JPS5999830A - Transmission system of digital sound signal - Google Patents

Abstract

PURPOSE:To reduce noise due to a bit error by detecting the maximum and minimum value of the amplitude of a sound signal included in a section of a constant length and transmitting the signal while specifying the range of level of the sound signal. CONSTITUTION:A signal input is given to an input terminal of a multiplex circuit 13 as a data signal via N-sample delay circuits 11, 12. Further, a data signal input is supplied also to a maximum level latch circuit 14 to hold the maximum level of the sound signal during N-sample period. The range of the level is discriminated 15, the result is given to an error correction coding circuit 16 for applying the coding of error correction of the range bit and the data signal is multiplexed 13 and transmitted. On the other hand, the range bit is extracted 21 from the data signal at the receiving side and error correction/ detection 22 is applied, then the range output is supplied to a range limit circuit 23 so as to limit the signal output level to the signal input within the range. Then, noise due to bit error is reduced.

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention relates to a digital audio transmission system that reduces the amount of noise caused by bit errors when reproducing digital audio signals via a communication channel and transmission medium that have bit errors. It is something.

Generally, when converting analog quantities into digital quantities, there are two methods: LL H encoding, in which the quantization steps are uniform at all levels, and companding encoding, in which the quantization steps are not uniform depending on the signal level (non-linear encoding). , f1 formula). Furthermore, there are two types of companding coding methods: a temporal companding coding method that sends information regarding companding at every sampling, and a quasi-instantaneous companding method that sends information about upcompression at regular time intervals. Encoding method Figure 1 to Figure 1? Figure L shows the relationship between the input/output level and the quantization level for each of these encoding methods.

In the linear encoding method, as seen in the example of the /kbittobar line phrase υ· in FIG. 1, there is a linear relationship between the input/output level and the kl child level at all levels.

In this method, the 7H (subset step is uniform, so unlike the companding coding method, the amount f step changes due to large fluctuations in the signal level, and the quantization noise causes a breathing phenomenon that is noticeable in the auditory sense.) You'll get pure, high-quality audio without any hassle.

The instantaneous companding encoding method is shown in Figure 2. When the level is large, the quantization step is increased and encoding is performed. - In the second method, the quantity conversion step is divided into several stages depending on the input/output level, so the relationship between the input/output level and h) prediction level is a broken line. When the level of the signal is large, the number of transmission bits can be reduced compared to the linear coding method by utilizing the property that quantization noise is not noticeable to the auditory sense even if the Rk conversion step is large.

No.! In the example of / Novisoto's instant companding method shown in the figure, J
Too many bits/! Sound quality close to that of the 7-bit linear encoding system can be obtained.

The quasi-instantaneous companding encoding method is not shown in Fig. 3. As shown in the example of /ku + 10-bit quasi-instantaneous companding, the relationship between the input/output level and the quantization level is divided into S types with different quantization steps. It is represented by a divided scale 匈■～■.

To determine which scale to use among these five scales, select the slur that represents the maximum input/output level within a certain period of time.

An example of a quasi-instantaneous companding encoder implementing the quasi-instantaneous companding encoding method is not shown in FIG. In this example, the Zanbruge frequency is set to 3 KEy, the binding, the distractor/? At the same time, the slur selection circuit determines the scale to be used within the 3.2 sample period and extends the data signal input by 3.2 samples (/ms).
range bit that represents? 1lJ-sample period Nearest number: I output and C transmit. /4' in the compression circuit 3 in the range selected according to this range bit output.
-) 70 bit compression is performed and its IT:.

