CN102522089B - Information embedding and extracting method for G.723.1 voice coder - Google Patents
Information embedding and extracting method for G.723.1 voice coder Download PDFInfo
- Publication number
- CN102522089B CN102522089B CN201110396225XA CN201110396225A CN102522089B CN 102522089 B CN102522089 B CN 102522089B CN 201110396225X A CN201110396225X A CN 201110396225XA CN 201110396225 A CN201110396225 A CN 201110396225A CN 102522089 B CN102522089 B CN 102522089B
- Authority
- CN
- China
- Prior art keywords
- bit
- parameter
- hiding
- embedding
- hiding parameter
- 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
Images
Abstract
The invention relates to an information embedding method for a G.723.1 voice coder, which comprises the following steps: obtaining a voice signal, carrying out G.723.1 coding on the voice signal so as to generate a voice frame, obtaining the s hidden parameters and the hidden bits/blocks thereof of the voice frame and the embedding capacity, the embedding operation and the multi-scale form of the hidden parameters, setting a counter k equal to 1, obtaining a secret information bit according to the embedding capacity of the k hidden parameter, judging whether the embedding operation of the k hidden parameter is in least significant bit replacement or not, carrying out multi-scale conversion on the secret information bit and the hidden parameter according to the multi-scale form if the embedding operation is not in the least significant bit replacement and carrying out the least significant bit replacement on the multi-scale of the hidden parameter according to secret information after multi-scale conversion. According to the information embedding method for the G.723.1 voice coder, the adjacent states of bits with strong noise resistance in frame parameters are fully utilized, so that the modification of embedding is centralized at the bits with smaller influence. The information embedding method for the G.723.1 voice coder is beneficial to distortion decreasing and is suitable for transmitting secret information under a voice communication system.
Description
Technical field
The invention belongs to field of communication security, particularly a kind of based on the information concealing method of speech coder G.723.1, be applicable to the covert communications take the low-bit-rate speech coding of parameter coding as carrier.
Background technology
Along with infotech extensively popularizing of the communication technology particularly, information maintain secrecy or safety transmits and also more and more is subject to popular attention, the Secure Application that combines with the various communication technologys also seems extremely more urgent.The transmission of important information all was encrypted processing, it is disorderly and unsystematic that the information after the encryption becomes in the past, yet but can't cover the fact of its " existence ", more attracts on the contrary saboteur's attention.Information Hiding Techniques provides a more hidden close information to preserve or load mode comparatively speaking, the secret information that it will need to transmit is hidden under the common carrier data, from having hidden in essence " existence of secret information " this fact, attack for secret information is become have no way of doing it.Although what Information hiding can not be as encryption technology is fully alternative, it provides secret information to preserve and another the important selection that transmits, and has a wide range of applications in the application such as hidden close communication, digital watermarking.
The application of current Information hiding still concentrates in the static carriers such as image, audio frequency, and for the Information Hiding Techniques of the voice flow carrier that is widely used in the communications field also seldom.In the voice communication Speech Coding at Low Bit Rates of commonplace use have ITU-T G.723.1, G.729, iLBC etc., the speech encoding rate of their common feature is low, redundant few, add the characteristic of the instantaneity of transmission of flow media data, thereby brought very large difficulty to Information hiding.Information Hiding Techniques for low-bit-rate speech coding mainly contains the methods such as quantization index modulation technology, quiet concealing technology and least significant bit (LSB) (Least Significant Bit is called for short LSB) replacement at present.Because the characteristics such as LSB replacement method algorithm is simple, applicability is strong and capacity is large are the most practical and effective mode under the low-bit-rate speech coding carrier at present.
In image-carrier, the LSB replacement method has been owing to having revised its statistical property, thereby produced a lot of mutation.Such as LSB coupling scheduling algorithm, be characterized in the pixel in the image is replaced simple replace Algorithm with+1 and-1 operation, can increase like this it can not apperceive characteristic.This method can be used in the low-bit-rate speech coding carrier too.
In Speech Coding at Low Bit Rates, the be encoded form of framing of voice, typically G.723.1 two-forty speech coder frame length is 192 bits.The Speech property testing is carried out in each position in the frame, and set certain threshold value, can obtain some least significant bits.Thereby with the typical difference of image-carrier be that it is low that the distortion that each least significant bit (LSB) of speech frame brings has height to have, and incomplete same, can only be according to selecting the least significant bit (LSB) of some to embed to the restriction of distortion.And general LSB replacement method adopts is the form of single binary digit conversion, and binary digit is independently revised.Its shortcoming is that the large and voice concealed space of distortion does not obtain fully effectively utilizing.
