CN109039500A - A kind of frequency spectrum sensing method, device, equipment, system and storage medium - Google Patents
A kind of frequency spectrum sensing method, device, equipment, system and storage medium Download PDFInfo
- Publication number
- CN109039500A CN109039500A CN201810989697.8A CN201810989697A CN109039500A CN 109039500 A CN109039500 A CN 109039500A CN 201810989697 A CN201810989697 A CN 201810989697A CN 109039500 A CN109039500 A CN 109039500A
- Authority
- CN
- China
- Prior art keywords
- signal
- frequency spectrum
- obtains
- matrix
- sampled signal
- 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.)
- Pending
Links
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04B—TRANSMISSION
- H04B17/00—Monitoring; Testing
- H04B17/30—Monitoring; Testing of propagation channels
- H04B17/382—Monitoring; Testing of propagation channels for resource allocation, admission control or handover
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04B—TRANSMISSION
- H04B17/00—Monitoring; Testing
- H04B17/30—Monitoring; Testing of propagation channels
- H04B17/391—Modelling the propagation channel
Landscapes
- Physics & Mathematics (AREA)
- Electromagnetism (AREA)
- Engineering & Computer Science (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Monitoring And Testing Of Transmission In General (AREA)
Abstract
The invention discloses a kind of frequency spectrum sensing methods, comprising: obtains radio sampled signal；Noise reduction process is carried out to sampled signal, obtains de-noising signal；IQ decomposition is carried out to de-noising signal, obtains same phase matrix and orthogonal matrix；The difference that maximum eigenvalue and the average energy value are calculated according to the characteristic value of same phase matrix and the covariance matrix of orthogonal matrix, obtains feature vector；Tagsort division is carried out to feature vector, obtains the classification results of radio sampled signal.This method can promote frequency spectrum detection performance, optimize frequency spectrum perception effect.The invention also discloses a kind of frequency spectrum sensing device, equipment, system and a kind of readable storage medium storing program for executing, have above-mentioned beneficial effect.
Description
Technical field
The present invention relates to radio art, in particular to a kind of frequency spectrum sensing method, device, equipment, system and one kind can
Read storage medium.
Background technique
Due to the fast development of wireless network and wireless device, the frequency spectrum distribution policy of original fixation is shown obviously
Disadvantage.And cognitive radio (Cognitive Radio, CR) technology, authorized user (Primary can not influenced
User, PU) under the premise of, idle frequency range is made full use of, the problem of frequency spectrum resource shortage is alleviated, becomes and solves frequency spectrum money
The promising technology of distribution policy problem caused by the shortage of source.
The critical issue of cognitive radio technology is exactly frequency spectrum perception.Traditional frequency spectrum perception technology includes classical energy inspection
It surveys, cyclo-stationary detection, matched filtering.Wherein cyclo-stationary detection accuracy is high, can distinguish signal modulation mode, but require master
User has cyclostationary characteristic, is only applicable to specific occasion.Matched filtering precision is high, and the time is short, but computation complexity is high,
And it is only applicable to the occasion that CR node knows prior information.Classical energy measuring does not need prior information, needs default judgement
Thresholding, the more difficult determination of thresholding；It is easily affected by noise, under the uncertain environment of low signal-to-noise ratio environment and noise, it is also easy to produce
Erroneous judgement sharply declines so as to cause detection performance, and detection time is long.
Therefore, frequency spectrum detection performance how is promoted, frequency spectrum perception effect is optimized, is that those skilled in the art need to solve
Technical problem.
Summary of the invention
The object of the present invention is to provide a kind of frequency spectrum sensing method, this method can promote frequency spectrum detection performance, optimization frequency
Compose perceived effect；It is a further object of the present invention to provide a kind of frequency spectrum sensing device, equipment, system and a kind of readable storage mediums
Matter has above-mentioned beneficial effect.
The present invention provides a kind of frequency spectrum sensing method, comprising:
Obtain radio sampled signal；
Noise reduction process is carried out to the sampled signal, obtains de-noising signal；
IQ decomposition is carried out to the de-noising signal, obtains same phase matrix and orthogonal matrix；
According to the characteristic value of the same phase matrix and the covariance matrix of the orthogonal matrix calculate maximum eigenvalue with
The difference of the average energy value, obtains feature vector；
Tagsort division is carried out to described eigenvector, obtains the classification results of the radio sampled signal.
Preferably, described to include: to sampled signal progress noise reduction process
By EMD empirical mode decomposition method to the sampled signal carry out signal decomposition, and to the signal after decomposition into
Row intrinsic mode functions signal extraction, obtains de-noising signal.
Preferably, carrying out tagsort division to described eigenvector includes:
To several feature vector structural matrixes of acquisition, eigenmatrix is obtained；
The eigenmatrix is input to frequency spectrum disaggregated model, obtains classification results；Wherein, the frequency spectrum disaggregated model is
The Gaussian Mixture Clustering Model obtained according to the training of sample characteristics matrix.
Preferably, the parameter optimization method of the frequency spectrum disaggregated model includes:
Optimization is iterated to model parameter according to Maximum-likelihood estimation and expectation maximization EM algorithm.
Preferably, the frequency spectrum sensing method, which is characterized in that further include:
False-alarm probability and detection probability are calculated to the classification results；
Perceptual performance analysis is carried out according to the false-alarm probability and detection probability being calculated.
The present invention discloses a kind of frequency spectrum sensing device, comprising:
Sampled signal acquiring unit, for obtaining radio sampled signal；
Noise reduction processing unit obtains de-noising signal for carrying out noise reduction process to the sampled signal；
IQ decomposition unit obtains same phase matrix and orthogonal matrix for carrying out IQ decomposition to the de-noising signal；
Feature vector computing unit, for the spy according to the same phase matrix and the covariance matrix of the orthogonal matrix
Value indicative calculates the difference of maximum eigenvalue and the average energy value, obtains feature vector；
Tagsort unit obtains the radio sampling letter for carrying out tagsort division to described eigenvector
Number classification results.
Preferably, the noise reduction processing unit specifically: EMD noise reduction processing unit, the EMD noise reduction processing unit are used for
Signal decomposition is carried out to the sampled signal by EMD empirical mode decomposition method, and eigen mode is carried out to the signal after decomposition
Function signal extracts, and obtains de-noising signal.
The present invention discloses a kind of frequency spectrum perception equipment, comprising:
Memory, for storing computer program；
Processor, the step of frequency spectrum sensing method is realized when for executing the computer program.
The present invention discloses a kind of frequency spectrum perception system, comprising:
Radio sample devices obtains radio sampled signal, and the radio is adopted for acquiring radio signal
Sample signal is sent to frequency spectrum perception equipment；
The frequency spectrum perception equipment, for obtaining radio sampled signal；Noise reduction process is carried out to the sampled signal, is obtained
To de-noising signal；IQ decomposition is carried out to the de-noising signal, obtains same phase matrix and orthogonal matrix；According to the same phase matrix
And the characteristic value of the covariance matrix of the orthogonal matrix calculates the difference of maximum eigenvalue and the average energy value, obtains feature
Vector；Tagsort division is carried out to described eigenvector, obtains the classification results of the radio sampled signal.
The present invention discloses a kind of readable storage medium storing program for executing, and program is stored on the readable storage medium storing program for executing, and described program is located
The step of reason device realizes the frequency spectrum sensing method when executing.
In order to solve the above technical problems, the present invention provides a kind of frequency spectrum sensing method, by wireless after noise reduction process
Electric sampled signal carries out IQ decomposition, by signal carry out with mutually and the analysis of orthogonal direction, realize to signal in the same direction and
The abundant digging utilization of orthogonal two direction character, logically increases the number of cognitive user, according to same phase matrix and just
It hands over the characteristic value of the covariance matrix of matrix to calculate the difference (MSE) of maximum eigenvalue and the average energy value, is obtained according to difference
Feature vector is analyzed by precise and stable property of the MSE to data fluctuations situation in matrix, can accurately extract the feature letter of matrix
Breath can preferably improve the detection performance of frequency spectrum in and the lower situation of signal-to-noise ratio less in cognitive user number, promote frequency
Detection performance is composed, frequency spectrum perception effect is optimized, realizes the precise classification to sampled signal.
The invention also discloses a kind of frequency spectrum sensing device, equipment, system and a kind of readable storage medium storing program for executing, have with above-mentioned
Beneficial effect, details are not described herein.
Detailed description of the invention
In order to more clearly explain the embodiment of the invention or the technical proposal in the existing technology, to embodiment or will show below
There is attached drawing needed in technical description to be briefly described, it should be apparent that, the accompanying drawings in the following description is only this
The embodiment of invention for those of ordinary skill in the art without creative efforts, can also basis
The attached drawing of offer obtains other attached drawings.
Fig. 1 is the flow chart of frequency spectrum sensing method provided in an embodiment of the present invention；
Fig. 2 is two kinds of signal processing effect diagrams of EMD+IQ provided in an embodiment of the present invention and IQ；
Fig. 3 is the structural block diagram of frequency spectrum sensing device provided in an embodiment of the present invention；
Fig. 4 is the structural block diagram of frequency spectrum perception equipment provided in an embodiment of the present invention；
Fig. 5 is the structural schematic diagram of frequency spectrum perception equipment provided in an embodiment of the present invention；
Fig. 6 is the structural block diagram of frequency spectrum perception system provided in an embodiment of the present invention.
Specific embodiment
Core of the invention is to provide a kind of frequency spectrum sensing method, and this method can promote frequency spectrum detection performance, optimization frequency
Compose perceived effect；Another core of the invention is to provide a kind of frequency spectrum sensing device, system and a kind of readable storage medium storing program for executing, has
Above-mentioned beneficial effect.
In order to make the object, technical scheme and advantages of the embodiment of the invention clearer, below in conjunction with the embodiment of the present invention
In attached drawing, technical scheme in the embodiment of the invention is clearly and completely described, it is clear that described embodiment is
A part of the embodiment of the present invention, instead of all the embodiments.Based on the embodiments of the present invention, those of ordinary skill in the art
Every other embodiment obtained without making creative work, shall fall within the protection scope of the present invention.
Referring to FIG. 1, Fig. 1 is the flow chart of frequency spectrum sensing method provided in an embodiment of the present invention；This method may include:
Step s110, radio sampled signal is obtained.