The line output is delayed by 3.2 samples using a delay device 1 and transmitted as a garbled output. In this way, the range bit, which is the companding information, needs to be sent for every sample A, and only needs to be transmitted seven times in a fixed period, so that the transmission ffl of the companding information can be extremely reduced. Therefore, in the example of quasi-instantaneous companding of /<7→/θ bits in Figure 3, the input/output level corresponding to /dabit in the linear encoding system is expressed as a 10-bit quantization level, and the companding information is excluded. For example, the number of transmission bits can be reduced by l bits. Here, in order to represent the three scales, three bits are required as range bits, so
[The exact number of transmitted bits including E-expansion information is approximately /θ per l sample. It is l bit. As described above, the quasi-instantaneous companding coding method is an excellent method when there is a limit on transmission capacity.

these,? Comparing the noise caused by bit errors among different types of encoding systems, in the linear and transverse companding encoding systems, when a bit error occurs, errors occur over the entire range of human output levels. For example, MSB
If a bit close to (upper bit of the shell) is incorrect, it will be heard as loud noise.

On the other hand, in the instant companding method, if a bit error occurs, the error is limited to the range in use unless the range bit is incorrect.

Therefore, if the input/output level is small,
The noise caused by the noise is also limited to a relatively small amount.

In other words, the quasi-instantaneous companding coding system has an excellent feature of reducing noise caused by bit errors.

Therefore, an object of the present invention is to appropriately apply part of the principles of the semi-temporary companding coding method to reduce noise caused by bit errors even in the linear coding method and the instantaneous companding coding method. The purpose of the present invention is to provide a digital sound seat culm feeding system.

In order to achieve such an object, the present invention provides that when a voice signal is converted into a digital signal corresponding to its amplitude and transmitted, the 'if voice call signal is divided into predetermined periods and Detecting the highest value, or the highest value and lowest value, of the amplitude of the included audio signal, specifying the level range of the audio signal according to the detection output, and providing an index signal indicating the specified level range, The present invention is characterized in that the digital signal converted in the level range is transmitted at a rate of 1.7 for each period.

The principle of reducing errors caused by bit errors in the quasi-companding/right-encoding method is to reduce errors by transmitting range information at regular intervals and limiting the range in which bit protection occurs to a specific scale. That is true.

In this invention! By applying the = principle to linear and instantaneous encoding systems, we obtain the effect of reducing noise due to bit errors.

The present invention will be described in detail below with reference to the following sections.

The setting of the range, which is the principle of the present invention, will be explained with reference to FIG. Here, the term "range" refers to a specific level range in which the input/output level is divided into several parts C4. Even if a bit error occurs, if it is outside the level range indicated by the range, it will be completely suppressed, and the resulting error will be limited to within that scale.

FIG. 6 shows an example of a circuit configuration for generating all of the bit valleys. Here, the signal input is supplied as a data signal to one input terminal of the multiplex circuit /3 via the N sample delay circuit and 7. The maximum level of the audio signal during the N sampling periods is determined by the range determination circuit/!f, and the range bit representing the determination result is corrected. The range bits are supplied to the encoding circuit/B, where the range bits are encoded for error correction, and then supplied to the other input terminal of the multiplexer circuit/3 to be multiplexed into the data signal.
．． to transmit.

On the receiving side, as shown in Part 7.1, the range bit extraction circuit 2/ extracts the range bit from the data signal input, and then the error correction/L~output circuit I extracts the range bit from the range bit. After correction and detection of 1-rj, the obtained 1: range high power is supplied to the range limiting circuit 3, which limits the signal output level for the distorted input to that range.Here, the bit Noise caused by errors is also suppressed within this range, thereby reducing the noise.

A first embodiment of the present invention is shown in FIG. g. This example applies the present invention to the //-piL linear encoding method.7. The range is the entire input/output level range (full scale).
(n+/) range of Ωn (in this example, n-θ, /, 2.
．． ? , 5 ranges of lI).

