CN105513599B - A kind of rate adaptation acoustic code communication means protected again based on non-grade - Google Patents
A kind of rate adaptation acoustic code communication means protected again based on non-grade Download PDFInfo
- Publication number
- CN105513599B CN105513599B CN201510823888.3A CN201510823888A CN105513599B CN 105513599 B CN105513599 B CN 105513599B CN 201510823888 A CN201510823888 A CN 201510823888A CN 105513599 B CN105513599 B CN 105513599B
- Authority
- CN
- China
- Prior art keywords
- data block
- vocoder
- jth
- bits
- channel
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Classifications
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS OR SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING; SPEECH OR AUDIO CODING OR DECODING
- G10L19/00—Speech or audio signals analysis-synthesis techniques for redundancy reduction, e.g. in vocoders; Coding or decoding of speech or audio signals, using source filter models or psychoacoustic analysis
- G10L19/005—Correction of errors induced by the transmission channel, if related to the coding algorithm
Landscapes
- Engineering & Computer Science (AREA)
- Computational Linguistics (AREA)
- Signal Processing (AREA)
- Health & Medical Sciences (AREA)
- Audiology, Speech & Language Pathology (AREA)
- Human Computer Interaction (AREA)
- Physics & Mathematics (AREA)
- Acoustics & Sound (AREA)
- Multimedia (AREA)
- Detection And Prevention Of Errors In Transmission (AREA)
- Compression, Expansion, Code Conversion, And Decoders (AREA)
Abstract
The invention discloses a kind of based on the non-equal rate adaptation acoustic code communication means protected again; characterized by comprising the following steps: passing voice signal under the microphone that (1) passes through sender, it is encoded by vocoder; vocoder data block is obtained, vocoder data block is packaged；It is modulated according to the structure for sending signal after the vocoder data block after encapsulation is successively carried out data arrangement, LDPC coding and interweaved according to the non-equal weight channel coding under the transmission rate of vocoder data block and is sent to recipient；(2) recipient synchronizes detection and demodulation, and resolves to the corresponding walsh sequence of vocoder data block；Calculate the signal-to-noise ratio of channel when sending vocoder data block；Vocoder data block after demodulation is successively subjected to balanced, deinterleaving and LDPC decoding, then LDPC decoding is obtained into a section voice signal, and be uploaded to the microphone of recipient after vocoder decodes.
Description
Technical field
The invention belongs to voice communication technology fields, are related to a kind of based on the non-equal rate adaptation acoustic code communication protected again
Method.
Background technique
Currently, realizing that channel coding when voice communication is all the parameter coding under certain distortion factor, so information source is compiled
The output of code is no longer equal-probability distribution, wherein still having remaining redundancy, and the difference in source decoding after voice coding
Bit is not also identical to the susceptibility of error code.Wherein, significant bits compare rdativery sensitive to error code, once error code occurs can be significantly
Reduce the quality of synthesis voice, and unessential bit is to error code relative insensitivity, once error code pairing occurs into the matter of voice
Amount influences smaller.So currently used equal recode all carries out identical channel coding to all bits, it is easy in the bit error rate
The quality of synthesis voice is seriously affected in higher situation.
Currently, mainly using modem to the transmission of acoustic code words, realized under fixed transmission rate.But it is solid
The acoustic code words transmission technology of constant speed rate can not be well adapted for actual channel situation.When channel conditions deteriorate, transmission wave
The ability of shape load information reduces, and the bit error rate just will increase, need the data to a large amount of loading error occurrings to retransmit, reduce language
The communication efficiency of sound communication.
Summary of the invention
In view of the above shortcomings of the prior art, it is an object of the invention to propose it is a kind of based on non-equal rate protected again from
Adapt to acoustic code communication means.This method is directed to low density parity check code (Low-Density Parity-Check, LDPC code),
By the channel quality of real-time judge communicating pair, single times, double and four times of three kinds of transmission rates, energy can be adaptively replaced
Enough under the premise of meeting intelligibility, the naturalness of voice is utmostly improved；Meanwhile the method for the present invention is for above-mentioned three kinds transmission
Rate devises corresponding non-equal weight channel coding, and the program carries out stronger channel coding protection to important bit, to not
Important bit carries out weaker channel coding protection or protects without channel coding, can be obviously improved voice coding
Error-resilient performance, and improve the quality of synthesis voice.
To realize the above-mentioned technical purpose, the present invention is achieved by the following scheme.