One cognitive radio system, can there is PU and M user SU of a primary user, and the sampling number of each SU is
N, adjudicating primary user according to sampled data whether there is.When with H0It is expressed as primary user's signal to be not present, H1It is expressed as primary user's letter
In the presence of number, under both states, secondary user's received signal xi(n) it is represented by the model of formula (1):
Wherein, wi(n) representing mean value is 0, variance δ2White Gaussian noise, si(n) the transmitting signal of primary user is represented.
Assuming that the sampled signal vector of i-th of SU may be expressed as: xi(n)=[xi(1), xi(2) ..., xi(N)].Sampling letter
Number matrix X can be expressed as M × N-dimensional matrix X shown in (formula 2):
By carrying out matrix analysis to sampled signal to know the matrix character information in sampled signal.
Step s120, noise reduction process is carried out to sampled signal, obtains de-noising signal.
It will include much noise signal in the sampled signal of acquisition, generally to avoid noise signal to frequency spectrum perception effect
It influences, carries out noise reduction process before carrying out spectrum analysis.Without limitation to the method for noise reduction process at this, for example Fourier becomes
It changes, wavelet transformation etc..
The sampled signal obtained in frequency spectrum perception at present is usually non-linear, unstable signal.And in traditional Fu
Leaf transformation, the method for other signal processings such as wavelet transformation, can only generally handle linear, stable signal, it is preferable that can be with
Signal is handled using Empirical Mode Decomposition algorithm, usually noise is more included in high frequency band, is made an uproar in low-frequency band
Sound is weaker.EMD algorithm passes through the noise of removal high frequency band, and the signal for reconstructing low-frequency band is realized to the noise reduction of whole-sample signal
Reason.
The process for then carrying out noise reduction process to sampled signal is specifically as follows: by EMD empirical mode decomposition method to adopting
Sample signal carries out signal decomposition, and carries out intrinsic mode functions signal extraction to the signal after decomposition, obtains de-noising signal.
Step s130, IQ decomposition is carried out to de-noising signal, obtains same phase matrix and orthogonal matrix.
In order to using time of signal, space, the correlation information between phase is received, increase logic to the greatest extent
The quantity of signal, by the reception signal of each user (CU)(i=1,2 ..., M) carries out IQ decomposition.
Assuming that being the vector of a 1 × N-dimensional by the i collected signals comprising N number of sampled point of user (CU) It is broken down into I (same to phase) and Q (orthogonal) two parts, is expressed as following form:
After IQ is decomposed, correspondence has obtained two new M × N-dimensional signal matrix, wherein Y1 is same phase matrix, and Y2 is
Orthogonal matrix, as follows:
It, can be to avoid base band amplitude after reception mixing by reception signal and sheet using IQ resolution process sampled signal matrix
The influence of vibration signal phase difference, makes full use of the feature of signal, increases the number of cognitive user in logic, improves frequency spectrum detection
Accuracy.
Step s140, according to the characteristic value of phase matrix and the covariance matrix of orthogonal matrix calculate maximum eigenvalue with
The difference of the average energy value, obtains feature vector.
It uses signal energy as feature in the present embodiment, uses the difference (Maximum of maximum eigenvalue and average energy
Eigenvalue Substract average Energy, MSE) algorithm obtains matrix character, it can examined by MSE algorithm
Consider and simply carries out accurate profile extraction under average energy and maximum eigenvalue difference, it can and noise less in cognitive user number
The detection performance of frequency spectrum can be preferably improved in the case where relatively low.
Covariance matrix with phase matrix and orthogonal matrix is respectively as follows:
According to covariance matrix, relevant eigenvalue λ is calculated separatelyu(u=1,2 ..., M) then calculates maximum feature
The difference (MSE) of value and the average energy value:
According to above method, T is obtained1And T2, constitute two-dimensional feature vector T=[T1, T2]TCome representation signal feature to
Amount.
Step s150, tagsort division is carried out to feature vector, obtains the classification results of radio sampled signal.
Without limitation according to the method for feature vector progress tagsort division, for example can be divided by clustering algorithm
Class, can also be using gauss hybrid models (GMM) cluster etc..Wherein, GMM clusters the probability value of available each case, i.e.,
Obtain H0And H1Probability value.Be sorted in by GMM model it is with high accuracy while easy to use, it is available well
Classifying quality.Preferably, tagsort division is carried out to feature vector to be specifically as follows: several feature vectors of acquisition are constructed
Matrix obtains eigenmatrix；Eigenmatrix is input to frequency spectrum disaggregated model, obtains classification results；Wherein, frequency spectrum disaggregated model
For the Gaussian Mixture Clustering Model obtained according to the training of sample characteristics matrix.
Specifically, Gaussian Mixture cluster (Gaussian Mixture Model, GMM) expresses cluster using probabilistic model
Prototype, Gaussian Mixture distribution can indicate are as follows:
The distribution is made of k blending constituent altogether, the corresponding Gaussian Profile of each blending constituent, in formula, αiRepresent i-th
The corresponding mixed coefficint of a Gaussian Profile,And αi>0。P(x|μi, ∑i) expression mean value be μi, covariance is ∑i
Gaussian Profile, probability density function can indicate are as follows:
Frequency spectrum perception problem can regard two classification problems as, and in the case where frequency spectrum perception, GMM is two Gausses point
The superposition of cloth.P(x|μ1, ∑1) it is to estimate that feature belongs to the probability density of the first kind, it represents primary user and is not present, channel is available.P
(x|μ2, ∑2) then indicate that feature belongs to the probability density of the second class, primary user's presence is represented, channel is unavailable.
Herein only to be introduced for carrying out the classifying and dividing of feature to feature vector by Gaussian Mixture clustering method,
Details are not described herein for other classification methods.
Based on the above-mentioned technical proposal, frequency spectrum sensing method provided by the present embodiment, by wireless after noise reduction process
Electric sampled signal carries out IQ decomposition, by signal carry out with mutually and the analysis of orthogonal direction, realize to signal in the same direction and
The abundant digging utilization of orthogonal two direction character, logically increases the number of cognitive user, according to same phase matrix and just
It hands over the characteristic value of the covariance matrix of matrix to calculate the difference (MSE) of maximum eigenvalue and the average energy value, is obtained according to difference
Feature vector is analyzed by precise and stable property of the MSE to data fluctuations situation in matrix, can accurately extract the feature letter of matrix
Breath can preferably improve the detection performance of frequency spectrum in and the lower situation of signal-to-noise ratio less in cognitive user number, promote frequency
Detection performance is composed, frequency spectrum perception effect is optimized, realizes the precise classification to sampled signal.
In above-described embodiment to sampled signal carry out noise reduction process method without limitation, in order to reduce noise to cognition nothing
The influence of line electric system, guarantee system can also obtain ideal perceived effect under the lower environment of signal-to-noise ratio, can pass through EMD
Empirical mode decomposition method carries out signal decomposition to sampled signal, and carries out intrinsic mode functions signal to the signal after decomposition and mention
It takes, obtains de-noising signal.Specifically, the present embodiment is introduced noise reduction process process to based on sight spot model decomposing method.
Empirical mode decomposition method (Empirical Mode Decomposition, EMD) is using Fourier transform as base
One important breakthrough of the linear and stable state spectrum analysis of plinth, this method are the time scale features of foundation data itself to carry out
Signal decomposition, without presetting any basic function.This point and establish apriority harmonic wave basic function and wavelet basis function
On Fourier decomposition and wavelet-decomposing method have essential difference.Just because of such feature, EMD method is in theory
On can be applied to the decomposition of any kind of signal, thus on processing non-stationary and nonlinear data, have clearly
Advantage, be suitable for analyzing non-linear, non-stationary signal sequence.
Sophisticated signal is decomposed into limited intrinsic mode functions (IMF) from high frequency to low frequency by EMD algorithm, is decomposited and
Each IMF component contain original signal different time scales local feature signal.Sampled signal xi(n) at through EMD algorithm
After reason, it is decomposed into following form,
Wherein, IMF indicates intrinsic mode functions part, and r (n) indicates residual error portion.According to continuous mean square deviation as follows
Criterion finds the critical point m of high frequency band and low-frequency band.
Sampled signal is after EMD algorithm process, signal after having obtained following noise reduction:
In above-described embodiment without limitation to the process of feature vector progress cluster feature analysis, GMM can be selected to cluster
Algorithm, to deepen to carry out GMM clustering algorithm the understanding of tagsort division, its detailed process is introduced in the present embodiment,
Key step is as follows:
Obtain enough signal characteristics, construction feature matrixGiven Gaussian mixture components number
k。
Initialize the model parameter { (α of Gaussian Mixture distributioni, μi, ∑i)|1≤i≤k}。
According to Bayesian formula, all sample signal T are calculatedjThe posterior probability generated by each blending constituent, it may be assumed that γji=
PM(Zj=i | Tj) (1≤i≤k), determine each sample TjCluster mark λj: λj=argmax γji。
Calculate new mean vector:
Calculate new covariance matrix:
Calculate new mixed coefficint:
By model parameter { (α i,, μi, ∑i) | 1≤i≤k } it is updated to { (αi', μi', ∑i′)|1≤i≤k}。
According to sample signal TjCluster mark λj, by TjIt is divided to corresponding cluster.
It exports cluster and divides C={ C1, C2..., Ck}。
The training process of obtained frequency spectrum disaggregated model is trained to can refer to current model training based on Gaussian Mixture cluster
Method is trained by sample characteristics matrix, is iterated optimization to model parameter, until training result reaches preset defeated
It requires that the disaggregated model that training is completed can be obtained out, if classifying quality fall flat during actual test
When, optimizing and revising for model can be re-started, it is not limited here.
Wherein, without limitation to the parameter optimization method of frequency spectrum disaggregated model, it is preferable that can be according to Maximum-likelihood estimation
And expectation maximization EM algorithm is iterated optimization to model parameter.
Model parameter { (αi, μi, ∑i) | 1≤i≤k } use Maximum-likelihood estimation, i.e. maximization log-likelihood:
Optimization is iterated to LL (D) and is asked using expectation maximization (Expectation Maximization, EM) algorithm
Solve model parameter { (αi, μi, ∑i)|1≤i≤k}。
Parameter accuracy can be improved while reducing the number of iterations by above-mentioned parameter optimization method.
In addition, in order to which the sensing results to frequency spectrum carry out Adaptability Evaluation, to be got information about to sensing results,
Preferably, false-alarm probability and detection probability can be calculated to classification results after carrying out frequency spectrum perception；According to what is be calculated
False-alarm probability and detection probability carry out perceptual performance analysis.
Specifically, to the false-alarm probability P of systemfWith detection probability PdIt is defined as follows:
Pf=P [H1|H0]；
Pd=P [H0|H1]。
Fig. 2 show two kinds of signal processing effect diagrams of EMD+IQ and IQ, two methods using MSE as statistic,
Classified with GMM clustering algorithm.Through ROC curve as can be seen that in the case where signal-to-noise ratio is -14DbB environment, when pre- alarm probability, EMD+
The detection probability of IQ method is obviously higher than IQ method.
Frequency spectrum sensing device provided by the invention is introduced below, referring to FIG. 3, Fig. 3 mentions for the embodiment of the present invention
The structural block diagram of the frequency spectrum sensing device of confession；The apparatus may include: sampled signal acquiring unit 310, noise reduction processing unit
320, IQ decomposition unit 330, feature vector computing unit 340 are with tagsort unit 350.
Wherein, sampled signal acquiring unit 310 is mainly used for obtaining radio sampled signal；
Noise reduction processing unit 320 is mainly used for carrying out noise reduction process to sampled signal, obtains de-noising signal；
IQ decomposition unit 330 is mainly used for carrying out IQ decomposition to de-noising signal, obtains same phase matrix and orthogonal matrix；
Feature vector computing unit 340 is mainly used for；According to the spy of same phase matrix and the covariance matrix of orthogonal matrix
Value indicative calculates the difference of maximum eigenvalue and the average energy value, obtains feature vector；
Tagsort unit 350 is mainly used for carrying out tagsort division to feature vector, obtains radio sampled signal
Classification results.
Preferably, noise reduction processing unit is specifically as follows EMD noise reduction processing unit, for passing through EMD empirical mode decomposition
Method carries out signal decomposition to sampled signal, and carries out intrinsic mode functions signal extraction to the signal after decomposition, obtains noise reduction letter
Number.
Tagsort unit can specifically include:
Matrix construction subelement obtains eigenmatrix for several feature vector structural matrixes to acquisition；
Gaussian classification subelement obtains classification results for eigenmatrix to be input to frequency spectrum disaggregated model；Wherein, frequency
Spectrum disaggregated model is the Gaussian Mixture Clustering Model obtained according to the training of sample characteristics matrix.
Wherein, above-mentioned Gaussian classification subelement optimizes subelement by Gaussian parameter and joins to Gaussian Mixture Clustering Model
Number optimization, specifically, Gaussian parameter optimize subelement and are used for according to Maximum-likelihood estimation and expectation maximization EM algorithm to mould
Shape parameter is iterated optimization.
Preferably, frequency spectrum sensing device can further include interpretation of result unit, be used for: calculate false-alarm probability to classification results
And detection probability；Perceptual performance analysis is carried out according to the false-alarm probability and detection probability being calculated.
It should be noted that each unit in frequency spectrum sensing device in the specific embodiment of the invention, worked
Journey please refers to the corresponding specific embodiment of frequency spectrum sensing method, and details are not described herein.
Frequency spectrum perception equipment provided by the invention is introduced below, specifically the introduction of frequency spectrum perception equipment can refer to
The step of above-mentioned frequency spectrum sensing method, Fig. 4 are the structural block diagram of frequency spectrum perception equipment provided in an embodiment of the present invention；The equipment can
To include:
Memory 400, for storing computer program；
Processor 401, when for executing computer program the step of realization frequency spectrum sensing method.
Referring to FIG. 5, the structural schematic diagram of frequency spectrum perception equipment provided in an embodiment of the present invention, which can
Bigger difference is generated because configuration or performance are different, may include one or more processors (central
Processing units, CPU) 322 (for example, one or more processors) and memory 332, one or more
Store the storage medium 330 (such as one or more mass memory units) of application program 342 or data 344.Wherein, it deposits
Reservoir 332 and storage medium 330 can be of short duration storage or persistent storage.The program for being stored in storage medium 330 may include
One or more modules (diagram does not mark), each module may include to the series of instructions behaviour in data processing equipment
Make.Further, central processing unit 322 can be set to communicate with storage medium 330, hold in frequency spectrum perception equipment 301
Series of instructions operation in row storage medium 330.
Frequency spectrum perception equipment 301 can also include one or more power supplys 326, one or more wired or nothings
Wired network interface 350, one or more input/output interfaces 358, and/or, one or more operating systems 341,
Such as Windows ServerTM, Mac OS XTM, UnixTM, LinuxTM, FreeBSDTM etc..
Step in frequency spectrum sensing method described in above figure 1 can be realized by the structure of frequency spectrum perception equipment.
Frequency spectrum perception system provided in an embodiment of the present invention is introduced below, frequency spectrum perception system described below with
Above-described frequency spectrum perception equipment can correspond to each other reference.
Fig. 6 is the structural block diagram of frequency spectrum perception system provided in an embodiment of the present invention；The system may include: that radio is adopted
Sample equipment 600 and frequency spectrum perception equipment 601.
Radio sample devices 600 is mainly used for acquiring radio signal, obtains radio sampled signal, and by radio
Sampled signal is sent to frequency spectrum perception equipment；
Frequency spectrum perception equipment 601 is mainly used for obtaining radio sampled signal；Noise reduction process is carried out to sampled signal, is obtained
De-noising signal；IQ decomposition is carried out to de-noising signal, obtains same phase matrix and orthogonal matrix；；According to same phase matrix and orthogonal
The characteristic value of the covariance matrix of matrix calculates the difference of maximum eigenvalue and the average energy value, obtains feature vector；To feature
Vector carries out tagsort division, obtains the classification results of radio sampled signal.
Readable storage medium storing program for executing provided in an embodiment of the present invention is introduced below, readable storage medium storing program for executing described below with
Above-described frequency spectrum sensing method can correspond to each other reference.
A kind of readable storage medium storing program for executing disclosed by the invention, is stored thereon with program, and frequency is realized when program is executed by processor
The step of composing cognitive method.
It is apparent to those skilled in the art that for convenience and simplicity of description, the device of foregoing description,
The specific work process of equipment, storage medium and unit, can refer to corresponding processes in the foregoing method embodiment, herein no longer
It repeats.
In several embodiments provided by the present invention, it should be understood that disclosed device, system, storage medium and
Method may be implemented in other ways.For example, apparatus embodiments described above are merely indicative, for example, single
Member division, only a kind of logical function partition, there may be another division manner in actual implementation, for example, multiple units or
Component can be combined or can be integrated into another system, or some features can be ignored or not executed.Another point is shown
The mutual coupling, direct-coupling or communication connection shown or discussed can be through some interfaces, between device or unit
Coupling or communication connection are connect, can be electrical property, mechanical or other forms.
Unit may or may not be physically separated as illustrated by the separation member, shown as a unit
Component may or may not be physical unit, it can and it is in one place, or may be distributed over multiple networks
On unit.It can some or all of the units may be selected to achieve the purpose of the solution of this embodiment according to the actual needs.
It, can also be in addition, the functional units in various embodiments of the present invention may be integrated into one processing unit
It is that each unit physically exists alone, can also be integrated in one unit with two or more units.Above-mentioned integrated list
Member both can take the form of hardware realization, can also realize in the form of software functional units.
It, can if integrated unit is realized in the form of SFU software functional unit and when sold or used as an independent product
To be stored in a mobile terminal.Based on this understanding, technical solution of the present invention is substantially in other words to the prior art
The all or part of the part to contribute or the technical solution can be embodied in the form of software products, which deposits
It stores up in one storage medium, including some instructions are used so that a mobile terminal (can be mobile phone or tablet computer
Deng) execute all or part of the steps of each embodiment method of the present invention.And storage medium above-mentioned includes: USB flash disk, moves firmly
Disk, read-only memory (Read-Only Memory, ROM), random access memory (Random Access Memory, RAM),
The various media that can store program code such as magnetic or disk.
Each embodiment is described in a progressive manner in specification, the highlights of each of the examples are with other realities
The difference of example is applied, the same or similar parts in each embodiment may refer to each other.For device disclosed in embodiment
Speech, since it is corresponded to the methods disclosed in the examples, so being described relatively simple, related place is referring to method part illustration
?.
Professional further appreciates that, unit described in conjunction with the examples disclosed in the embodiments of the present disclosure
And algorithm steps, it can be realized with the combination of electronic hardware, terminal or the two, in order to clearly demonstrate hardware and software
Interchangeability generally describes each exemplary composition and step according to function in the above description.These functions are studied carefully
Unexpectedly it is implemented in hardware or software, the specific application and design constraint depending on technical solution.Professional technique people
Member can use different methods to achieve the described function each specific application, but this realization is it is not considered that super
The scope of the present invention out.
The step of method described in conjunction with the examples disclosed in this document or algorithm, can directly be held with hardware, processor
The combination of capable software module or the two is implemented.Software module can be placed in random access memory (RAM), memory, read-only deposit
Reservoir (ROM), electrically programmable ROM, electrically erasable ROM, register, hard disk, moveable magnetic disc, CD-ROM or technology
In any other form of storage medium well known in field.
Frequency spectrum sensing method provided by the present invention, device, equipment, system and readable storage medium storing program for executing have been carried out in detail above
It is thin to introduce.Used herein a specific example illustrates the principle and implementation of the invention, and above embodiments are said
It is bright to be merely used to help understand method and its core concept of the invention.It should be pointed out that for the ordinary skill of the art
, without departing from the principle of the present invention, can be with several improvements and modifications are made to the present invention for personnel, these improvement
It is also fallen within the protection scope of the claims of the present invention with modification.
Claims (10)
1. a kind of frequency spectrum sensing method characterized by comprising
Obtain radio sampled signal；
Noise reduction process is carried out to the sampled signal, obtains de-noising signal；
IQ decomposition is carried out to the de-noising signal, obtains same phase matrix and orthogonal matrix；
According to the characteristic value of the same phase matrix and the covariance matrix of orthogonal matrix calculating maximum eigenvalue and averagely
The difference of energy value, obtains feature vector；
Tagsort division is carried out to described eigenvector, obtains the classification results of the radio sampled signal.
2. frequency spectrum sensing method as described in claim 1, which is characterized in that described to carry out noise reduction process to the sampled signal
Include:
Signal decomposition is carried out to the sampled signal by EMD empirical mode decomposition method, and this is carried out to the signal after decomposition
Modular function signal extraction is levied, de-noising signal is obtained.
