CN115208722B - Novel frame synchronous scrambling code descrambling method - Google Patents
Novel frame synchronous scrambling code descrambling method Download PDFInfo
- Publication number
- CN115208722B CN115208722B CN202210897758.4A CN202210897758A CN115208722B CN 115208722 B CN115208722 B CN 115208722B CN 202210897758 A CN202210897758 A CN 202210897758A CN 115208722 B CN115208722 B CN 115208722B
- Authority
- CN
- China
- Prior art keywords
- descrambling
- scrambling
- layer
- network model
- training
- 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
- 238000000034 method Methods 0.000 title claims abstract description 37
- 230000001360 synchronised effect Effects 0.000 title claims abstract description 32
- 238000012549 training Methods 0.000 claims abstract description 35
- 238000013135 deep learning Methods 0.000 claims abstract description 18
- 230000006870 function Effects 0.000 claims description 15
- 230000004913 activation Effects 0.000 claims description 13
- 238000010606 normalization Methods 0.000 claims description 8
- 238000005070 sampling Methods 0.000 claims description 4
- ORILYTVJVMAKLC-UHFFFAOYSA-N Adamantane Natural products C1C(C2)CC3CC1CC2C3 ORILYTVJVMAKLC-UHFFFAOYSA-N 0.000 claims description 3
- 238000004891 communication Methods 0.000 abstract description 13
- 238000012937 correction Methods 0.000 abstract description 6
- 238000005516 engineering process Methods 0.000 abstract description 4
- 239000007787 solid Substances 0.000 abstract description 3
- 230000009466 transformation Effects 0.000 abstract description 3
- 230000008569 process Effects 0.000 description 10
- 238000004422 calculation algorithm Methods 0.000 description 6
- 238000010586 diagram Methods 0.000 description 6
- 238000004088 simulation Methods 0.000 description 6
- 230000000694 effects Effects 0.000 description 4
- 238000010200 validation analysis Methods 0.000 description 4
- 230000005540 biological transmission Effects 0.000 description 3
- 238000004364 calculation method Methods 0.000 description 2
- 238000002474 experimental method Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 210000002569 neuron Anatomy 0.000 description 2
- 238000012545 processing Methods 0.000 description 2
- 238000007619 statistical method Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 238000013144 data compression Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 238000012546 transfer Methods 0.000 description 1
- 238000012795 verification Methods 0.000 description 1
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L25/00—Baseband systems
- H04L25/02—Details ; arrangements for supplying electrical power along data transmission lines
- H04L25/03—Shaping networks in transmitter or receiver, e.g. adaptive shaping networks
- H04L25/03006—Arrangements for removing intersymbol interference
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L25/00—Baseband systems
- H04L25/02—Details ; arrangements for supplying electrical power along data transmission lines
- H04L25/03—Shaping networks in transmitter or receiver, e.g. adaptive shaping networks
- H04L25/03006—Arrangements for removing intersymbol interference
- H04L25/03178—Arrangements involving sequence estimation techniques
- H04L25/03248—Arrangements for operating in conjunction with other apparatus
- H04L25/0328—Arrangements for operating in conjunction with other apparatus with interference cancellation circuitry
Landscapes
- Engineering & Computer Science (AREA)
- Power Engineering (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Computational Linguistics (AREA)
- General Health & Medical Sciences (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- Artificial Intelligence (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Health & Medical Sciences (AREA)
- Synchronisation In Digital Transmission Systems (AREA)
Abstract
The invention discloses a new frame synchronous scrambling and descrambling method, which comprises the following steps: s1, generating signal samples scrambled by frame synchronous scrambling sequences of different primitive polynomials; s2, constructing a frame synchronous scrambling and descrambling network model based on deep learning; s3, setting training super parameters of a frame synchronous scrambling and descrambling network model; s4, inputting the signal sample into a descrambling network model, wherein the scrambling code is used as input data of the network, the original data is used as a label, and training is carried out to obtain a final form of the frame synchronization scrambling code descrambling network model. The invention combines the deep learning with the communication field, uses the deep learning technology to descramble the simulated scrambled communication signals, uses the network model in the deep learning to automatically extract the scrambling characteristics of the input signals, learns the transformation relation between scrambling codes and original signals, improves the accuracy of scrambling code identification and descrambling, and lays a solid foundation for the subsequent works such as de-interleaving, error correction coding and decoding.
Description
Technical Field
The invention belongs to the technical field of communication, and particularly relates to a novel frame synchronous scrambling code descrambling method.
Background
Digital communication systems, which transfer information in digital signals, are the mainstay of contemporary communication technology. The transmission process of the digital signal in the digital communication system generally comprises source coding, error correction coding, interleaving, scrambling, modulation, channel passing, descrambling, deinterleaving, error correction code decoding, source decoding and the like, wherein the source coding mainly plays a role in data compression to complete analog-to-digital conversion; error correction coding and interleaving are mainly aimed at combating random errors and burst errors of the channel. In practice, normal transmission of information data cannot be guaranteed by performing coding processes such as source coding, error correction coding, interleaving and the like, and as 0 or 1 connection may occur in a transmitted data stream, extraction of timing information is difficult, and inter-code crosstalk may also occur. At this time, the randomization processing is needed to be carried out on the data, so that the problems of the signals of 0 and 1 are solved, inter-code crosstalk is reduced, meanwhile, the data is encrypted and protected, and the reliability of data transmission is effectively improved. The process of randomizing the data is a scrambling process, and the de-randomizing process opposite to the scrambling process is de-scrambling. Scrambling has received increasing attention from researchers in recent years as one of the key technologies in the field of digital communications.
Non-cooperative communication as an access communication mode of non-regular or unauthorized acquisition, a third party can only obtain little or no a priori information of the sender, such as coding parameters. Under the condition, how to correctly and effectively identify the scrambling parameters of the scrambling code from the intercepted information sequence and realize descrambling directly influences whether the subsequent decoding can be correctly performed is a key for recovering the information data, and has important significance in the non-cooperative field.
Scrambling codes are classified into linear and nonlinear according to the nature of feedback logic by using a feedback shift register as a basic component. At present, linear scrambling codes with linear homogeneous polynomials as generator polynomials are used in practical communications. According to the different synchronous modes of the Linear Feedback Shift Registers (LFSR) of the receiving and transmitting sides, the scrambling codes are divided into frame synchronous scrambling codes and self-synchronous scrambling codes.
Hao Shiji et al in the article "a new blind pseudo-random scrambling code identification method" determine the order of the synchronous scrambling code generator polynomial based on the autocorrelation of the synchronous scrambling code, the algorithm is computationally inexpensive, but the threshold is set by using the exact prior information of the source. Yang Zhongli and Liu Yujun in the article "study of comprehensive algorithm of self-synchronizing scrambling sequences" solve the system of error-containing equations by using Walsh-Hadamard transform to recover the generator polynomial, but the use of this method requires a known scrambling number of stages, and the calculation amount of the algorithm increases exponentially with the increase of the scrambling number of stages. Zhang Yongguang et al introduce a separate conquer attack algorithm (DC algorithm) of stream ciphers into blind recognition of scrambling codes in the article of "a blind recognition method of an auto-synchronous scrambling code generating polynomial", descramble and re-scramble a scrambling code sequence with a possible polynomial, determine a generating polynomial of an auto-synchronous scrambling code according to the correlation degree of a newly generated ciphertext sequence and an original scrambling code sequence, the algorithm has a large limitation on the order and the term number of the scrambling code generating polynomial, and the setting of a complex calculation threshold is difficult.
Disclosure of Invention
The invention aims to overcome the defects of the prior art and provides a frame synchronous scrambling and descrambling method which combines deep learning with the communication field, can automatically extract scrambling characteristics of an input signal by using a network model in the deep learning, learns the transformation relation between scrambling codes and original signals and improves the accuracy of scrambling code identification and descrambling.
The aim of the invention is realized by the following technical scheme: a new frame synchronization scrambling and descrambling method comprising the steps of:
s1, generating signal samples scrambled by frame synchronous scrambling sequences of different primitive polynomials;
s2, constructing a frame synchronous scrambling and descrambling network model based on deep learning;
s3, setting training super parameters of a frame synchronous scrambling and descrambling network model;
s4, inputting the signal sample into a descrambling network model, wherein the scrambling code is used as input data of the network, the original data is used as a label, and training is carried out to obtain a final form of the frame synchronization scrambling code descrambling network model.
Further, the frame synchronous scrambling and descrambling network model based on the deep learning in the step S2 sequentially comprises an input layer, an encoder, a decoder and an output layer;
the encoder comprises a plurality of dimension-reducing convolution blocks, wherein each dimension-reducing convolution block sequentially comprises a two-dimensional convolution layer, a normalization layer and an activation function layer according to the sequence;
the decoder comprises a plurality of dimension-lifting convolution blocks, wherein each dimension-lifting convolution block sequentially comprises a two-dimensional convolution layer, a normalization layer, an activation function layer and an up-sampling layer;
the output layer comprises two layers, wherein the first layer is a dimension-reducing convolution block, and the second layer is a two-dimensional convolution layer.
The activation function layer is further followed by a two-dimensional convolution layer.
Further, the specific implementation method of the step S3 is as follows: the network selects Adam as an optimizer; the sample number batch_size of each time of input network is set to 256 during training; the maximum iteration number epoch of training is 200; network training using custom learning rate, setting initial learning rate lr init Is 10 -4 The learning rate is set as follows;
further, the specific implementation method of the step S4 is as follows: inputting the signal samples into a frame synchronous scrambling and descrambling network model based on deep learning for training, stopping training when the training times reach the set maximum iteration times epoch, obtaining the optimal network parameters, and storing the network model formed by the optimal network parameters.
The beneficial effects of the invention are as follows: the invention combines the deep learning with the communication field, uses the deep learning technology to descramble the simulated scrambled communication signals, and compared with the scrambling parameter identification and descrambling using the statistical method and algebraic method, the network model in the deep learning can automatically extract the scrambling characteristics of the input signals, learn the transformation relation between the scrambling and the original signals, improve the accuracy of scrambling identification and descrambling, and lay a solid foundation for the subsequent works of de-interleaving, error correction coding and decoding and the like.
Drawings
FIG. 1 is a flow chart of a frame synchronous scrambling and descrambling method of the present invention;
fig. 2 is a diagram of simulation results of the present embodiment.
Detailed Description
The technical scheme of the invention is further described below with reference to the accompanying drawings.
As shown in fig. 1, a new frame synchronization scrambling and descrambling method of the present invention includes the following steps:
s1, scrambling signals by using m frame synchronous scrambling sequences of different primitive polynomials in an n-level register, and generating signal samples scrambled by using the frame synchronous scrambling sequences of different primitive polynomials; the embodiment of the invention adopts a 7-stage shift register, and 18 primitive polynomials are adopted. In this example, 5 ten thousand samples are generated under each primitive polynomial, the length dataLength of each data sample is set to 2048 bits, and the total data set size is 90 ten thousand samples. 80% is extracted from the total data set as training set and 20% is validation set.
S2, constructing a frame synchronous scrambling and descrambling network model based on deep learning; the frame synchronous scrambling and descrambling network model based on the deep learning sequentially comprises an input layer, an encoder, a decoder and an output layer;
input layer: dimension information of input signal samples, namely datalength×1×1;
an encoder: the number of the neurons is reduced along with the increase of the number of layers of the encoder, and signal data of an input layer are compressed; the system comprises a plurality of dimension-reducing convolution blocks, wherein each dimension-reducing convolution block sequentially comprises a two-dimensional convolution layer, a normalization layer and an activation function layer according to the sequence; in the encoder, an optional two-dimensional convolution layer is further arranged behind the activation function layer, wherein the convolution kernel size in the first two-dimensional convolution layer is 3 multiplied by 1, padding is set to be in a same mode, and the input length and the output length of the convolution layer are kept consistent in a patching mode; the normalization layer adopts Batchnormalization; the activation function layer adopts a relu function; the second two-dimensional convolution layer is different from the first convolution layer in that the layer uses the parameters with the step length of (2, 1) to achieve the effect of reducing the input data length to half of the original value and achieve the coding effect.
In the invention, the encoder is provided with k=6, namely the encoder is composed of 6 dimension-reducing convolution blocks, and the condition that the descrambling effect is poor due to too few obtained characteristic points after the input data is subjected to dimension-reducing processing for many times is considered, the number of the output characteristic diagrams of each convolution block is increased to be optimized, namely the number of the output characteristic diagrams is 128+32×ix, wherein the value range of ix is 0 to 5, namely the number of layers of the encoder is 6 layers of dimension-reducing convolution blocks.
A decoder: the number of the neurons is increased along with the increase of the layer number of the decoder, and the last layer of the encoder is decompressed; the system comprises a plurality of dimension-increasing convolution blocks, wherein each dimension-increasing convolution block sequentially comprises a two-dimensional convolution layer, a normalization layer, an activation function layer and an up-sampling layer; the up-sampling layer parameter is set to be (2, 1), so as to achieve the dimension-increasing effect of restoring the original length of the input data. The coding process and the decoding process are in a corresponding relation, so that the same parameter value k, namely 6 upwarp dimension convolution blocks, should be used in the module to construct a decoding network, and the number of the output characteristic diagrams of each layer is adjusted to be consistent with the number of the output characteristic diagrams of the encoder symmetrically, namely the number of the output characteristic diagrams is 128+32× (k-ix-1), wherein k is 6, and the range of ix is 0 to 5. The output data length of the final decoder is the same as the input data length of the input layer.
Output layer: the output signal length is consistent with the input time and is dataLength multiplied by 1; the method comprises two layers, wherein the first layer is a dimension-reducing convolution block (sequentially comprises a two-dimensional convolution layer, a normalization layer and an activation function layer according to the sequence), and the second two-dimensional convolution layer is not selected, namely the output layer does not halve the data length; the second layer is a two-dimensional convolution layer, the number of output feature images is 1, the activation function is sigmoid, and the input and output lengths of the layer are kept unchanged by adopting a patching mode.
S3, setting training super parameters of a frame synchronous scrambling and descrambling network model; the specific implementation method comprises the following steps: the network selects Adam as an optimizer; the sample number batch_size of each time of input network is set to 256 during training; the maximum iteration number epoch of training is 200; network training using custom learning rate, setting initial learning rate lr init Is 10 -4 The learning rate is set as follows;
s4, inputting the signal sample into a descrambling network model, wherein a scrambling code is used as input data of a network, and the original data is used as a label to train so as to obtain a final form of the frame synchronization scrambling code descrambling network model; the specific implementation method comprises the following steps: inputting the signal samples into a frame synchronous scrambling and descrambling network model based on deep learning for training, stopping training when the training times reach the set maximum iteration times epoch, obtaining the optimal network parameters, and storing the network model formed by the optimal network parameters.
The scrambling and descrambling method of the present invention is verified by simulation.
Simulation conditions: according to the simulation experiment, under the Nvidia GeForce RTX 3080Ti 12GB and Ubuntu Server 20.04 systems, MATLAB R2020a and a Keras framework taking TensorFlow as a rear end are used for completing the descrambling experiment based on a self-coding network of the frame synchronous scrambling code generated by 18 different primitive polynomials of 7-level registers.
Simulation experiment contents: the simulation experiment of the invention is to firstly use MATLAB to generate a data set for network training. The invention concerns frame synchronous scrambling code, which is generated by selecting 7-level linear shift register, wherein 18 primitive polynomials which meet the condition of generating scrambling code sequence are provided, and the initial value of the shift register is set as 0, 1 bit of 7-bit random, except for the full 0 sequence (0,0,0,0,0,0,0). And performing modulo-2 sum on the synchronous scrambling sequence generated under each primitive polynomial and the original information to finish the scrambling process and generate a corresponding scrambling code. Each primitive polynomial generates 5 ten thousand data samples, and the total number of data sets is 90 ten thousand, 80% of which are used as training set samples, and 20% of which are used as verification set samples. Inputting the data set into the constructed network model for network training, iteratively training 200 epochs, and storing the obtained best network model as a final model of the experiment.
The accuracy curve and the Loss curve trained by this example are shown in fig. 2. The upper graph of fig. 2 is the accuracy of the training set and the validation set during the training process, and the lower graph is the loss function value of the training set and the validation set. The curve with the reference "x" in the figure represents the resulting curve of the validation set, the solid curveIs a training set result curve. From the result graph, it can be seen that the invention can effectively descramble the scrambling code, the loss value gradually converges in the training process, and tends to 10 -9 . Compared with the method for identifying and descrambling the scrambling code by using an algebraic method and a statistical method, the method can obviously reduce the computational complexity and improve the accuracy of blind identification of the scrambling code and the accuracy of descrambling.
Those of ordinary skill in the art will recognize that the embodiments described herein are for the purpose of aiding the reader in understanding the principles of the present invention and should be understood that the scope of the invention is not limited to such specific statements and embodiments. Those of ordinary skill in the art can make various other specific modifications and combinations from the teachings of the present disclosure without departing from the spirit thereof, and such modifications and combinations remain within the scope of the present disclosure.
Claims (4)
1. A new frame synchronization scrambling and descrambling method, comprising the steps of:
s1, generating signal samples scrambled by frame synchronous scrambling sequences of different primitive polynomials;
s2, constructing a frame synchronous scrambling and descrambling network model based on deep learning; the frame synchronous scrambling and descrambling network model based on the deep learning sequentially comprises an input layer, an encoder, a decoder and an output layer;
the encoder comprises a plurality of dimension-reducing convolution blocks, wherein each dimension-reducing convolution block sequentially comprises a two-dimensional convolution layer, a normalization layer and an activation function layer according to the sequence;
the decoder comprises a plurality of dimension-lifting convolution blocks, wherein each dimension-lifting convolution block sequentially comprises a two-dimensional convolution layer, a normalization layer, an activation function layer and an up-sampling layer;
the output layer comprises two layers, wherein the first layer is a dimension-reducing convolution block, and the second layer is a two-dimensional convolution layer;
s3, setting training super parameters of a frame synchronous scrambling and descrambling network model;
s4, inputting the signal sample into a descrambling network model, wherein the scrambling code is used as input data of the network, the original data is used as a label, and training is carried out to obtain a final form of the frame synchronization scrambling code descrambling network model.
2. A new frame sync scrambling and descrambling method according to claim 1, wherein the activation function layer further comprises a two-dimensional convolution layer after the activation function layer in the encoder.
3. The method for descrambling the new frame synchronization scrambling code according to claim 1, wherein the specific implementation method of step S3 is as follows: the network selects Adam as an optimizer; the sample number batch_size of each time of input network is set to 256 during training; the maximum iteration number epoch of training is 200; network training using custom learning rate, setting initial learning rate lr init Is 10 -4 The learning rate is set as follows;
4. the method for descrambling the new frame synchronization scrambling according to claim 1, wherein the specific implementation method of step S4 is as follows: inputting the signal samples into a frame synchronous scrambling and descrambling network model based on deep learning for training, stopping training when the training times reach the set maximum iteration times epoch, obtaining the optimal network parameters, and storing the network model formed by the optimal network parameters.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210897758.4A CN115208722B (en) | 2022-07-28 | 2022-07-28 | Novel frame synchronous scrambling code descrambling method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210897758.4A CN115208722B (en) | 2022-07-28 | 2022-07-28 | Novel frame synchronous scrambling code descrambling method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN115208722A CN115208722A (en) | 2022-10-18 |
| CN115208722B true CN115208722B (en) | 2024-03-01 |
Family
ID=83584817
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210897758.4A Active CN115208722B (en) | 2022-07-28 | 2022-07-28 | Novel frame synchronous scrambling code descrambling method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115208722B (en) |
Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101610122A (en) * | 2009-07-03 | 2009-12-23 | 中兴通讯股份有限公司 | Scrambling device that a kind of concurrent frame is synchronous and descrambling code device thereof |
| CN101809997A (en) * | 2007-07-02 | 2010-08-18 | Lg电子株式会社 | Broadcasting receiver and broadcast signal processing method |
| US9836484B1 (en) * | 2015-12-30 | 2017-12-05 | Google Llc | Systems and methods that leverage deep learning to selectively store images at a mobile image capture device |
| US10372991B1 (en) * | 2018-04-03 | 2019-08-06 | Google Llc | Systems and methods that leverage deep learning to selectively store audiovisual content |
| CN110430013A (en) * | 2019-07-15 | 2019-11-08 | 华中科技大学 | A kind of RCM method based on deep learning |
| CN110460402A (en) * | 2019-07-15 | 2019-11-15 | 哈尔滨工程大学 | A kind of end-to-end communication system method for building up based on deep learning |
| CN111245563A (en) * | 2019-12-30 | 2020-06-05 | 湖南迈克森伟电子科技有限公司 | GTX (GTX) no-coding transmission method and device |
| CN112214929A (en) * | 2020-09-27 | 2021-01-12 | 电子科技大学 | Radar interference suppression method for intermittent sampling repeated forwarding type interference |
| CN113630130A (en) * | 2021-07-31 | 2021-11-09 | 西南电子技术研究所（中国电子科技集团公司第十研究所） | End-to-end digital communication demodulation method |
| CN114697183A (en) * | 2022-03-31 | 2022-07-01 | 中国人民解放军国防科技大学 | Channel synchronization method based on deep learning |
-
2022
- 2022-07-28 CN CN202210897758.4A patent/CN115208722B/en active Active
Patent Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101809997A (en) * | 2007-07-02 | 2010-08-18 | Lg电子株式会社 | Broadcasting receiver and broadcast signal processing method |
| CN101610122A (en) * | 2009-07-03 | 2009-12-23 | 中兴通讯股份有限公司 | Scrambling device that a kind of concurrent frame is synchronous and descrambling code device thereof |
| US9836484B1 (en) * | 2015-12-30 | 2017-12-05 | Google Llc | Systems and methods that leverage deep learning to selectively store images at a mobile image capture device |
| US10372991B1 (en) * | 2018-04-03 | 2019-08-06 | Google Llc | Systems and methods that leverage deep learning to selectively store audiovisual content |
| CN110430013A (en) * | 2019-07-15 | 2019-11-08 | 华中科技大学 | A kind of RCM method based on deep learning |
| CN110460402A (en) * | 2019-07-15 | 2019-11-15 | 哈尔滨工程大学 | A kind of end-to-end communication system method for building up based on deep learning |
| CN111245563A (en) * | 2019-12-30 | 2020-06-05 | 湖南迈克森伟电子科技有限公司 | GTX (GTX) no-coding transmission method and device |
| CN112214929A (en) * | 2020-09-27 | 2021-01-12 | 电子科技大学 | Radar interference suppression method for intermittent sampling repeated forwarding type interference |
| CN113630130A (en) * | 2021-07-31 | 2021-11-09 | 西南电子技术研究所（中国电子科技集团公司第十研究所） | End-to-end digital communication demodulation method |
| CN114697183A (en) * | 2022-03-31 | 2022-07-01 | 中国人民解放军国防科技大学 | Channel synchronization method based on deep learning |
Non-Patent Citations (1)
| Title |
|---|
| 基于深度学习神经网络技术的数字电视监测平台告警模型的研究;韦坚;有线电视技术;全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN115208722A (en) | 2022-10-18 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106161012B (en) | A kind of quantum key distribution after-treatment system and method based on polarization code error correction | |
| CN103476026B (en) | Self-adaption covert communication method based on satellite channel coding | |
| Chen et al. | A novel digital watermarking based on general non-negative matrix factorization | |
| CN104751401B (en) | A kind of novel Piecewise linear chaotic map image encryption and coding method | |
| CN110120079B (en) | Color image encryption method based on three-dimensional Logistic mapping and generalized Cat mapping | |
| CN110677236B (en) | OFDM encryption method based on composite chaotic system and XOR operation | |
| CN107292184B (en) | Image encryption method, device and key stream generating method and key stream generator | |
| CN106027230A (en) | Method for carrying out Hamming error correction for error code in quantum secret key distribution postprocessing | |
| CN113810172A (en) | Low-redundancy physical layer encryption method and system for polarization code quantum noise stream | |
| Al-Hazaimeh et al. | Chaotic based multimedia encryption: a survey for network and internet security | |
| CN112351151A (en) | FCN-based image information encryption method and device | |
| Sun et al. | Lightweight internet of things device authentication, encryption, and key distribution using end-to-end neural cryptosystems | |
| CN115208722B (en) | Novel frame synchronous scrambling code descrambling method | |
| Molato et al. | Lsb-based random embedding image steganography technique using modified collatz conjecture | |
| Mansour et al. | Steganography-Based Transmission of Medical Images Over Unsecure Network for Telemedicine Applications. | |
| WO2020156641A1 (en) | Device and method for processing data of a quantum key distribution system | |
| Agradriya et al. | Audio watermarking technique based on Arnold transform | |
| CN1889428A (en) | Biological characteristic data safety hidden transmitting method under network environment | |
| CN111082921B (en) | Method for improving second-order power consumption analysis of password chip through first-order leakage model | |
| CN107959650A (en) | A kind of blind frame synchornization method towards low bit- rate BCH code | |
| Mishra et al. | Security enhanced digital image steganography based on successive Arnold transformation | |
| Günlü et al. | Reliable secret-key binding for physical unclonable functions with transform coding | |
| Hamad | Hiding Text Information in a Digital Image Based on Entropy Function. | |
| Yan et al. | Research on the capacity of error-correcting codes-based information hiding | |
| CN113037781A (en) | RNN-based voice information encryption method and device |
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 |