An example of a configuration that generates such a range bit is
Shown in IJ. Here the sampling frequency is 41gK
It is a platform for Uz, latches the maximum level in the tag sample period (/m11), and selects a range for the period outside the latched level range. In this example, two channels such as stereo are configured, and the L side and R'
How many samples in each data signal input in each channel delayer J/L and 3.2L and 3/R and 3.2
It is supplied to the multiplex circuit 33 via R, and also to the maximum level latch circuits 37/L and 3'7-R, respectively.
% pay. Maximum level latch circuit 39 L and 3'+
R (well, each JLStI & the maximum level in the sample is held) and their latch outputs are sent to the range judgment circuit 3.!
iL and? , tR, and IJ constant covers the range in which the maximum level falls. The range bit of one channel determined in this way is supplied to the Hamming circuit 3A and encoded using a Hamming (7+3) code, thereby providing strong protection against errors. The output is supplied to a multiplexer (1) path 33, where the data signals of both channels are multiplexed with Hamming encoded range bits P, taken out, and transmitted.

The range in this first embodiment is set to ±4" (n: 0./, 2.+?-) of the entire input/output level range, and the smaller the input/output level, the more concentrated the range is. However, the range becomes narrower.This method is effective for transmitting ordinary audio signals for the following three reasons.

(1) The probability distribution of instantaneous values of audio signals is generally represented by a negative exponential distribution. That is, the probability of occurrence decreases as the signal level increases.

(2) When recording audio, usually θ~JOdB
Since the peak margin is set, many of the signal levels are concentrated at low levels.

(3) When the signal level of noise is low, it is not masked by the signal and is easily detected audibly.

For the above reasons, it is more effective in terms of noise reduction to divide the range narrowly in parts where the signal level is low.

Next, an example of the configuration of a noise suppression circuit on the receiving side will be shown. First, the data input signal is subjected to correction and detection by the first error correction detection circuit q/, and then the range bit is extracted by the range bit extraction circuit q/. For the extracted range bits, a range bit error correction/detection circuit Q performs error correction/detection of the range bits themselves. The correct range output from circuit DA 3 is then supplied to range limiting circuit 411 to limit the signal level of the correct data signal from circuit 1if/ to within that range. As a result, noise caused by bit manipulation is suppressed, resulting in a noise reduction effect.

Another advantage of this method is that it is possible to determine the upper 7 to g bits from the range bit information, and the output from the range limiting circuit is transferred to the error correction/detection circuit 9.8'.
By correcting and detecting the remaining lower bits again to obtain a data signal output, the error correction effect can be further improved.

A second embodiment of the invention is shown in Figure 1/. In this example, the range dividing point is the equal dividing point n/S (n:/
, 2, 3. Da..., S: number of divisions, i.e. number of ranges)
It is set to . Such a range division method is particularly effective when the probability distribution of instantaneous values of the signal level is not concentrated in a small portion of the signal level. For example, the audio signal may already be compressed in the analog domain and the small signal level may be amplified. Particularly, noise reduction effect devices for analog signals perform such compression.

Furthermore, the average signal level also increases when a limiter or the like is used during audio recording. In this way, it may be effective to distribute range settings evenly over the entire input/output level range.

An example of the configuration of a circuit for suppressing noise on the receiving side using this method is shown in FIG. 7.2. Here, the data coefficient input is supplied to the range bit extraction circuit S/ to obtain the range bit? The range bit output is supplied to the range bit error correction/detection circuit tacho, which detects the range bit error vIT correct/
Perform detection. On the other hand, the data signal input is also supplied to the upper 3 bit extraction circuit S3 to extract the upper 3 bits of the data signal. These range bits and the upper three bits of the signal are supplied to a comparison circuit 3tl to compare them. If the signal level exceeds the level range indicated by the range bit, the interpolated signal obtained by passing through the data signal interpolation circuit S is selected by the switch 56 and output as a data signal. Conversely, if the level of the data signal is within the level range indicated by the range bit, the switch 56 selects the data signal input as it is and outputs the data signal.

In this way, noise due to bit errors is limited to the level range indicated by the range bit, and the noise can be reduced.

The third embodiment of the present invention is shown in FIG. Namely S
n/SOm/S(n:/p-2t J ・
"pm: 0./, 2..., s: number of divisions) are distinguished as ranges.

In the case of the first and second embodiments shown above, when the signal level is large, a wide level range is used, and the noise becomes relatively large.

However, in this third embodiment, even if the signal level is large, a narrow level range can be used as long as the signal level fluctuation is small, and errors caused by bit errors are limited to a narrow level range. 2 sounds can be roughly reduced. An audio signal is usually formed by adding a harmonic component having a small amplitude Φ1.1 at a high frequency to a fundamental wave having a large amplitude at a low frequency. Therefore, by dividing the range into many sections to create a multi-range as in this example, it is possible to further increase the noise reduction effect of the sound signal.

An example of the structure of the shield 7 of the noise suppression circuit on the receiving side in this system is shown in FIG. Here, the data signal input is supplied to the range bit extraction circuit B1 to extract the range bit.

The extracted range bit is supplied to the range bit correction/detection circuit 62 to correct and detect the range bit. On the other hand, the data signal input is supplied to the 1st and 3rd bit extraction circuit 63. The 3 bits in the data signal are output. The upper level and lower level indicated by the range bit mentioned above and the upper 3 bits of the data signal are connected to the upper and lower level comparison circuits &4t and 6, respectively.
j- is supplied to compare the upper 3 bits of the -F level relevel signal among the range bits, and the lower level of the range bit and the upper 3 bits of the signal, respectively. Or game these comparison circuit reviews and the comparison output of 6.3') 4
If the level of the data signal exceeds the upper and lower level range indicated by the range bit, the interpolated signal obtained by passing the data signal through the interpolation circuit 6g is output. Select by switch A7. Conversely, if the level of the data signal is within the upper and lower level ranges, the signal is selected by the switch 67 as is. In this way, noise due to bit errors is limited to the upper and lower level ranges indicated by the range bits, and the noise is effectively reduced. In this example, as the number of ranges increases,
Although the number of range bits also increases, the increase is at most about 1-copy, and is only a very small increase when considering the total number of transmission bits.

As shown in the above embodiments, if \range bits are also transmitted, noise due to bit errors can be reduced.

The effect varies depending on the nature of the signal and the range bit transmission interval, but assuming that there are almost no signal peak values that change the upper 3 bits of the quantization bits, the effect will be lower than when the range bit is not sent. ,
Calculationally, a noise improvement effect of /l dBB10 can be obtained.

As can be seen from the above, by implementing the present invention, when transmitting an audio signal through a communication path and transmission medium with bit errors, the range bit indicating the level range in which the signal exists during a certain period can be In short, by applying li) coding with a large correction effect to this range bit and transmitting it, and limiting the noise caused by bit errors to within that range, it is possible to reduce the noise. .

In the first embodiment, the range is the entire input/output level range.
, 2" (n: 0./, 2.J=), so the smaller the signal level, the narrower the range. When the signal level is small, the noise due to bit y4 suddenly decreases. This method uses the property that audio signals are concentrated at a small level, and has a large noise reduction effect on normal audio signals.

In the second embodiment, the range is set to n/of the total input/output level range.
S (n: l, 2, 3 PII..., S: number of divisions),
Since the range is equally distributed over the entire input and output level range, it is suitable for noise reduction when the signal level is not concentrated at a low level, such as when the audio signal has already been compressed.

In the third embodiment, the range is further divided into sections, and even if the signal level is high, if the fluctuation of the signal during a certain period is small, the noise can be limited within the narrow range. Compared to the first and second embodiments, this example has a noise reduction effect in a portion where the signal level is high.

As explained above, the present invention has the effect of reducing noise caused by focus errors, and this fact can be utilized in the fields of broadcasting and communications that use transmission paths with bit errors, or in the field of bit errors. It can be effectively applied to fields such as package manufacturers for code and data that use error-prone transmission media.

[Brief explanation of the drawing]

Figure 1 is a characteristic diagram showing all the relationships between input/output levels and predicted levels using the linear coding method, n,! The figure is a characteristic diagram showing the relationship between the input/output level and the h1 childization level using the 7.7-bit instantaneous companding method. FIG. 9 is a block diagram showing an example of the circuit configuration of a conventional quasi-instantaneous compander T4 unit, and FIG. 5 is a characteristic diagram showing the relationship between levels. FIG. Figure 6 is a block diagram showing an example of the circuit configuration of the transmission system of the present invention, and Figure 7 is a block diagram showing an example of a circuit diagram for noise suppression on the receiving side using the transmission system of the present invention. Figure 9 is a characteristic diagram showing a specific example of the input/output level versus quantization level according to the transmission method of the present invention, and Figure 9 is a block diagram showing an example of the circuit configuration for generating the range bit according to Figure 9. , 1st
Figure O is a block diagram showing an example of the configuration of the ecstatic noise suppression circuit on the receiving side of transmission of range bits according to Figure g, and Figure 1/Figure 1 shows the relationship between input and output levels versus idleness levels according to the transmission method of the present invention. Characteristic diagram showing a specific example, Figure 7.2 is a block diagram showing a configuration example of a noise suppression circuit on the receiving side of range bit transmission according to Figure 1/Figure 1, Figures 1 and 7 are based on the method of the present invention 1 Characteristic diagram showing a specific example of input/output level versus quantization level,
FIG. 141 is a 4-block diagram showing a configuration example of a noise suppression circuit on the receiving side of range bit transmission according to FIG. 73. /..., 7-ripple delay device, -...scale selection circuit 1. 3...Bit compression circuit, R...3) N-sample delay device, //, /,)...N sample delay device, /3...Multiple circuit, /II...Maximum level latch circuit , /S...range determination circuit, /B...error correction/encoding circuit 1. 2/... Range bit extraction circuit, 〃... Error correction/detection circuit, J3... Range limiting circuit, J/I, , ? /R, J2L, ? , 2R... How many jumble delay devices, 33... Multiplex circuit, J', IL, 3R... Maximum level latch circuit, 3!
rL,,BR...Range judgment circuit, 3B...Hamming circuit, l/...1st error correction/detection circuit, Q...Range bit extraction circuit, 13...Range bit error NJ' jE/detection circuit,
Dac...Range limiting circuit, S...Error correction/detection circuit 1. 1/-...Range bit extraction circuit, Error correction/detection circuit, 3...Higher 3 bit extraction circuit 1, Tag...Comparison circuit, 3゛, (...Interpolation circuit, S Otsu...Switch. Enter/Urikashiku Nre 2nd person/King level 51210242048 4096
Figure 4 Figure 6 Figure 175 Figure 7 Figure 9 Figure 8 Figure 10 Figure 12 Figure 11

Claims (1)

[Claims] /) When converting an audio signal into a digital signal corresponding to its amplitude and transmitting it, the audio signal is divided into predetermined periods in advance, and the maximum value of the amplitude of the audio signal included in that period is determined. , or detect the maximum value and the minimum value, specify the level range of the voice signal according to the detection output, and convert the converted digital signal in the level range into an index signal indicating the specified level range. A digital audio signal transmission method characterized in that a signal is appended to the signal for each period and transmitted. , 2. In the digital audio signal transmission system according to claim 1, the level range is equal to the total input/output level range of the amplitude of the audio signal.
3...), and transmits range bits indicating each level range together with the digital signal. 3) In the digital audio signal transmission system according to claim 1, the level range is equal to the total input/output level of the amplitude of the audio signal '5:S points n/S.
(n: / , , 2, , ? , 11...
4t) In the digital audio signal transmission system according to claim 1, a range bit indicating each level range determined by the digital signal is transmitted together with the digital signal. , the level range is a point n/S obtained by equally dividing the total input/output level of the amplitude of the audio signal into eight parts.
(n : θv 'p 2p "") and rQ//S(m
=/,j,,? , da...) is determined by all combinations that connect the openings with range picks that indicate the range of each level.
l- A digital audio signal transmission method characterized in that the digital audio signal is transmitted together with the digital signal.