3. frequency spectrum sensing method as described in claim 1, which is characterized in that carry out tagsort division to described eigenvector
Include:
To several feature vector structural matrixes of acquisition, eigenmatrix is obtained；
The eigenmatrix is input to frequency spectrum disaggregated model, obtains classification results；Wherein, according to the frequency spectrum disaggregated model
The Gaussian Mixture Clustering Model that the training of sample characteristics matrix obtains.
4. frequency spectrum sensing method as claimed in claim 3, which is characterized in that the parameter optimization method of the frequency spectrum disaggregated model
Include:
Optimization is iterated to model parameter according to Maximum-likelihood estimation and expectation maximization EM algorithm.
5. frequency spectrum sensing method as described in claim 1, which is characterized in that further include:
False-alarm probability and detection probability are calculated to the classification results；
Perceptual performance analysis is carried out according to the false-alarm probability and detection probability being calculated.
6. a kind of frequency spectrum sensing device characterized by comprising
Sampled signal acquiring unit, for obtaining radio sampled signal；
Noise reduction processing unit obtains de-noising signal for carrying out noise reduction process to the sampled signal；
IQ decomposition unit obtains same phase matrix and orthogonal matrix for carrying out IQ decomposition to the de-noising signal；
Feature vector computing unit, for the characteristic value according to the same phase matrix and the covariance matrix of the orthogonal matrix
The difference for calculating maximum eigenvalue and the average energy value, obtains feature vector；
Tagsort unit obtains the radio sampled signal for carrying out tagsort division to described eigenvector
Classification results.
7. frequency spectrum sensing device as claimed in claim 6, which is characterized in that the noise reduction processing unit specifically: EMD noise reduction
Processing unit, the EMD noise reduction processing unit are used to carry out signal to the sampled signal by EMD empirical mode decomposition method
It decomposes, and intrinsic mode functions signal extraction is carried out to the signal after decomposition, obtain de-noising signal.
8. a kind of frequency spectrum perception equipment characterized by comprising
Memory, for storing computer program；
Processor, realizing the frequency spectrum sensing method as described in any one of claim 1 to 5 when for executing the computer program
Step.
9. a kind of frequency spectrum perception system characterized by comprising
Radio sample devices obtains radio sampled signal, and the radio is sampled and is believed for acquiring radio signal
Number it is sent to frequency spectrum perception equipment；
The frequency spectrum perception equipment, for obtaining radio sampled signal；Noise reduction process is carried out to the sampled signal, is dropped
Noise cancellation signal；IQ decomposition is carried out to the de-noising signal, obtains same phase matrix and orthogonal matrix；According to the same phase matrix and
The characteristic value of the covariance matrix of the orthogonal matrix calculates the difference of maximum eigenvalue and the average energy value, obtain feature to
Amount；Tagsort division is carried out to described eigenvector, obtains the classification results of the radio sampled signal.
10. a kind of readable storage medium storing program for executing, which is characterized in that be stored with program on the readable storage medium storing program for executing, described program is located
It manages and is realized when device executes as described in any one of claim 1 to 5 the step of frequency spectrum sensing method.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810989697.8A CN109039500A (en) | 2018-08-28 | 2018-08-28 | A kind of frequency spectrum sensing method, device, equipment, system and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810989697.8A CN109039500A (en) | 2018-08-28 | 2018-08-28 | A kind of frequency spectrum sensing method, device, equipment, system and storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109039500A true CN109039500A (en) | 2018-12-18 |
Family
ID=64625785
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810989697.8A Pending CN109039500A (en) | 2018-08-28 | 2018-08-28 | A kind of frequency spectrum sensing method, device, equipment, system and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109039500A (en) |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110071773A (en) * | 2019-04-28 | 2019-07-30 | 广东工业大学 | A kind of frequency spectrum sensing method, system and device |
| CN110460401A (en) * | 2019-07-05 | 2019-11-15 | 广东工业大学 | A kind of cooperative frequency spectrum sensing method clustered based on matrix decomposition and particle group optimizing |
| WO2020143300A1 (en) * | 2019-01-07 | 2020-07-16 | 哈尔滨工业大学（深圳） | Auditory attention state arousal level recognition method and apparatus, and storage medium |
| CN112073131A (en) * | 2020-07-29 | 2020-12-11 | 北京邮电大学 | Spectrum sensing method based on phase difference distribution curve analytic expression and related equipment |
| CN112073130A (en) * | 2020-07-29 | 2020-12-11 | 北京邮电大学 | Frequency spectrum sensing method based on three-point shaping of phase difference distribution curve and related equipment |
| CN112235061A (en) * | 2020-09-08 | 2021-01-15 | 北京邮电大学 | Frequency spectrum sensing method based on phase difference robust statistical characteristics and related equipment |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20160112080A1 (en) * | 2014-10-17 | 2016-04-21 | Intel Corporation | Communication device and method for processing a received signal |
| CN106911410A (en) * | 2017-05-02 | 2017-06-30 | 广东工业大学 | One kind communication primary user's cognitive method and system |
| CN107426736A (en) * | 2017-06-07 | 2017-12-01 | 广东工业大学 | The frequency spectrum sensing method and system of a kind of cognitive radio |
| CN108400826A (en) * | 2018-05-23 | 2018-08-14 | 北京科技大学 | A kind of frequency spectrum sensing method based on circulant matrix eigenvalue |
-
2018
- 2018-08-28 CN CN201810989697.8A patent/CN109039500A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20160112080A1 (en) * | 2014-10-17 | 2016-04-21 | Intel Corporation | Communication device and method for processing a received signal |
| CN106911410A (en) * | 2017-05-02 | 2017-06-30 | 广东工业大学 | One kind communication primary user's cognitive method and system |
| CN107426736A (en) * | 2017-06-07 | 2017-12-01 | 广东工业大学 | The frequency spectrum sensing method and system of a kind of cognitive radio |
| CN108400826A (en) * | 2018-05-23 | 2018-08-14 | 北京科技大学 | A kind of frequency spectrum sensing method based on circulant matrix eigenvalue |
Non-Patent Citations (2)
| Title |
|---|
| GOUNOU CHARLES SOBABE等: "A Cooperative Spectrum Sensing Algorithm Based on Unsupervised Learning", 《2017 10TH INTERNATIONAL CONGRESS ON IMAGE AND SIGNAL PROCESSING, BIOMEDICAL ENGINEERING AND INFORMATICS (CISP-BMEI)》 * |
| 宋昱辉: "认知无线电中频谱检测算法研究", 《中国优秀硕士学位论文全文数据库信息科技辑》 * |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2020143300A1 (en) * | 2019-01-07 | 2020-07-16 | 哈尔滨工业大学（深圳） | Auditory attention state arousal level recognition method and apparatus, and storage medium |
| CN110071773A (en) * | 2019-04-28 | 2019-07-30 | 广东工业大学 | A kind of frequency spectrum sensing method, system and device |
| CN110460401A (en) * | 2019-07-05 | 2019-11-15 | 广东工业大学 | A kind of cooperative frequency spectrum sensing method clustered based on matrix decomposition and particle group optimizing |
| CN110460401B (en) * | 2019-07-05 | 2022-02-11 | 广东工业大学 | Cooperative spectrum sensing method based on matrix decomposition and particle swarm optimization clustering |
| CN112073131A (en) * | 2020-07-29 | 2020-12-11 | 北京邮电大学 | Spectrum sensing method based on phase difference distribution curve analytic expression and related equipment |
| CN112073130A (en) * | 2020-07-29 | 2020-12-11 | 北京邮电大学 | Frequency spectrum sensing method based on three-point shaping of phase difference distribution curve and related equipment |
| CN112235061A (en) * | 2020-09-08 | 2021-01-15 | 北京邮电大学 | Frequency spectrum sensing method based on phase difference robust statistical characteristics and related equipment |
| CN112235061B (en) * | 2020-09-08 | 2021-10-22 | 北京邮电大学 | Frequency spectrum sensing method based on phase difference robust statistical characteristics and related equipment |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109039500A (en) | A kind of frequency spectrum sensing method, device, equipment, system and storage medium | |
| Zhou et al. | Direction-of-arrival estimation for coprime array via virtual array interpolation | |
| Qiao et al. | Gridless line spectrum estimation and low-rank Toeplitz matrix compression using structured samplers: A regularization-free approach | |
| CN109307862A (en) | A kind of target radiation source individual discrimination method | |
| Hory et al. | Spectrogram segmentation by means of statistical features for non-stationary signal interpretation | |
| CN103559888A (en) | Speech enhancement method based on non-negative low-rank and sparse matrix decomposition principle | |
| CN108462544B (en) | Spectrum sensing method and device | |
| Soro et al. | Joint time-frequency RSSI features for convolutional neural network-based indoor fingerprinting localization | |
| Zeng et al. | Improving SVM classification with imbalance data set | |
| Satija et al. | Cyclostationary features based modulation classification in presence of non Gaussian noise using sparse signal decomposition | |
| CN111010356A (en) | Underwater acoustic communication signal modulation mode identification method based on support vector machine | |
| Huang et al. | Design of learning engine based on support vector machine in cognitive radio | |
| Roth et al. | Reliable recovery of hierarchically sparse signals for Gaussian and Kronecker product measurements | |
| Wang et al. | A cooperative spectrum sensing method based on empirical mode decomposition and information geometry in complex electromagnetic environment | |
| CN109309538A (en) | A kind of frequency spectrum sensing method, device, equipment, system and storage medium | |
| Du et al. | SVM-assisted adaptive kernel power density clustering algorithm for millimeter wave channels | |
| Li et al. | Single-channel speech enhancement based on adaptive low-rank matrix decomposition | |
| Tan et al. | Multivariate statistical models for image denoising in the wavelet domain | |
| CN109525339A (en) | Frequency spectrum sensing method, device, equipment and the storage medium of cognitive radio | |
| Liu et al. | A novel signal separation algorithm for wideband spectrum sensing in cognitive networks | |
| CN110866609B (en) | Method, device, server and storage medium for acquiring interpretation information | |
| Parker et al. | Nonlinear time series classification using bispectrum‐based deep convolutional neural networks | |
| CN116776181A (en) | Terminal side load identification method, medium and system based on improved fuzzy clustering | |
| Orlitsky et al. | Algorithms for modeling distributions over large alphabets | |
| Huang et al. | Novel modulation recognizer for frequency-hopping signals based on persistence diagram |
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 | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20181218 |