CN112564834A - Intelligent cognition and interference method and system for wireless communication system - Google Patents
Intelligent cognition and interference method and system for wireless communication system Download PDFInfo
- Publication number
- CN112564834A CN112564834A CN202011240132.3A CN202011240132A CN112564834A CN 112564834 A CN112564834 A CN 112564834A CN 202011240132 A CN202011240132 A CN 202011240132A CN 112564834 A CN112564834 A CN 112564834A
- Authority
- CN
- China
- Prior art keywords
- wireless communication
- interference
- level
- signal
- learning
- 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.)
- Granted
Links
Images
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04B—TRANSMISSION
- H04B17/00—Monitoring; Testing
- H04B17/30—Monitoring; Testing of propagation channels
- H04B17/309—Measuring or estimating channel quality parameters
- H04B17/345—Interference values
Abstract
The invention discloses an intelligent cognition and interference method and system for a wireless communication system, wherein a closed intelligent cognition and interference flow is formed by designing wireless communication signal characteristic extraction, wireless communication signal parameter identification, wireless communication signal reverse analysis, countermeasure decision, interference generation based on gene programming and the like. The method can be used for intelligently recognizing and interfering the wireless communication system on the carriers such as unmanned aerial vehicles, unmanned ships and the like, and solves the problems of incompleteness, no system, poor environmental adaptability, single means and the like of the intelligent interference system in the links of feature extraction, signal identification, reverse analysis, interference decision, interference generation and the like.
Description
Technical Field
The invention belongs to the technical field of wireless communication, and particularly relates to an intelligent cognition and interference method and system for a wireless communication system.
Background
With the rapid development of cognitive radio technology, wireless communication systems gradually have adaptive capacity to the environment, and the difficulty of effectively interfering the wireless communication systems is increasing. The interference mode of the traditional adaptive interference system is relatively fixed, and the bandwidth, the frequency point and the like are mostly used as the basis for mode switching; the agility is insufficient, and the change adjustment of the signal of the wireless communication system is difficult to detect or the reaction is delayed; interference is of limited effectiveness, and only signal level interference, but not information level interference, can be achieved by means of power throttling and simple deceptive interference.
Disclosure of Invention
The technical problem solved by the invention is as follows: the method and the system for intelligent cognition and interference for the wireless communication system have the advantages of higher cognition speed, higher accuracy and various interference modes, and have stronger comprehensive interference capability.
The purpose of the invention is realized by the following technical scheme: an intelligent cognition and interference method oriented to a wireless communication system, the method comprises the following steps: (1) extracting the characteristics of the wireless communication signals in a sample library, processing the wireless communication signals from the angles of multiple levels, multiple dimensions and multiple dimensions to obtain multiple-level characteristics, multiple-dimension characteristics and multiple-dimension characteristics, and constructing a characteristic set by the multiple-level characteristics, the multiple-dimension characteristics and the multiple-dimension characteristics; (2) performing signal level parameter identification on the received wireless communication signal through transfer learning or on-line semi-supervised learning according to signal level characteristics in multi-level characteristics in a characteristic set in combination with multi-dimensional characteristics and multi-scale characteristics to obtain signal level parameters; (3) based on the signal level parameters, according to the information level characteristics in the multi-level characteristics in the characteristic set, combining the corresponding multi-dimensional characteristics and multi-scale characteristics, performing information level reverse analysis on the wireless communication signals received this time in a mode of combining expert experience and transfer learning, realizing corresponding coding, encryption and protocol analysis, and reversely deducing part or all information level parameters of the wireless communication signals received this time; (4) the signal level parameters, part or all of the information level parameters and the last interference effect evaluation condition are used as input to obtain the most effective interference mode for the wireless communication signals received at this time; (5) the signal level parameters, some or all of the information level parameters are added to the feature set to enrich the feature set.
In the above intelligent cognition and interference method for a wireless communication system, in step (1), the multi-level features include signal level features and information level features; the signal level characteristics comprise frequency points, bandwidth, system and modulation characteristics, and the information level characteristics comprise coding, encryption and protocol characteristics.
In the above intelligent cognition and interference method for the wireless communication system, in the step (2), the migration learning is that the previous learning experience is stored in a migration learning experience library, and an optimal migration algorithm required by the current environment is selected by using a learning migration algorithm, so that the efficiency of algorithm selection is improved, and the rapid adaptation of the model is realized.
In the above intelligent cognition and interference method for the wireless communication system, in the step (2), the online semi-supervised learning is measured by Euclidean distance by using data similarity in a transformation space, and the sample library and the neural network model are updated online, so that the operation complexity is controlled, and the training efficiency is improved.
In the above intelligent cognition and interference method for the wireless communication system, in the step (3), the combination of the expert experience and the transfer learning is to extract knowledge in an expert experience base to form feature description about a wireless communication signal, and the feature description is used as a small amount of labeled data to be transferred and learned to a current target domain to participate in training of a current target domain recognition model; wherein, the expert experience base is used for storing the knowledge and experience of the expert.
In the above intelligent cognition and interference method for the wireless communication system, in step (4), a gene programming mode is adopted to generate interference to the wireless communication signal received this time.
An intelligent cognitive and interference system oriented to a wireless communication system, comprising: the signal feature extraction module is used for extracting the wireless communication signal features in the sample library, processing the wireless communication signals from the angles of multiple levels, multiple dimensions and multiple dimensions to obtain multiple-level features, multiple-dimension features and multiple-dimension features, and constructing a feature set by the multiple-level features, the multiple-dimension features and the multiple-dimension features; the signal parameter identification module is used for carrying out signal level parameter identification on the wireless communication signal received this time through transfer learning or on-line semi-supervised learning according to the combination of signal level characteristics in multi-level characteristics in a characteristic set and multi-dimensional characteristics and multi-scale characteristics to obtain signal level parameters; the signal reverse analysis module is used for performing information-level reverse analysis on the received wireless communication signal in a mode of combining expert experience and transfer learning according to information-level characteristics in multi-level characteristics in a characteristic set by taking the signal-level parameters as a basis and combining corresponding multi-dimensional and multi-scale characteristics, so that corresponding coding, encryption and protocol analysis are realized, and part or all of the information-level parameters of the received wireless communication signal are reversely conjectured; the interference generation module is used for taking the signal level parameters, part or all of the information level parameters and the last interference effect evaluation condition as input to obtain the most effective interference mode for the wireless communication signals received this time; and the feature-rich set matching module is used for adding the signal level parameters and part or all of the information level parameters into the feature set to enrich the feature set.
In the intelligent cognition and interference system facing the wireless communication system, the multi-level characteristics comprise signal level characteristics and information level characteristics; the signal level characteristics comprise frequency points, bandwidth, system and modulation characteristics, and the information level characteristics comprise coding, encryption and protocol characteristics.
In the intelligent cognition and interference system oriented to the wireless communication system, the migration learning is that the previous learning experience is stored in a migration learning experience library, and the learning migration algorithm is used for selecting the optimal migration algorithm required by the current environment, so that the algorithm selection efficiency is improved, and the rapid adaptation of the model is realized.
In the intelligent cognition and interference system oriented to the wireless communication system, the on-line semi-supervised learning is measured by Euclidean distance by utilizing data similarity in a transformation space, and the sample library and the neural network model are updated on line, so that the operation complexity is controlled, and the training efficiency is improved.
Compared with the prior art, the invention has the following beneficial effects:
(1) compared with the traditional interference system, the method has the advantages of higher cognitive speed, higher accuracy, various interference modes and stronger comprehensive interference capability;
(2) the invention provides five execution links comprising feature extraction, signal identification, reverse analysis, countermeasure decision and interference generation, and a complete closed loop can be formed by the interference evaluation of a wireless communication system;
(3) the invention provides a method for realizing deep deconstruction of wireless communication signals by aiming at signal level characteristics (frequency point, bandwidth, system, modulation and the like) and information level characteristics (coding, encryption, decryption, protocol and the like) of the wireless communication signals from two levels of signal identification and reverse analysis;
(4) the interference generation of the invention adopts a gene programming mode, fully utilizes the signal level characteristics and the information level characteristics for the interference waveform generation, and can support more effective interference modes such as deception, combined type, invasion and the like.
Drawings
Various other advantages and benefits will become apparent to those of ordinary skill in the art upon reading the following detailed description of the preferred embodiments. The drawings are only for purposes of illustrating the preferred embodiments and are not to be construed as limiting the invention. Also, like reference numerals are used to refer to like parts throughout the drawings. In the drawings:
FIG. 1 is a general flow diagram of an intelligent cognitive and interference method of the present invention;
FIG. 2 is a flow chart of wireless communication signal feature extraction according to the present invention;
FIG. 3 illustrates a wireless communication signal identification and reverse parsing process according to the present invention;
FIG. 4 is a flow diagram of an intelligent confrontation decision process according to the present invention;
FIG. 5 shows the interference generation process based on gene programming according to the present invention.
Detailed Description
Exemplary embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. While exemplary embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure may be embodied in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the disclosure to those skilled in the art. It should be noted that the embodiments and features of the embodiments may be combined with each other without conflict. The present invention will be described in detail below with reference to the embodiments with reference to the attached drawings.
Fig. 1 is a general flow of the intelligent cognitive and interference method of the present invention. As shown in fig. 1, the intelligent cognitive and interference method for a wireless communication system includes the following steps:
(1) extracting the characteristics of the wireless communication signals in a sample library, processing the wireless communication signals from the angles of multiple levels, multiple dimensions and multiple dimensions to obtain multiple-level characteristics, multiple-dimension characteristics and multiple-dimension characteristics, and constructing a characteristic set by the multiple-level characteristics, the multiple-dimension characteristics and the multiple-dimension characteristics;
(2) performing signal level parameter identification on the received wireless communication signal through transfer learning or on-line semi-supervised learning according to signal level characteristics in multi-level characteristics in a characteristic set in combination with multi-dimensional characteristics and multi-scale characteristics to obtain signal level parameters;
(3) based on the signal level parameters, according to the information level characteristics in the multi-level characteristics in the characteristic set, combining the corresponding multi-dimensional characteristics and multi-scale characteristics, performing information level reverse analysis on the wireless communication signals received this time in a mode of combining expert experience and transfer learning, realizing corresponding coding, encryption and protocol analysis, and reversely deducing part or all information level parameters of the wireless communication signals received this time;
(4) the signal level parameters, part or all of the information level parameters and the last interference effect evaluation condition are used as input to obtain the most effective interference mode for the wireless communication signals received at this time;
(5) the signal level parameters, some or all of the information level parameters are added to the feature set to enrich the feature set.
In the step (4), interference is generated by adopting a gene programming mode, and the signal level and information level parameters of the interference can be adjusted according to the requirement and act on a wireless communication system; and finally, the characteristic change condition of the wireless communication signal at the next time is used as the evaluation of the interference effect at this time and as the input of the interference decision at the next time, so that the single flow closed loop is completed.
And repeating the process from feature extraction to interference effect evaluation on the received wireless communication signals each time, and realizing iterative cycle of the whole process.
Through five links, an interference signal is generated, and the interference signal acts on the wireless communication system, so that the performance of the wireless communication system is reduced or the adaptability of the wireless communication signal is adjusted. The whole intelligent cognition and interference method facing the wireless communication system and the wireless communication system with the function form a complete closed loop.
The method comprises five parts of wireless communication signal feature extraction, wireless communication signal parameter identification, wireless communication signal reverse analysis, countermeasure decision and interference generation based on gene programming.
The method comprises the steps of extracting, screening and establishing a set, wherein the steps of extracting, screening and establishing the set are performed by multi-level, multi-dimension, multi-scale signal level and information level fine and accurate characteristic extraction.
The multi-level characteristics comprise signal level characteristics such as frequency points, bandwidth, system and modulation, and information level characteristics such as coding, encryption and protocol. The multi-dimensional features comprise time domain, frequency domain, wavelet domain, fractional Fourier domain and other dimensional features. The multi-scale features comprise features after transformation processing such as multi-scale wavelet transformation, multi-partition transformation and the like.
The extraction step mainly extracts characteristics of multiple layers, multiple dimensions and multiple scales of the wireless communication signals. The screening step is mainly to evaluate the feature subset formed by the features, and obtain the features meeting the requirements by utilizing various selection algorithms. The step of constructing the set mainly carries out structural description on the features, and classifies and induces the features to generate the feature set.
The wireless communication signal parameter identification is used for improving the rapid adaptation and efficient training of the cognitive model to a new sample, a new environment and new equipment through technologies such as transfer learning and on-line semi-supervised learning and the like aiming at the problems that the existing cognitive model is poor in robustness to changes of unknown environments and sensing equipment and low in training efficiency and the like, and provides input for subsequent confrontation decisions.
The migration learning method comprises the steps of storing the past learning experience into a migration learning experience library, and selecting the optimal migration algorithm required by the current environment by using the learning migration algorithm, so that the algorithm selection efficiency is improved, and the rapid adaptation of the model is realized. On-line semi-supervised learning measures by Euclidean distance by using data similarity in a transformation space, and controls operation complexity and improves training efficiency by updating a sample library and a neural network model on line.
The wireless communication signal reverse analysis method aims at the problems that the existing cognitive model is lack of insufficient information level identification capability and the like, improves the reverse analysis capability of the cognitive model on information through a mode of combining expert experience and transfer learning, and provides input for subsequent countermeasures.
And the mode of combining the expert experience with the transfer learning is used for extracting knowledge in an expert experience library to form feature description about the wireless communication signal, and the feature description is used as a small amount of labeled data to be transferred and learned to a current target domain to participate in training a current target domain recognition model. The expert experience base is mainly used for storing the knowledge and experience of the experts and systematically expressing the knowledge and experience.
And (3) countermeasure decision, aiming at the influence of information incompleteness, environmental uncertainty and the like on the decision, realizing automatic generation of countermeasure strategies, defining interference modes and parameters of interference signals, and providing input for subsequent interference generation based on gene programming.
The automatic generation of the countermeasure is realized by the intersection of two processes of perception-decision-implementation and perception-feedback-improvement: the perception is that the parameter identification and reverse analysis result of the wireless communication signal is used as the input of decision; the decision is to select the optimal interference strategy according to an interference strategy library generated by intelligent countermeasure training and the requirement of an interference task; "implement" is the subsequent generation of interference based on genetic programming and implementation to the target wireless communication system; "feedback" evaluates the "perceived" condition of the continuous wireless communication signal; and improving, and dynamically feeding back an evaluation result as an input of decision.
Interference generation based on gene programming aims at the problems of simple interference pattern, single means and the like, and flexibly configures the interference pattern, parameters and the like through a gene programming strategy to meet different requirements of signal-level and information-level interference.
The gene programming strategy firstly determines interference patterns including suppression, deception, invasion and the like; then, parameter configuration is carried out on different interference patterns; and finally, optimizing the interference in detail to generate usable interference.
Fig. 2 is a flow chart of wireless communication signal feature extraction according to the present invention. Firstly, extracting the characteristics of a received signal, and successively extracting from multiple levels (including frequency points, bandwidths, systems, adjustment and the like of signal levels, coding, encryption, protocols and the like of information levels), multiple dimensions (including time domain, frequency domain, wavelet domain, fractional Fourier domain and the like), multiple dimensions (including multiple-dimension wavelet transform, multiple-dimension parting transform and the like); then screening the extracted features, generating a feature subset and evaluating the feature subset by using a feature screening method such as RELIEF-F, F Score, Laplacian Score and the like, and finding out the minimum feature combination which can describe the category most at tolerable cost in an allowable time; and finally, constructing a feature set, dividing the feature set into large layers according to the layers, dimensions and dimensions, and specifically describing the sub-layers to form the structural representation of the features.
Fig. 3 is a flow chart of wireless communication signal identification and reverse analysis according to the present invention. The identification and reverse analysis of the wireless communication signals follow the same flow, and the characteristic extraction is firstly carried out on the wireless communication signals in the source domain environment; then, the expert system extracts knowledge from an expert experience base, and the feature description about the target to be detected is obtained after recombination; in the transfer learning link, a transfer learning algorithm (including feature space based, classifier based, deep neural network based and the like) is learned according to sign extraction and expert feature description; obtaining a transfer learning algorithm which is most suitable for the new environment of the target domain according to the new environment of the target domain; and finally, updating the model of the target domain in an online semi-supervised learning link, updating the feature set, and feeding back the feature set to a feature extraction link to form a closed flow loop.
Fig. 4 is a flow of an intelligent confrontation decision according to the present invention. And (3) realizing the cross of two sub-processes of perception-decision-implementation and perception-feedback-improvement. Signals of the wireless communication system enter a countermeasure decision link through sensing (including feature extraction, signal identification, reverse analysis and the like), a decision is generated (including an interference pattern, a signal level, information level parameter configuration and the like), and the interference generation link correspondingly implements to generate corresponding interference to act on the wireless communication system. Signals of the wireless communication system enter an evaluation link through feedback, and the change conditions of signal systems, parameters, signal quality and the like caused by the influence of the last interference on the wireless communication system are evaluated, so that countermeasure decisions are fed back in an improved mode.
FIG. 5 shows the interference generation process based on gene programming according to the present invention. Firstly, determining interference types including suppression, deception, composition and the like according to a countermeasure decision; genetic programming of the interference waveform is then performed: for suppressing interference, selecting a waveform from a signal level waveform library according to a signal level parameter, and calibrating the signal level interference according to a radio frequency parameter; for deception interference, on the basis of a corresponding signal level waveform, adding a corresponding information part from an information level waveform library according to an information level parameter, and calibrating an information level interference waveform according to a radio frequency parameter; for the composite interference, a combination of signal level interference and information level interference is generated, and the composite interference is calibrated according to the radio frequency parameters.
The embodiment also provides an intelligent cognition and interference system facing the wireless communication system, which comprises a signal feature extraction module, a signal feature extraction module and a signal feature extraction module, wherein the signal feature extraction module is used for extracting the wireless communication signal features in a sample library, processing the wireless communication signal from the angles of multiple levels, multiple dimensions and multiple scales to obtain multiple levels of features, multiple dimensions of features and multiple scales of features, and constructing a feature set by the multiple levels of features, the multiple dimensions of features and the multiple scales of features; the signal parameter identification module is used for carrying out signal level parameter identification on the wireless communication signal received this time through transfer learning or on-line semi-supervised learning according to the combination of signal level characteristics in multi-level characteristics in a characteristic set and multi-dimensional characteristics and multi-scale characteristics to obtain signal level parameters; the signal reverse analysis module is used for performing information-level reverse analysis on the received wireless communication signal in a mode of combining expert experience and transfer learning according to information-level characteristics in multi-level characteristics in a characteristic set by taking the signal-level parameters as a basis and combining corresponding multi-dimensional and multi-scale characteristics, so that corresponding coding, encryption and protocol analysis are realized, and part or all of the information-level parameters of the received wireless communication signal are reversely conjectured; the interference generation module is used for taking the signal level parameters, part or all of the information level parameters and the last interference effect evaluation condition as input to obtain the most effective interference mode for the wireless communication signals received this time; and the feature-rich set matching module is used for adding the signal level parameters and part or all of the information level parameters into the feature set to enrich the feature set.
In the above embodiment, the multi-level features include signal level features and information level features; the signal level characteristics comprise frequency points, bandwidth, system and modulation characteristics, and the information level characteristics comprise coding, encryption and protocol characteristics.
In the embodiment, the migration learning is to store the past learning experience into the migration learning experience library, and select the optimal migration algorithm required by the current environment by using the learning migration algorithm, so as to improve the efficiency of algorithm selection and realize the rapid adaptation of the model. The on-line semi-supervised learning is to measure by Euclidean distance by utilizing data similarity in a transformation space, and to control the operation complexity and improve the training efficiency by updating a sample library and a neural network model on line.
Compared with the traditional interference system, the method has the advantages of higher cognitive speed, higher accuracy, various interference modes and stronger comprehensive interference capability; the invention provides five execution links comprising feature extraction, signal identification, reverse analysis, countermeasure decision and interference generation, and a complete closed loop can be formed by the interference evaluation of a wireless communication system; the invention provides a method for realizing deep deconstruction of wireless communication signals by aiming at signal level characteristics (frequency point, bandwidth, system, modulation and the like) and information level characteristics (coding, encryption, decryption, protocol and the like) of the wireless communication signals from two levels of signal identification and reverse analysis; the interference generation of the invention adopts a gene programming mode, fully utilizes the signal level characteristics and the information level characteristics for the interference waveform generation, and can support more effective interference modes such as deception, combined type, invasion and the like.
Although the present invention has been described with reference to the preferred embodiments, it is not intended to limit the present invention, and those skilled in the art can make variations and modifications of the present invention without departing from the spirit and scope of the present invention by using the methods and technical contents disclosed above.
Claims (10)
1. An intelligent cognition and interference method oriented to a wireless communication system is characterized by comprising the following steps:
(1) extracting the characteristics of the wireless communication signals in a sample library, processing the wireless communication signals from the angles of multiple levels, multiple dimensions and multiple dimensions to obtain multiple-level characteristics, multiple-dimension characteristics and multiple-dimension characteristics, and constructing a characteristic set by the multiple-level characteristics, the multiple-dimension characteristics and the multiple-dimension characteristics;
(2) performing signal level parameter identification on the received wireless communication signal through transfer learning or on-line semi-supervised learning according to signal level characteristics in multi-level characteristics in a characteristic set in combination with multi-dimensional characteristics and multi-scale characteristics to obtain signal level parameters;
(3) based on the signal level parameters, according to the information level characteristics in the multi-level characteristics in the characteristic set, combining the corresponding multi-dimensional characteristics and multi-scale characteristics, performing information level reverse analysis on the wireless communication signals received this time in a mode of combining expert experience and transfer learning, realizing corresponding coding, encryption and protocol analysis, and reversely deducing part or all information level parameters of the wireless communication signals received this time;
(4) the signal level parameters, part or all of the information level parameters and the last interference effect evaluation condition are used as input to obtain the most effective interference mode for the wireless communication signals received at this time;
(5) the signal level parameters, some or all of the information level parameters are added to the feature set to enrich the feature set.
2. The intelligent cognitive and interference method oriented to wireless communication systems of claim 1, wherein: in step (1), the multi-level features comprise signal-level features and information-level features; the signal level characteristics comprise frequency points, bandwidth, system and modulation characteristics, and the information level characteristics comprise coding, encryption and protocol characteristics.
3. The intelligent cognitive and interference method oriented to wireless communication systems of claim 1, wherein: in the step (2), the migration learning is to store the past learning experience into a migration learning experience library, and select the optimal migration algorithm required by the current environment by using the learning migration algorithm, so as to improve the efficiency of algorithm selection and realize the rapid adaptation of the model.
4. The intelligent cognitive and interference method oriented to wireless communication systems of claim 1, wherein: in the step (2), the on-line semi-supervised learning is to measure by Euclidean distance by utilizing data similarity in a transformation space, and to control the operation complexity and improve the training efficiency by updating the sample library and the neural network model on line.
5. The intelligent cognitive and interference method oriented to wireless communication systems of claim 1, wherein: in the step (3), the knowledge in the expert experience base is extracted to form a feature description about the wireless communication signal, and the feature description is taken as a small amount of labeled data to be migrated and learned to the current target domain to participate in training of the current target domain recognition model; wherein, the expert experience base is used for storing the knowledge and experience of the expert.
6. The intelligent cognitive and interference method oriented to wireless communication systems of claim 1, wherein: in the step (4), interference is generated on the wireless communication signal received this time by adopting a gene programming mode.
7. An intelligent cognitive and interference system oriented to a wireless communication system, characterized by comprising:
the signal feature extraction module is used for extracting the wireless communication signal features in the sample library, processing the wireless communication signals from the angles of multiple levels, multiple dimensions and multiple dimensions to obtain multiple-level features, multiple-dimension features and multiple-dimension features, and constructing a feature set by the multiple-level features, the multiple-dimension features and the multiple-dimension features;
the signal parameter identification module is used for carrying out signal level parameter identification on the wireless communication signal received this time through transfer learning or on-line semi-supervised learning according to the combination of signal level characteristics in multi-level characteristics in a characteristic set and multi-dimensional characteristics and multi-scale characteristics to obtain signal level parameters;
the signal reverse analysis module is used for performing information-level reverse analysis on the received wireless communication signal in a mode of combining expert experience and transfer learning according to information-level characteristics in multi-level characteristics in a characteristic set by taking the signal-level parameters as a basis and combining corresponding multi-dimensional and multi-scale characteristics, so that corresponding coding, encryption and protocol analysis are realized, and part or all of the information-level parameters of the received wireless communication signal are reversely conjectured;
the interference generation module is used for taking the signal level parameters, part or all of the information level parameters and the last interference effect evaluation condition as input to obtain the most effective interference mode for the wireless communication signals received this time;
and the feature-rich set matching module is used for adding the signal level parameters and part or all of the information level parameters into the feature set to enrich the feature set.
8. The intelligent cognitive and interference system oriented to wireless communication systems of claim 7, wherein: the multi-level features include signal level features and information level features; the signal level characteristics comprise frequency points, bandwidth, system and modulation characteristics, and the information level characteristics comprise coding, encryption and protocol characteristics.
9. The intelligent cognitive and interference system oriented to wireless communication systems of claim 7, wherein: the migration learning is that the previous learning experience is stored in a migration learning experience library, and an optimal migration algorithm required by the current environment is selected by using a learning migration algorithm, so that the efficiency of algorithm selection is improved, and the rapid adaptation of the model is realized.
10. The intelligent cognitive and interference system oriented to wireless communication systems of claim 7, wherein: the on-line semi-supervised learning is to measure by Euclidean distance by utilizing data similarity in a transformation space, and to control the operation complexity and improve the training efficiency by updating a sample library and a neural network model on line.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011240132.3A CN112564834B (en) | 2020-11-09 | 2020-11-09 | Intelligent cognition and interference method and system for wireless communication system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011240132.3A CN112564834B (en) | 2020-11-09 | 2020-11-09 | Intelligent cognition and interference method and system for wireless communication system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112564834A true CN112564834A (en) | 2021-03-26 |
| CN112564834B CN112564834B (en) | 2023-02-03 |
Family
ID=75041679
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011240132.3A Active CN112564834B (en) | 2020-11-09 | 2020-11-09 | Intelligent cognition and interference method and system for wireless communication system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112564834B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116484180A (en) * | 2023-06-21 | 2023-07-25 | 中国人民解放军国防科技大学 | System and method for extracting communication signal gene |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20080293353A1 (en) * | 2007-03-08 | 2008-11-27 | Mody Apurva N | Cognitive radio methodology, physical layer policies and machine learning |
| CN104468252A (en) * | 2013-09-23 | 2015-03-25 | 重庆康拜因科技有限公司 | Intelligent network service identification method based on positive transfer learning |
| CN108508411A (en) * | 2018-03-22 | 2018-09-07 | 天津大学 | Passive radar external sort algorithm signal recognition method based on transfer learning |
| US20180357552A1 (en) * | 2016-01-27 | 2018-12-13 | Bonsai AI, Inc. | Artificial Intelligence Engine Having Various Algorithms to Build Different Concepts Contained Within a Same AI Model |
-
2020
- 2020-11-09 CN CN202011240132.3A patent/CN112564834B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20080293353A1 (en) * | 2007-03-08 | 2008-11-27 | Mody Apurva N | Cognitive radio methodology, physical layer policies and machine learning |
| CN104468252A (en) * | 2013-09-23 | 2015-03-25 | 重庆康拜因科技有限公司 | Intelligent network service identification method based on positive transfer learning |
| US20180357552A1 (en) * | 2016-01-27 | 2018-12-13 | Bonsai AI, Inc. | Artificial Intelligence Engine Having Various Algorithms to Build Different Concepts Contained Within a Same AI Model |
| CN108508411A (en) * | 2018-03-22 | 2018-09-07 | 天津大学 | Passive radar external sort algorithm signal recognition method based on transfer learning |
Non-Patent Citations (4)
| Title |
|---|
| MARC LICHTMAN等: "一种通信干扰分类方法", 《通信对抗》 * |
| 王莉萍: "迁移学习相关知识介绍", 《电脑迷》 * |
| 逄天洋等: "通信电子干扰的分类与发展", 《通信技术》 * |
| 陈康等: "大数据时代机器学习的新趋势", 《电信科学》 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116484180A (en) * | 2023-06-21 | 2023-07-25 | 中国人民解放军国防科技大学 | System and method for extracting communication signal gene |
| CN116484180B (en) * | 2023-06-21 | 2023-09-22 | 中国人民解放军国防科技大学 | System and method for extracting communication signal gene |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112564834B (en) | 2023-02-03 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Zhang et al. | Walking on the edge: Fast, low-distortion adversarial examples | |
| US6865582B2 (en) | Systems and methods for knowledge discovery in spatial data | |
| Zaknich | Introduction to the modified probabilistic neural network for general signal processing applications | |
| CN111199127B (en) | Radar interference decision method based on deep reinforcement learning | |
| CN111767848A (en) | Radiation source individual identification method based on multi-domain feature fusion | |
| Bacanin et al. | RFID network planning by ABC algorithm hybridized with heuristic for initial number and locations of readers | |
| KR102284356B1 (en) | Data imbalance solution method using Generative adversarial network | |
| CN109444831B (en) | Radar interference decision method based on transfer learning | |
| CN108924847B (en) | Cognitive radio frequency spectrum prediction method and device based on ANN | |
| CN112564834B (en) | Intelligent cognition and interference method and system for wireless communication system | |
| CN113406579A (en) | Camouflage interference waveform generation method based on deep reinforcement learning | |
| Gong et al. | Multi-task based deep learning approach for open-set wireless signal identification in ISM band | |
| Wei et al. | Automatic modulation recognition using neural architecture search | |
| Zhang et al. | Limited data spectrum sensing based on semi-supervised deep neural network | |
| CN113128584B (en) | Mode-level unsupervised sorting method of multifunctional radar pulse sequence | |
| Misra et al. | Image segmentation using clustering with fireworks algorithm | |
| CN115705393A (en) | Radar radiation source grading identification method based on continuous learning | |
| Dablain et al. | Efficient augmentation for imbalanced deep learning | |
| Strumberger et al. | Bare bones fireworks algorithm for the rfid network planning problem | |
| CN110503205A (en) | A kind of multi-modal data processing method and system | |
| CN114118151A (en) | Intelligent spectrum sensing method with environment adaptive capacity | |
| Gonçalves et al. | Online learning in estimation of distribution algorithms for dynamic environments | |
| Perez et al. | Face Patches Designed through Neuroevolution for Face Recognition with Large Pose Variation | |
| Venkatesan et al. | Comparative performance investigation of supervised and unsupervised learning outlines applied in cognitive radio systems | |
| Wang et al. | Augmented convolutional neural networks with transformer for wireless interference identification |
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 |