CN105306098B - A kind of method and device of Second Generation Wavelets Kernel - Google Patents
A kind of method and device of Second Generation Wavelets Kernel Download PDFInfo
- Publication number
- CN105306098B CN105306098B CN201510768324.4A CN201510768324A CN105306098B CN 105306098 B CN105306098 B CN 105306098B CN 201510768324 A CN201510768324 A CN 201510768324A CN 105306098 B CN105306098 B CN 105306098B
- Authority
- CN
- China
- Prior art keywords
- signal
- kernel function
- generation
- frequency hopping
- identified
- 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
Landscapes
- Complex Calculations (AREA)
Abstract
The invention discloses a kind of method and device of Second Generation Wavelets Kernel.This method includes:Obtain the discrete signal after sampling;Second Generation Wavelet Transformation is carried out to the discrete signal and obtains high dimensional signal;The high dimensional signal is converted to obtain Second Generation Wavelets kernel function according to higher-order spectrum kernel function;The Frequency Hopping Signal in signal to be identified is identified using the Second Generation Wavelets kernel function.Kernel function is constructed by using Second Generation Wavelets, Frequency Hopping Signal is mapped to higher dimensional space from lower dimensional space, not only lifts the accuracy rate of Frequency Hopping Signal identification, also improves the efficiency of identification.
Description
Technical field
The present invention relates to wavelet transformation technique field, more particularly to a kind of method and dress of Second Generation Wavelets Kernel
Put.
Background technology
Frequency Hopping Signal is a kind of typical non-stationary signal, it is necessary to use time frequency analysis to it, while obtain Frequency Hopping Signal
Arrival time, carrier parameter etc..Frequency hopping analysis refers to analyze unknown Frequency Hopping Signal with processing, including is made an uproar to being mingled in
Frequency Hopping Signal in sound is detected, the frequency hopping rate of the Frequency Hopping Signal then come detected by estimation, timing offset, frequency hopping pattern
Etc. parameter, identification and sorting of Frequency Hopping Signal etc. are finally completed.
The big measure feature that high-order statistic is carried due to it, therefore be also widely applied in field of signal identification, can
In the method by constructing higher-order spectrum kernel function, the inseparable sample of low-dimensional is mapped to higher dimensional space, so as to effectively reduce branch
The number of vector is held, improves the separability of sample, reaches the purpose for improving discrimination.On the basis of higher-order spectrum kernel function, draw
Enter first generation Wavelet Kernel Function and use it for the identification to Frequency Hopping Signal.Eventually through simulation results show first generation small echo
Kernel function can improve the accuracy of Frequency Hopping Signal identification but the space of lifting is still suffered from terms of recognition efficiency.
The content of the invention
In view of this, the embodiment of the present invention provides a kind of method and device of Second Generation Wavelets Kernel, to solve
The low technical problem of first generation wavelet recognition efficiency in the prior art.
In a first aspect, the embodiments of the invention provide a kind of method of Second Generation Wavelets Kernel, including:
Obtain the discrete signal after sampling;
Second Generation Wavelet Transformation is carried out to the discrete signal and obtains high dimensional signal;
The high dimensional signal is converted to obtain Second Generation Wavelets kernel function according to higher-order spectrum kernel function;
The Frequency Hopping Signal in signal to be identified is identified using the Second Generation Wavelets kernel function.
Second aspect, the embodiment of the present invention additionally provide a kind of device of Second Generation Wavelets Kernel, including:
Acquisition module, for obtaining the discrete signal after sampling;
Conversion module, high dimensional signal is obtained for carrying out Second Generation Wavelet Transformation to the discrete signal;
Kernel module is small for being converted to obtain the second generation to the high dimensional signal according to higher-order spectrum kernel function
Ripple kernel function;
Identification module, for identifying the Frequency Hopping Signal in signal to be identified using the Second Generation Wavelets kernel function.
A kind of method and device of Second Generation Wavelets Kernel provided in an embodiment of the present invention, by acquisition from
Scattered signal carries out Second Generation Wavelet Transformation and obtains high dimensional signal, and the high dimensional signal is converted according to higher-order spectrum kernel function
Second Generation Wavelets kernel function is obtained, the Frequency Hopping Signal in signal to be identified is identified using the kernel function of construction.By using above-mentioned
The method and device of Second Generation Wavelets Kernel, frequency hopping letter can be lifted while Frequency Hopping Signal recognition accuracy is improved
Number recognition efficiency.
Brief description of the drawings
By reading the detailed description made to non-limiting example made with reference to the following drawings, of the invention is other
Feature, objects and advantages will become more apparent upon:
Fig. 1 is a kind of flow chart of the method for Second Generation Wavelets Kernel that the embodiment of the present invention one provides;
Fig. 2 is a kind of flow chart of the method for Second Generation Wavelets Kernel that the embodiment of the present invention two provides;
Fig. 3 is the flow chart for the method that a kind of SVM that the embodiment of the present invention two provides identifies Frequency Hopping Signal;
Fig. 4 is a kind of schematic diagram of the device for Second Generation Wavelets Kernel that the embodiment of the present invention three provides.
Embodiment
The present invention is described in further detail with reference to the accompanying drawings and examples.It is understood that this place is retouched
The specific embodiment stated is used only for explaining the present invention, rather than limitation of the invention.It also should be noted that in order to just
Part related to the present invention rather than full content are illustrate only in description, accompanying drawing.
Embodiment one
Fig. 1 is a kind of flow chart of the method for Second Generation Wavelets Kernel that the embodiment of the present invention one provides.This reality
Applying the method for example can be performed by the device of Second Generation Wavelets Kernel, and the device can be by software and/or hardware Lai real
It is existing, it is integrated in the terminal that can construct Second Generation Wavelets kernel function.As shown in figure 1, this method includes:
S110, obtain the discrete signal after sampling.
Exemplary, the discrete signal can be obtained after the standard signal sent to signal transmitter device samples
Discrete signal.The discrete signal is 2D signal, and the specific features of the discrete signal can be carried out according to actual conditions
Limit.
S120, high dimensional signal is obtained to discrete signal progress Second Generation Wavelet Transformation.
Exemplary, Second Generation Wavelets are also referred to as Lifting Wavelet, are the small echos using Second Generation Wavelet Transformation method construct.
After obtaining discrete signal, Second Generation Wavelet Transformation is carried out to discrete signal using wavelet basis and obtains high dimensional signal, is completed discrete
Signal is from low dimensional to high-dimensional mapping.Wherein wavelet basis can be selected according to actual conditions, in the present embodiment preferably
To use the small echos of Le Gall 5/3 as wavelet basis.
Further, Second Generation Wavelet Transformation is carried out according to preset model to the sampled signal and obtains high dimensional signal.
The preset model is Second Generation Wavelet Transformation model, including divides, predicts and update three steps.Division refers to
Discrete signal is divided into odd element sequence and even element sequence.Prediction refers to utilize between odd element sequence and even element sequence
Correlation, certain elemental signals is predicted by the closing signal of certain elemental signals.Renewal refers to the result come according to predicting, right
The odd element sequence and even element sequence obtained after division is updated, wherein the sequence obtained after renewal can be according to different
Updating factor obtains different renewal results, is preferably to be used as more using the wavelet basis of the small echos of Le Gall 5/3 in the present embodiment
The new factor.The high dimensional signal obtained after renewal keeps the characteristic of discrete signal.Realized after the completion of three steps to discrete signal
Second Generation Wavelet Transformation, that is, discrete signal is realized from low dimensional to high-dimensional mapping, obtains high dimensional signal, high dimensional signal can
To improve the separability of sample, be easy to low-dimensional can not sub-signal make a distinction and identify.Preferably, the preset model is:R
(n)=(R (2n), R (2n+1)), wherein, R (n) represents the high dimensional signal exported after Second Generation Wavelet Transformation, R (2n) and R
(2n+1) represents the even element sequence and odd element sequence of the high dimensional signal R (n) of output respectively, wherein,AndWherein
S (n) represents the discrete signal of input, S (2n) and S (2n+1) represent respectively the discrete signal S (n) of input even element sequence with
Odd element sequence.
S130, according to higher-order spectrum kernel function the high dimensional signal is converted to obtain Second Generation Wavelets kernel function.
Exemplary, the higher-order spectrum kernel function is the convolution of real function x (t) He y (t) the k rank higher-order spectrums of the two.According to
Higher-order spectrum kernel function is to high dimensional signal R1And high dimensional signal R (n)2(n) the Second Generation Wavelets kernel function K (x, y) that line translation obtains is entered
For Wherein, wherein K
(x, y) is Second Generation Wavelets kernel function, R1And R (n)2(n) it is respectively two high dimensional signals, S1And S (n)2(n) it is respectively R1(n)
And R2(n) discrete signal corresponding to, i.e., to being rolled up again to it after Second Generation Wavelet Transformation is carried out to the discrete signal of input
Product calculates, and obtained signal is also high dimensional signal.Can be to the discrete letter that arbitrarily inputs with obtained Second Generation Wavelets kernel function
Number enter line translation obtain corresponding to high dimensional signal.
S140, utilize the Frequency Hopping Signal in Second Generation Wavelets kernel function identification signal to be identified.
Exemplary, after Second Generation Wavelets kernel function is obtained, the Second Generation Wavelets kernel function is preserved, and described in utilization
Second Generation Wavelets kernel function identifies the Frequency Hopping Signal in signal to be identified.The Second Generation Wavelets kernel function is preferably put into branch
Support in vector machine, the Frequency Hopping Signal in signal to be identified is identified using the SVM.Support vector function reflects vector
It is mapped in a higher dimensional space, is easy to classification and regression analysis.Preferably signal to be identified is mapped to using SVM
High dimensional signal, and it is classified, identify the Frequency Hopping Signal in signal to be identified.Wherein treat identification signal and be mapped to height
Dimensional signal is exactly to be mapped to obtain using the Second Generation Wavelets kernel function being put into SVM.
A kind of method for Second Generation Wavelets Kernel that the embodiment of the present invention one provides, passes through the discrete letter to acquisition
Number carry out Second Generation Wavelet Transformation obtain high dimensional signal, high dimensional signal is converted to obtain the second generation according to higher-order spectrum kernel function
Wavelet Kernel Function, treat the Frequency Hopping Signal in identification signal using obtained Second Generation Wavelets kernel function and be identified.Using upper
Method is stated, Frequency Hopping Signal is identified by constructing Second Generation Wavelets kernel function, can be while Frequency Hopping Signal discrimination be improved
Also lift the recognition efficiency of Frequency Hopping Signal.
Embodiment two
Fig. 2 is a kind of flow chart of the method for Second Generation Wavelets Kernel that the embodiment of the present invention two provides.This reality
It is on the basis of embodiment one to apply example, and after Second Generation Wavelets kernel function is constructed, the Frequency Hopping Signal treated in identification signal enters
Further restriction has been done in row identification.
Further, with the addition of and the Second Generation Wavelets kernel function be put into SVM, using it is described support to
Amount machine identifies the Frequency Hopping Signal in signal to be identified.Specific steps include:
S210, obtain the discrete signal after sampling.
S220, high dimensional signal is obtained to discrete signal progress Second Generation Wavelet Transformation.
S230, according to higher-order spectrum kernel function the high dimensional signal is converted to obtain Second Generation Wavelets kernel function.
S240, the Second Generation Wavelets kernel function is put into SVM, treated using SVM identification
Frequency Hopping Signal in identification signal.
SVM (Support Vector Machine, SVM) is known for a kind of pattern based on Statistical Learning Theory
Other method model, available for classifying and returning, its key is kernel function.
Exemplary, after Second Generation Wavelets kernel function is constructed, Second Generation Wavelets kernel function is put into SVM
In, as the kernel function of SVM, the Frequency Hopping Signal in signal to be identified is identified using the SVM.Fig. 3 is
A kind of flow chart of the method for SVM identification Frequency Hopping Signal that the embodiment of the present invention two provides.As shown in figure 3, S240 has
Body includes:
S241, by the Second Generation Wavelets kernel function be stored in SVM in.
Exemplary, after Second Generation Wavelets kernel function is constructed, Second Generation Wavelets kernel function is stored in SVM
In, the kernel function as SVM.
S242, to two standard Frequency Hopping Signals after the SVM input sample.
Exemplary, the standard Frequency Hopping Signal can be that the carrier frequency that Frequency hopping transmissions device is sent becomes according to certain rule
The Frequency Hopping Signal of change, specific rule can be set according to actual conditions, not limited here.Two standard Frequency Hopping Signals
Characteristic quantity differ, preferably frequency changing rule differs.
S243, the SVM are according to the Second Generation Wavelets kernel function respectively to two standard Frequency Hopping Signals
Enter line translation and obtain two training sample signals, and preserve two training sample signals.
Exemplary, SVM is after two standard Frequency Hopping Signals are received, according to deposit SVM
Second Generation Wavelets kernel function enters line translation to two standard Frequency Hopping Signals respectively and obtains two training sample signals, and preserves
The training sample signal is in SVM.The training sample signal is high dimensional signal.
S244, line translation is entered to the signal to be identified according to the Second Generation Wavelets kernel function obtain higher-dimension letter to be identified
Number.
Abovementioned steps can be to prepare work in advance, and this step starts, and obtain signal to be identified, utilize SVM pair
Frequency Hopping Signal in signal to be identified is identified.
Exemplary, the signal to be identified differs including at least two class subsignals, the characteristic quantity of all kinds of subsignals, together
The characteristic quantity of a kind of subsignal is identical.Signal preferably to be identified includes the subsignal that two quefrency rules of conversion differ, institute
It is sub- Frequency Hopping Signal to state subsignal.After SVM receives signal to be identified, according to Second Generation Wavelets kernel function pair
The signal to be identified enters line translation and obtains higher-dimension signal to be identified, that is, completes low-dimensional the reflecting to higher-dimension of signal to be identified
Penetrate, make can not sub-signal become to divide.
S245, higher-dimension signal to be identified is compared with two training sample signals respectively, obtains two classes
Subsignal is respectively as two class Frequency Hopping Signals, the two classes Frequency Hopping Signal comparison result with corresponding training sample signal respectively
Higher than default similarity.
Exemplary, higher-dimension signal to be identified is believed with two training samples being stored in SVM respectively
Number it is compared, a kind of subsignal of default similarity will be higher than with the comparison result of one of training sample signal as one
Class Frequency Hopping Signal, the another kind of subsignal of default similarity will be higher than with the comparison result of another training sample signal as another
A kind of Frequency Hopping Signal, two class Frequency Hopping Signals are respectively outputted in corresponding interface, that is, have distinguished two classes in signal to be identified and jumped
Frequency signal.Alignments can be to be compared with least one characteristic of standard signal, and default similarity is higher than to similarity
Subsignal carry out Classification and Identification, the characteristic can be frequency transformation characteristic or phse conversion characteristic etc..Default similarity can
To be set according to actual conditions, do not limit here.
The method for a kind of Second Generation Wavelets Kernel that the embodiment of the present invention two provides, by the way that the second generation is small
Ripple kernel function is put into SVM, and SVM enters the signal to be identified received using Second Generation Wavelets kernel function
Line translation obtains higher-dimension signal to be identified so that the feature of signal to be identified is more obvious, by higher-dimension signal to be identified and storage
Two training sample signals be compared, obtain two class subsignals as two class Frequency Hopping Signals, the two classes Frequency Hopping Signal point
Default similarity is not higher than it with the comparison result of corresponding training sample signal.Using the above method, signal to be identified is passed through
Second Generation Wavelets kernel function in SVM is mapped to high dimensional signal, makes signal characteristic more obvious, is easy to distinguish and knows
Not, the recognition accuracy and recognition efficiency of Frequency Hopping Signal can be improved.
Embodiment three
Fig. 4 is a kind of schematic diagram of the device for Second Generation Wavelets Kernel that the embodiment of the present invention three provides.Specifically
Including:Acquisition module 401, conversion module 402, Kernel module 403 and identification module 404.
Wherein, acquisition module 401, for obtaining the discrete signal after sampling;Conversion module 402, for described discrete
Signal carries out Second Generation Wavelet Transformation and obtains high dimensional signal;Kernel module 403, for according to higher-order spectrum kernel function to institute
High dimensional signal is stated to be converted to obtain Second Generation Wavelets kernel function;Identification module 404, for utilizing the Second Generation Wavelets core letter
Number identifies the Frequency Hopping Signal in signal to be identified.
Further, the identification module also includes:SVM, for being stored in the Second Generation Wavelets kernel function,
And identify the Frequency Hopping Signal in signal to be identified.
Preferably, described device also includes:Sending module, for being marked to two after the SVM input sample
Quasi- Frequency Hopping Signal.
Further, the SVM also includes:Wavelet Kernel Function access unit, sample signal memory cell, treat
Identification signal converter unit and recognition unit.
Wherein, Wavelet Kernel Function access unit, for being stored in the Second Generation Wavelets kernel function;Sample signal storage is single
Member, for receive sending module send two standard Frequency Hopping Signals after according to the Second Generation Wavelets kernel function respectively to two
The individual standard Frequency Hopping Signal enters line translation and obtains two training sample signals, and preserves two training sample signals;Treat
Identification signal converter unit, higher-dimension is obtained for entering line translation to the signal to be identified according to the Second Generation Wavelets kernel function
Signal to be identified;Recognition unit, for higher-dimension signal to be identified to be compared with two sample signals respectively, obtain
To two class subsignals respectively as two class Frequency Hopping Signals, the two classes Frequency Hopping Signal ratio with corresponding training sample signal respectively
It is higher than default similarity to result.
The conversion module has and is used on the basis of above-described embodiment:The discrete signal is entered according to preset model
Row Second Generation Wavelet Transformation obtains high dimensional signal, and the preset model is:R (n)=(R (2n), R (2n+1)), wherein, R (n) tables
Show the high dimensional signal exported after Second Generation Wavelet Transformation, R (2n) and R (2n+1) represent the high dimensional signal R (n) of output respectively
Even element sequence and odd element sequence, wherein,With
AndWherein S (n) represents the discrete signal of input, S
(2n) and S (2n+1) represent the discrete signal S (n) of input even element sequence and odd element sequence respectively.
Further, the Second Generation Wavelets kernel function is: Wherein K (x, y)
For Second Generation Wavelets kernel function, R1And R (n)2(n) it is respectively two high dimensional signals, S1And S (n)2(n) it is respectively R1And R (n)2
(n) discrete signal corresponding to.
A kind of device for Second Generation Wavelets Kernel that the embodiment of the present invention three provides, passes through the discrete letter to acquisition
Number carry out Second Generation Wavelet Transformation obtain high dimensional signal, high dimensional signal is converted to obtain the second generation according to higher-order spectrum kernel function
Wavelet Kernel Function, treat the Frequency Hopping Signal in identification signal using obtained Second Generation Wavelets kernel function and be identified.Using upper
Device is stated, Frequency Hopping Signal is identified using the Second Generation Wavelets kernel function of construction, Frequency Hopping Signal identification standard can improved
The recognition efficiency of Frequency Hopping Signal is lifted while true rate.
The device for the Second Generation Wavelets Kernel that the embodiment of the present invention is provided carries for performing the embodiment of the present invention
The method of the Second Generation Wavelets Kernel of confession, possess corresponding function and beneficial effect.
Pay attention to, above are only presently preferred embodiments of the present invention and institute's application technology principle.It will be appreciated by those skilled in the art that
The invention is not restricted to specific embodiment described here, can carry out for a person skilled in the art various obvious changes,
Readjust and substitute without departing from protection scope of the present invention.Therefore, although being carried out by above example to the present invention
It is described in further detail, but the present invention is not limited only to above example, without departing from the inventive concept, also
Other more equivalent embodiments can be included, and the scope of the present invention is determined by scope of the appended claims.
Claims (6)
- A kind of 1. method of Second Generation Wavelets Kernel, it is characterised in that including:Obtain the discrete signal after sampling;Second Generation Wavelet Transformation is carried out to the discrete signal and obtains high dimensional signal;The high dimensional signal is converted to obtain Second Generation Wavelets kernel function according to higher-order spectrum kernel function;The Frequency Hopping Signal in signal to be identified is identified using the Second Generation Wavelets kernel function;It is described to identify that the Frequency Hopping Signal in signal to be identified includes using the Second Generation Wavelets kernel function:The Second Generation Wavelets kernel function is put into SVM, identified using the SVM in signal to be identified Frequency Hopping Signal;The signal to be identified includes at least two class subsignals, described that the Second Generation Wavelets kernel function is put into SVM In, identify that the Frequency Hopping Signal in signal to be identified includes using the SVM:The Second Generation Wavelets kernel function is stored in SVM;To two standard Frequency Hopping Signals after the SVM input sample;The SVM enters line translation to two standard Frequency Hopping Signals respectively according to the Second Generation Wavelets kernel function Two training sample signals are obtained, and preserve two training sample signals;Line translation is entered to the signal to be identified according to the Second Generation Wavelets kernel function and obtains higher-dimension signal to be identified;Higher-dimension signal to be identified is compared with two training sample signals respectively, obtains two class subsignals difference As two class Frequency Hopping Signals, the two classes Frequency Hopping Signal is higher than default phase with the comparison result of corresponding training sample signal respectively Like degree.
- 2. according to the method for claim 1, it is characterised in that Second Generation Wavelet Transformation is carried out to the discrete signal and obtained High dimensional signal includes:Second Generation Wavelet Transformation is carried out according to preset model to the discrete signal and obtains high dimensional signal, the preset model is:R (n)=(R (2n), R (2n+1)), wherein, R (n) represents the high dimensional signal exported after Second Generation Wavelet Transformation, R (2n) and R (2n+1) represents the even element sequence and odd element sequence of the high dimensional signal R (n) of output respectively, wherein,AndWherein S (n) represents the discrete signal of input, S (2n) and S (2n+1) represent respectively the discrete signal S (n) of input even element sequence with Odd element sequence.
- 3. according to the method for claim 2, it is characterised in that the Second Generation Wavelets kernel function is Wherein K (x, Y) it is Second Generation Wavelets kernel function, R1And R (n)2(n) it is respectively two high dimensional signals, S1And S (n)2(n) it is respectively R1And R (n)2 (n) discrete signal corresponding to.
- A kind of 4. device of Second Generation Wavelets Kernel, it is characterised in that including:Acquisition module, for obtaining the discrete signal after sampling;Conversion module, high dimensional signal is obtained for carrying out Second Generation Wavelet Transformation to the discrete signal;Kernel module, for being converted to obtain Second Generation Wavelets core to the high dimensional signal according to higher-order spectrum kernel function Function;Identification module, for identifying the Frequency Hopping Signal in signal to be identified using the Second Generation Wavelets kernel function;The identification module also includes:SVM, for being stored in the Second Generation Wavelets kernel function, and identify the Frequency Hopping Signal in signal to be identified;Described device includes:Sending module, for two standard Frequency Hopping Signals after the SVM input sample;The SVM also includes:Wavelet Kernel Function access unit, for being stored in the Second Generation Wavelets kernel function;Sample signal memory cell, for receive sending module send two standard Frequency Hopping Signals after according to the second generation Wavelet Kernel Function enters line translation to two standard Frequency Hopping Signals and obtains two training sample signals respectively, and preserves two institutes State training sample signal;Signal conversion unit to be identified, obtained for entering line translation to the signal to be identified according to the Second Generation Wavelets kernel function To higher-dimension signal to be identified;Recognition unit, for higher-dimension signal to be identified to be compared with two training sample signals respectively, obtain Two class subsignals are respectively as two class Frequency Hopping Signals, the two classes Frequency Hopping Signal comparison with corresponding training sample signal respectively As a result higher than default similarity.
- 5. device according to claim 4, it is characterised in that conversion module is specifically used for:Second Generation Wavelet Transformation is carried out according to preset model to the discrete signal and obtains high dimensional signal, the preset model is:R (n)=(R (2n), R (2n+1)), wherein, R (n) represents the high dimensional signal exported after Second Generation Wavelet Transformation, R (2n) and R (2n+1) represents the even element sequence and odd element sequence of the high dimensional signal R (n) of output respectively, wherein,And Wherein S (n) represents the discrete signal of input, and S (2n) and S (2n+1) represent the discrete signal S (n) of input even element sequence respectively Row and odd element sequence.
- 6. device according to claim 5, it is characterised in that the Second Generation Wavelets kernel function is Wherein K (x, Y) it is Second Generation Wavelets kernel function, R1And R (n)2(n) it is respectively two high dimensional signals, S1And S (n)2(n) it is respectively R1And R (n)2 (n) discrete signal corresponding to.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201510768324.4A CN105306098B (en) | 2015-11-12 | 2015-11-12 | A kind of method and device of Second Generation Wavelets Kernel |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201510768324.4A CN105306098B (en) | 2015-11-12 | 2015-11-12 | A kind of method and device of Second Generation Wavelets Kernel |
Publications (2)
Publication Number | Publication Date |
---|---|
CN105306098A CN105306098A (en) | 2016-02-03 |
CN105306098B true CN105306098B (en) | 2018-01-05 |
Family
ID=55202928
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201510768324.4A Active CN105306098B (en) | 2015-11-12 | 2015-11-12 | A kind of method and device of Second Generation Wavelets Kernel |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN105306098B (en) |
Families Citing this family (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN106952272B (en) * | 2017-03-08 | 2019-09-10 | 江苏理工学院 | A kind of anti-noise Otsu image segmentation method based on two generation wavelet transformations |
Family Cites Families (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101832471B (en) * | 2010-04-19 | 2013-02-06 | 哈尔滨工程大学 | Signal identification and classification method |
-
2015
- 2015-11-12 CN CN201510768324.4A patent/CN105306098B/en active Active
Also Published As
Publication number | Publication date |
---|---|
CN105306098A (en) | 2016-02-03 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN110926782B (en) | Circuit breaker fault type judgment method and device, electronic equipment and storage medium | |
Alexandrov et al. | A review of some modern approaches to the problem of trend extraction | |
CN112001306A (en) | Electroencephalogram signal decoding method for generating neural network based on deep convolution countermeasure | |
Steinley | Profiling local optima in K-means clustering: developing a diagnostic technique. | |
CN110147760B (en) | Novel efficient electric energy quality disturbance image feature extraction and identification method | |
CN109039503A (en) | A kind of frequency spectrum sensing method, device, equipment and computer readable storage medium | |
CN108052863A (en) | Electrical energy power quality disturbance recognition methods based on the maximum variance method of development | |
CN109598175A (en) | It is a kind of based on before multi-wavelet bases function and transothogonal to the Time-Frequency Analysis Method of recurrence | |
CN109798920A (en) | Based on the MEMS gyro stochastic error modeling filtering method for improving EMD | |
CN113343801B (en) | Automatic wireless signal modulation and identification method based on lightweight convolutional neural network | |
CN105306098B (en) | A kind of method and device of Second Generation Wavelets Kernel | |
CA2330878A1 (en) | Pre-processing and post-processing for enhancing knowledge discovery using support vector machines | |
CN109309538A (en) | A kind of frequency spectrum sensing method, device, equipment, system and storage medium | |
CN112766340A (en) | Depth capsule network image classification method and system based on adaptive spatial mode | |
CN107894837A (en) | Dynamic sentiment analysis model sample processing method and processing device | |
CN114052675B (en) | Pulse abnormality judging method and system based on fused attention mechanism | |
CN114358162A (en) | Falling detection method and device based on continuous wavelet transform and electronic equipment | |
CN116430317A (en) | Radiation source modulation pattern and individual identification method and system | |
CN109614999A (en) | A kind of data processing method, device, equipment and computer readable storage medium | |
AU2021100368A4 (en) | Odour analysis method and device | |
US11640558B2 (en) | Unbalanced sample classification method and apparatus | |
CN109614587A (en) | A kind of intelligence relationship among persons method for analyzing and modeling, terminal device and storage medium | |
Chen et al. | Adaptive multi-scale TF-net for high-resolution time–frequency representations | |
CN116188834B (en) | Full-slice image classification method and device based on self-adaptive training model | |
CN112529035B (en) | Intelligent identification method for identifying individual types of different radio stations |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |