JPS60144037A - Multiplex simultaneous transmission system between voice signal and data signal - Google Patents

Abstract

PURPOSE:To attain simultaneous transmission while a voice signal and a data signal are subjected to multiplex and keeping excellent voice quality by changing an encoded bit number of the voice signal and the data signal in response to the transmitting state of the data signal at that time. CONSTITUTION:An analog voice signal A inputted from a voice input terminal 1 is converted into a digital voice signal B by an analog digital converter 2, and inputted to a filter group 4 comprising n-set of filters 31, 32,...3n whose pass band is changed optionally. Output signals C1-Cn of the filter group 4 are inputted respectively to voice coders 51, 52,...5n and they are encoded at each band so that the encoded bit number per one frame is Ns-bit in total. Output signals E1-En of the voice coders 51-5n are subjected to multiplex by a multiplex circuit 8 together with a data signal F inputted from a data input terminal 7, and outputted to an output terminal 9 as a multiplex signal G of Nt-bit per one frame.

Description

[Detailed Description of the Invention] The present invention relates to a digital transmission system in which audio signals and data are digitally encoded, multiplexed with a fixed frame length, and transmitted simultaneously.1. <Prior Art> Conventionally, singing signals 1) (lVt (pulse F1 modulation), 8P
CM (J response CIVI), J) P CM (
difference i'cIV]), ADPCM (adaptation 1) P CM
, ), the audio signal is encoded with I) N S bits per frame and encoded with pNd bits per frame.
Multiplexed with Nishi No., Ns + Nd (=
A method of transmitting it as a code of Nt ) bits has been considered.

Conventionally, the coded bits for the audio signal are 4y N
S and the number of bits Nd for Cheetai No. 1 are fixed once they are determined, and the audio signal is transmitted as a code of N's bits even when there is no data signal to be transmitted at the same time. For example, in a conference call, when data of a hand-drawn figure drawn on a blackboard is multiplexed and transmitted at the same time as an audio signal, the conventional method requires a large number of bits to transmit the data of the hand-drawn figure. (data speed when drawing at the highest speed) is applied to one egg for the data signal, and the number of remaining bits in one frame N t bits N
's for audio signals. For this reason, the number of bits encoded for the audio signal is small, resulting in poor transmission quality of the audio signal. However, hand-drawn figures must be transmitted at the same time, and even when the hand-drawn figure data is not being transmitted, in the conventional system, the audio signal is sent to bit 111 of the remaining bits. The signal was encoded and transmitted using Ns, and the transmission path was not A.1. If you repeatedly draw graphics by hand, you will end up allocating an unnecessarily large number of bits to the hand-drawn graphics data. .

<Summary of the Invention> This invention eliminates these drawbacks of performance and function mJ7,
While effectively using the transmission path and maintaining good quality and quality,
The voice signal and data (No. 9) are multiplexed and transmitted simultaneously. Therefore, the number of encoded bits of the M voice signal and the data signal is changed depending on the transmission status of Take No. 16 at that time.

That is, in the present invention, an audio signal is divided into a number of frequency subbands, and each frequency subband is encoded. It is known that the energy distribution of the speech spectrum is uneven depending on the frequency band, and that the degree of contribution to clarity and naturalness differs depending on the frequency band. For this reason,
When encoding audio No. 16 with NS bits per frame, the audio signal is divided into multiple frequency sub-bands and encoded between the sub-bands depending on the degree of 71J required for the audio quality of each sub-band. The bit sand is distributed non-uniformly and the sub-band VL-, J: 1. By allocating more bits, the voice quality is improved compared to the case where band division coding is not performed.

Furthermore, corresponding to the N's immortal of the gods, the number of a band divisions,
5 frequency bandwidth, 31n number of encoding bits in each C band
Set the value of the f IN parameter in advance,
For Ns (=Nt-Nd), which is determined according to the change in the number of bits N'd required for encoding the data signal, the values of the three parameters J are determined from those set in advance by the previous day.21f-L, 7° based on this? By encoding the voice code, it has a function that allows you to arbitrarily set the number of encoding bits between the voice signal and data (i4 code), and reduces voice quality deterioration due to the reduction of bits for data signal transmission. Figure 1 shows an embodiment of the multiplex and simultaneous transmission system for audio signals and data signals according to the present invention. The analog audio signal input from the voice input terminal 1 is converted to the digital voice signal 1N by the analog-to-digital converter 2, and from this, the range can be changed arbitrarily. The output signal C of the filter group 4 is inputted to a filter group 4 consisting of n filters 31, 32°, . . . 3n.
1 to Cn are coders 51 and 521, respectively.
5n, and is encoded for each band so that the total number of encoded bits per frame is gtNs bits. The number of filters (i.e., the number of band component divisions) used for speech encoding in the filter group 4, the passband of each filter, and the number of bits encoded in the prayer signifiers 51 to 5n are stored in the control circuit 6 in advance, The control signal is based on information about the number of filters, the passband of each filter, and the number of coding bits in each band, compared to the number of bits allocated to the harmonization of the voice (N8 bits per frame). - adaptively selected by i[i].

The output signals E1 to En of the audio encoders 51 to 5n are multiplexed by the multiplexing circuit 8 together with the data signal 1+1 input from the data input Δ1)
It is output to the output terminal 9 as a multiplexed signal G of t bits.

In the above processing, the audio encoding methods in each band include adaptive PCM (APCM) and adaptive differential PCM.
A known method such as M (ADPCM) may be used, and all three types of parameters described above are not changed in response to the bridge formation of Ns and Nd, for example, the number of band divisions and the bandwidth are fixed. f? to change only the number of encoding bits in each band? ii It is also possible to simplify. The method of this invention can be implemented using a hard-wired logic configuration using variable band digital filters, quantizers, predictors, etc., or can be implemented using a programmable signal processing processor. Furthermore, the sampling frequency in each frequency band is set to 8 i (1-17
, and by setting the frame length Nt to 8 bits, the 8K H7 clock provided by the network can be used for frame identification. Compared to the one-segment method, the control is simple, and the frame rate is 1z jjj
It is possible to avoid quality deterioration due to errors in different bits during transmission.

As an example of the multiplex simultaneous transmission system for voice signals and data one by one according to the present invention, an H
? is divided into two frequencies, 0 to 4 KHz and 4 to 8 Ktlz, and in each band, p,
I) Audio is encoded using the P CIV1 method, and supports cases such as no data signal transmission at this time, ■8Kb/s 8Kb/S data signal transmission, (■simultaneous transmission with 16Kb/S data signal) Then, the number of encoded bits for one frame (8
bits) r 1 to 1 for low-frequency audio signals, high-frequency audio signals, and data signals, 6 bits, 2 pins, and 0 bits for each of ■, 5 bits, 2 bits, and 1 bit for each of ■, 4 bits for each of ■, A method of dividing into 2 bits, 2 bits, and 1] is mentioned.

Methods for identifying that the number of bits assigned to each voice signal and data signal have changed include a method of transmitting identification information on separate channels, a method of always providing bits for identification within one frame, and a method by Pi] Steel. There are many possible methods. Compared to these methods, stone 7'L1 item No. 1
An identification method that is effective in reducing the transmission bit rate and transmission bit rate will be described below. The lowest pitch of the audio code in the preset frequency subband for each category in which the number of encoded bits of the reputation signal and the data signal are different) (L
S
1'·J: and use this bit to construct an identification bit pattern. In such identification hit binding, it is sufficient to use the LSB of a code that causes less deterioration of voice quality due to an error in p-1) L S t+. For example, the code with the longest code length among the B-voice codes of each subband may be used.

As an example, when audio in the 7Kllz band of [)l;71 is ADPCM encoded by dividing the frequency band into two, the LSB of the low frequency audio code is divided into 8 bits of one frame as shown in Figure 2. Placement shift at the same bit position of the code (the hatched bit at the right end in Figure 2), and the bit pattern of the bit here is, for example, O], 0101・e・
Fist , 0011001100 ・ ・ # Book , (1
00], I 1 0 0 0], 1. ], 00
(In each case of 1..., 2゛1ko21!XI
A.

B and C are added to represent no data signal transmission, simultaneous transmission with 8 Kb/S data signal, and simultaneous transmission with 16 Kb/S data signal. When the bit allocation is changed, this identification bit pattern only needs to be transmitted for a period of one ship, and the quality of the audio signal will deteriorate for that song, but 1. /+J
Since the lower bits are used, there is no risk of significant deterioration.3 In the method, the identification signal is not transmitted on a separate channel, the transmission bit rate increases accordingly, and the bits for each job are always provided in six frames. However, the above-mentioned identification method makes it possible to increase the bit rate for identification. When there are no identification bits and the number of identification bits is too low to supply, the sign of the audio signal is changed to a strange bit, and the bits that have little influence on the bit distribution are selected and used.
Deterioration of voice quality can be reduced. For the identification signal, not only the least significant bit, but also a plurality of bits including this and its higher order bits (1111 bits) may be used.

Note that such changes in bit allocation can be made, for example, in the above-mentioned conference'
On the telephone, when a speaker sends hand-drawn graphic data, a switch indicating this is turned on so that bits are allocated accordingly. The bandwidth and the number of band divisions may also be controlled. For example, when simultaneously transmitting 8Kb/s data No. 11, 1 bit is allocated to this and the audio signal is divided into 0 to 4K.
Hz band and 4-8K) +7. 4 bits and 3 bits are allocated to each band, and 16Kb/
When transmitting simultaneously with the data signal of s, allocate 2 bits to the data signal and transfer the audio signal to
Divide into 3 bands from 16 to 8KH7, and assign 3 bits, 2 bits, and 1 bit to each band.0 If you want to do this with hardware, each of these ij
i\J'J1.1f, Ij1 (prepare 4theta waveformers and use them selectively.Also, for bit assignment to data signals, for example, store data signals once in a waveform and then multiplex them. It can be supplied to circuit 8 and made to take place depending on the fullness of its buffer.

<Effects> As described above, according to the present invention, when audio signals and data signals are multiplexed and transmitted simultaneously, effective bit allocation is performed by band division, and when data signals are not transmitted, In order to allocate bits to the audio signal as much as possible when the data signal speed is slow or low, the performance and functionality allows the number of encoding bits of the audio signal and data signal to be arbitrarily changed while maintaining good audio quality. Simultaneous voice and data transmission can be achieved with the same features as a dog.

As mentioned above, an example of an application field of this invention is, for example, in a conference telephone, where a wideband voice signal of 0 to 7Kt(z) is multiplexed with a handwritten character graphic signal and transmitted simultaneously at a total bit rate of 64Kb/s. It is extremely effective in improving audio quality (wideband) and functionality in conference telephone services.

[Brief explanation of the drawing]

FIG. 1 shows -'fl! multiple simultaneous transmission system for voice signals and Qi-tai b according to the present invention. 1) A block diagram of an example of 1) Figure 2: 1. Bit arrangement of one frame (frame length: 8 bits) in an embodiment of the multiplex simultaneous transmission system for a voice signal and a data signal according to the present invention. FIG. 1 is a diagram showing an example. 1: Audio input terminal, 2: Analog-digital converter, 3
: band variable filter, 4: filter group, 5: audio encoder, 6: ffft1 circuit, 7: data input & jM child, 8: multiplexing circuit, 9: output terminal. 'l''F ll・Entry/exit; 1↓ person To 1 City Shinichi 1 story Public corporation representative Person Takuo Jono 1 Figure

Claims (1)

[Claims] (1) In a method in which audio signals and data signals are digitally encoded, multiplexed with a frame length of Nt bits, and transmitted simultaneously, some data per frame (, corresponding to changes in the number of encoded bits Nd of jM, Divide the voice signal into two or more frequency bands L7, A band 1) L number, BL closed wave number band 11, 4, C each; iii; Among the three conditions of the number of encoding bits per frame in the range The frame of the audio signal is changed by changing at least some conditions of N s (N5=Nt
-N'd) bits, thereby changing the number of encoded bits Ns of the audio signal per frame and the encoded bit 1YNd of the data signal and transmitting the same. and data signal multiplex simultaneous transmission method 1,
(2) Each divided frequency': the 8 bits of the audio signal in the t'i range. I (transmission formula 9゜(3) The lowest 1 bit or the highest bit of the prayer code in a predetermined frequency band): The mantissa bit from the most significant bit to the most significant bit, i.e. 7L signal and data signal encoding. Pit F = N The same piti of 4 bows for beginners with different combinations of s and Nd) 't',
f41 for 1ift. A patent claim constituting a pattern that identifies the difference in the number of bits between a voice signal and a data signal in a multiplexed code. jp (4th circle) A multiplex simultaneous transmission system for audio and data signals mounted on the tomb.