A kind of rate adaptation acoustic code communication means protected again based on non-grade, which comprises the following steps:
Step 1, the channel coding of setting acoustic code communication uses LDPC code；Then, for single times, double and four times of three kinds of biographies
Defeated rate, non-equal weight channel coding under conditions of vocoder delay is fixed delay, under every kind of transmission rate of design；
Step 2, setting sends the structure of signal: sending signal and is made of the sequence of four different function, respectively transmits
Layer control/automatic growth control (Transmit Level Control/Automatic Gain Control, TLC/AGC) sequence
Column, synchronization header (Preamble, PRE) sequence, walsh sequence and data (DATA) sequence；
Step 3, firstly, passing paragraph 1 voice signal under the microphone for passing through sender, and the biography of paragraph 1 voice signal is set
Defeated rate is single times of transmission rate；Then, it selects using single times of transmission rate to be the vocoder of code rate to paragraph 1 voice signal
It is encoded, obtains the 1st vocoder data block, the 1st vocoder data block is packaged by sender, the 1st acoustic code after being encapsulated
Data block；Then, the 1st vocoder data block after encapsulation is successively carried out according to the non-equal weights channel coding under single times of transmission rate
Data arrangement, LDPC coding and intertexture, obtain the 1st coded data to be sent；Finally, by the 1st coded data to be sent according to hair
The structure for the number of delivering letters is modulated, and is sent to recipient with single times of transmission rate；
Step 4, recipient synchronizes detection, and demodulates to the data detected are synchronized, and the after being demodulated the 1st
Vocoder data block；The corresponding walsh sequence of the 1st vocoder data block is resolved again, obtains the transmission speed of the 1st vocoder data block
The initial valuation of rate and channel；Then, recipient is according to the initial of the corresponding synchronization header of the 1st vocoder data block and channel
Channel is estimated in valuation, and the signal-to-noise ratio of channel when sending the 1st vocoder data block is calculated；Finally, recipient will demodulate
The 1st vocoder data block afterwards successively carries out balanced, deinterleaving and LDPC decoding, then LDPC decoding is decoded by vocoder
Afterwards, paragraph 1 voice signal is obtained, and paragraph 1 voice signal is uploaded to the microphone of recipient；
If recipient need to sender reply, recipient according to send the 1st vocoder data block when channel noise
Than the transmission rate for determining the 2nd vocoder data block, the transmission rate of the 2nd vocoder data block is fed back to sender by recipient；If
The block number j for determining code data block is 2, goes to step 5；
If recipient does not need to reply to sender, acoustic code communication process terminates；
Step 5, firstly, passing jth section voice signal under the microphone for passing through sender；Then, selection is with jth vocoder data block
Transmission rate be code rate vocoder jth section voice signal is encoded, obtain jth vocoder data block, sender
Jth vocoder data block is packaged according to the frame structure of setting, the jth vocoder data block after being encapsulated；Then, according to
Non- equal weight channel coding under the transmission rate of j vocoder data block, successively carries out data row to the jth vocoder data block after encapsulation
Column, LDPC coding and intertexture, obtain jth coded data to be sent；Finally, by jth coded data to be sent according to transmission signal
Structure be modulated, recipient is sent to the transmission rate of jth vocoder data block；
Step 6, recipient synchronizes detection, and demodulates to the data detected are synchronized, the jth after being demodulated
Vocoder data block；The corresponding walsh sequence of jth vocoder data block is resolved again, obtains the transmission speed of jth vocoder data block
- 1 valuation of rate and the jth of channel；Then, recipient is according to the jth-of jth vocoder data block corresponding synchronization header and channel
Channel is estimated in 1 valuation, and the signal-to-noise ratio of channel when sending jth vocoder data block is calculated；Finally, recipient will demodulate
Jth vocoder data block afterwards successively carries out balanced, deinterleaving and LDPC decoding, then LDPC decoding is decoded by vocoder
Afterwards, jth section voice signal is obtained, and jth section voice signal is uploaded to the microphone of recipient；
If recipient need to sender reply, recipient according to send jth vocoder data block when channel noise
Than the transmission rate of determining+1 vocoder data block of jth, the transmission rate of jth vocoder data block is fed back to sender by recipient；
The block number j of vocoder data block is enabled to increase by 1, return step 5；
If recipient does not need to reply to sender, acoustic code communication process terminates.
The features of the present invention and further improvement lies in that:
(1) in step 1, the list times, double and four times of three kinds of transmission rates be respectively 600bps, 1200bps and
2400bps, the fixed delay are 300ms.
(2) in step 2, the load that the walsh sequence carries is 4bit, wherein preceding 2bit indicates transmission rate, after
The valuation of 2bit expression channel；
The data sequence indicates vocoder data block, and every piece of vocoder data block includes n+1 data packet, is followed successively by data packet
EDataPKT0 to data packet EDataPKTn, wherein data packet EDataPKTi is encapsulated by 3 frame data, and every frame data wrap
Unknown data and given data are included, unknown data is the vocoder data to be transmitted, it is known that data are training sequence, and i takes 0 to n, n
It is natural number.
(3) in step 3, the 1st vocoder data block is packaged according to the frame structure of setting, the 1st sound after being encapsulated
Code data block, wherein the frame structure of the setting are as follows:
To single times of transmission rate, 576 bits are exported after non-equal recodification, form 576 BPSK by Gray code
Symbol splits into 3 frames and is sent, and 192 BPSK symbols of every frame, the training sequence being inserted into thereafter is 48 BPSK symbols, is
Known array, then 600bps transmission rate a data segment, length is 720 BPSK symbols；
For double transmission rate, 1152 bits are exported after non-equal recodification, form 576 by Gray code
QPSK symbol splits into 3 frames and is sent, 192 QPSK symbols of every frame, and the training sequence being inserted into thereafter is 48 QPSK symbols
Number, it is known array, then 1200bps transmission rate a data segment, length is 720 QPSK symbols；
For four times of transmission rates, 1152 bits are exported after non-equal recodification, form 576 by Gray code
QPSK symbol splits into 3 frames and is sent, 192 QPSK symbols of every frame, and the training sequence being inserted into thereafter is 48 QPSK symbols
Number, it is known array, then 1200bps transmission rate a data segment, length is 720 QPSK symbols.
(4) in step 4, the transmission of the 2nd vocoder data block is determined according to the signal-to-noise ratio of channel when sending the 1st vocoder data block
Rate, detailed process are as follows:
The list times, double and four times of three kinds of transmission rates are respectively 600bps, 1200bps and 2400bps, it is known that
The optimum signal-noise ratio range of 600bps, 1200bps and 2400bps transmission rate is respectively as follows: as SNR < 2dB, and channel can be protected
Demonstrate,prove 600bps reliable transmission；As 2dB≤SNR < 6dB, channel can guarantee 1200bps reliable transmission；As SNR >=6dB,
Channel can guarantee 2400bps reliable transmission；By transmission rate corresponding to the signal-to-noise ratio of channel when sending the 1st vocoder data block
It is determined as the transmission rate of the 2nd vocoder data block.
(5) in step 4, the 1st vocoder data block by after demodulation successively carries out balanced, deinterleaving and LDPC decoding,
Its detailed process are as follows:
(b1) setting i indicates the number of iterations, enables initial value i=1；
(b2) the Turbo equalization algorithm based on least mean-square error is used, is calculated by the 1st vocoder data block after demodulation
Data sequence x in 1st vocoder data block out1I-th equilibrium estimated valueAnd it is deinterleaved；
(b3) by the i-th equilibrium estimated value after deinterleaving after LDPC is decoded, i-th LDPC decoding estimated value is obtained
(b4) if ldpc decoder determines that i-th LDPC decodes estimated valueIt is incorrect, by the 1st vocoder data block
In data sequence x1It is updated toAnd the number of iterations i is enabled to increase by 1, return step (b2)；
If ldpc decoder determines that i-th LDPC decodes estimated valueIt is for correct or the number of iterations i=I, I
The maximum equalization iteration number of setting, then enable the data sequence in the 1st vocoder data blockBy the 1st vocoder data block
In data sequence x1Vocoder is sent into be decoded.
(6) the non-equal weight channel coding in step 1, under the single times of 600bps transmission rate are as follows: encoded for 600bps
The coding of the vocoder of rate, 75ms frame exports 45 bits, and 180 ratio of coding output of 300ms superframe is made of 4 75ms frames
It is special；The LDPC for carrying out 4/13 code rate plus the CRC check position of 12 bits to 156 bits important in 300ms superframe encodes output
546 bits, the CRC check position to 24 bit remaining in 300ms superframe plus 6 bits are directly exported without channel coding,
Each 300ms superframe exports 576 bits altogether；
Non- equal weight channel coding under the double 1200bps transmission rate are as follows: for the acoustic code of 1200bps code rate
The coding of device, 50ms frame exports 60 bits, exports 360 bits by the coding that 6 50ms frames form 300ms superframe；It is super to 300ms
264 important bits carry out LDPC coding 1056 bits of output of 1/4 code rate in frame, to 96 bit remaining in 300ms superframe
It is directly exported without channel coding, each 300ms superframe exports 1152 bits altogether；
Non- equal weight channel coding under four times of 2400bps transmission rates are as follows: for the acoustic code of 2400bps code rate
The coding of device, 25ms frame exports 60 bits, exports 720 bits by the coding that 12 25ms frames form 300ms superframe；To 300ms
384 important bits carry out LDPC coding 816 bits of output of 1/2 code rate plus the CRC check position of 24 bits in superframe, right
Remaining 336 bit is directly exported without channel coding in 300ms superframe, and each 300ms superframe exports 1152 bits altogether.
The invention has the benefit that the present invention is capable of the channel quality of real-time judge communicating pair, adaptively replace
Dan Bei, double and four times of three kinds of transmission rates utmostly improve the naturalness of voice under the premise of meeting intelligibility；This hair
It is bright to wait weight channel channel coding that carry out stronger channel to important bit for the non-of above-mentioned three kinds of transmission rates design
Coding protection carries out weaker channel coding protection even without protection to unessential bit, can be obviously improved voice
The error-resilient performance of coding.
Detailed description of the invention
Invention is further described in detail with specific embodiment for explanation with reference to the accompanying drawing.
Fig. 1 is the transmitting and receiving flow chart of the method for the present invention；
Fig. 2 is the structure chart of transmission signal of the invention；
Fig. 3 is the inferior heavy channel coding of 600bps transmission rate and weight channel coding important ratio under identical signal-to-noise ratio such as non-
Special residual BER comparison diagram, wherein abscissa is signal-to-noise ratio, and unit is decibel (dB), and ordinate is residual BER；
Fig. 4 is that the inferior heavy channel coding of 1200bps transmission rate and the weight channel coding such as non-are important under identical signal-to-noise ratio
The residual BER comparison diagram of bit, wherein abscissa is signal-to-noise ratio, and unit is decibel (dB), and ordinate is residual BER；
Fig. 5 is that the inferior heavy channel coding of 2400bps transmission rate and the weight channel coding such as non-are important under identical signal-to-noise ratio
The residual BER comparison diagram of bit, wherein abscissa is signal-to-noise ratio, and unit is decibel (dB), and ordinate is residual BER.
Specific embodiment
Below in conjunction with Detailed description of the invention, technical scheme in the embodiment of the invention is clearly and completely described, shows
So, described embodiments are only a part of the embodiments of the present invention, instead of all the embodiments.Based on the reality in the present invention
Example is applied, every other embodiment obtained by those of ordinary skill in the art without making creative efforts all belongs to
In the scope of protection of the invention.
Of the invention is a kind of based on the non-equal rate adaptation acoustic code communication means protected again, comprising the following specific steps
Step 1, the channel coding of setting acoustic code communication uses LDPC code；Then, for single times, double and four times of three kinds of biographies
Defeated rate, non-equal weight channel coding under conditions of vocoder delay is fixed delay, under every kind of transmission rate of design.
In embodiments of the present invention, single times of setting, double and four times of three kinds of transmission rates are respectively 600bps, 1200bps
And 2400bps, the fixed delay are 300ms.
In embodiments of the present invention, the non-equal weight channel coding under Dan Bei 600bps transmission rate are as follows: compiled for 600bps
The coding of the vocoder of bit rate, 75ms frame exports 45 bits, and 180 ratio of coding output of 300ms superframe is made of 4 75ms frames
It is special；The LDPC for carrying out 4/13 code rate plus the CRC check position of 12 bits to 156 bits important in 300ms superframe encodes output
546 bits, the CRC check position to 24 bit remaining in 300ms superframe plus 6 bits are directly exported without channel coding,
Each 300ms superframe exports 576 bits altogether；
Non- equal weight channel coding under double 1200bps transmission rate are as follows: for the vocoder of 1200bps code rate,
The coding of 50ms frame exports 60 bits, exports 360 bits by the coding that 6 50ms frames form 300ms superframe；To 300ms superframe
In important 264 bits carry out LDPC coding 1056 bits of output of 1/4 code rate, not to 96 bit remaining in 300ms superframe
It carries out channel coding directly to export, each 300ms superframe exports 1152 bits altogether；
Non- equal weight channel coding under four times of 2400bps transmission rates are as follows: for the vocoder of 2400bps code rate,
The coding of 25ms frame exports 60 bits, exports 720 bits by the coding that 12 25ms frames form 300ms superframe；To 300ms superframe
In important 384 bits LDPC coding 816 bits of output of 1/2 code rate are carried out plus the CRC check position of 24 bits, to 300ms
Remaining 336 bit is directly exported without channel coding in superframe, and each 300ms superframe exports 1152 bits altogether.
Step 2, referring to Fig. 2, setting sends the structure of signal: it sends signal and is made of the sequence of four different function, point
Not Wei transport layer control/automatic growth control (Transmit Level Control/Automatic Gain Control,
TLC/AGC) sequence, synchronization header (Preamble, PRE) sequence, walsh sequence and data (DATA) sequence.
In the present invention is implemented, the TLC/AGC stage is a string of leader sequences, it is therefore an objective to send data and receive data
Sender can be made to reach the operating power of requirement in time；PRE sequence, which is used to provide to recipient in synchronous detection data, to be sentenced
Disconnected information；The load that walsh sequence carries is 4bit, wherein preceding 2bit indicates transmission rate, and rear 2bit indicates estimating for channel
Value；The DATA stage indicates vocoder data block, and every piece of vocoder data block includes n+1 data packet, is followed successively by data packet EDataPKT0
To data packet EDataPKTn, wherein data packet EDataPKTi is encapsulated by 3 frame data, and every frame data include unknown number
According to and given data, unknown data be the vocoder data to be transmitted, it is known that data are training sequence, and it is natural number that i, which takes 0 to n, n,.
Step 3, referring to Fig.1, firstly, passing paragraph 1 voice signal under the microphone for passing through sender, and paragraph 1 voice is set
The transmission rate of signal is single times of transmission rate；Then, it selects using single times of transmission rate to be the vocoder of code rate to paragraph 1
Voice signal is encoded, and obtains the 1st vocoder data block, the 1st vocoder data block is packaged by sender, after being encapsulated
1st vocoder data block；Then, according to the non-equal weights channel coding under single times of transmission rate to the 1st vocoder data block after encapsulation
It successively carries out data arrangement, LDPC coding and interweaves, obtain the 1st coded data to be sent；Finally, by the 1st coded number to be sent
It is modulated according to according to the structure for sending signal, recipient is sent to single times of transmission rate.
In embodiments of the present invention, the 1st vocoder data block is packaged according to the frame structure of setting, after being encapsulated
1st vocoder data block, wherein the frame structure of setting are as follows:
To single times of transmission rate, 576 bits are exported after non-equal recodification, form 576 BPSK by Gray code
Symbol splits into 3 frames and is sent, and 192 BPSK symbols of every frame, the training sequence being inserted into thereafter is 48 BPSK symbols, is
Known array, then 600bps transmission rate a data segment, length is 720 BPSK symbols；
For double transmission rate, 1152 bits are exported after non-equal recodification, form 576 by Gray code
QPSK symbol splits into 3 frames and is sent, 192 QPSK symbols of every frame, and the training sequence being inserted into thereafter is 48 QPSK symbols
Number, it is known array, then 1200bps transmission rate a data segment, length is 720 QPSK symbols；
For four times of transmission rates, 1152 bits are exported after non-equal recodification, form 576 by Gray code
QPSK symbol splits into 3 frames and is sent, 192 QPSK symbols of every frame, and the training sequence being inserted into thereafter is 48 QPSK symbols
Number, it is known array, then 1200bps transmission rate a data segment, length is 720 QPSK symbols.
Step 4, recipient synchronizes detection, and demodulates to the data detected are synchronized, and the after being demodulated the 1st
Vocoder data block；The corresponding walsh sequence of the 1st vocoder data block is resolved again, obtains the transmission speed of the 1st vocoder data block
The initial valuation of rate and channel；Then, recipient is according to the initial of the corresponding synchronization header of the 1st vocoder data block and channel
Channel is estimated in valuation, and the signal-to-noise ratio of channel when sending the 1st vocoder data block is calculated；Finally, recipient will demodulate
The 1st vocoder data block afterwards successively carries out balanced, deinterleaving and LDPC decoding, then LDPC decoding is decoded by vocoder
Afterwards, paragraph 1 voice signal is obtained, and paragraph 1 voice signal is uploaded to the microphone of recipient；
If recipient need to sender reply, recipient according to send the 1st vocoder data block when channel noise
Than the transmission rate for determining the 2nd vocoder data block, the transmission rate of the 2nd vocoder data block is fed back to sender by recipient；If
The block number j for determining code data block is 2, goes to step 5；
If recipient does not need to reply to sender, acoustic code communication process terminates.
In embodiments of the present invention, the 2nd vocoder data block is determined according to the signal-to-noise ratio of channel when sending the 1st vocoder data block
Transmission rate, detailed process are as follows: the optimum signal-noise ratio range of known 600bps, 1200bps and 2400bps transmission rate point
Not are as follows: as SNR < 2dB, channel can guarantee 600bps reliable transmission；As 2dB≤SNR < 6dB, channel can guarantee
1200bps reliable transmission；As SNR >=6dB, channel can guarantee 2400bps reliable transmission；The 1st vocoder data block will be sent
Transmission rate corresponding to the signal-to-noise ratio of Shi Xindao is determined as the transmission rate of the 2nd vocoder data block.
In embodiments of the present invention, the 1st vocoder data block after demodulation is successively subjected to balanced, deinterleaving and LDPC is translated
Code, detailed process are as follows:
(b1) setting i indicates the number of iterations, enables initial value i=1；
(b2) the Turbo equalization algorithm based on least mean-square error is used, is calculated by the 1st vocoder data block after demodulation
Data sequence x in 1st vocoder data block out1I-th equilibrium estimated valueAnd it is deinterleaved；
(b3) by the i-th equilibrium estimated value after deinterleaving after LDPC is decoded, i-th LDPC decoding estimated value is obtained
(b4) if ldpc decoder determines that i-th LDPC decodes estimated valueIt is incorrect, by the 1st vocoder data block
In data sequence x1It is updated toAnd the number of iterations i is enabled to increase by 1, return step (b2)；
If ldpc decoder determines that i-th LDPC decodes estimated valueIt is for correct or the number of iterations i=I, I
The maximum equalization iteration number of setting, then enable the data sequence in the 1st vocoder data blockBy the 1st vocoder data block
In data sequence x1Vocoder is sent into be decoded.
Step 5, firstly, passing jth section voice signal under the microphone for passing through sender；Then, selection is with jth vocoder data block
Transmission rate be code rate vocoder jth section voice signal is encoded, obtain jth vocoder data block, sender
Jth vocoder data block is packaged according to the frame structure of setting, the jth vocoder data block after being encapsulated；Then, according to
Non- equal weight channel coding under the transmission rate of j vocoder data block, successively carries out data row to the jth vocoder data block after encapsulation
Column, LDPC coding and intertexture, obtain jth coded data to be sent；Finally, by jth coded data to be sent according to transmission signal
Structure be modulated, recipient is sent to the transmission rate of jth vocoder data block.
Step 6, recipient synchronizes detection, and demodulates to the data detected are synchronized, the jth after being demodulated
Vocoder data block；The corresponding walsh sequence of jth vocoder data block is resolved again, obtains the transmission speed of jth vocoder data block
- 1 valuation of rate and the jth of channel；Then, recipient is according to the jth-of jth vocoder data block corresponding synchronization header and channel
Channel is estimated in 1 valuation, and the signal-to-noise ratio of channel when sending jth vocoder data block is calculated；Finally, recipient will demodulate
Jth vocoder data block afterwards successively carries out balanced, deinterleaving and LDPC decoding, then LDPC decoding is decoded by vocoder
Afterwards, jth section voice signal is obtained, and jth section voice signal is uploaded to the microphone of recipient；
If recipient need to sender reply, recipient according to send jth vocoder data block when channel noise
Than the transmission rate of determining+1 vocoder data block of jth, the transmission rate of jth vocoder data block is fed back to sender by recipient；
The block number j of vocoder data block is enabled to increase by 1, return step 5；
If recipient does not need to reply to sender, acoustic code communication process terminates.
Effect of the invention can be further illustrated by following emulation experiment:
1) simulation parameter:
Being set in multipath delay is 300ms.
It sets and waits weight channel coding under 600bps transmission rate are as follows: the coding of 75ms frame exports 45 bits, by 4 75ms
The coding that frame forms 300ms superframe exports 180 bits, and 180 bits of reciprocity weight carry out 1/3 plus the CRC check position of 12 bits
LDPC coding 576 bits of output of code rate.
It sets and waits weight channel coding under 1200bps transmission rate are as follows: the coding of 50ms frame exports 60 bits, by 6
The coding that 50ms frame forms 300ms superframe exports 360 bits, and 360 bits of reciprocity weight are carried out plus the CRC check position of 24 bits
LDPC coding 1152 bits of output of 1/3 efficiency.
It sets and waits weight channel coding under 2400bps transmission rate are as follows: the coding of 25ms frame exports 60 bits, by 12
The coding that 25ms frame forms 300ms superframe exports 720 bits；720 bits of reciprocity weight are to carry out plus the CRC check of 48 bits
LDPC coding 1152 bits of output of 2/3 efficiency.
2) emulation experiment content and interpretation of result:
It compares under 600bps, 1200bps and 2400bps transmission rate respectively, waits weight channel coding and non-equal weight channel volume
The residual BER of code significant bits under identical signal-to-noise ratio.
It draws the inferior heavy channel coding of 600bps transmission rate and non-equal weighs channel coding important ratio under identical signal-to-noise ratio
Special residual BER comparison diagram, as shown in Figure 3；In Fig. 3, black dot curve indicates the inferior heavy channel of 600bps transmission rate
The residual BER of the significant bits of coding, black plus sige curve indicate the non-weight for waiting weight channel coding under 600bps transmission rate
Want the residual BER of bit.From figure 3, it can be seen that under identical residual BER, the noise of non-equal weight channel coding
Than the raising that the signal-to-noise ratio compared to equal weight channel coding has 0.5dB.
It draws the inferior heavy channel coding of 1200bps transmission rate and non-equal weighs channel coding important ratio under identical signal-to-noise ratio
Special residual BER comparison diagram, as shown in Figure 4；In Fig. 4, black dot curve indicates the inferior heavy letter of 1200bps transmission rate
The residual BER of the significant bits of road coding, black plus sige curve indicate non-under 1200bps transmission rate and wait weight channel coding
Significant bits residual BER.Figure 4, it is seen that under identical residual BER, non-equal weight channel coding
Signal-to-noise ratio has the raising of 1db or more compared to the signal-to-noise ratio of equal weight channel coding.
It draws the inferior heavy channel coding of 2400bps transmission rate and non-equal weighs channel coding important ratio under identical signal-to-noise ratio
Special residual BER comparison diagram, as shown in Figure 5；In Fig. 5, black dot curve indicates the inferior heavy letter of 2400bps transmission rate
The residual BER of the significant bits of road coding, black plus sige curve indicate non-under 2400bps transmission rate and wait weight channel coding
Significant bits residual BER.From figure 5 it can be seen that under identical residual BER, non-equal weight channel coding
Signal-to-noise ratio has the raising of 1.5dB or more compared to the signal-to-noise ratio of equal weight channel coding.
From the analysis above, we can see that the non-equal weight channel coding that the present invention designs can be obviously improved the non fouling of voice coding
Can, and improve the quality of synthesis voice.
Obviously, various changes and modifications can be made to the invention without departing from essence of the invention by those skilled in the art
Mind and range.In this way, if these modifications and changes of the present invention belongs to the range of the claims in the present invention and its equivalent technologies
Within, then the present invention is also intended to include these modifications and variations.
Claims (7)
1. a kind of based on the non-equal rate adaptation acoustic code communication means protected again, which comprises the following steps:
Step 1, the channel coding of setting acoustic code communication uses LDPC code；Then, for single times, double and four times of three kinds of transmission speed
Rate, non-equal weight channel coding under conditions of vocoder delay is fixed delay, under every kind of transmission rate of design；
Step 2, setting sends the structure of signal: it sends signal and is made of the sequence of four different function, respectively transport layer control
System/automatic growth control sequence, synchronization header sequence, walsh sequence and data sequence；
Step 3, firstly, passing paragraph 1 voice signal under the microphone for passing through sender, and the transmission speed of paragraph 1 voice signal is set
Rate is single times of transmission rate；Then, it selects to carry out paragraph 1 voice signal using single times of transmission rate as the vocoder of code rate
Coding, obtains the 1st vocoder data block, the 1st vocoder data block is packaged by sender, the 1st vocoder data after being encapsulated
Block；Then, data are successively carried out to the 1st vocoder data block after encapsulation according to the non-equal weights channel coding under single times of transmission rate
Arrangement, LDPC coding and intertexture, obtain the 1st coded data to be sent；Finally, l coded data to be sent is believed according to transmission
Number structure be modulated, recipient is sent to single times of transmission rate；
Step 4, recipient synchronizes detection, and demodulates to the data detected are synchronized, the 1st acoustic code after being demodulated
Data block；The corresponding walsh sequence of the 1st vocoder data block is resolved again, obtains the transmission rate of the 1st vocoder data block,
And the initial valuation of channel；Then, recipient is according to the initial valuation of the 1st vocoder data block corresponding synchronization header and channel
Channel is estimated, the signal-to-noise ratio of channel when sending the 1st vocoder data block is calculated；Finally, recipient will be after demodulation
1st vocoder data block successively carries out balanced, deinterleaving and LDPC decoding, then LDPC decoding is obtained after vocoder decodes
To paragraph 1 voice signal, and paragraph 1 voice signal is uploaded to the microphone of recipient；
If recipient needs to reply to sender, recipient is true according to the signal-to-noise ratio of channel when sending the 1st vocoder data block
The transmission rate of 2nd vocoder data block is fed back to sender by the transmission rate of fixed 2nd vocoder data block, recipient；Set code
The block number j of data block is 2, goes to step 5；
If recipient does not need to reply to sender, acoustic code communication process terminates；
Step 5, firstly, passing jth section voice signal under the microphone for passing through sender；Then, selection is with the biography of jth vocoder data block
Defeated rate is that the vocoder of code rate encodes jth section voice signal, obtains jth vocoder data block, and sender is by jth
Vocoder data block is packaged according to the frame structure of setting, the jth vocoder data block after being encapsulated；Then, according to jth acoustic code
Non- equal weight channel coding under the transmission rate of data block, to the jth vocoder data block after encapsulation successively carry out data arrangement,
LDPC coding and intertexture, obtain jth coded data to be sent；Finally, by jth coded data to be sent according to the knot for sending signal
Structure is modulated, and is sent to recipient with the transmission rate of jth vocoder data block；
Step 6, recipient synchronizes detection, and demodulates to the data detected are synchronized, the jth acoustic code after being demodulated
Data block；The corresponding walsh sequence of jth vocoder data block is resolved again, obtains the transmission rate of jth vocoder data block,
And -1 valuation of jth of channel；Then, recipient estimates according to the jth -1 of the corresponding synchronization header of jth vocoder data block and channel
Value estimates channel, and the signal-to-noise ratio of channel when sending jth vocoder data block is calculated；Finally, after recipient will demodulate
Jth vocoder data block successively carry out it is balanced, deinterleave and LDPC decoding, then by LDPC decoding after vocoder decodes,
Jth section voice signal is obtained, and jth section voice signal is uploaded to the microphone of recipient；
If recipient needs to reply to sender, recipient is true according to the signal-to-noise ratio of channel when sending jth vocoder data block
Determine the transmission rate of+1 vocoder data block of jth, the transmission rate of jth vocoder data block is fed back to sender by recipient；Enable sound
The block number j of code data block increases by 1, return step 5；
If recipient does not need to reply to sender, acoustic code communication process terminates.
2. as described in claim 1 a kind of based on the non-rate adaptation acoustic code communication means for waiting and protecting again, which is characterized in that
In step 1, the list times, double and four times of three kinds of transmission rates are respectively 600bps, 1200bps and 2400bps, the fixation
Time delay is 300ms.
3. as described in claim 1 a kind of based on the non-rate adaptation acoustic code communication means for waiting and protecting again, which is characterized in that
In step 2, the load that the walsh sequence carries is 4bit, wherein preceding 2bit indicates transmission rate, and rear 2bit indicates channel
Valuation；
The data sequence indicates vocoder data block, and every piece of vocoder data block includes n+1 data packet, is followed successively by data packet
EDataPKT0 to data packet EDataPKTn, wherein data packet EDataPKTi is encapsulated by 3 frame data, and every frame data wrap
Unknown data and given data are included, unknown data is the vocoder data to be transmitted, it is known that data are training sequence, and i takes 0 to n, n
It is natural number.
4. as described in claim 1 a kind of based on the non-rate adaptation acoustic code communication means for waiting and protecting again, which is characterized in that
In step 3, the 1st vocoder data block is packaged according to the frame structure of setting, the 1st vocoder data block after being encapsulated,
In, the frame structure of the setting are as follows:
To single times of transmission rates, 576 bits are exported after non-equal recodification, form 576 BPSK symbols by Gray code,
It splits into 3 frames to be sent, 192 BPSK symbols of every frame, the training sequence being inserted into thereafter is 48 BPSK symbols, is known sequence
Column, then 600bps transmission rate a data segment, length is 720 BPSK symbols；
For double transmission rate, 1152 bits are exported after non-equal recodification, form 576 QPSK symbols by Gray code
Number, it splits into 3 frames and is sent, 192 QPSK symbols of every frame, the training sequence being inserted into thereafter is 48 QPSK symbols, for
Know sequence, then 1200bps transmission rate a data segment, length is 720 QPSK symbols；
For four times of transmission rates, 1152 bits are exported after non-equal recodification, form 576 QPSK symbols by Gray code
Number, it splits into 3 frames and is sent, 192 QPSK symbols of every frame, the training sequence being inserted into thereafter is 48 QPSK symbols, for
Know sequence, then 2400bps transmission rate a data segment, length is 720 QPSK symbols.
5. as claimed in claim 2 a kind of based on the non-rate adaptation acoustic code communication means for waiting and protecting again, which is characterized in that
In step 4, the transmission rate of the 2nd vocoder data block is determined according to the signal-to-noise ratio of channel when sending the 1st vocoder data block, it is specific
Process are as follows:
The list times, double and four times of three kinds of transmission rates are respectively 600bps, 1200bps and 2400bps, it is known that 600bps,
The optimum signal-noise ratio range of 1200bps and 2400bps transmission rate is respectively as follows: as SNR < 2dB, and channel can guarantee 600bps
Reliable transmission；As 2dB≤SNR < 6dB, channel can guarantee 1200bps reliable transmission；As SNR >=6dB, channel can be protected
Demonstrate,prove 2400bps reliable transmission；Transmission rate corresponding to the signal-to-noise ratio of channel when sending the 1st vocoder data block is determined as the 2nd
The transmission rate of vocoder data block.
6. as described in claim 1 a kind of based on the non-rate adaptation acoustic code communication means for waiting and protecting again, which is characterized in that
In step 4, the 1st vocoder data block by after demodulation successively carries out balanced, deinterleaving and LDPC decoding, detailed process
Are as follows:
(b1) setting i indicates the number of iterations, enables initial value i=1；
(b2) the Turbo equalization algorithm based on least mean-square error is used, calculates the by the 1st vocoder data block after demodulation
Data sequence x in 1 vocoder data block1I-th equilibrium estimated valueAnd it is deinterleaved；
(b3) by the i-th equilibrium estimated value after deinterleaving after LDPC is decoded, i-th LDPC decoding estimated value is obtained
(b4) if ldpc decoder determines that i-th LDPC decodes estimated valueIt is incorrect, by the number in the 1st vocoder data block
According to sequence x1It is updated toAnd the number of iterations i is enabled to increase by 1, return step (b2)；
If ldpc decoder determines that i-th LDPC decodes estimated valueIt is setting for correct or the number of iterations i=I, I
Maximum equalization iteration number, then enable the data sequence in the 1st vocoder data blockIt will be in the 1st vocoder data block
Data sequence x1Vocoder is sent into be decoded.
7. as claimed in claim 2 a kind of based on the non-rate adaptation acoustic code communication means for waiting and protecting again, which is characterized in that
In step 1,
Non- equal weight channel coding under the single times of 600bps transmission rate are as follows: for the vocoder of 600bps code rate,
The coding of 75ms frame exports 45 bits, exports 180 bits by the coding that 4 75ms frames form 300ms superframe；To 300ms superframe
In important 156 bits LDPC coding 546 bits of output of 4/13 code rate are carried out plus the CRC check position of 12 bits, to 300ms
Remaining 24 bit is directly exported plus the CRC check position of 6 bits without channel coding in superframe, and each 300ms superframe is total
Export 576 bits；
Non- equal weight channel coding under the double 1200bps transmission rate are as follows: for the vocoder of 1200bps code rate,
The coding of 50ms frame exports 60 bits, exports 360 bits by the coding that 6 50ms frames form 300ms superframe；To 300ms superframe
In important 264 bits carry out LDPC coding 1056 bits of output of 1/4 code rate, not to 96 bit remaining in 300ms superframe
It carries out channel coding directly to export, each 300ms superframe exports 1152 bits altogether；
Non- equal weight channel coding under four times of 2400bps transmission rates are as follows: for the vocoder of 2400bps code rate,
The coding of 25ms frame exports 60 bits, exports 720 bits by the coding that 12 25ms frames form 300ms superframe；To 300ms superframe
In important 384 bits LDPC coding 816 bits of output of 1/2 code rate are carried out plus the CRC check position of 24 bits, to 300ms
Remaining 336 bit is directly exported without channel coding in superframe, and each 300ms superframe exports 1152 bits altogether.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510823888.3A CN105513599B (en) | 2015-11-24 | 2015-11-24 | A kind of rate adaptation acoustic code communication means protected again based on non-grade |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510823888.3A CN105513599B (en) | 2015-11-24 | 2015-11-24 | A kind of rate adaptation acoustic code communication means protected again based on non-grade |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN105513599A CN105513599A (en) | 2016-04-20 |
| CN105513599B true CN105513599B (en) | 2019-05-21 |
Family
ID=55721526
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510823888.3A Active CN105513599B (en) | 2015-11-24 | 2015-11-24 | A kind of rate adaptation acoustic code communication means protected again based on non-grade |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105513599B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114039705B (en) * | 2022-01-10 | 2022-03-22 | 中国电子科技集团公司第五十四研究所 | Method for transmitting low-speed voice data by using E1 line |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP1028555A2 (en) * | 1999-02-10 | 2000-08-16 | Nec Corporation | Error detection encoder with unequal error protection |
| KR20000076730A (en) * | 1999-02-26 | 2000-12-26 | 루센트 테크놀러지스 인크 | Coupled error code protection for multi-mode vocoders |
| US6240386B1 (en) * | 1998-08-24 | 2001-05-29 | Conexant Systems, Inc. | Speech codec employing noise classification for noise compensation |
| CN1870132A (en) * | 2006-06-26 | 2006-11-29 | 南京邮电大学 | Didigal audio waterprint method based on low density parity check code |
| CN101086844A (en) * | 2007-07-19 | 2007-12-12 | 清华大学 | Voice coding transmission method for resisting bad channel and packet loss and accompanied error code |
| CN203242333U (en) * | 2013-03-08 | 2013-10-16 | 中国计量学院 | LDPC code-based anti-interference voice signal transmission system |
-
2015
- 2015-11-24 CN CN201510823888.3A patent/CN105513599B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6240386B1 (en) * | 1998-08-24 | 2001-05-29 | Conexant Systems, Inc. | Speech codec employing noise classification for noise compensation |
| EP1028555A2 (en) * | 1999-02-10 | 2000-08-16 | Nec Corporation | Error detection encoder with unequal error protection |
| KR20000076730A (en) * | 1999-02-26 | 2000-12-26 | 루센트 테크놀러지스 인크 | Coupled error code protection for multi-mode vocoders |
| CN1870132A (en) * | 2006-06-26 | 2006-11-29 | 南京邮电大学 | Didigal audio waterprint method based on low density parity check code |
| CN101086844A (en) * | 2007-07-19 | 2007-12-12 | 清华大学 | Voice coding transmission method for resisting bad channel and packet loss and accompanied error code |
| CN203242333U (en) * | 2013-03-08 | 2013-10-16 | 中国计量学院 | LDPC code-based anti-interference voice signal transmission system |
Also Published As
| Publication number | Publication date |
|---|---|
| CN105513599A (en) | 2016-04-20 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US7813453B2 (en) | Data detection for a hierarchical coded data transmission | |
| US11901929B2 (en) | Communication system and communication method | |
| CN111510199B (en) | Code modulation method and system in satellite mobile communication | |
| Lau et al. | Variable rate adaptive trellis coded QAM for high bandwidth efficiency applications in Rayleigh fading channels | |
| CN106130687B (en) | Polar code puncturing method based on decoding bit reliability under fading channel | |
| EP1489774A2 (en) | System and method to determinate a bit error probability of received data within a cellular wireless network | |
| CN112713966B (en) | Code modulation switching method based on likelihood estimation correction signal-to-noise ratio | |
| CA2609850A1 (en) | Transmitting apparatus, receiving apparatus and spatial multiplex number controlling method | |
| JP2003501859A (en) | Optimal turbo decoding architecture and method using constant or quasi-constant signal-to-noise ratio | |
| CN107528671A (en) | A kind of System Frame Number detection method for arrowband Internet of Things NB IoT | |
| JP2014123956A (en) | Methods, apparatus, and articles of manufacture for adaptive power control | |
| CN102739322A (en) | Remote underwater acoustic communication method based on soft-demodulation soft-decoding joint iteration | |
| CN105049388A (en) | Short-wave communication system based on fountain codes and single carrier frequency domain equalization and transmission method thereof | |
| CN104486754B (en) | Cooperation communication system relay selection method based on safety of physical layer technology | |
| CN101848174B (en) | Transmission preprocessing method and data demodulation method of transmission preprocessing signals | |
| CN110519191A (en) | A kind of high spectrum effect single carrier communication method of time-frequency two-dimensional compression | |
| CN106533620A (en) | Log-likelihood-ratio-threshold-based adaptive demodulation method for high-order QAM modulation symbol | |
| CN105513599B (en) | A kind of rate adaptation acoustic code communication means protected again based on non-grade | |
| CN103929386B (en) | A kind of air interference removing method | |
| CN104284375A (en) | Wireless network rate adaptive adjusting method | |
| CN107979453B (en) | Weighting soft decoding method based on single-carrier frequency division multiple access access system | |
| CN109873686A (en) | A kind of 5G broadcast channel merging method of reseptance based on polarization code | |
| CN104135455B (en) | Iterative receiving method for communication system | |
| CN106559366A (en) | Multipath fading signal diversity based on multidiameter fading channel merges method of reseptance | |
| AU657031B2 (en) | Voice signal communication with burst error reduction |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |