Data classification and identification method based on unstable periodic orbit of chaotic system

Through the data classification identification method based on the unstable periodic track of the chaotic system, the problem of excessive manual participation and low classification accuracy in the existing technology is solved, and higher data classification accuracy and reliability are achieved, which is suitable for a variety of complex Data processing tasks.

CN118940106AActive Publication Date: 2024-11-12CHINA INFOMRAITON CONSULTING & DESIGNING INST CO LTD
View PDF 9 Cites 0 Cited by

Patent Information

Application Number
CN202410978692.0
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-07-22
Publication Date
2024-11-12
Estimated Expiration
2044-07-22

AI Technical Summary

Technical Problem

Existing data classification methods have excessively participated in manual processing, low classification accuracy and low portability, making it difficult to effectively process large-scale and complex data.

Method used

Using data classification and identification method based on unstable periodic orbits of chaotic system, the initial chaotic pulse signal is generated by building a chaotic mapping system, and the timing control signal is added to obtain a stable periodic orbit, and the symbolic dynamics is used to map it into a symbol sequence, and control is formulated. Rules for pattern recognition.

Benefits of technology

It improves the accuracy and reliability of data classification, enhances information processing capabilities in complex environments, improves parallel technology efficiency and data processing capabilities, and is suitable for big data hierarchical classification, image recognition, voice processing and other fields.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118940106A_ABST
    Figure CN118940106A_ABST
Patent Text Reader

Abstract

The invention discloses a data classification and identification method based on an unstable periodic orbit of a chaotic system, and the method comprises the steps: collecting to-be-classified original data in a computer system, and building a data set; constructing a chaotic mapping system, and generating an initial chaotic pulse signal for mapping the data set; adding a time sequence control signal into the chaotic mapping system, and controlling the mapped data set to obtain a stable periodic orbit; mapping the stable periodic orbit into a symbol sequence by adopting a symbol dynamics method; making a control rule, and adjusting the symbol sequence into a final chaotic pulse signal; and performing pattern classification identification on the data set based on the final chaotic pulse signal. The method is high in universality and can be used in scenes of big data grading classification, image recognition classification, voice processing classification, bioinformatics and the like.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The invention relates to a data classification and identification method, in particular to a data classification and identification method based on an unstable periodic orbit of a chaotic system. Background Art

[0002] This section merely provides background information related to the present disclosure and is not necessarily prior art.

[0003] The technical research background of the present invention mainly stems from the application of chaos theory in neuroscience and computational science. Chaos theory, as a nonlinear scientific theory, studies the complex behavior of systems under uncertain and nonlinear conditions. Chaotic systems are highly complex and sensitive, and small changes in their initial conditions can lead to huge differences in long-term behavior. This characteristic makes chaotic systems show unique advantages in data processing and storage.

[0004] In the study of chaotic dynamic characteristics of the human brain, chaos is regarded as a possible physical mechanism of self-referential logic. The activity of brain neurons exhibits complex nonlinear dynamic behavior, which is considered to be the basis for the brain to process and store information. Chaos theory provides a new perspective and method for understanding and simulating the complex behavior of brain neural networks.

[0005] In the field of computational science, traditional data processing and storage methods face many challenges when dealing with large-scale and complex data. The data classification method based on the unstable periodic orbit of chaotic systems can greatly improve the system's data classification capabilities for large data sets by utilizing the controllable and pattern-recognizable characteristics of the unstable periodic orbit of chaotic systems, providing a new way to solve these challenges.

[0006] However, the classification methods in the prior art involve too much manual effort, and have the problems of low classification accuracy and poor portability.

[0007] It should be noted that the information disclosed in the above background technology section is only used to enhance the understanding of the background of the present disclosure, and therefore may include information that does not constitute the prior art known to ordinary technicians in the field. Summary of the invention

[0008] Purpose of the invention: The technical problem to be solved by the present invention is to provide a data classification and identification method based on unstable periodic orbits of chaotic systems in view of the deficiencies in the prior art.

[0009] In order to solve the above technical problems, the present invention discloses a data classification and identification method based on unstable periodic orbits of chaotic systems, comprising the following steps:

[0010] Step 1, collect the original data to be classified in the computer system and establish a data set;

[0011] Step 2, constructing a chaotic mapping system to generate an initial chaotic pulse signal for mapping the data set;

[0012] Step 3, adding a timing control signal to the chaotic mapping system to control the mapped data set to obtain a stable periodic orbit;

[0013] Step 4, using a symbol dynamics method to map the stable periodic orbit into a symbol sequence;

[0014] Step 5, formulate control rules to adjust the symbol sequence to the final chaotic pulse signal;

[0015] Step 6: Based on the final chaotic pulse signal, pattern classification and recognition are performed on the data set of step 1 to complete the data classification and recognition based on the unstable periodic orbit of the chaotic system.

[0016] Furthermore, the construction of the chaotic mapping system described in step 2 specifically includes:

[0017] Step 2-1, constructing a chaotic mapping system, that is, using a dimensionless mathematical model to construct an impulse signal;

[0018] Step 2-2, using recursive mapping to describe the dynamic behavior of the chaotic mapping system constructed above;

[0019] Step 2-3, perform periodic mapping correspondence;

[0020] Step 2-4, using different reference signals b(τ) according to the recursive mapping description in step 2-2 to generate different mappings.

[0021] Furthermore, the dimensionless mathematical model described in step 2-1 is as follows:

[0022]

[0023] where y is the internal state of the dataset, is the state that changes over time, x is the input, τ is the normalized time unit, b(τ) is the reference signal with a period of 1, and λ is a prime or composite parameter.

[0024] Furthermore, the recursive mapping described in step 2-2 is specifically expressed as follows:

[0025] θ(n+1)=F(θ(n))=θ(n)-b(θ(n))

[0026] Where n is the number of units of the chaotic pulse signal, θ(n) is the state vector, F(θ(n)) represents the periodic point of the θ(n) state vector mapping F, and b(θ(n)) represents the periodic point of the θ(n) state vector mapping b.

[0027] Furthermore, the periodic mapping described in step 2-3 is as follows:

[0028] F: S 1 ->S′

[0029] Among them, F represents the mapping relationship, S 1 - > S' represents the mapping unit circle.

[0030] Furthermore, the generation of different mappings described in steps 2-4 specifically includes:

[0031] Select λ=1, b(τ)=-τ-θ0, θ0>0 to control the mapping, then the recursive mapping is expressed as follows:

[0032] θ(n+1)=F(θ(n))=2θ(n)+θ0(mod1)

[0033] Among them, θ0 represents recursive mapping, and mod1 represents the decimal operation.

[0034] Furthermore, the timing control signal is added to the chaotic mapping system in step 3, that is, the dimensionless mathematical model in step 2-1 is adjusted, as follows:

[0035]

[0036] Among them, c(τ) is the added timing control signal, and z is the control switch;

[0037] After adjustment, the output of the chaotic mapping system is a stable periodic orbit.

[0038] Furthermore, the step 4 of mapping the periodic orbit into a symbol sequence specifically includes:

[0039] Step 4-1, mapping the mapping relationship F into a symbol sequence z using a mapping equation n , each state vector θ(n)∈z n The mapping equation is as follows:

[0040]

[0041] Among them, Δ represents all integers;

[0042] Step 4-2, the dimensionless mathematical model in the chaotic mapping system is finally optimized as follows:

[0043]

[0044] Where i = 1, 2, 3...N, N is infinite; z = (θ(n), ω, λ) is the control switch, which is represented by the state vector θ(n) = (θ1(n), θ2(n), ...θ N (n)), window vector set ω and category key λ.

[0045] Furthermore, the formulation of control rules described in step 5 specifically includes:

[0046] The following control rules are formulated for the control switch z:

[0047] When λm≤τ<λm+1, θ(n)∈ω, then z=z(θ(n),ω,λ) is set to 1 when λm≤τ<λ(m+1), where m=1, 2, 3…λ, and λ is a prime number or a composite number.

[0048] Furthermore, the pattern classification and recognition described in step 6 specifically includes:

[0049] Use the category key λ as the basis for judgment to classify the final chaotic pulse signal described in step 5, that is, when λ≠λ j and λ=λ j , the data set is identified as different types, where λ j is λ to the jth power, where j is a random number, either a prime number or a composite number.

[0050] Beneficial effects:

[0051] The present invention obtains a stable periodic orbit by controlling the chaotic mapping system, and exhibits strong robustness when processing fuzzy, incomplete or noisy data, which enhances its information processing capabilities in complex environments; symbolic dynamics are used to map the periodic orbit into a symbol sequence, and the complex chaotic pulses are simplified into a symbol sequence that is easier to process, so that the system behavior can be better understood and the computing efficiency can be improved; the pattern recognition classification algorithm uses the characteristics of the chaotic system's sensitivity to initial conditions to enable it to distinguish between similar patterns, which helps to improve the accuracy and reliability of data classification. The data classification and identification method designed by the present invention based on the unstable periodic orbit of the chaotic system can not only improve the efficiency of parallel technology and data processing capabilities, but also bring significant beneficial effects in data storage, knowledge discovery, etc., so it can be widely used in big data classification, image recognition classification, speech processing classification, bioinformatics and other fields. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] The present invention will be further described in detail below in conjunction with the accompanying drawings and specific embodiments, and the above and / or other advantages of the present invention will become more clear.

[0053] Figure 1 It is a schematic diagram of the overall process of the present invention.

[0054] Figure 2 The trajectory of the pulse signal and the reference signal diagram of the chaotic mapping system.

[0055] Figure 3 It is a periodic mapping correspondence diagram.

[0056] Figure 4 This is a diagram of signal changes caused by regulation.

[0057] Figure 5 A sound waveform diagram in a data set in one embodiment.

[0058] Figure 6 It is a chaotic plane projection diagram after chaotic mapping of a data set in an embodiment.

[0059] Figure 7 A bifurcation diagram for data set pattern recognition in one embodiment. DETAILED DESCRIPTION

[0060] The present invention discloses a data classification and identification method based on an unstable periodic orbit of a chaotic system. In the process of implementing the method, not only a chaotic pulse signal is provided, the chaotic pulse signal is mapped into a symbol sequence by using symbol dynamics, and the unstable periodic orbit is controlled by a timing control signal, but also a pattern recognition algorithm for the unstable periodic orbit is provided. Firstly, a chaotic pulse signal is constructed, and the chaotic pulse signal is mapped into a symbol sequence by using symbol dynamics, and the unstable periodic orbit is controlled by a timing control signal so that it has the conditions for pattern recognition, and finally the pattern recognition method is used to classify the data. The present invention solves the problems of excessive manual participation, low classification accuracy and poor portability of existing classification methods.

[0061] The data classification described in the present invention refers to the classification of data in the system in the computer field. The specific technical solution is as follows: Figure 1 As shown, including:

[0062] Data set preparation, construction of chaotic mapping system, control to obtain stable periodic orbits, mapping periodic orbits into symbol sequences, formulating rules to adjust chaotic pulse signals, and using pattern recognition algorithms to classify data sets;

[0063] Among them, data set preparation: used to prepare test data for classification;

[0064] Construct a chaotic mapping system: to generate chaotic pulse signals for mapping data sets (input);

[0065] Control to obtain a stable periodic orbit: Add a timing control signal to the chaotic mapping system to control it to obtain a stable periodic orbit;

[0066] Mapping periodic orbits into symbol sequences: In order to distinguish unstable periodic orbits, symbol dynamics is used to map chaotic pulse periodic orbits into symbol sequences;

[0067] Formulate rules to adjust chaotic pulse signals: Use control rules to control chaotic pulse signals so that they have the computational conditions for pattern recognition;

[0068] Perform pattern recognition and classification on data sets: Use pattern recognition algorithms to classify data sets.

[0069] The method for data classification of the present invention specifically includes:

[0070] Step 1: Prepare the data set and participate in the subsequent classification algorithm;

[0071] Step 2, constructing a chaotic mapping system to generate a chaotic pulse signal for mapping (input) the data set;

[0072] Step 3, by adding a timing control signal, a stable periodic orbit is obtained;

[0073] Step 4, mapping the chaotic pulse periodic orbit into a symbol sequence through symbol dynamics;

[0074] Step 5, control the chaotic pulse signal by formulating rules to make it have the calculation conditions for pattern recognition;

[0075] Step 6: Use pattern recognition algorithms to classify the data.

[0076] The construction of the chaotic mapping system described in step 2 preliminarily gives a chaotic pulse signal, and the dimensionless mathematical model is as follows:

[0077]

[0078] Where y is an internal state, τ is a normalized time unit, b(τ) is a reference signal with a period of 1, and λ is a parameter. When λ = 1, the trajectory of the impulse signal of this chaotic mapping system and the reference signal are shown in the figure below: Figure 2 ;

[0079] The chaotic mapping system described in step 2 is constructed as follows: Figure 2, when τ=τ(l), the trajectory reaches the threshold value y=0, and the next step is that the system will immediately jump to the reference signal b(τ(l)), and then the system will output another pulse signal, and repeat this process continuously. Therefore, the dynamic behavior of the system can be described as θ(n+1)=F(θ(n))=θ(n)-b(θ(n)), where n is the number of chaotic pulse neural network units, and θ(n) is a state vector. Such a periodic F:S 1 ->The mapping relationship of S′ is as follows Figure 3 .

[0080] Using different reference signals b(τ) will produce different mappings according to the above formula. In order to make the mapping controllable, a simple case of λ=1, b(τ)=--τ-θ0, θ0>0 is selected;

[0081] When p=1, b(τ)=--τ-θ0, and the chaotic pulse signal for 0≤τ<1, θ0>0 can be expressed as: θ(n+1)=F(θ(n))=2θ(n)+θ0(mod l). The above is the chaotic mapping system constructed by the present invention.

[0082] The control described in step 3 obtains a stable periodic orbit. The chaotic mapping system obtained in step 2 has two important characteristics (1) j is any integer; (2) It is not a constant value. This means that the unstable periodic orbits on S′ are all periodic orbits with a period of Q, and their corresponding periods in the time domain are always different. In order to obtain a stable periodic orbit, the present invention adopts a timing control signal to control the chaotic mapping system;

[0083] In step 3, the chaotic mapping system is controlled to obtain a stable periodic orbit. Here, the present invention adjusts the infinite equation to:

[0084]

[0085] c(τ) is the timing control signal, and z is the control switch. When θ0=0, c0=Δ / 2, the present invention can derive the recursive mapping period T of the system on the unit circle under control, and its expression is: θ(n+1)=T(θ(n))=2qΔ-Δ / 2, θ(n)∈w q , where w q =((q-1)Δ, qΔ), q=1, 2, 3...2 q -1. Figure 4 Given Q=2 is an example of a mapping T, where the dotted line represents the mapping F Δ / 2 , that is, the mapping F with θ0 = Δ / 2.

[0086] The use of symbol dynamics in step 4 to map the periodic orbit into a symbol sequence is mainly to use the mapping F Equations are mapped into a specific sequence of symbols. Here each The initial periodic orbits all use a unique symbol vector z n =(s1, s2, ...s q ) to indicate.

[0087] The chaotic pulse signal is controlled by formulating rules in step 5, and is optimized by optimizing the infinite formula:

[0088]

[0089] Where N is the number of grid cells, z = (θ(n), ω, λ) is a control switch, and z is represented by the state vector θ(n) = (θ1(n), θ2(n), ...θ N (n)), the window vector set ω and is a function of the class key λ;

[0090] The control rules described in step 5 need to meet the calculation condition that all elements in the same window vector must represent unstable periodic orbits with the same time period in order to ensure that the chaotic pulse signal has pattern recognition, and this time period is a prime number, that is, The periodic key function is used to participate in the calculation, where λ n is a prime number, i, j = 1, 2, 3... N. The following control rule needs to be formulated for the control switch z: suppose for a certain n, when λm≤τ<λm+1, θ(n)∈ω, then z=z(θ(n),ω,λ) is set to 1 when λm≤τ<λ(m+1), where m=0, 1..., λ is a prime number.

[0091] The pattern recognition algorithm in step 6 classifies the data. After the control rules in step 5, the system will work in the following two ways:

[0092] When θ(n)∈ω j ,λm≤τ<λm+1 andλ≠λ j According to the control rule of step 4, when λm≤τ<λm+1, z is equal to 1. At this time, all units are controlled and the system trajectory changes from The unstable cycle starts, but since λ≠λ j Moreover, λ is a prime number. When λ(m+1)≤τ<λ(m+1)+1, z is equal to 0, and the system trajectory cannot always be stable on the periodic orbit.

[0093] When θ(n)∈ω j , λm≤τ<λm+1 and λ=λ jAccording to the control rule of step 4, when λm≤τ<λ(m+1), z is equal to 1, and when λ(m+1)≤τ<λ(m+1), the network is always in a controlled state, and the system trajectory is always stable from In the initial stable period, different categories in the data set can be stably restored.

[0094] After the above two identification processes, the data is divided into two categories by different time periods of unstable periods in the time domain. In this way, according to the time domain period of the above system trajectory as the system feature, as long as the time domain periods of the two types of patterns are different, the control rules of step 5 can be used to control the chaotic pulse signal, so that two different types of patterns can be identified. If multiple patterns need to be identified, the parallel identification calculation method can be adopted, that is, grouping in pairs and identifying multiple nodes at the same time, thereby speeding up the efficiency of data classification.

[0095] Embodiment 1:

[0096] One embodiment of the present invention proposes a data classification and identification method based on unstable periodic orbits of chaotic systems, see Figure 1 , the data classification process includes the following steps:

[0097] Step 1: Input the data to be processed and establish a data set for data classification based on the data.

[0098] Step 2 constructs a chaotic mapping system to generate a chaotic pulse signal for mapping (input) a data set. The present invention uses a dimensionless mathematical model Construct a chaotic system to generate chaotic impulse signals.

[0099] Step 3 adds a timing control signal to obtain a stable periodic orbit. Based on the two important characteristics of chaotic pulse signals (1) j is any integer; (2) It is not a constant value. It can be deduced that unstable periodic orbits on S′ are all periodic orbits with a period of Q. Their corresponding periods in the time domain are always different. In order to obtain stable periodic orbits, the present invention adopts a timing control signal added to the chaotic mapping system for control.

[0100] Step 4 maps the chaotic pulse periodic orbit into a symbol sequence through symbol dynamics. Here, the equation is used:

[0101]

[0102] The chaotic pulse F is mapped into a specific symbol sequence.

[0103] Step 5 controls the chaotic pulse signal by formulating rules to make it meet the calculation conditions for pattern recognition. In order to ensure that the chaotic pulse signal has pattern recognition, it is necessary to meet the calculation conditions that all elements in the same window vector must represent unstable periodic orbits with the same time period, and this time period is a prime number, that is, The periodic key function is used to participate in the calculation, where λ n is a prime number, i, j = 1, 2, 3... N. The following control rule needs to be formulated for the control switch z: suppose for a certain n, when λm≤τ<λm+1, θ(n)∈ω, then z=z(θ(n),ω,λ) is set to 1 when λm≤τ<λ(m+1), where m=0, 1..., λ is a prime number.

[0104] Step 6 uses pattern recognition algorithms to classify the data. After the control of the control switch z, the system will work in the following two ways:

[0105] (1) When θ(n)∈ω j ,λm≤τ<λm+1 andλ≠λ j , when λm≤τ<λm+1, z is equal to 1, at this time all units are controlled, and the system trajectory changes from However, since λ≠λj and λ is a prime number, when λ(m+1)≤τ<λ(m+1)+1, z is equal to 0, and the system trajectory cannot always be stable on the periodic orbit.

[0106] (2) When θ(n)∈ω j , λm≤τ<λm+1 and λ=λ j , when λm≤τ<λ(m+1), z is equal to 1, and when λ(m+1)≤τ<λ(m+1), the chaotic pulse signal is always in a controlled state, and the system trajectory is always stable from In the initial stable period, different categories in the data set can be stably restored.

[0107] After the above two identification processes, the data is divided into two categories by different time periods of unstable periods in the time domain. In this way, according to the time domain period of the above system trajectory as the system feature, as long as the time domain periods of the two types of patterns are different, the control rules of step 5 can be used to control the chaotic pulse signal, so that two different types of patterns can be identified. If multiple patterns need to be identified, the parallel identification calculation method can be adopted, that is, grouping in pairs and identifying multiple nodes at the same time, thereby speeding up the efficiency of data classification.

[0108] Embodiment 2:

[0109] One embodiment of the present invention proposes a data classification and identification method based on unstable periodic orbits of chaotic systems, see Figure 1 , the data classification process includes the following steps:

[0110] Step 1 Input the data to be processed, that is, an audio data set 1.wav, whose basic waveform is as follows Figure 5 .

[0111] After steps 2-5, the audio dataset 1.wav is mapped as follows Figure 6 on a chaotic periodic orbit and at the same time is well controlled.

[0112] Step 6, such as Figure 7 , under the control of the control switch, but when z=1, λ will be divided into λ≠λ j and λ=λ j The two forks perform pattern recognition on the data set 1.wav, and the same type of sound will continue to be classified. For example, the unstable periodic orbit is initially identified as a human voice by pattern recognition; after a round of pattern recognition, it is found that the periodic orbit of the chaotic pulse signal can still be controlled, so the pattern recognition continues to be a male or female voice until the chaotic pulse signal cannot be controlled.

[0113] In a specific implementation, the present application provides a computer storage medium and a corresponding data processing unit, wherein the computer storage medium can store a computer program, and when the computer program is executed by the data processing unit, the invention content of the data classification and identification method based on the unstable periodic orbit of a chaotic system provided by the present invention and some or all steps in each embodiment can be run. The storage medium can be a disk, an optical disk, a read-only memory (ROM) or a random access memory (RAM), etc.

[0114] Those skilled in the art can clearly understand that the technical solutions in the embodiments of the present invention can be implemented by means of computer programs and their corresponding general hardware platforms. Based on such an understanding, the technical solutions in the embodiments of the present invention can be essentially or partly contributed to the prior art in the form of computer programs, i.e., software products, which can be stored in a storage medium and include several instructions for enabling a device including a data processing unit (which can be a personal computer, a server, a single-chip microcomputer, an MCU or a network device, etc.) to execute the methods described in various embodiments of the present invention or certain parts of the embodiments.

[0115] The present invention provides a method and idea for data classification and identification based on unstable periodic orbits of chaotic systems. There are many methods and approaches to implement the technical solution. The above is only a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present invention. These improvements and modifications should also be considered as the scope of protection of the present invention. All components not specified in this embodiment can be implemented using existing technologies.

Claims

1. A data classification and identification method based on unstable periodic orbits of chaotic systems, characterized in that: The steps include: Step 1, collect the original data to be classified in the computer system and establish a data set; Step 2, constructing a chaotic mapping system to generate an initial chaotic pulse signal for mapping the data set; Step 3, adding a timing control signal to the chaotic mapping system to control the mapped data set to obtain a stable periodic orbit; Step 4, using a symbol dynamics method to map the stable periodic orbit into a symbol sequence; Step 5, formulate control rules to adjust the symbol sequence to the final chaotic pulse signal; Step 6: Based on the final chaotic pulse signal, pattern classification and recognition are performed on the data set of step 1 to complete the data classification and recognition based on the unstable periodic orbit of the chaotic system.

2. According to claim 1, a data classification and identification method based on unstable periodic orbits of chaotic systems is characterized in that: The construction of the chaotic mapping system described in step 2 specifically includes: Step 2-1, constructing a chaotic mapping system, that is, using a dimensionless mathematical model to construct an impulse signal; Step 2-2, using recursive mapping to describe the dynamic behavior of the chaotic mapping system constructed above; Step 2-3, perform periodic mapping correspondence; Step 2-4, using different reference signals b(τ) according to the recursive mapping description in step 2-2 to generate different mappings.

3. The data classification and identification method based on unstable periodic orbits of chaotic systems according to claim 2 is characterized in that: The dimensionless mathematical model described in step 2-1 is as follows: where y is the internal state of the dataset, is the state that changes over time, x is the input, τ is the normalized time unit, b(τ) is the reference signal with a period of 1, and λ is a prime or composite parameter.

4. The data classification and identification method based on unstable periodic orbits of chaotic systems according to claim 3 is characterized in that: The recursive mapping described in step 2-2 is specifically expressed as follows: θ(n+1)=F(θ(n))=θ(n)-b(θ(n)) Where n is the number of units of the chaotic pulse signal, θ(n) is the state vector, F(θ(n)) represents the periodic point of the θ(n) state vector mapping F, and b(θ(n)) represents the periodic point of the θ(n) state vector mapping b.

5. The data classification and identification method based on unstable periodic orbits of chaotic systems according to claim 4 is characterized in that: The periodic mapping described in step 2-3 is as follows: F:S 1 ->S′ Among them, F represents the mapping relationship, S 1 ->S′ represents the mapping unit circle.

6. The data classification and identification method based on unstable periodic orbits of chaotic systems according to claim 5 is characterized in that: The different mappings described in steps 2-4 include: Select λ=1, b(τ)=-τ-θ0, θ0>0 to control the mapping, then the recursive mapping is expressed as follows: θ(n+1)=F(θ(n))=2θ(n)+θ0(mod1) Among them, θ0 represents recursive mapping, and mod1 represents the decimal operation.

7. The data classification and identification method based on unstable periodic orbits of chaotic systems according to claim 6 is characterized in that: Adding a timing control signal to the chaotic mapping system in step 3 is to adjust the dimensionless mathematical model in step 2-1, as follows: Among them, c(τ) is the added timing control signal, and z is the control switch; After adjustment, the output of the chaotic mapping system is a stable periodic orbit.

8. The data classification and identification method based on unstable periodic orbits of chaotic systems according to claim 7 is characterized in that: The mapping of the periodic orbit into a symbol sequence in step 4 specifically includes: Step 4-1, mapping the mapping relationship F into a symbol sequence z using a mapping equation n , each state vector θ(n)∈z n The mapping equation is as follows: Among them, Δ represents all integers; Step 4-2, the dimensionless mathematical model in the chaotic mapping system is finally optimized as: Where i = 1, 2, 3…N, N is infinite; z = (θ(n), ω, λ) is the control switch, which is represented by the state vector θ(n) = (θ1(n), θ2(n), ...θ N (n)), window vector set ω and category key λ.

9. The data classification and identification method based on unstable periodic orbits of chaotic systems according to claim 8 is characterized in that: The formulation of control rules described in step 5 specifically includes: The following control rules are formulated for the control switch z: When λm≤τ<λm+1, θ(n)∈ω, then z=z(θ(n),ω,λ) is set to 1 when λm≤τ<λ(m+1), where m=1,2,3…λ, and λ is a prime number or a composite number.

10. The data classification and identification method based on unstable periodic orbits of chaotic systems according to claim 9 is characterized in that: The pattern classification and recognition described in step 6 specifically includes: Use the category key λ as the basis for judgment to classify the final chaotic pulse signal described in step 5, that is, when λ≠λ j and λ=λ j , the data set is identified as different types, where λ j is λ to the jth power, where j is a random number, either a prime number or a composite number.

Citation Information

Patent Citations

  • Data classification method based on chaos depth wavelet network

    CN104463205A

  • Image defect recognition method and system, electronic equipment and storage medium

    CN107886509A

  • Orthogonal broadband modulation and demodulation method based on chaotic sequence

    CN110224958A

  • Sound big data analysis and calculation imaging system based on cloud side end

    CN113539298A

  • Chaotic neural network with complex value weight and application thereof in electrocardiogram classification

    CN113837139A