CN114070467B - Information source encryption privacy protection method facing deep joint information source channel coding - Google Patents
Information source encryption privacy protection method facing deep joint information source channel coding Download PDFInfo
- Publication number
- CN114070467B CN114070467B CN202111186217.2A CN202111186217A CN114070467B CN 114070467 B CN114070467 B CN 114070467B CN 202111186217 A CN202111186217 A CN 202111186217A CN 114070467 B CN114070467 B CN 114070467B
- Authority
- CN
- China
- Prior art keywords
- signal
- module
- information
- source
- network
- 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
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L1/00—Arrangements for detecting or preventing errors in the information received
- H04L1/0001—Systems modifying transmission characteristics according to link quality, e.g. power backoff
- H04L1/0009—Systems modifying transmission characteristics according to link quality, e.g. power backoff by adapting the channel coding
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L1/00—Arrangements for detecting or preventing errors in the information received
- H04L1/0001—Systems modifying transmission characteristics according to link quality, e.g. power backoff
- H04L1/0014—Systems modifying transmission characteristics according to link quality, e.g. power backoff by adapting the source coding
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L1/00—Arrangements for detecting or preventing errors in the information received
- H04L1/0001—Systems modifying transmission characteristics according to link quality, e.g. power backoff
- H04L1/0023—Systems modifying transmission characteristics according to link quality, e.g. power backoff characterised by the signalling
- H04L1/0026—Transmission of channel quality indication
Abstract
The invention provides a source encryption privacy protection method for depth joint source channel coding. The method comprises the following steps: transmitting the first information to a first module and a second module of a transmitting end, and a third module and a fourth module of a receiving end; the first module encrypts the first signal according to the first information to obtain a second signal, and transmits the second signal to a second module serving as a DJSCC encoding module at the transmitting end; the second module encodes the second signal and the first information into a third signal, the third signal becomes a fourth signal after passing through a wireless channel, the fourth signal reaches a receiving end and is input into the third module serving as a DJSCC decoding module, and the third module decodes and demodulates the fourth signal according to the first information to obtain a fifth signal and transmits the fifth signal to the fourth module; and the fourth module decrypts the fifth signal according to the first information to obtain a sixth signal. The invention can successfully protect the privacy information of the information source and can ensure the good performance of end-to-end reconstruction between the decrypted information source and the original information source.
Description
Technical Field
The present invention relates to the field of communications technologies, and in particular, to an encryption privacy protection method for a DJSCC (Deep Joint Source Channel Coding) transmission technology.
Background
The modularized design principle based on the shannon separation theorem is a foundation of modern communication systems, and has achieved great success in the development of wireless communication. However, the assumptions of infinite codeword length, infinite delay, and infinite Coding complexity in disjoint scheduling cannot be guaranteed in a practical wireless communication environment, which leads to suboptimal performance of disjoint Source Channel Coding (SSCC). In the last decades, Joint Source Channel Coding (JSCC) is theoretically proved to have a better error index than SSCC, and some specific JSCC Coding schemes are developed based on expert design, but the schemes have lower performance gain, higher complexity and poorer universality and cannot be applied to a wireless communication system in a large scale.
Recently, with the spread of Deep Learning (DL) in computer vision and natural language processing, DL-based methods are introduced into the fields of source coding and channel coding, exhibiting remarkable effects. This prompted researchers to consider combining DL with JSCC. Compared to SSCC schemes (e.g., JPEG/JPEG2000 for image coding and LDPC for channel coding), djscc (dl based jscc) techniques have better source recovery quality at low signal-to-noise ratios and low channel bandwidths. More remarkably, DJSCC is not affected by the "cliff effect" which provides good performance degradation when the channel signal-to-noise ratio is varied relative to the signal-to-noise ratio assumed during training.
As a wireless transmission technology with a wide application prospect, DJSCC can support various services such as autopilot, telemedicine, smart manufacturing, and military communications. However, when the source owner transmits an unencrypted source to the DJSCC service provider over an unsecured channel (e.g., an internet channel) and the DJSCC service provider transmits the reconstructed source to the image receiver over an unsecured channel after completing the service, eavesdropping and recovery are possible, which pose a significant threat to the data privacy and security of the source. Therefore, before transmitting services using DJSCC, the source should be privacy protected by encryption techniques.
An encryption module of a modern communication system is usually located after a source coding module, and encrypts bit information after source coding, and common encryption technologies include: DES (data encryption standard), AES (advanced encryption standard), RSA (Rivest-Shamir-Adleman), and the like. However, the encryption of joint coding for deep source and channel requires direct encryption of the source, i.e.: the information source information before the information source coding is encrypted, and the existing encryption technology designed based on the modern communication system cannot be used. Taking an image information source as an example, the image privacy protection oriented encryption technology can directly encrypt the image information source. The encryption technology encrypts the image by methods such as pixel scrambling, bit flipping, key encryption and the like, destroys the structural information of the plaintext image, and cannot support the deep learning-based algorithm to perform subsequent processing on the ciphertext image.
At present, a processing flow of performing a subsequent encryption algorithm (LE encryption algorithm, PE encryption algorithm) on a ciphertext image based on deep learning in the prior art is shown in fig. 1.
The following encryption algorithm for the ciphertext image based on the deep learning in the prior art has the following disadvantages: the method is applied to DJSCC transmission, the quality of the picture obtained after decryption at a receiving end is greatly reduced, and the corresponding image encryption effect of the method is shown in figure 2.
Disclosure of Invention
The embodiment of the invention provides a source encryption privacy protection method facing deep joint source channel coding, so as to realize effective encryption privacy transmission of a source.
In order to achieve the purpose, the invention adopts the following technical scheme.
A source encryption privacy protection method facing deep joint source channel coding comprises the following steps:
step S1: a sending end acquires a first signal and first information;
step S2: transmitting the first information to a first module and a second module of a transmitting end, and a third module and a fourth module of a receiving end;
and transmitting the first signal to a first module serving as an encryption module of the transmitting end, and encrypting the first signal by the first module according to the first information to obtain a second signal. The first module transmits the second signal to a second module which is used as a DJSCC coding module at the transmitting end;
step S3: the second module encodes the second signal and the first information into a third signal, the third signal becomes a fourth signal after passing through a wireless channel and reaches a receiving end, the fourth signal is input into a third module serving as a DJSCC decoding module, and the third module decodes and demodulates the fourth signal according to the first information to obtain a fifth signal;
step S4: the third module transmits the fifth signal to a fourth module of the receiving end;
step S5: and the fourth module decrypts the fifth signal according to the first information to obtain a sixth signal.
Preferably, the first information comprises one or more of the following combinations:
(1) channel State Information (CSI) information of communication channels of a sending end and a receiving end;
(2) information collected by a second module sensor of the sending end;
(3) and receiving the information acquired by the third module sensor.
Preferably, the first information is acquired by one or more of the following combinations:
(1) acquiring the first information through a reference signal;
(2) acquiring the first information through a sensor of the sending end;
(3) acquiring the first information through a sensor of the receiving end;
(4) and acquiring the first information according to the service index requirement of an application layer.
Preferably, the first signal comprises information generated by one or more of the following sources:
(1) a text source;
(2) a rich text source;
(3) an audio source;
(4) an image information source;
(5) a video source;
(6) an Augmented Reality (AR) source;
(7) a Virtual Reality (VR) information source;
(8) a holographic source.
Preferably, the second signal is a privacy-preserving signal corresponding to the first signal, and the scale of the second signal is the same as or different from that of the first signal;
the third signal satisfies at least one of:
(1) the third signal carries bit information;
(2) the third signal carries modulation symbol information;
the fourth signal is a third signal interfered by a wireless channel;
the fifth signal is a privacy protection signal obtained by the receiving end after DJSCC decoding, and the scale of the privacy protection signal is the same as or different from that of the second signal;
the sixth signal is information generated by one or more of the following original information sources reconstructed by the receiving end after decryption, and the signal scale is the same as that of the first signal:
(1) a text source;
(2) a rich text source;
(3) an audio source;
(4) an image information source;
(5) a video source;
(6) an Augmented Reality (AR) source;
(7) a Virtual Reality (VR) information source;
(8) a holographic source.
Preferably, the structure of the first module, the second module, the third module and the fourth module comprises one or more of the following in combination:
(1) a fully connected network;
(2) a convolutional network;
(3) a circulating network;
(4) a residual network.
Preferably, the method further comprises a training process, specifically comprising:
training step 1: selecting a training data set to construct a first signal;
training step 2: building an encryption and decryption coding and decoding network structure based on DJSCC transmission, and building a first module, a second module, a third module and a fourth module;
training step 3: constructing first information and channel data;
(1) acquiring background information of a transmitting end and/or a receiving end corresponding to channel data, wherein the background information comprises position information, temperature information, humidity information, time information, audio information, video information and image information;
(2) modeling the channel using either channel simulation data or real channel measurement data;
training step 4: constructing a first module of an encryption network, a second module of an encoding network, a third module of a decoding network and a fourth module of a decryption network which are used as sending ends;
training step 5: processing the first signal and the first information by using a first module to obtain a second signal;
training step 6: processing the second signal and the first information by using a second module to obtain a third signal;
training step 7: processing the fourth signal and the first information by using a third module to obtain a fifth signal;
training step 8: processing the fifth signal and the first information by using a fourth module to obtain a sixth signal;
a training step 9: processing the first signal by using a fifth module to obtain a seventh signal; processing the second signal by using a fifth module to obtain an eighth signal; processing the fifth signal by using a fifth module to obtain a ninth signal;
a training step 10: processing the first signal, the sixth signal, the seventh signal, the eighth signal and the ninth signal to obtain a tenth signal;
a training step 11: according to the tenth signal, carrying out gradient back propagation and updating the network parameters of the first module, the second module, the third module and the fourth module;
a training step 12: iterative training steps 5, 6, 7, 8, 9, 10 and 11, wherein the network is trained;
a training step 13: setting training termination conditions, and stopping training after the termination conditions are met;
and after the training is stopped, obtaining the trained first module, second module, third module and fourth module.
Preferably, the fifth module comprises one or more of the following combinations:
(1) a fully connected network;
(2) a convolutional network;
(3) a circulating network;
(4) a residual network.
It can be seen from the technical solutions provided by the embodiments of the present invention that the method according to the embodiments of the present invention can successfully protect the privacy information of the information source, and decrypt the received encrypted information source after the service is transmitted through the DJSCC, and the method can ensure good performance of end-to-end reconstruction between the decrypted information source and the original information source.
Additional aspects and advantages of the invention will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings needed to be used in the description of the embodiments are briefly introduced below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and it is obvious for those skilled in the art to obtain other drawings based on these drawings without creative efforts.
FIG. 1 is a flowchart illustrating a process for performing a subsequent encryption algorithm on a ciphertext image based on deep learning in the prior art;
FIG. 2 is a schematic diagram of an image encryption effect of a subsequent encryption algorithm for a ciphertext image based on deep learning in the prior art;
fig. 3 is a schematic application scenario diagram of a DJSCC-oriented information source encryption privacy protection method according to an embodiment of the present invention;
fig. 4 is a processing flow chart of a DJSCC-oriented information source encryption privacy protection method according to an embodiment of the present invention;
FIG. 5 is a schematic diagram of a training process according to an embodiment of the present invention;
fig. 6 is a schematic diagram illustrating performance simulation and comparison of DJSCC-oriented encryption and privacy protection technologies, DJSCC-based LE encryption technologies, and DJSCC-based PE encryption technologies in an embodiment of the present application;
fig. 7 is a schematic diagram illustrating comparison of privacy protection visual performance between DJSCC-oriented encryption and privacy protection technology and DJSCC-based LE encryption technology and DJSCC-based PE encryption technology in the embodiment of the present application.
Detailed Description
Reference will now be made in detail to embodiments of the present invention, examples of which are illustrated in the accompanying drawings, wherein like reference numerals refer to the same or similar elements or elements having the same or similar function throughout. The embodiments described below with reference to the accompanying drawings are illustrative only for the purpose of explaining the present invention, and are not to be construed as limiting the present invention.
As used herein, the singular forms "a", "an", "the" and "the" are intended to include the plural forms as well, unless the context clearly indicates otherwise. It will be further understood that the terms "comprises" and/or "comprising," when used in this specification, specify the presence of stated features, integers, steps, operations, elements, and/or components, but do not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components, and/or groups thereof. It will be understood that when an element is referred to as being "connected" or "coupled" to another element, it can be directly connected or coupled to the other element or intervening elements may also be present. Further, "connected" or "coupled" as used herein may include wirelessly connected or coupled. As used herein, the term "and/or" includes any and all combinations of one or more of the associated listed items.
It will be understood by those skilled in the art that, unless otherwise defined, all terms (including technical and scientific terms) used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs. It will be further understood that terms, such as those defined in commonly used dictionaries, should be interpreted as having a meaning that is consistent with their meaning in the context of the prior art and will not be interpreted in an idealized or overly formal sense unless expressly so defined herein.
For the convenience of understanding of the embodiments of the present invention, the following detailed description will be given by way of example with reference to the accompanying drawings, and the embodiments are not limited to the embodiments of the present invention.
The embodiment of the invention provides a DJSCC-oriented information source encryption and privacy protection method, which is used for protecting the privacy information of an information source, decrypting a received encrypted information source after the information source transmits service through DJSCC and ensuring the good performance of end-to-end reconstruction between the decrypted information source and an original information source. The application scenario of this method is shown in fig. 3, where the source owner intends to transmit the source to the source receiver through the wireless transmission service, while the source must be sent to the wireless service provider through an untrusted channel. To protect the source privacy information, the source owner encrypts/privacy protects the source before providing it to the wireless service provider. The encrypted/privacy protected source is then transmitted by the wireless service provider via a DJSCC transmission. After wireless transmission by the DJSCC, the received compromised encryption/privacy protected source is transmitted over an untrusted channel to the source receiver. The source receiver decrypts the compromised encryption/privacy protection to recover the original source information. In this scenario, even if eavesdropping or leakage occurs in transmission through an untrusted channel, an eavesdropper cannot directly acquire the privacy information of the source.
The processing flow of the information source encryption privacy protection method facing the depth joint information source channel coding provided by the embodiment of the invention is shown in fig. 4, and comprises the following processing steps:
step S1: the transmitting end acquires a first signal and first information.
Step S2: and transmitting the first information to a first module and a second module of the transmitting end, and a third module and a fourth module of the receiving end.
And transmitting the first signal to a first module serving as an encryption module of the transmitting end, and encrypting the first signal by the first module according to the first information to obtain a second signal. The first module transmits the second signal to a second module of the transmitting end as a DJSCC encoding module.
Step S3: the second module encodes the second signal and the first information into a third signal, the third signal becomes a fourth signal after passing through a wireless channel and reaches a receiving end, the fourth signal is input into the third module serving as a DJSCC decoding module, and the third module decodes and demodulates the fourth signal according to the first information to obtain a fifth signal.
Step S4: the third module transmits the fifth signal to the fourth module at the receiving end.
Step S5: and the fourth module decrypts the fifth signal according to the first information to obtain a sixth signal.
Wherein the first information may include one or more of the following:
(1) CSI (Channel State Information) Information of communication channels of a transmitting end and a receiving end; optionally, the information of CSI includes one or more of the following combinations: path loss of a channel, noise, RSRP, RSRQ, interference (intra-cell interference, inter-cell interference), received signal-to-noise ratio, received signal-to-interference-and-noise ratio, frequency selection characteristics, time-varying, doppler, etc.
(2) Information collected by a second module sensor of the sending end;
(3) and receiving the information collected by the third module sensor.
In the embodiment of the present application, the first information is obtained by one or more of the following combinations:
(1) acquiring the first information through a reference signal;
(2) acquiring the first information through a sensor of the sending end;
(3) acquiring the first information through a sensor of the receiving end;
(4) and acquiring the first information according to the service index requirement of an application layer.
The first signal includes information generated by one or more of the following sources:
(1) a text source;
(2) a rich text source;
(3) an audio source;
(4) an image source;
(5) a video source;
(6) augmented Reality (AR) sources;
(7) a Virtual Reality (VR) source;
(8) a holographic source, etc.
The second signal is a privacy protection signal corresponding to the first signal, and the second signal has the same or different scale as the first signal.
The third signal satisfies at least one of:
(1) the third signal carries bit information;
(2) the third signal carries modulation symbol information;
the fourth signal is a third signal subject to interference from a wireless channel.
And the fifth signal is a privacy protection signal acquired by the receiving end after DJSCC decoding, and the scale of the privacy protection signal is the same as or different from that of the second signal.
The sixth signal is information generated by one or more of the following original information sources reconstructed by the receiving end after decryption, and the signal scale is the same as that of the first signal:
(1) a text source;
(2) a rich text source;
(3) an audio source;
(4) an image information source;
(5) a video source;
(6) an Augmented Reality (AR) source;
(7) a Virtual Reality (VR) source;
(8) a holographic source, etc.
Wherein the structure of the first module, the second module, the third module, and the fourth module comprises one or more of the following in combination:
(1) a fully connected network;
(2) a convolutional network;
(3) a circulating network;
(4) a residual network.
In the embodiment of the present application, the training process is shown in fig. 5, and the fifth module only exists in the training stage:
training step 1: selecting a training data set to construct a first signal;
training step 2: building an encryption and decryption coding and decoding network structure based on DJSCC transmission, namely building a first module, a second module, a third module and a fourth module;
training step 3: constructing first information and channel data;
(1) acquiring background information of a transmitting end and/or a receiving end corresponding to channel data, wherein the background information comprises position information, temperature information, humidity information, time information, audio information, video information, image information and the like;
(2) modeling the channel using either channel simulation data or real channel measurement data;
training step 4: constructing an encryption network (a first module) of a sending end, an encoding network (a second module), a decoding network (a third module) of a receiving end and a decryption network (a fourth module);
training step 5: processing the data (first signal) and the first information by using a first module to obtain a second signal;
training step 6: and processing the second signal and the first information by using a second module to obtain a third signal.
Training step 7: and processing the fourth signal and the first information by using a third module to obtain a fifth signal.
Training step 8: and processing the fifth signal and the first information by using a fourth module to obtain a sixth signal.
A training step 9: processing the first signal by using a fifth module to obtain a seventh signal; processing the second signal by using a fifth module to obtain an eighth signal; and processing the fifth signal by using a fifth module to obtain a ninth signal.
A training step 10: and processing the first signal, the sixth signal, the seventh signal, the eighth signal and the ninth signal to obtain a tenth signal.
A training step 11: and performing gradient back propagation according to the tenth signal, and updating the network parameters of the first module, the second module, the third module and the fourth module.
A training step 12: and (5) iteratively training the steps 5, 6, 7, 8, 9, 10 and 11 to train the network.
A training step 13: and setting training termination conditions, and stopping training after the termination conditions are met.
And after the training is stopped, obtaining the trained first module, second module, third module and fourth module.
In the embodiment of the present application, the structure of the fifth module includes one or more of the following combinations:
(1) a fully connected network;
(2) a convolutional network;
(3) a circulating network;
(4) a residual network.
In this embodiment of the present application, the fifth module is a pre-training module, and only participates in processing the first signal, the second signal, and the fifth signal, and the parameters thereof are not updated in the training phase.
Fig. 6 is a schematic diagram illustrating performance simulation and comparison of DJSCC-oriented encryption and privacy protection technologies, DJSCC-based LE encryption technologies, and DJSCC-based PE encryption technologies in an embodiment of the present application; fig. 7 is a schematic diagram illustrating comparison of privacy protection visual performance between DJSCC-oriented encryption and privacy protection technology and DJSCC-based LE encryption technology and DJSCC-based PE encryption technology in the embodiment of the present application. Fig. 6 compares the DJESCC (Deep Joint Encryption and Source-channel Coding) method with the LE Encryption method based on DJSCC and the PE Encryption method based on DJSCC under an Additive White Gaussian Noise (AWGN) channel when the bandwidth ratio R is 1/6. When the signal-to-noise ratio is increased from 0dB to 20dB, the performance of DJSCC _ PE is slightly improved, about 18.5 to 20.5dB, which is far lower than that of the DJSCC-oriented encryption and privacy protection method DJESCC. Although the performance of DJSCC PE is better than DJSCC _ LE, when SNRtest is 0dB, the performance of DJSCC _ PE is still 5dB lower than DJESCC, and the performance gap between DJSCC _ PE and DJESCC is further expanded as the signal-to-noise ratio increases. Fig. 7 shows the corresponding visual performance of different encryption methods when the SNR is 0, 10, 20 dB. The encrypted images of the DJSCCE _ LE method and the DJSCCE _ PE method are displayed as noise images, unlike the black and white lattice characteristics of the encrypted images displayed in the DJESCC method. At signal-to-noise ratios of 0, 10, 20dB, the decrypted image of the DJSCC LE method is disturbed by some noise pixels, whereas at signal-to-noise ratios of 0, 10, 20, the decrypted image of the DJSCC PE method looks blurred.
In summary, the method of the embodiment of the present invention can successfully protect the privacy information of the information source, and decrypt the received encrypted information source after the DJSCC transmission service, and the method can ensure good performance of end-to-end reconstruction between the decrypted information source and the original information source.
Those of ordinary skill in the art will understand that: the figures are schematic representations of one embodiment, and the blocks or processes shown in the figures are not necessarily required to practice the present invention.
From the above description of the embodiments, it is clear to those skilled in the art that the present invention can be implemented by software plus necessary general hardware platform. Based on such understanding, the technical solutions of the present invention may be embodied in the form of software products, which may be stored in a storage medium, such as ROM/RAM, magnetic disk, optical disk, etc., and include instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute the method according to the embodiments or some parts of the embodiments.
All the embodiments in the present specification are described in a progressive manner, and the same and similar parts among the embodiments are referred to each other, and each embodiment focuses on the differences from other embodiments. In particular, apparatus or system embodiments, which are substantially similar to method embodiments, are described in relative ease, and reference may be made to some descriptions of method embodiments for related points. The above-described embodiments of the apparatus and system are merely illustrative, and the units described as separate parts may or may not be physically separate, and parts displayed as units may or may not be physical units, may be located in one position, or may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of the present embodiment. One of ordinary skill in the art can understand and implement it without inventive effort.
While the invention has been described with reference to specific preferred embodiments, it will be understood by those skilled in the art that various changes and modifications may be made without departing from the spirit and scope of the invention as defined in the following claims. Therefore, the protection scope of the present invention shall be subject to the protection scope of the claims.
Claims (8)
1. A source encryption privacy protection method facing deep joint source channel coding is characterized by comprising the following steps:
step S1: a sending end acquires a first signal and first information;
step S2: transmitting the first information to a first module and a second module of a transmitting end, and a third module and a fourth module of a receiving end;
the first module transmits the first signal to a first module serving as an encryption module of the transmitting end, the first module encrypts the first signal according to the first information to obtain a second signal, and the first module transmits the second signal to a second module serving as a DJSCC encoding module of the transmitting end;
step S3: the second module encodes the second signal and the first information into a third signal, the third signal becomes a fourth signal after passing through a wireless channel and reaches a receiving end, the fourth signal is input into a third module serving as a DJSCC decoding module, and the third module decodes and demodulates the fourth signal according to the first information to obtain a fifth signal;
step S4: the third module transmits the fifth signal to a fourth module of the receiving end;
step S5: and the fourth module decrypts the fifth signal according to the first information to obtain a sixth signal.
2. The method of claim 1, wherein the first information comprises one or more of the following:
(1) channel State Information (CSI) information of communication channels of a sending end and a receiving end;
(2) information collected by a second module sensor of the sending end;
(3) and receiving the information collected by the third module sensor.
3. The method of claim 1, wherein the first information is obtained by one or more of the following combinations:
(1) acquiring the first information through a reference signal;
(2) acquiring the first information through a sensor of the sending end;
(3) acquiring the first information through a sensor of the receiving end;
(4) and acquiring the first information according to the service index requirement of an application layer.
4. The method of claim 1, wherein the first signal comprises information generated by one or more of the following sources:
(1) a text source;
(2) a rich text source;
(3) an audio source;
(4) an image information source;
(5) a video source;
(6) an Augmented Reality (AR) source;
(7) a Virtual Reality (VR) information source;
(8) a holographic source.
5. The method of claim 1, wherein the second signal is a privacy preserving signal corresponding to the first signal, and the scale of the second signal is the same as or different from that of the first signal;
the third signal satisfies at least one of:
(1) the third signal carries bit information;
(2) the third signal carries modulation symbol information;
the fourth signal is a third signal interfered by a wireless channel;
the fifth signal is a privacy protection signal obtained by the receiving end after DJSCC decoding, and the scale of the privacy protection signal is the same as or different from that of the second signal;
the sixth signal is information generated by one or more of the following original information sources reconstructed by the receiving end after decryption, and the signal scale is the same as that of the first signal:
(1) a text source;
(2) a rich text source;
(3) an audio source;
(4) an image source;
(5) a video source;
(6) an Augmented Reality (AR) source;
(7) a Virtual Reality (VR) information source;
(8) a holographic source.
6. The method of claim 1, wherein the structure of the first module, the second module, the third module, and the fourth module comprises one or more of the following in combination:
(1) a fully connected network;
(2) a convolutional network;
(3) a circulating network;
(4) a residual network.
7. The method according to any one of claims 1 to 6, characterized in that the method further comprises a training procedure, in particular comprising:
training step 1: selecting a training data set to construct a first signal;
training step 2: building an encryption, decryption and coding network structure based on DJSCC transmission, and building a first module, a second module, a third module and a fourth module;
training step 3: constructing first information and channel data;
(1) acquiring background information of a transmitting end and/or a receiving end corresponding to channel data, wherein the background information comprises position information, temperature information, humidity information, time information, audio information, video information and image information;
(2) modeling the channel using either channel simulation data or real channel measurement data;
and 4, training: constructing a first module of an encryption network, a second module of an encoding network, a third module of a decoding network and a fourth module of a decryption network which are used as sending ends;
training step 5: processing the first signal and the first information by using a first module to obtain a second signal;
training step 6: processing the second signal and the first information by using a second module to obtain a third signal;
training step 7: processing the fourth signal and the first information by using a third module to obtain a fifth signal;
training step 8: processing the fifth signal and the first information by using a fourth module to obtain a sixth signal;
a training step 9: processing the first signal by using a fifth module to obtain a seventh signal; processing the second signal by using a fifth module to obtain an eighth signal; processing the fifth signal by using a fifth module to obtain a ninth signal;
a training step 10: processing the first signal, the sixth signal, the seventh signal, the eighth signal and the ninth signal to obtain a tenth signal;
a training step 11: according to the tenth signal, carrying out gradient back propagation and updating the network parameters of the first module, the second module, the third module and the fourth module;
a training step 12: iterative training steps 5, 6, 7, 8, 9, 10 and 11 are carried out, and the network is trained;
a training step 13: setting training termination conditions, and stopping training after the termination conditions are met;
and after the training is stopped, obtaining the trained first module, second module, third module and fourth module.
8. The method of claim 7, wherein the fifth module comprises one or more of the following in combination:
(1) a fully connected network;
(2) a convolutional network;
(3) a circulating network;
(4) a residual network.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111186217.2A CN114070467B (en) | 2021-10-12 | 2021-10-12 | Information source encryption privacy protection method facing deep joint information source channel coding |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111186217.2A CN114070467B (en) | 2021-10-12 | 2021-10-12 | Information source encryption privacy protection method facing deep joint information source channel coding |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN114070467A CN114070467A (en) | 2022-02-18 |
| CN114070467B true CN114070467B (en) | 2022-09-30 |
Family
ID=80234280
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111186217.2A Active CN114070467B (en) | 2021-10-12 | 2021-10-12 | Information source encryption privacy protection method facing deep joint information source channel coding |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114070467B (en) |
Family Cites Families (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| GB2576702A (en) * | 2018-08-15 | 2020-03-04 | Imperial College Sci Tech & Medicine | Joint source channel coding of information sources using neural networks |
| CN109525299B (en) * | 2018-10-24 | 2020-08-04 | 清华大学 | Satellite communication system and communication method for adaptive coding modulation optimization |
| CN109379120A (en) * | 2018-12-11 | 2019-02-22 | 深圳大学 | Chain circuit self-adaptive method, electronic device and computer readable storage medium |
| CN113472479B (en) * | 2020-03-31 | 2022-11-22 | 维沃移动通信有限公司 | Transmission processing method and equipment |
-
2021
- 2021-10-12 CN CN202111186217.2A patent/CN114070467B/en active Active
Also Published As
| Publication number | Publication date |
|---|---|
| CN114070467A (en) | 2022-02-18 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Zhang et al. | Embedding cryptographic features in compressive sensing | |
| US7512238B2 (en) | Method and apparatus for detection of loss of cipher synchronization | |
| US6182214B1 (en) | Exchanging a secret over an unreliable network | |
| US20070297612A1 (en) | Method, device and system of encrypted wireless communication | |
| Hussain et al. | Security enhancement for video transmission via noise aggregation in immersive systems | |
| CN112104454B (en) | Data secure transmission method and system | |
| CN103634100A (en) | Encryption method and device | |
| US20080317243A1 (en) | Low complexity encryption method for content that is coded by a rateless code | |
| JPWO2020059535A1 (en) | Transmitter and transmit method, and receiver and receive method | |
| CN111756690A (en) | Data processing system, method and server | |
| CN108924594B (en) | Video encryption method based on hyperchaotic system | |
| Sethi et al. | Chaos-based uncompressed frame level video encryption | |
| Taneja et al. | Chaos based cryptosystem for still visual data | |
| Khattabi et al. | Revisiting lightweight encryption for IoT applications: Error performance and throughput in wireless fading channels with and without coding | |
| CN114070467B (en) | Information source encryption privacy protection method facing deep joint information source channel coding | |
| CN111641500B (en) | Encryption and decryption method for wireless video transmission safety of unmanned aerial vehicle | |
| EP1994672B1 (en) | Method and apparatus for synchronous stream cipher encryption with reserved codes | |
| CN107888611B (en) | Communication method and device | |
| El-Bakary et al. | DRPE encryption with chaotic interleaving for video communication | |
| CN114337849B (en) | Physical layer confidentiality method and system based on mutual information quantity estimation neural network | |
| RU2534925C2 (en) | Security method, decoding method, data medium and terminal for security method | |
| Ayad et al. | Secure Images Transmission Through OFDM System | |
| KR100797106B1 (en) | Method for encrypting and decrypting transmmited and received packet in wireless lan | |
| Guan et al. | Enhanced physical layer security of MIMO-SDM systems through information scrambling | |
| US10873773B2 (en) | Countermeasure for cryptographic cribs |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |