive prediction means on the transmission side.
A PCM receiving device, wherein a gain calculation means for calculating a gain of a prediction means in the adaptive prediction means on the reception side and a calculated value of the gain calculation means are input,
a comparison means for comparing the calculated value with a predetermined value; an output value of the comparison means is inputted to the adaptive prediction means on the reception side; and means for reducing or stopping the correction speed of the prediction coefficient value in the adaptive prediction means on the receiving side at a sample time exceeding the value.
An adaptive differential PCM transmitter/receiver including a PCM receiver.