Summary of the invention
The present invention proposes a kind of information concealing method of the multi-system under speech coder G.723.1, under the prerequisite of introducing a small amount of distortion, improves the capacity of Information hiding, takes full advantage of simultaneously the embedding position of speech frame, is conducive to increase the security of Information hiding.
The present invention is achieved by the following technical solutions:
A kind of for the information embedding method of speech coder G.723.1, may further comprise the steps: obtain voice signal, G.723.1, voice signal is encoded, to generate speech frame, obtain the individual hiding parameter of s and the hidden bit/piece thereof of speech frame, hide the embedding capacity of parameter, embedding operation and multi-system form, set counter k=1, embedding capacity according to k hiding parameter obtains the secret information bit, whether the embedding operation of judging k hiding parameter is that least significant bit (LSB) is replaced, if not being least significant bit (LSB), embedded operation do not replace, then according to the multi-system form secret information bit and hiding parameter are carried out the multi-system conversion, according to the secret information after the multi-system conversion multi-system of hiding parameter being carried out least significant bit (LSB) replaces, hiding parameter after replacing is converted into scale-of-two, with the former hiding parameter in the hiding parameter replacement speech frame after transforming, judges that whether counter k is greater than s, if counter k greater than s, then obtains comprising the close speech frame that carries of secret information bit.
Information embedding method of the present invention also comprises step: replace if embedding operation is least significant bit (LSB), then replace the hidden bit of k parameter with the secret information bit; And forward step with the former hiding parameter in the hiding parameter replacement speech frame after transforming to, if counter k is not more than s, k=k+1 is set then, and returns the step of obtaining the secret information bit according to the embedding capacity of k hiding parameter.
Obtain speech frame s hiding parameter and hidden bit/piece thereof, hide the embedding capacity of parameter, the step of embedding operation and multi-system form comprises following substep: the scale-of-two upset is carried out in the position that the permission of speech frame is revised, speech frame after utilizing the subjective speech quality assessment algorithm to upset is tested, to obtain one group of subjective speech quality assessment value, choose in the subjective voice assessed value the corresponding position greater than threshold value, preserve the index set P of corresponding position in speech frame, and index set P affiliated frame parameter in speech frame is gathered Q, and the number of parameters of setting among the frame parameter set Q is s, set counter i=1, judge the bit number N of i hiding parameter
iWhether equal 1, if bit number N
iBe not equal to 1, then set counter j=1, whether index corresponding to j bit of judging i hiding parameter belongs to index set P, if j index corresponding to bit belongs to index set P, whether index corresponding to j+1 bit of then judging i hiding parameter belongs to index set P, if j+1 index corresponding to bit belongs to index set P, then under the decimal system, i hiding parameter added respectively 2
J-1With subtract 2
J-1Computing, with the raw parameter in the i after the computing the hiding parameter replacement speech frame, and with the subjective speech quality assessment algorithm speech frame after replacing is tested, to obtain one group of subjective speech quality assessment value of j bit in i the hiding parameter, j=j+1 is set, judges that whether j is greater than N
iIf j is greater than N
i, obtain then that the adjacent and corresponding index in position belongs to m the bit of index set P in i the hiding parameter, and the embedding operation that will hide parameter is labeled as the 2m+1 system and embeds, embed the embedding capacity of determining the individual hiding parameter of i according to the 2m+1 system and be
Whether bit judges i less than s, if i is not less than s, then preserves s hiding parameter and hidden bit/piece thereof, embedding capacity, embedding operation and the multi-system form of s hiding parameter.
Obtain speech frame s hiding parameter and hidden bit/piece thereof, the step of hiding embedding capacity, embedding operation and the multi-system form of parameter also comprises following substep: if bit number N
iEqual 1, then the embedding operation with i hiding parameter is labeled as the least significant bit (LSB) replacement, its embedding capacity is 1 bit, then enter and judge that i is whether less than the step of s, if j index corresponding to bit do not belong to index set P, then enter the step that j=j+1 is set, if j+1 index corresponding to bit belongs to index set P, then under the decimal system, i parameter carried out+2 respectively
J-1,-2
J-1,+2
j,-2
jComputing, then enter with the raw parameter in the i after the computing the hiding parameter replacement speech frame, and with the subjective speech quality assessment algorithm speech frame after replacing is tested, to obtain the step of one group of subjective speech quality assessment value of j bit in i the hiding parameter, if j is not more than N
i, then return the step whether index corresponding to j bit of judging i hiding parameter belongs to index set P, if i less than s, then arranges i=i+1, and return the bit number N that judges i hiding parameter
iWhether equal 1 step.
A kind of for the information extracting method of speech coder G.723.1, may further comprise the steps: obtain and carry close speech frame, obtain the individual hiding parameter of the s that carries close speech frame and hidden bit/piece, embedding operation and multi-system form, extract and carry s the hiding parameter that contains secret information in the close speech frame, set counter l=1, whether the embedding operation of judging the 1st parameter is that least significant bit (LSB) is replaced, if not being least significant bit (LSB), do not replace embedding operation, then the 1st parameter is converted into the multi-system form of parameter, extract the least significant bit (LSB) under the multi-system form of the 1st parameter, the least significant bit (LSB) that extracts is converted into binary mode, it is the secret information bit, whether judge counter l greater than s, if all secret information bits that counter l greater than s, then obtains in the frame to be hidden.
Breath extracting method of the present invention also comprises step: replace if embedding operation is least significant bit (LSB), the least significant bit (LSB) of extracting parameter then, it is the secret information bit, and forward to and judge that counter l is whether greater than the step of s, if counter l is not more than s, l=l+1 then is set, and returns whether the embedding operation of judging the 1st parameter is the step that least significant bit (LSB) is replaced.
Different from the embedding grammar of least significant bit (LSB) under the traditional G.723.1 scrambler, information embedding method of the present invention fully takes into account all adjacent states of lowest order, and the coding method of employing multi-system, make these state sets can effectively apply to the embedding leaching process.Information concealing method of the present invention utilizes the inconsistent characteristic of each bit distortion of speech frame, takes full advantage of the adjacent states of the least significant bit (LSB) with higher subjective speech quality assessment value, and the bit of low distortion is fully used.Generally speaking, the present invention has the following advantages:
1. in the situation that the identical hiding capacity of increase is introduced less distortion;
2. be similar to cryptographic Kerckhoffs criterion, Information hiding should not depend on maintaining secrecy of algorithm, and should take the form of key, the introducing of multi-system algorithm, so that available hiding figure place has obtained increase, thereby enlarged the key space based on the Information hiding of key, security performance has obtained enhancing;
3. multi-system embedding grammar of the present invention has utilized the characteristic of the least significant bit (LSB) of low-bit-rate speech coding frame, and is independent mutually with concrete encryption algorithm, thereby is applicable to other similar speech coding algorithm.
Description of drawings
Fig. 1 is that the present invention is used for the G.723.1 process flow diagram of the information embedding method of speech coder.
Fig. 2 is the refinement process flow diagram of step in the information embedding method of the present invention (3).
Fig. 3 is that the present invention is used for the G.723.1 process flow diagram of the information extracting method of speech coder.
Fig. 4 is the refinement process flow diagram of step in the information extracting method of the present invention (2).
Embodiment
Below technical terms more of the present invention are defined and explain:
Speech frame: the unit of transfer after voice are encoded through scrambler.
Frame parameter: the parameter that speech frame comprises specifically comprises linear predictive coding (Linear Predictive Coding is called for short LPC) index, self-adapting code book hysteresis, code book gain, pulse position, impulse code and mesh index.Wherein except the LPC index, other parameter all comprises four parts, corresponds respectively to four subframes that frame comprises.
Parameter bit: the binary bits of component frame parameter.
Hide parameter: be used for the parameter of the speech frame of Information hiding, specifically comprise the code book gain of LPC index, four subframes, the pulse position of four subframes, the impulse code of four subframes and the mesh index of four subframes.
Hidden bit: be used for the bit that the parameter of the speech frame of Information hiding comprises.
Hide piece: hide position continuous in the parameter and form a hiding piece.
Embedding capacity: the bit number that allows embedding.
Embedding operation: the arithmetic plus-minus or the logic xor operation that carry out when embedding secret.
The multi-system form: the expansion of digital system, scale-of-two, the decimal system etc. all belongs to a kind of of multi-system.
Secret information bit: the binary digit that forms secret information.
Least significant bit (LSB) is replaced: the Information hiding embedding grammar of replacing the least significant bit (LSB) of carrier with the secret information bit.
Carry close speech frame: embedded the speech frame behind the secret information.
Subjective speech quality assessment algorithm: Perceptual Evaluation of Speech Quality, be called for short PESQ, it is a kind of testing algorithm to voice, obtain a subjective speech quality assessment value after the test, estimate the quality of voice quality according to the size of value, general value is larger, and voice quality is better.
Subjective speech quality assessment value: carry out the value that obtains behind the subjective speech quality assessment algorithm.
As shown in Figure 1, G.723.1 the present invention is used for that the information embedding method of speech coder may further comprise the steps:
(1) obtains voice signal;
(2) G.723.1 voice signal is encoded, generate speech frame;
(3) obtain s hiding parameter and the hidden bit/piece thereof of speech frame, hide embedding capacity, embedding operation and the multi-system form of parameter, specifically comprise following substep:
(4) set counter k=1;
(5) embedding capacity according to k hiding parameter obtains the secret information bit;
Whether the embedding operation of (6) judging k hiding parameter is that least significant bit (LSB) is replaced, and then changes step (7) over to if least significant bit (LSB) is replaced, otherwise changes step (8) over to;
(7) replace the hidden bit of k hiding parameter with the secret information bit, forward step (11) to;
(8) according to the multi-system form secret information bit and hiding parameter are carried out the multi-system conversion;
(9) according to the secret information after the multi-system conversion multi-system of hiding parameter being carried out least significant bit (LSB) replaces;
(10) the hiding parameter after will replacing is converted into scale-of-two;
(11) with the former hiding parameter in the hiding parameter replacement speech frame after transforming;
(12) whether judge counter k greater than s, if counter k, then changes step (14) over to greater than s, otherwise change step (13) over to;
(13) k=k+1 is set, and returns step (5);
(14) obtain comprising the close speech frame that carries of secret information bit.
As shown in Figure 2, the step (3) in the information embedding method of the present invention specifically comprises following substep:
The scale-of-two upset is carried out in the position of (3-1) permission of speech frame being revised;
(3-2) utilize the subjective speech quality assessment algorithm that the speech frame after overturning is tested, to obtain one group of subjective speech quality assessment value;
(3-3) choose in the subjective voice assessed value corresponding position greater than threshold value, preserve the index set P of corresponding position in speech frame, and index set P affiliated frame parameter in speech frame is gathered Q, and the number of parameters of setting among the frame parameter set Q is s, threshold value can be set according to different embedding demands, in the present embodiment, the threshold value of setting is 3.4;
(3-4) set counter i=1;
(3-5) the bit number N of i hiding parameter of judgement
iWhether equal 1, if bit number N
iEqual 1, then change step (3-6) over to, otherwise change step (3-7) over to;
(3-6) embedding operation with i hiding parameter is labeled as the least significant bit (LSB) replacement, and its embedding capacity is 1 bit, changes step (3-17) over to
(3-7) set counter j=1;
Whether index corresponding to j bit of (3-8) judging i hiding parameter belongs to index set P, gathers P if j index corresponding to bit belongs to index, then changes step (3-9) over to, otherwise changes step (3-13) over to;
Whether index corresponding to j+1 bit of (3-9) judging i hiding parameter belongs to index set P, gathers P if j+1 index corresponding to bit belongs to index, then changes step (3-10) over to, otherwise changes step (3-11) over to;
(3-10) under the decimal system, i hiding parameter added respectively 2
J-1With subtract 2
J-1Computing changes step (3-12) over to;
(3-11) under the decimal system, i parameter carried out+2 respectively
J-1,-2
J-1,+2
j,-2
jComputing
(3-12) with the raw parameter in the i after the computing the hiding parameter replacement speech frame, and with the subjective speech quality assessment algorithm speech frame after replacing is tested, to obtain one group of subjective speech quality assessment value of j bit in i the hiding parameter;
(3-13) j=j+1 is set;
(3-14) judge that whether j is greater than N
iIf j is greater than N
i, then change step (3-15) over to, otherwise change step (3-8) over to;
(3-15) obtain that the adjacent and corresponding index in position belongs to m the bit of index set P in i the hiding parameter, and the embedding operation that will hide parameter is labeled as the embedding of 2m+1 system;
(3-16) embedding the embedding capacity of determining i hiding parameter according to the 2m+1 system is
(symbol
Expression rounds downwards) bit;
(3-17) whether judge i less than s, if i, then changes step (3-18) over to less than s, otherwise change step (3-19) over to;
(3-18) i=i+1 is set, changes step (3-5) over to;
(3-19) preserve s hiding parameter and hidden bit/piece thereof, embedding capacity, embedding operation and the multi-system form of s hiding parameter.
As shown in Figure 3, G.723.1 the present invention is used for that the information extracting method of speech coder may further comprise the steps:
(1) obtains a year close speech frame;
(2) obtain the individual hiding parameter of the s that carries close speech frame and hidden bit/piece, embedding operation and multi-system form;
(3) extract s the hiding parameter that contains secret information in year close speech frame;
(4) set counter l=1;
Whether the embedding operation of (5) judging the 1st parameter is that least significant bit (LSB) is replaced, and replaces if embedding operation is least significant bit (LSB), then changes step (6) over to, otherwise changes step (7) over to;
(6) least significant bit (LSB) of extracting parameter, namely the secret information bit changes step (10) over to;
(7) the 1st parameter is converted into the multi-system form of parameter;
(8) least significant bit (LSB) under the multi-system form of the 1st parameter of extraction;
(9) least significant bit (LSB) that extracts is converted into binary mode, i.e. the secret information bit;
(10) whether judge counter l greater than s, if greater than s, then change step (12) over to, otherwise change step (11) over to;
(11) l=l+1 is set, changes step (5) over to;
(12) all secret information bits that obtain in the frame being hidden.
As shown in Figure 4, the step (2) in the information extracting method of the present invention specifically comprises following substep:
The scale-of-two upset is carried out in the position of (2-1) permission of carrying close speech frame being revised;
(2-2) utilize the subjective speech quality assessment algorithm that the speech frame after overturning is tested, to obtain one group of subjective speech quality assessment value;
(2-3) choose in the subjective voice assessed value corresponding position greater than threshold value, preserve the index set P of corresponding position in speech frame, and index set P affiliated frame parameter in speech frame is gathered Q, and the number of parameters of setting among the frame parameter set Q is s, threshold value can be set according to different embedding demands, in the present embodiment, the threshold value of setting is 3.4;
(2-4) set counter i=1;
(2-5) the bit number N of i hiding parameter of judgement
iWhether equal 1, if bit number N
iEqual 1, then change step (2-6) over to, otherwise change step (2-7) over to;
(2-6) embedding operation with i hiding parameter is labeled as the least significant bit (LSB) replacement, and its embedding capacity is 1 bit, changes step (2-17) over to;
(2-7) set counter j=1;
Whether index corresponding to j bit of (2-8) judging i hiding parameter belongs to index set P, gathers P if j index corresponding to bit belongs to index, then changes step (2-9) over to, otherwise changes step (2-13) over to;
Whether index corresponding to j+1 bit of (2-9) judging i hiding parameter belongs to index set P, gathers P if j+1 index corresponding to bit belongs to index, then changes step (2-10) over to, otherwise changes step (2-11) over to;
(2-10) under the decimal system, i hiding parameter added respectively 2
J-1With subtract 2
J-1Computing changes step (2-12) over to;
(2-11) under the decimal system, i parameter carried out+2 respectively
J-1,-2
J-1,+2
j,-2
jComputing;
(2-12) with the raw parameter in the i after the computing the hiding parameter replacement speech frame, and with the subjective speech quality assessment algorithm speech frame after replacing is tested, to obtain one group of subjective speech quality assessment value of j bit in i the hiding parameter;
(2-13) j=j+1 is set;
(2-14) judge that whether j is greater than N
iIf j is greater than N
i, then change step (2-15) over to, otherwise return step (2-8);
(2-15) obtain that the adjacent and corresponding index in position belongs to m the bit of index set P in i the hiding parameter, and the embedding operation that will hide parameter is labeled as the embedding of 2m+1 system;
(2-16) embedding the embedding capacity of determining i hiding parameter according to the 2m+1 system is
(symbol
Expression rounds downwards) bit;
(2-17) whether judge i less than s, if i, then changes step (2-18) over to less than s, otherwise change step (2-19) over to;
(2-18) i=i+1 is set, and returns step (2-5);
(2-19) preserve s hiding parameter and hidden bit/piece thereof, embedding capacity, embedding operation and the multi-system form of s hiding parameter.
Claims (6)
1. one kind is used for the G.723.1 information embedding method of speech coder, may further comprise the steps:
Obtain voice signal;
G.723.1, described voice signal is encoded, to generate speech frame;
Obtain the individual hiding parameter of s and the hidden bit/piece thereof of described speech frame, the embedding capacity of described hiding parameter, embedding operation and multi-system form comprise following substep:
The scale-of-two upset is carried out in the position that the permission of described speech frame is revised;
Speech frame after utilizing the subjective speech quality assessment algorithm to upset is tested, to obtain one group of subjective speech quality assessment value;
Choose in the described subjective speech quality assessment value the corresponding position greater than threshold value, preserve the index set P of described corresponding position in described speech frame, and the frame parameter set Q of described index set P under in described speech frame, and the number of parameters of setting among the described frame parameter set Q is s;
Set counter i=1;
Judge the bit number N of i hiding parameter
iWhether equal 1;
If described bit number N
iBe not equal to 1, then set counter j=1;
Whether index corresponding to j bit of judging described i hiding parameter belongs to described index set P;
If described j index corresponding to bit belongs to described index set P, judge then whether the index corresponding to j+1 bit of described i hiding parameter belongs to described index set P;
If described j+1 index corresponding to bit belongs to described index set P, then under the decimal system, described i hiding parameter added respectively 2
J-1With subtract 2
J-1Computing;
With the raw parameter in the described speech frame of hiding parameter replacement of the i after the computing, and with the subjective speech quality assessment algorithm speech frame after replacing is tested, to obtain one group of subjective speech quality assessment value of j bit in the individual hiding parameter of described i; J=j+1 is set;
Judge that whether j is greater than N
i
Adjacent and the corresponding index in position belongs to m the bit that P gathered in described index in described i the hiding parameter if j greater than Ni, then obtains, and the embedding operation of described hiding parameter is labeled as the embedding of 2m+1 system;
Embedding the embedding capacity of determining described i hiding parameter according to described 2m+1 system is
Bit, symbol
Expression rounds downwards;
Judge that whether i is less than s;
If i is not less than s, then preserve described s hiding parameter and hidden bit/piece thereof, embedding capacity, embedding operation and the multi-system form of described s hiding parameter;
Set counter k=1;
Embedding capacity according to k hiding parameter obtains the secret information bit;
Whether the embedding operation of judging described k hiding parameter is that least significant bit (LSB) is replaced;
Replace if described embedding operation is not least significant bit (LSB), then according to described multi-system form described secret information bit and described hiding parameter are carried out the multi-system conversion;
According to the secret information after the multi-system conversion multi-system of described hiding parameter being carried out least significant bit (LSB) replaces;
Hiding parameter after replacing is converted into scale-of-two;
With the former hiding parameter in the described speech frame of hiding parameter replacement after transforming;
Judge that whether described counter k is greater than s;
If described counter k greater than s, then obtains comprising the close speech frame that carries of described secret information bit.
2. information embedding method according to claim 1 is characterized in that:
Replace if described embedding operation is least significant bit (LSB), then replace the hidden bit of described k hiding parameter with described secret information bit; And forward step with the former hiding parameter in the described speech frame of hiding parameter replacement after transforming to;
If described counter k is not more than s, k=k+1 then is set, and returns the step that described embedding capacity according to k hiding parameter obtains the secret information bit.
3. information embedding method according to claim 1 is characterized in that, described described speech frame s hiding parameter and the hidden bit/piece thereof of obtaining, and the step of the embedding capacity of described hiding parameter, embedding operation and multi-system form also comprises following substep:
If described bit number N
iEqual 1, then the embedding operation of described i hiding parameter is labeled as least significant bit (LSB) and replaces, its embedding capacity is 1 bit, then enters the described i of judgement whether less than the step of s;
If index corresponding to described j bit do not belong to described index set P, the step that then enters the described j=j+1 of setting;
If described j+1 index corresponding to bit belongs to described index set P, then under the decimal system, described i parameter carried out+2 respectively
J-1,-2
J-1,+2
j,-2
jComputing, then enter described with the raw parameter in the i after the computing the described speech frame of hiding parameter replacement, and with the subjective speech quality assessment algorithm speech frame after replacing is tested, to obtain the step of one group of subjective speech quality assessment value of j bit in described i the hiding parameter;
If j is not more than N
i, then return and describedly judge whether the index corresponding to j bit of described i hiding parameter belongs to the step that P gathered in described index;
If i less than s, then arranges i=i+1, and return the bit number N of described judgement i hiding parameter
iWhether equal 1 step.
4. one kind is used for the G.723.1 information extracting method of speech coder, may further comprise the steps:
Obtain and carry close speech frame;
Obtain the individual hiding parameter of s and hidden bit/piece, embedding operation and the multi-system form of described year close speech frame, comprise following substep:
The scale-of-two upset is carried out in the position that the permission of described speech frame is revised;
Speech frame after utilizing the subjective speech quality assessment algorithm to upset is tested, to obtain one group of subjective speech quality assessment value;
Choose in the described subjective speech quality assessment value the corresponding position greater than threshold value, preserve the index set P of described corresponding position in described speech frame, and the frame parameter set Q of described index set P under in described speech frame, and the number of parameters of setting among the described frame parameter set Q is s;
Set counter i=1;
Judge the bit number N of i hiding parameter
iWhether equal 1;
If described bit number N
iBe not equal to 1, then set counter j=1;
Whether index corresponding to j bit of judging described i hiding parameter belongs to described index set P;
If described j index corresponding to bit belongs to described index set P, judge then whether the index corresponding to j+1 bit of described i hiding parameter belongs to described index set P;
If described j+1 index corresponding to bit belongs to described index set P, then under the decimal system, described i hiding parameter added respectively 2
J-1With subtract 2
J-1Computing;
With the raw parameter in the described speech frame of hiding parameter replacement of the i after the computing, and with the subjective speech quality assessment algorithm speech frame after replacing is tested, to obtain one group of subjective speech quality assessment value of j bit in the individual hiding parameter of described i;
J=j+1 is set;
Judge that whether j is greater than N
i
Adjacent and the corresponding index in position belongs to m the bit that P gathered in described index in described i the hiding parameter if j greater than Ni, then obtains, and the embedding operation of described hiding parameter is labeled as the embedding of 2m+1 system;
Embedding the embedding capacity of determining described i hiding parameter according to described 2m+1 system is
Bit, symbol
Expression rounds downwards;
Judge that whether i is less than s;
If i is not less than s, then preserve described s hiding parameter and hidden bit/piece thereof, embedding capacity, embedding operation and the multi-system form of described s hiding parameter; Extract s the hiding parameter that contained secret information in described year in the close speech frame;
Set counter l=1;
Whether the embedding operation of judging l parameter is that least significant bit (LSB) is replaced;
Do not replace if described embedding operation is not least significant bit (LSB), then l parameter is converted into the multi-system form of described parameter;
Extract the least significant bit (LSB) under the multi-system form of described l parameter;
The described least significant bit (LSB) that extracts is converted into binary mode, i.e. the secret information bit;
Judge that whether described counter l is greater than s;
If described counter l is greater than s, all secret information bits that then obtain being hidden in the described frame.
5. information extracting method according to claim 4 is characterized in that, also comprises step:
Replace if described embedding operation is least significant bit (LSB), then extract the least significant bit (LSB) of described parameter, i.e. secret information bit, and forward to and judge that described counter l is whether greater than the step of s;
If described counter l is not more than s, l=l+1 then is set, and whether return the described embedding operation of judging l parameter be the step that least significant bit (LSB) is replaced.
6. information extracting method according to claim 4 is characterized in that, the step of the described s that obtains a described year close speech frame hiding parameter and hidden bit/piece, embedding operation and multi-system form also comprises following substep:
If described bit number N
iEqual 1, then the embedding operation of described i hiding parameter is labeled as least significant bit (LSB) and replaces, its embedding capacity is 1 bit, then enters the described i of judgement whether less than the step of s;
If index corresponding to described j bit do not belong to described index set P, the step that then enters the described j=j+1 of setting;
If described j+1 index corresponding to bit belongs to described index set P, then under the decimal system, described i parameter carried out+2 respectively
J-1,-2
J-1,+2
j,-2
jComputing, then enter described with the raw parameter in the i after the computing the described speech frame of hiding parameter replacement, and with the subjective speech quality assessment algorithm speech frame after replacing is tested, to obtain the step of one group of subjective speech quality assessment value of j bit in described i the hiding parameter;
If j is not more than N
i, then return and describedly judge whether the index corresponding to j bit of described i hiding parameter belongs to the step that P gathered in described index;
If i less than s, then arranges i=i+1, and return the bit number N of described judgement i hiding parameter
iWhether equal 1 step.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201110396225XA CN102522089B (en) | 2011-12-02 | 2011-12-02 | Information embedding and extracting method for G.723.1 voice coder |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201110396225XA CN102522089B (en) | 2011-12-02 | 2011-12-02 | Information embedding and extracting method for G.723.1 voice coder |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN102522089A CN102522089A (en) | 2012-06-27 |
| CN102522089B true CN102522089B (en) | 2013-03-27 |
Family
ID=46292986
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201110396225XA Active CN102522089B (en) | 2011-12-02 | 2011-12-02 | Information embedding and extracting method for G.723.1 voice coder |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN102522089B (en) |
Families Citing this family (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103106901B (en) * | 2013-01-22 | 2015-05-20 | 宁波大学 | Audio digital steganography and extraction method in compressed domain based on index values |
| CN103986721B (en) * | 2014-05-27 | 2017-02-22 | 华侨大学 | Least significant bit grading steganography method for voice over IP |
| CN104091597B (en) * | 2014-06-26 | 2017-03-01 | 华侨大学 | A kind of ip voice steganography method based on rate modulation |
| CN106161021B (en) * | 2015-03-30 | 2020-03-20 | 重庆邮电大学 | Private data sending method and device and private data receiving method and device |
| CN107993669B (en) * | 2017-11-20 | 2021-04-16 | 西南交通大学 | Voice content authentication and tampering recovery method based on modification of least significant digit weight |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP1089550A2 (en) * | 1999-09-30 | 2001-04-04 | Kabushiki Kaisha Toshiba | Code generating method and unit thereof, code detecting method and unit thereof, and watermark embedding unit and watermark detecting unit |
| CN101206861A (en) * | 2007-12-25 | 2008-06-25 | 宁波大学 | Method for imbedding digital music production authentication information and method for authentication of said production |
| CN101478391A (en) * | 2009-01-13 | 2009-07-08 | 华中科技大学 | IP voice hidden communication method based on stream encryption |
| CN102169691A (en) * | 2011-04-06 | 2011-08-31 | 北京航空航天大学 | Reversible audio watermarking method based on bidirectional prediction and generalized expansion |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR100644627B1 (en) * | 2004-09-14 | 2006-11-10 | 삼성전자주식회사 | Method for encoding a sound field control information and method for processing therefor |
-
2011
- 2011-12-02 CN CN201110396225XA patent/CN102522089B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP1089550A2 (en) * | 1999-09-30 | 2001-04-04 | Kabushiki Kaisha Toshiba | Code generating method and unit thereof, code detecting method and unit thereof, and watermark embedding unit and watermark detecting unit |
| CN101206861A (en) * | 2007-12-25 | 2008-06-25 | 宁波大学 | Method for imbedding digital music production authentication information and method for authentication of said production |
| CN101478391A (en) * | 2009-01-13 | 2009-07-08 | 华中科技大学 | IP voice hidden communication method based on stream encryption |
| CN102169691A (en) * | 2011-04-06 | 2011-08-31 | 北京航空航天大学 | Reversible audio watermarking method based on bidirectional prediction and generalized expansion |
Also Published As
| Publication number | Publication date |
|---|---|
| CN102522089A (en) | 2012-06-27 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN102522089B (en) | Information embedding and extracting method for G.723.1 voice coder | |
| CN108346125B (en) | Airspace image steganography method and system based on generation countermeasure network | |
| CN111640444B (en) | CNN-based adaptive audio steganography method and secret information extraction method | |
| CN101577605A (en) | Speech LPC hiding and extraction algorithm based on filter similarity | |
| CN104010193B (en) | Video steganographic method based on macro block partition mode disturbance | |
| CN103986721A (en) | Least significant bit steganography method for voice over IP | |
| CN104934036A (en) | Audio Encoding Device, Method And Program, And Audio Decoding Device, Method And Program | |
| CN103106901B (en) | Audio digital steganography and extraction method in compressed domain based on index values | |
| CN104091597B (en) | A kind of ip voice steganography method based on rate modulation | |
| CN103456308B (en) | A kind of recoverable ciphertext domain voice content authentication method | |
| CN107689226A (en) | A kind of low capacity Methods of Speech Information Hiding based on iLBC codings | |
| CN104269175B (en) | A kind of ip voice steganography method matched based on Best similarity degree | |
| Lu et al. | Watermarking combined with CELP speech coding for authentication | |
| CN103050120A (en) | High-capacity digital audio reversible watermark processing method | |
| CN102664014B (en) | Blind audio watermark implementing method based on logarithmic quantization index modulation | |
| Djebbar et al. | Lightweight noise resilient steganography scheme for Internet of Things | |
| Wu et al. | Adaptive speech information hiding method based on K-means | |
| CN101901325A (en) | Copyright protection method | |
| CN103731654A (en) | Information embedding system and information extracting system using 2D/3D videos | |
| CN102194459B (en) | Information hiding method based on audio video coding standard (AVS) audio coding | |
| Kadhim | A new audio steganography system based on auto-key generator | |
| Jiao et al. | Compressed domain perceptual hashing for MELP coded speech | |
| CN104485109B (en) | A kind of compressed domain audio digital extraction method based on index value | |
| CN111787332B (en) | HEVC video data hiding method based on residual coefficient pairing | |
| CN110085242A (en) | A kind of adaptive steganography method in SILK fundamental tone domain based on minimum distortion cost |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant |