JPS61158227A - Method for pcm transmission and its device - Google Patents

Abstract

PURPOSE:To attain PCM transmission with a high S/N and a wide transmission frequency band at a low transmission bit rate, by discriminating a substantial highest frequency of the primary component contained in an analog original signal and producing the transmission data with said discriminating information to transmit this data. CONSTITUTION:The analog signal is converted into the digital data of 14 bits by an A/D converter 2 and applied to the 1st and 2nd data converters 3 and 4 as well as to a frequency discriminator 5. Both converters 3 and 4 use the comprementary bit numbers 7 and 14 and sample frequencies fS and fS/2 to perform the conversion of data so as to obtain the same total number of bits after conversion for each data block between both converters 3 and 4. A frequency discriminator 5 discriminates a high or low level according to whether the substantial highest frequency of the primary component contained in an analog original signal is higher than fS/4 or not. If a high level is decided, a selector 6 is switched to transmit the output of the converter 3 to a transmission line through a terminal 7 as the transmission data together with the discriminating information.

Description

[Detailed Description of the Invention] [Technical Field of the Invention] The present invention relates to pCM (pulse code mod
(pulse code modulation) transmission method,
In particular, the present invention relates to a PCM transmission method and apparatus for realizing high S/N (signal-to-noise ratio) PCM transmission with high efficiency and wideband. In this case, "transmission" means transmission in a broad sense, not only through wireless or wired communication lines, but also simply through modulation/demodulation systems, recording/playback systems, etc. '' means a system including a communication/signal line, a modulation/demodulation system, a recording/reproduction system, etc., or a combination thereof.

[Technical background of the invention]

The theoretical S/N in PCM transmission is generally uniquely determined by the number of quantization bits, that is, the number n of bits of sample data that is substantially transmitted by PCM transmission, and is expressed by the following equation.

S/N=nx6+2 [dB] Therefore, in order to obtain a high S/N in PCM transmission, the number n of quantization bits must be increased.

Of course, even if the number of quantization bits is increased, if the sampling frequency fs is lowered, the limited bit rate (
= fSXn), but in this case, the reproduction frequency is limited by fs/2, which creates a new problem of narrowing the frequency range of the transmitted information, that is, the reproduction frequency range. .

Therefore, for example, when transmitting audio signals under limited transmission conditions, the bit rate can be lowered by increasing the efficiency of the transmission method without narrowing the transmission frequency range to increase the S/N. It is necessary to plan.

[Purpose of the invention]

The purpose of the present invention is to improve the efficiency of transmission,
It is an object of the present invention to provide a PCM transmission system and its apparatus that can realize transmission with excellent S/N and transmission frequency band with a low transmission bit rate.

[Summary of the invention]

In the present invention, the frequency components included in the analog original signal are substantially monitored to determine the substantial highest frequency in the analog original signal, and when the highest frequency is low, the actual sampling frequency is lowered and transmitted. The effective number of quantization bits of data is increased, and when the maximum frequency is high, the effective sampling frequency is increased and the effective number of quantization bits of transmission data is decreased.

Therefore, at low frequencies, a high number of quantization bits is transmitted at the cost of lowering the actual sampling frequency, resulting in high S/N and high precision; at high frequencies, instead of lowering the number of quantization bits, Since transmission is performed at a substantially high sampling frequency of 10-wavenumbers, a high bandwidth and high efficiency can be obtained.

At this time, at low frequencies, the effective sampling frequency is lowered, and at high frequencies, the number of quantization bits is lowered, but at low frequencies, the importance of the number of quantization bits becomes relatively high, and at higher In,
Since the importance of sampling frequency is relatively high,
It is easy to keep the substantial reduction in the sampling frequency at low frequencies and the reduction in the number of quantization bits at high frequencies to a range that does not pose a practical problem.

Note that the above-mentioned changes in the actual sampling frequency and number of quantization bits are performed for each data block consisting of multiple samples, and discrimination information, such as a discrimination code, corresponding to the sampling frequency of the corresponding data block is added before each block data. Send to. On the receiving side, based on the discrimination information, the sampling frequency in D/A conversion during demodulation is changed in accordance with the actual sampling frequency during transmission, and an analog demodulated signal (reproduced signal) is obtained. At the same time, the cutoff frequency of the low-pass filter provided on the receiving side to remove harmonic components generated in connection with sampling is changed in accordance with the actual sampling frequency, resulting in good analog reproduction over a wide frequency band. Get a signal.

That is, the first invention according to the present invention is an invention of a method, which substantially monitors frequency components included in an analog original signal to determine the substantial highest frequency of the main components included in the analog original signal. , In accordance with this discrimination information, the sampling frequency and the number of quantization bits of the digital transmission data based on the analog original signal are controlled in a complementary manner for each data block consisting of a plurality of samples, and as a result of this control, the transmission The data and the discrimination information are transmitted, the data is received, and the sampling frequency in D/A conversion for decoding and demodulation is controlled according to the discrimination information, and the data is transmitted at the sampling frequency corresponding to the transmission data. This is an F'CM transmission system characterized by decoding and demodulating data.

The second invention according to the present invention is an invention of a product directly used for carrying out the method of the first invention, which substantially monitors the frequency components of the analog original signal and detects the main components contained in the analog original signal. Discrimination means for determining the practical highest frequency;
A/D converting the analog original signal, and variably controlling the sampling frequency and the number of quantization bits in the A/D conversion for each data block consisting of a plurality of samples in response to the discrimination information of the discrimination means; Variable A/D conversion means for obtaining digital transmission data obtained by A/D converting the analog original signal, and sending a transmission signal containing the transmission data obtained by the variable A/D conversion means and the discrimination information to the transmission system. This is a transmitting device characterized by comprising a transmitting means for transmitting.

The third invention according to the present invention is an invention of a product directly used for carrying out the method of the first invention, in which the analog While the actual sampling frequency and the number of quantization bits for the original signal are variably controlled for each data block consisting of a plurality of samples, the analog original signal is A/D converted to determine whether the analog original signal corresponds to digital transmission data and the sampling frequency. In the receiving device for receiving a transmission signal containing information and demodulating the transmission signal to obtain a reproduced signal, a demodulation means for D/A converting the transmission data in the transmission signal to obtain a demodulated signal; , further comprising control means for controlling the demodulation means according to the discrimination information in the transmission signal so that the sampling frequency in the demodulation means corresponds to the substantial sampling frequency of the transmission data. receiving device.

[Embodiments of the invention]

FIG. 1 and FIG. 2 respectively show the configurations of a transmitting device and a receiving device in an embodiment of the present invention.

In FIG. 1, an analog signal applied to an input terminal 1 is sampled by an A/D converter 2 at a sampling frequency of "S" (for example, about 40kHz),
It is converted into 4-bit digital data. This A/D
The original sample data output from the converter 2, that is, the original P'CM data, is provided to each of a first data converter 3, a second data converter 4, and a frequency discriminator 5.

The frequency discriminator 5 receives the original sample data, analyzes and monitors the frequency components included in the analog original signal, and analyzes the actual frequency components included in the analog original signal for each data block consisting of a plurality of original sample data. Determine whether the highest frequency is high or low.

In this case, the frequency is determined by determining which of two levels, high and low, it corresponds to, and whether it is higher or lower than a preset value. The frequency discriminator 5 can be configured to use, for example, a digital filter to determine the magnitude of a specific frequency component and determine whether the highest frequency of the main component is high or low.

That is, in this case, the frequency is fs/4 or more and fs
/2 or less (as this is the sampling frequency fs of the original sampling data, components of fs/2 or more are meaningless ~ For example, signal components of fS72 or more are removed by a low-pass filter before being input to the A/D converter 2. ) is determined to be included (by comparison with the overall signal size or components in other frequency ranges), the
If it is determined that it is not included, it is determined to be "low". Therefore, this frequency discriminator 5 analyzes and monitors the frequency components included in the analog original signal corresponding to each data block, and determines that the substantial highest frequency of the main components included in the analog original signal is fs/4 Information corresponding to "high" or "low" is output depending on whether it is above or below. The first data converter 3 has a bit length of 1
Upon receiving the above-mentioned original sample data of 4 bits and sampling frequency fs, the bit length of each data is shortened while keeping the sampling frequency fs unchanged, and is converted into, for example, 7 bit data.

Further, the second data converter 4 receives the original sample data and converts the bit length of each data, which is 14 bits, as it is.
The number of samples is reduced to 1/2 and the actual sampling frequency, that is, the sampling frequency is set to fs/2, for example. Here, the first and second data converters 3.4 each unit time,
In this case, data conversion is performed by controlling the bit length and sampling frequency in a complementary manner so that the total number of converted bits for each data block is equal. The outputs of the first and second data converters 3.4 are applied to a selector 6 which selects one of the 100 outputs. The selector 6 is comprised of a changeover switch circuit that selects the input from both data converters 3.4 according to the output of the frequency discriminator 5, and if the frequency discriminator 5 determines "high", that is, fs/4 or more. , the output of the first data converter 3 is selected, and when the determination by the frequency discriminator 5 is "low", that is, less than fs/4, the output of the second data converter 4 is selected. . The data selected by the selector 6 becomes the transmission data, and during transmission, this transmission data and the discrimination data of the frequency discriminator 5 are sent out from the transmission output terminal 7 to the transmission line.

In this way, the transmission data sent to the transmission path is data in which data with a bit length of 7 bits at the sampling frequency fs and data with a bit length of 14 bits at the sampling frequency fs/2 are combined in data blocks. , the discrimination data is data indicating "high" and "low". Therefore, the discrimination data only needs to have at least one pit, and since it is easy to transmit, it is also easy to time-division combine the discrimination data and the transmission data and send the transmission signal over a single transmission path. However, it is also easy to provide the discrimination data and the transmission data to different transmission lines, and to send the transmission signals through two parallel transmission lines.

If each of the above-mentioned data is schematically shown in FIGS. 3(a) to (C), the original sampling data output from the A/D converter 2 has a sampling frequency as shown in FIG. 3(a). fs
The bit length is 14 bits.

Then, the discrimination data and the transmission data are time-division synthesized,
When the transmission signal is sent through a single transmission path, the transmission signal is as shown in FIG. 3(b) when the first data converter 3 is selected.
As shown in FIG. 3, for example, 4-bit discrimination data is inserted in each data block in addition to the 7-bit transmission data at the sampling frequency fs, and when the second data converter 4 is selected, the second data converter 4 is selected. As shown in FIG. 3C, for example, 4-bit discrimination data is inserted in each data block in addition to the 14-bit transmission data at a sampling frequency of fs/2. In any of these cases, it is clear that the total amount of data transmitted is less than the original sampled data shown in Figure 3(a), but even with such transmission important information of the original signal is lost. Most can be transmitted. If the number of bits of the output data of the first data converter 3 is set to 8 bits or more, or if the sampling frequency of the output data of the second data converter 4 is set to a value exceeding fs/2, the amount of information to be transmitted can be reduced. will further improve. That is, if an amount of data equivalent to transmitting the original sampled data as is is transmitted, information comparable to data sampled with a substantially larger bit length or a substantially higher sampling frequency can be transmitted.

In FIG. 2, a reception input terminal 11 is given a transmission signal from a transmission system, that is, a transmission path. The transmission signal inputted to the reception input terminal 11 is applied to a D/A converter 12, which has an accuracy of 14 bits in this case, and a detection controller 13 that detects discrimination data in the transmission signal. D/
The output of the A converter 12 is applied to an analog low-pass filter 14 with a variable cutoff frequency for removing unnecessary high frequency components such as harmonic components caused by sampling and quantization. The detection controller 13 detects discrimination data from the received signal, and controls the sampling frequency (input data latch frequency) of the D/A converter 12 and the cutoff frequency of the low-pass filter 14 according to the discrimination data. . That is, in this case, when the discrimination data "high" is detected, the D/A converter 12 is operated at the sampling frequency fs, and the sampling frequency fs/
The low-pass filter 14 is operated at a cutoff frequency of 2. Also, when discrimination data "low" is detected,
The D/A converter 12 has a sampling frequency of fs/2.
is operated, and the low-pass filter 14 is operated at a cutoff frequency of fs/4. Here, the cutoff frequency of the low-pass filter 14 is controlled because the high-frequency components to be removed by the low-pass filter 14 change depending on the actual sampling frequency.

In transmission using such a transmission method, the S/
As in the N vs. frequency characteristic, high frequency components of fs/4 or higher, where the actual sampling frequency is important, are transmitted with 7 bits, so the S/ interval is slightly worse, but transmission at the original sample frequency fs The low frequency components below fs/4, where S/ and accuracy are very important, are transmitted in 14 bits.At this time, the actual sampling frequency is fs/2, but the frequency of the original signal is low. So there is no problem.

That is, in an actual audio signal such as a music signal or a speech, the period during which the main component is a component in a high frequency range such as fs/4 to fs/2 is very short. For other analog signals, if the original sampling frequency fs is set appropriately, generally fs/4 to fs/
2 frequency component appears only for a very short time. Therefore, for most of the time during which transmission is performed, transmission is performed with high S/N with 14-bit accuracy, for example.

Note that if the original signal contains noise in the frequency range of fs/4 to fs/2, if the frequency discriminator 5 simply discriminates only the highest frequency, the frequency discriminator 5 will always discriminate as "high". As a result, the transmitted data is always 7 bits, resulting in a poor S/interval. Therefore, if there is such a risk, even if a signal in the above frequency range is included, the absolute level of the component in question is determined, and if the level is lower than a predetermined value, it is determined to be "low". For example, it is possible to prevent the S/interval from deteriorating due to noise during actual use.

The present invention is not limited to the embodiments described above and shown in the drawings, but can be implemented with various modifications without changing the gist thereof.

For example, FIG. 5 shows the configuration of a receiving device according to another embodiment of the present invention.

FIG. 5 shows an embodiment in which a digital filter is used as a low-pass filter for removing harmonic components caused by sampling and quantization. That is, in FIG. 5, the digital low-pass filter 21 is naturally inserted before the D/A converter 12. And the detection controller 13
controls the cutoff frequency of this digital filter 21 and the sampling (latch) frequency of the D/A converter 12. Furthermore, an analog low-pass filter 22 for removing unnecessary high frequencies is inserted after the D/A converter 12, and the output of this analog filter 22 is led out to the output terminal 15. The digital filter 21 first converts an input digital signal into a high frequency digital signal, that is, a digital signal with a data frequency about four times the original frequency in the normal case, and then performs digital filter processing. The output of this digital filter 21 is sent to the D/A converter 1.
By performing D/A conversion in step 2, an effectively filtered signal can be obtained, but in this state, high frequency components based on the data frequency remain. Therefore, an analog filter 22 is provided after the D/A converter 12 to remove the high frequency components. Analog filter 22
Since it is necessary to remove only the above-mentioned high frequency component of about four times the original data frequency, a good effect can be obtained with a simple filter, and the digital signal -22- Since it is also possible to obtain filter characteristics, very good results can be obtained overall.

Further, in the above embodiment, the sampling frequency and bit length are changed in two steps, but they may be changed in three steps or more, and as long as complementary control can be performed as a whole. , the number of change stages may be different between the two.

Furthermore, the frequency of the analog original signal may be determined directly from the analog original signal using an analog filter or the like. As means for determining the frequency, whether directly or indirectly, means similar to means used for FM (frequency modulation) demodulation or means used for so-called frequency analysis can be used. The means for variably controlling the effective sampling frequency and the number of quantization bits for the analog original signal for each data block consisting of multiple samples is to extract the upper significant bits of the original sample data and to thin out the original sample data. Therefore, instead of switching between multiple data converters, it is easy to configure a single data converter in which the actual sampling frequency and number of quantization bits for the analog original signal are variable.

Furthermore, the transmission data is not limited to normal so-called linear PCM data, but may also be PCM data subjected to quasi-instantaneous companding, that is, PCM data of quasi-instantaneous companding PCM method, or differential PCM data. Also, differential PCM data that has been subjected to quasi-instantaneous companding, so to speak, differential quasi-instantaneous companding P
It may be PCM data of CM system.

The quasi-instantaneous companding PCM method consists of a data block made up of multiple pieces of original (linear) PCM data, and by sending scale data indicating the effective digit position of each data block to each data block, the PCM data is compressed into fewer bits. In this method, the received PCM data is compressed into a number and transmitted, and the receiving side expands the received PCM data according to the scale data to obtain a reproduced signal.

In addition, the differential quasi-instantaneous companding PCM method is a method in which quasi-instantaneous companding is applied to differential PCM data. A data block is composed of multiple pieces of original differential PCM data, and a scale indicating the effective digit position of each data block is used. By sending data block by data block, the original differential PCM data is compressed to a small number of bits and transmitted, and on the receiving side, after expanding the received data according to the scale data, integration processing for differential PCM decoding etc. This method obtains a reproduced signal by performing the following steps.

〔Effect of the invention〕

According to the present invention, high S/N with low transmission bit rate
Furthermore, it is possible to provide a PCM transmission method that can realize efficient and excellent transmission having a wide transmission frequency band, and a transmitting device and a receiving device for direct use in its implementation.

[Brief explanation of the drawing]

FIG. 1 is a block diagram schematically showing a schematic configuration of a transmitting device according to an embodiment of the present invention, FIG. 2 is a block diagram schematically showing a schematic configuration of a receiving device in the same embodiment, and FIG. 4 is a diagram schematically showing various data in the same example.
The figure shows the S/N ratio for explaining the transmission characteristics in the same example.
Frequency characteristic diagram, =25- FIG. 5 is a block diagram schematically showing the schematic configuration of a receiving device in another embodiment of the present invention. 2... A/D converter, 3.4... Data converter, 5
...Frequency discriminator, 6...Selector, 12...D
/A converter, 13... detection control device, 14.22...
Analog low-pass filter, 21...digital low-pass filter.

Claims (18)

[Claims] (1) Substantially monitor the frequency components included in the analog original signal to determine the substantial highest frequency of the main components included in the analog original signal, and according to this determination information, a data block consisting of multiple samples In each case, the sampling frequency and number of quantization bits of the digital transmission data based on the analog original signal are complementarily set in the direction of increasing the sampling frequency and lowering the number of quantization bits when the maximum frequency of the analog original signal is high. At the same time, it transmits the transmission data and the discrimination information obtained as a result of this control, receives them, and decodes and decodes them according to the discrimination information.
A PCM transmission system characterized in that the sampling frequency in D/A (digital-to-analog) conversion for demodulation is controlled, and the transmission data is decoded and demodulated using a frequency corresponding to the transmission data. (2) The PCM transmission method according to claim 1, characterized in that the frequency components included in the analog original signal are determined by monitoring original sampling data obtained by sampling and digitizing the analog original signal. . (3) The PCM transmission method according to claim 1 or 2, wherein data processing according to a quasi-instantaneous companding PCM method is performed when obtaining digital transmission data. (4) When obtaining digital transmission data, differential PCM
3. The PCM transmission method according to claim 1 or 2, wherein data processing is performed according to the quasi-instantaneous companding PCM method. (5) When obtaining digital transmission data, differential PCM
The PCM transmission method according to claim 1 or 2, wherein data processing is performed according to the method. (6) Discrimination means for substantially monitoring the frequency components of the analog original signal to determine the substantial highest frequency of the main components included in the analog original signal;
D (analog-to-digital) conversion, and in response to the discrimination information of the discrimination means, substantially the sampling frequency and the number of quantization bits in the A/D conversion are variably controlled for each data block consisting of a plurality of samples, and the analog When the highest frequency included in the original signal is relatively high, the analog original signal has a relatively high sampling frequency and a relatively low number of quantization bits.
A variable A/D conversion means for obtaining D-converted digital transmission data, and a transmission means for sending a transmission signal containing the transmission data obtained by the variable A/D conversion means and the above-mentioned discrimination information to a transmission system. A transmitting device characterized by: (7) The variable A/D conversion means includes an A/D converter that A/D converts the analog original signal with a sufficiently high sampling frequency and a sufficiently high number of quantization bits;
Claim 1, characterized in that it is equipped with variable conversion means for converting the output data of the converter substantially in accordance with the determination result of the determination means and complementarily varying the sampling frequency and the number of quantization bits. The transmitting device according to item 6. (8) The variable conversion means converts the output data of the A/D converter so that at least one of the sampling frequency and the number of quantization bits changes, and obtains converted data having a different sampling frequency and a different number of quantization bits. 8. The transmitting device according to claim 7, comprising a plurality of converting means and a selection control means for selectively enabling the converting means in accordance with the output of the determining means. (9) The transmitting device according to claim 7, wherein the determining means is means for determining the highest frequency based on the output of the A/D converter. (10) The transmitting device according to claim 9, wherein the variable conversion means includes means for converting the A/D converted data according to a quasi-instantaneous companding PCM method. (11) The transmitting device according to claim 9, wherein the variable conversion means includes means for converting the A/D converted data according to the differential PCM method. (12) The transmitting device according to claim 9, wherein the variable conversion means includes means for converting the A/D converted data according to the differential PCM method and the quasi-instantaneous companding PCM method. (13) The actual sampling frequency and number of quantization bits for the analog original signal are variably controlled for each data block consisting of a plurality of samples, depending on the actual highest frequency of the main component included in the analog original signal. , receives a transmission signal including digital transmission data obtained by A/D (analog-digital) conversion of the analog original signal and discrimination information corresponding to the sampling frequency, and demodulates the transmission signal to obtain a reproduced signal. In the receiving device, the transmission data in the transmission signal is D/
demodulation means for obtaining a demodulated signal through A (digital-to-analog) conversion; 1. A receiving device comprising: a control means for controlling the means. (14) The demodulation means performs D/A conversion of the transmission data.
The control means includes an A converter and a low-pass filter provided after the D/A converter to remove harmonic components generated in connection with sampling, and the control means is configured to control the D/A converter.
14. The receiving device according to claim 13, wherein the receiving device is configured to control the sampling frequency of the A converter and the cutoff frequency of the low-pass filter. (15) The demodulation means performs D/A conversion of the transmission data.
A converter, a digital low-pass filter provided before the D/A converter for removing harmonic components generated in connection with sampling, and a D/A converter provided after the D/A converter. an analog low-pass filter for removing unnecessary high-frequency components of the converter output, and the control means is configured to control a sampling frequency of the D/A converter and a cutoff frequency of the digital low-pass filter. The receiving device according to claim 12, characterized in that: (16) The receiving device according to claim 13, wherein the demodulating means includes decoding means according to a quasi-instantaneous companding PCM method. (17) The receiving device according to claim 13, wherein the demodulating means includes a decoding means according to a differential PCM method. (18) The receiving device according to claim 13, wherein the demodulating means includes decoding means according to the differential PCM method and the quasi-instantaneous companding PCM method.