CN107395301A - A kind of frequency spectrum sensing method based on K mean algorithms - Google Patents
A kind of frequency spectrum sensing method based on K mean algorithms Download PDFInfo
- Publication number
- CN107395301A CN107395301A CN201710706974.5A CN201710706974A CN107395301A CN 107395301 A CN107395301 A CN 107395301A CN 201710706974 A CN201710706974 A CN 201710706974A CN 107395301 A CN107395301 A CN 107395301A
- Authority
- CN
- China
- Prior art keywords
- class
- mme
- signal
- central point
- features
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 238000001228 spectrum Methods 0.000 title claims abstract description 56
- 238000000034 method Methods 0.000 title claims abstract description 27
- 238000012549 training Methods 0.000 claims abstract description 64
- 238000012360 testing method Methods 0.000 claims abstract description 55
- 230000008447 perception Effects 0.000 claims abstract description 52
- 238000001514 detection method Methods 0.000 claims abstract description 31
- 239000011159 matrix material Substances 0.000 claims description 110
- 238000004364 calculation method Methods 0.000 claims description 23
- 230000015572 biosynthetic process Effects 0.000 claims 1
- 230000000694 effects Effects 0.000 abstract description 3
- 238000005070 sampling Methods 0.000 description 10
- 238000000354 decomposition reaction Methods 0.000 description 6
- 241001269238 Data Species 0.000 description 5
- 230000001149 cognitive effect Effects 0.000 description 3
- 238000010276 construction Methods 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 238000002474 experimental method Methods 0.000 description 2
- 230000006978 adaptation Effects 0.000 description 1
- 238000000205 computational method Methods 0.000 description 1
- 239000012141 concentrate Substances 0.000 description 1
- 230000007423 decrease Effects 0.000 description 1
- 230000007812 deficiency Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000008676 import Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000002360 preparation method Methods 0.000 description 1
- 230000000750 progressive effect Effects 0.000 description 1
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
Landscapes
- Physics & Mathematics (AREA)
- Electromagnetism (AREA)
- Engineering & Computer Science (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Radar Systems Or Details Thereof (AREA)
Abstract
The invention discloses a kind of frequency spectrum sensing method based on K mean algorithms, obtains training signal, and the MME features of the training signal are calculated；Grader is obtained using K mean cluster algorithm and the MME features；Test signal is obtained, the test MME features of the test signal are calculated；Classification results are obtained to the test MME tagsorts using the grader, detection probability is calculated using the classification results.It can be seen that, a kind of frequency spectrum sensing method based on K mean algorithms provided in an embodiment of the present invention, the MME features of training signal are calculated, then obtain grader using K mean algorithms, are classified to obtain classification results by the test MME features importing grader of test signal again after training grader.Combined using MME features with K mean algorithms, improve the detection performance of frequency spectrum perception.The invention also discloses a kind of frequency spectrum sensing device based on K mean algorithms, above-mentioned technique effect can be equally realized.
Description
Technical field
The present invention relates to technical field of radio, more specifically to a kind of frequency spectrum perception side based on K mean algorithms
Method.
Background technology
As the progress and development of radiotechnics, all trades and professions are more and more to the demand of radio frequency, frequency spectrum money
The problem of source scarcity getting worse.At present, the management mode that frequency spectrum resource is authorized using unification of the motherland distribution, is divided into two by frequency spectrum
Type：Authorize frequency range and unauthorized frequency range.Wherein, frequency range is authorized in occupation of most of frequency spectrum resource, such as television broadcast band,
It is but many to authorize frequency range to be in idle condition；The open unauthorized frequency range used accounts for a seldom part for whole frequency spectrum resource, such as
The wireless networks such as WLAN, wireless MAN are being worked using unauthorized frequency range mostly, and the user in the frequency range is a lot, industry
Business amount is also very big, and radio frequency band tends to saturation substantially.Thus, the scarcity of frequency spectrum resource is not frequency truly
Spectrum resource deficiency, is more as caused by the availability of frequency spectrum is too low.
Want the utilization rate of raising frequency spectrum, the method for first having to find accurate frequency spectrum detection.Frequency more classical at present
Spectrum detection method has energy detection algorithm, still, due to the more difficult determination of decision threshold in energy detection algorithm, in low noise
It is easier than under environment to be influenceed to produce erroneous judgement by incorrect noise so as to cause detection performance drastically to decline, even can be certainly
The threshold value of adaptation, detector can not make correct judgement, therefore the testing result of frequency spectrum is not accurate enough.
Therefore, accurate frequency spectrum detecting result how is obtained, is that those skilled in the art need to solve the problems, such as.
The content of the invention
It is an object of the invention to provide a kind of frequency spectrum sensing method based on K mean algorithms, to obtain accurate frequency spectrum
Testing result.
To achieve the above object, the embodiments of the invention provide following technical scheme：
A kind of frequency spectrum sensing method based on K mean algorithms, including：
Training signal is obtained, the MME features of the training signal are calculated；
Grader is obtained using K mean cluster algorithm and the MME features；
Test signal is obtained, the test MME features of the test signal are calculated；
Classification results are obtained to the test MME tagsorts using the grader, calculated using the classification results
Detection probability.
Wherein, the acquisition training signal, the MME features of the training signal are calculated, including：
The first signal characteristic and secondary signal feature are gathered, first signal characteristic is formed with the secondary signal feature
Training signal；Wherein described first signal characteristic is the signal characteristic that in the presence of primary user, secondary user collects, second letter
When number being characterized as that the primary user is not present, signal characteristic that secondary user collects；
Multiple secondary users in a detecting period section are calculated using the training signal first perceives matrix；
Described first perception matrix is divided into the second of the first predetermined number and perceives matrix；
The each second MME features for perceiving matrix are calculated.
Wherein, grader is obtained using K mean cluster algorithm and the MME features, including：
S301, concentrated in the data object being made up of the MME features, the random MME features for determining the second predetermined number
Class central point of the object as the second predetermined number, form the initial classes of the second predetermined number；Wherein, each class is included in class
Heart point and non-class central point；
S302, the non-class central point in each initial classes is calculated to the distance of each class central point, will each non-class center
Point is assigned to the class where the nearest class central point of each non-class central point of distance, forms target class；
S303, calculate target class in all MME feature objects average value, judge the average value whether with the target
The class central point of class is identical, if so, then stopping, determining grader；If it is not, the class central point of the target class is then reduced to institute
The non-class central point of target class is stated, the class central point using the average value as the target class, using the target class as just
Beginning class, return to S302.
Wherein, after the described first perception matrix being divided into the second perception matrix of predetermined number, in addition to：
The signal vector of described second perception matrix is decomposed into I, Q two parts, obtains the 3rd perception matrix；
The then MME features that each second perception matrix is calculated, including：
The each 3rd MME features for perceiving matrix are calculated.
Wherein, after the calculating detection probability, in addition to：
Calculate false-alarm probability.
A kind of frequency spectrum sensing device based on K mean algorithms, including：
MME feature calculation modules, for obtaining training signal, the MME features of the training signal are calculated；
Classifier calculated module, for obtaining grader using K mean cluster algorithm and the MME features；
MME feature calculation modules are tested, for obtaining test signal, the test MME that the test signal is calculated is special
Sign；
Sort module, for obtaining classification results to the test MME tagsorts using the grader, using described
Classification results calculate detection probability.
Wherein, the MME feature calculation modules, including：
Training signal collecting unit, for gathering the first signal characteristic and secondary signal feature, first signal characteristic
With the secondary signal feature composing training signal；Wherein described first signal characteristic is secondary user's collection in the presence of primary user
The signal characteristic arrived, when the secondary signal is characterized as that the primary user is not present, signal characteristic that secondary user collects；
First perceives matrix calculation unit, multiple in a detecting period section for being calculated using the training signal
The first of secondary user perceives matrix；
Grouped element, matrix is perceived for the described first perception matrix to be divided into the second of the first predetermined number；
MME feature calculation units, for each second MME features for perceiving matrix to be calculated.
Wherein, the classifier calculated module, including：
Class center determining unit, it is random to determine that second is default for being concentrated in the data object being made up of the MME features
Class central point of the MME feature objects of number as the second predetermined number, form the initial classes of the second predetermined number；Wherein, often
Individual class includes class central point and non-class central point；
Target class forms unit, for calculating the non-class central point in each initial classes to the distance of each class central point,
The class each non-class central point being assigned to where the nearest class central point of each non-class central point of distance, forms target class；
Judging unit, calculate target class in all MME feature objects average value, judge the average value whether with it is described
The class central point of target class is identical, if so, then stopping, determining grader；If it is not, then the class central point of the target class is reduced
For the non-class central point of the target class, the class central point using the average value as the target class, the target class is made
For initial classes, continue to call the target class to form unit.
Wherein, in addition to：
Decomposing module, after the described first perception matrix is divided into the second perception matrix of predetermined number, by described in
The signal vector of second perception matrix is decomposed into I, Q two parts, obtains the 3rd perception matrix；
The then MME feature calculation units, are specifically used for：
The each 3rd MME features for perceiving matrix are calculated.
Wherein, in addition to：
False-alarm probability computing module, after calculating detection probability, calculate false-alarm probability.
By above scheme, the present invention provides a kind of frequency spectrum sensing method based on K mean algorithms, obtains training letter
Number, the MME features of the training signal are calculated；Grader is obtained using K mean cluster algorithm and the MME features；Obtain
Test signal is taken, the test MME features of the test signal are calculated；Using the grader to the test MME features
Classification obtains classification results, and detection probability is calculated using the classification results.
It can be seen that a kind of frequency spectrum sensing method based on K mean algorithms provided in an embodiment of the present invention, calculates training signal
MME features, grader then is obtained using K mean algorithms, again lead the test MME features of test signal after training grader
Enter grader to be classified to obtain classification results.Combined using MME features with K mean algorithms, improve the detection of frequency spectrum perception
Performance.
Brief description of the drawings
In order to illustrate more clearly about the embodiment of the present invention or technical scheme of the prior art, below will be to embodiment or existing
There is the required accompanying drawing used in technology description to be briefly described, it should be apparent that, drawings in the following description are only this
Some embodiments of invention, for those of ordinary skill in the art, on the premise of not paying creative work, can be with
Other accompanying drawings are obtained according to these accompanying drawings.
Fig. 1 is a kind of frequency spectrum sensing method flow chart based on K mean algorithms disclosed in the embodiment of the present invention；
Fig. 2 is a kind of flow chart for calculating training signal MME features disclosed in the embodiment of the present invention；
Fig. 3 is a kind of flow chart for calculating grader disclosed in the embodiment of the present invention；
Fig. 4 is a kind of specifically frequency spectrum sensing method experiment effect based on K mean algorithms disclosed in the embodiment of the present invention
Figure；
Fig. 5 is a kind of frequency spectrum sensing device structural representation based on K mean algorithms disclosed in the embodiment of the present invention.
Embodiment
Below in conjunction with the accompanying drawing in the embodiment of the present invention, the technical scheme in the embodiment of the present invention is carried out clear, complete
Site preparation describes, it is clear that described embodiment is only part of the embodiment of the present invention, rather than whole embodiments.It is based on
Embodiment in the present invention, those of ordinary skill in the art are obtained every other under the premise of creative work is not made
Embodiment, belong to the scope of protection of the invention.
The embodiment of the invention discloses a kind of frequency spectrum sensing method based on K mean algorithms, is examined with obtaining accurate frequency spectrum
Survey result.
Referring to Fig. 1, a kind of frequency spectrum sensing method based on K mean algorithms provided in an embodiment of the present invention, specifically include：
S101, training signal is obtained, the MME features of the training signal are calculated；
Specifically, in training, first in the presence of primary user (PU), signal data is gathered using secondary user (SU), so
Afterwards when primary user is not present, signal data is gathered using secondary user, both signal datas constitute training signal.
Assuming that it is N to have K SU and each SU sampling number in cognitive radio networks.We assume that H0It is expressed as
Signal is not present, H1It is expressed as signal presence.Therefore training signal can be expressed from the next under assuming at two kinds：
Wherein si(n) primary user (PU) signal, w are representedi(n) it is 0 to represent average, variance σ2White Gaussian noise signal.
Using training signal, the perception matrix in a detecting period is obtained, so as to calculate covariance matrix, utilizes association
The ratio between eigenvalue of maximum and minimal eigenvalue of variance matrix obtain MME features.
S102, grader is obtained using K mean cluster algorithm and the MME features；
Specifically, matrix will be perceived and is divided into multigroup, so as to obtain multiple MME features, a data are determined using MME features
Object set, final grader is obtained using set of data objects and K mean algorithms.
S103, test signal is obtained, the test MME features of the test signal are calculated；
Specifically, after getting test signal, the MME features of test signal, and construction feature matrix are determined.
S104, classification results are obtained to the test MME tagsorts using the grader, utilize the classification results
Calculate detection probability.
Specifically, eigenmatrix is imported into grader, obtains classification results, and detection probability is calculated using classification results,
Namely calculate detection performance index.
At the same time it can also further calculate classification results, its false-alarm probability is calculated, to know that primary user is not present
When, probability existing for primary user is thought in system mistake.
By above scheme, a kind of frequency spectrum sensing method based on K mean algorithms provided in an embodiment of the present invention, meter
The MME features of training signal are calculated, then obtain grader using K mean algorithms, are trained after grader again by test signal
Test MME features import grader and are classified to obtain classification results.Combined using MME features with K mean algorithms, improve frequency
Compose the detection performance perceived.
The embodiment of the present invention provides a kind of frequency spectrum sensing method specifically based on K mean algorithms, is different from above-mentioned implementation
Example, the embodiment of the present invention have made specific restriction and explanation, other step contents and above-mentioned implementation to S101 in above-described embodiment
Example is roughly the same, and particular content may be referred to above-described embodiment, and here is omitted.Specifically, specifically wrapped referring to Fig. 2, S101
Include：
S201, the first signal characteristic of collection and secondary signal feature, first signal characteristic are special with the secondary signal
Levy composing training signal；Wherein described first signal characteristic is the signal characteristic that in the presence of primary user, secondary user collects, described
When secondary signal is characterized as that the primary user is not present, signal characteristic that secondary user collects；
Specifically, in training, first in the presence of primary user (PU), signal data is gathered using secondary user (SU), so
Afterwards when primary user is not present, signal data is gathered using secondary user, both signal datas constitute training signal.Training letter
Number it can be expressed asWherein si(n) primary user (PU) signal, w are representedi
(n) it is 0 to represent average, variance σ2White Gaussian noise signal.H0It is expressed as signal to be not present, H1It is expressed as signal presence；M is
The number of secondary user, the sampling number of each user is N.
S202, multiple secondary users in a detecting period section are calculated using the training signal first perceive square
Battle array；
Specifically, Xi=[xi(1)xi(2)....xi(N) i-th of SU sampling matrix] is represented, then in a detecting period
The perception matrix of M time users can be expressed as in section：
S203, the described first perception matrix is divided into the second of the first predetermined number and perceives matrix；
Specifically, by user grouping, the perception matrix after being grouped.Packet number is set according to actual conditions.Assuming that
There are 9 user collaborations to perceive, sampled point 1000 is default to be divided into 3 groups；The data of so this 9 user's collections are exactly one
The matrix X that individual 9 row 1000 arranges,It is divided into three groups：
Group one：
Group two：
Group three：
S204, each second MME features for perceiving matrix are calculated.
Specifically, each second covariance matrix for perceiving matrix is obtained using each second perception matrix, so as to obtain
The MME features of every group of userWherein it is λmaxThe eigenvalue of maximum of covariance matrix, λminFor minimal eigenvalue.
The embodiment of the present invention provides a kind of frequency spectrum sensing method specifically based on K mean algorithms, is different from above-mentioned implementation
Example, the embodiment of the present invention have made specific restriction and explanation, other step contents and above-mentioned implementation to S102 in above-described embodiment
Example is roughly the same, and particular content may be referred to above-described embodiment, and here is omitted.Specifically, specifically wrapped referring to Fig. 3, S102
Include：
S301, concentrated in the data object being made up of the MME features, the random MME features for determining the second predetermined number
Class central point of the object as the second predetermined number, form the initial classes of the second predetermined number；Wherein, each class is included in class
Heart point and non-class central point；
Specifically, MME features constitute a set of data objects, concentrate K object conduct of selection in data object at random
Class central point, so as to obtain K class, there is a central point in class, and remaining point is non-central point.
S302, the non-class central point in each initial classes is calculated to the distance of each class central point, will each non-class center
Point is assigned to the class where the nearest class central point of each non-class central point of distance, forms target class；
Specifically, it is necessary to which the non-class central point in all classes is separately dispensed into apart from it most in K mean cluster algorithm
In class where near class central point, target class is formed after being assigned.
S303, calculate target class in all MME feature objects average value, judge the average value whether with the target
The class central point of class is identical, if so, then stopping, determining grader；If it is not, the class central point of the target class is then reduced to institute
The non-class central point of target class is stated, the class central point using the average value as the target class, using the target class as just
Beginning class, return to S302.
Specifically, the average value of all MME feature objects in target class is calculated, judges that the average value in current goal class is
No is exactly current class center, if it is not, then need class central point being reduced to non-class central point, and using average value as new
Class central point, be then back to S302 continue calculate distance and distribute to form new target class, until the average value in target class
When being exactly its central point, stop circulation and determine grader.
In order to improve classifying quality, so as to further improve the accuracy of frequency spectrum perception, the embodiment of the present invention provides a kind of
Frequency spectrum sensing method specifically based on K mean algorithms, based on above-described embodiment, the embodiment of the present invention also includes after S203：
The signal vector of described second perception matrix is decomposed into I, Q two parts, obtains the 3rd perception matrix；
The then MME features that each second perception matrix is calculated, including：
The each 3rd MME features for perceiving matrix are calculated.
Specifically, it is assumed that the signal vector of the sampling of i-th of SU user is Xi=[xi(1)xi(2)...xi(N)].By XiPoint
Solution is as follows into I and Q two parts, expression formula：
Wherein fcIt is to carry frequency, fsIt is sample frequency.So the perception matrix of the group one after being decomposed in above-described embodiment can
To be expressed as：
Group two, group three are also equally decomposed, after decomposition
Perception matrix obtain new covariance matrix, so as to obtain MME features.It ensure that reception signal is deposited in new covariance matrix
Time, the maximum of signal correlation between space or phase related (I and Q component) utilize, therefore obtained after decomposing
MME characteristic values can more reflect the specifying information of current demand signal, improve classifying quality, so as to improve the performance of detection.
With reference to figure 4, for the experiment effect figure of the present embodiment, wherein PdFor detection probability, PfFor false-alarm probability.As can be seen that
Under the conditions of SNR=-12, the Detection results by the use of RMET as feature are better than the Detection results by the use of energy as feature.And add
After upper signal decomposition, its detection performance has further lifting.
A kind of frequency spectrum sensing method specifically based on K mean algorithms provided is provided to the present invention below, specifically included：
1st step, obtain training signal；
In training, first in the presence of primary user (PU), signal data is gathered using secondary user (SU), then primary
When family is not present, signal data is gathered using secondary user, both signal datas constitute training signal.
Specifically, it is assumed that it is N to have K SU and each SU sampling number in cognitive radio networks.H0It is expressed as
Signal is not present, H1It is expressed as signal presence.Therefore the training signal received can be byTable
Show.
Wherein si(n) primary user (PU) signal, w are representedi(n) it is 0 to represent average, variance σ2White Gaussian noise signal.
2nd step, determine the first perception matrix；
Specifically, it is assumed that Xi=[xi(1)xi(2)....xi(N) sampling matrix of i-th of SU user] is represented, then at one
Perception matrix in detecting period section can be expressed as
3rd step, by secondary user grouping, so as to obtain the MME features of every group user.
It should be noted that covariance matrix can be obtained according to matrix is perceived, so as to obtain MME features.
The covariance matrix of above-mentioned perception matrix can be expressed asChooseAs signal
MME features.Wherein λmaxFor the eigenvalue of maximum of covariance matrix, λminFor minimal eigenvalue.
After secondary user grouping, every group of MME feature T, particular user group and T corresponding relation are obtained using the above method
As shown in table 1.
Table 1
It should be noted that can be by above-mentioned vector of samples Xi=[xi(1)xi(2)....xi(N) IQ decomposition] is carried out,
Expand dimension, expanded so as to the dimension of matrix.Specifically, to Xi=[xi(1)xi(2)....xi(N)] the IQ that vector is carried out
Point solution's expression is
Wherein fcIt is to carry frequency, fsIt is sample frequency.So the perception matrix after decomposing can be expressed as：
Compared with classical sample covariance matrix R (N), new sample covariance matrix RY(N) ensure that in reception signal
The maximum utilization of signal correlation between existing time, space or phase related (I and Q component).
The method decomposed by above-mentioned IQ,, can be by each matrix after obtaining multiple perception matrixes by user grouping
Vector carries out IQ decomposition, so that matrix is perceived after being decomposed using IQ obtains new covariance matrix and new MME features, specifically
Corresponding relation between user's group and T and new T is as shown in table 2, and wherein GMME represents to carry out the characteristic value obtained after IQ decomposition.
Table 2
4th step, grader is determined using K mean algorithms；
Computational methods specifically may be referred to the S301 to S303 in above-described embodiment, and here is omitted.
Finally give formulaJudge A if the formula is met*=1, sentence if being unsatisfactory for
Determine A*=0.
It should be noted that S=1 and S=0 is allowed to represent PU state respectively.The availability A of channel can be defined as：When A=0 represents that channel can not use, A=1 represents that channel can use.
5th step, determine the characteristic value of test matrix；
Specifically, the collecting test signal in unknown network, the test matrix and characteristic value of test signal are determined, so as to
To characteristic vector, wherein the step of calculating characteristic value of the characteristic value of test signal with calculating training signal is substantially the same, specifically
The 1st step to the 3rd step is may be referred to, here is omitted.According to characteristic value, can define test matrix isWherein, L represents the number of training characteristics.
6th step calculates testing result；
Test matrix is imported in grader, so as to obtain classification results, utilizes formulaIt is determined that
Detection probability PdWith false-alarm probability Pfa。
A kind of frequency spectrum sensing device based on K mean algorithms provided in an embodiment of the present invention is introduced below, referred to
Fig. 5, the embodiment of the present invention specifically include：
MME feature calculation modules 401, for obtaining training signal, the MME features of the training signal are calculated；
Specifically, in training, first in the presence of primary user (PU), signal data is gathered using secondary user (SU), so
Afterwards when primary user is not present, signal data is gathered using secondary user, both signal datas constitute training signal.
Assuming that it is N to have K SU and each SU sampling number in cognitive radio networks.We assume that H0It is expressed as
Signal is not present, H1It is expressed as signal presence.Therefore training signal can be expressed from the next under assuming at two kinds：
Wherein si(n) primary user (PU) signal, w are representedi(n) it is 0 to represent average, variance σ2White Gaussian noise signal.
MME feature calculation modules 401 utilize training signal, the perception matrix in a detecting period are obtained, so as to calculate
Go out covariance matrix, MME features are obtained using the ratio between the eigenvalue of maximum of covariance matrix and minimal eigenvalue.
Classifier calculated module 402, for obtaining grader using K mean cluster algorithm and the MME features；
Specifically, classifier calculated module 402 will perceive matrix be divided into it is multigroup, so as to obtain multiple MME features, utilize
MME features determine a set of data objects, and final grader is obtained using set of data objects and K mean algorithms.
MME feature calculation modules 403 are tested, for obtaining test signal, the test MME of the test signal is calculated
Feature；
Specifically, after getting test signal, test MME feature calculation modules 403 determine the MME features of test signal,
And construction feature matrix.
Sort module 404, for obtaining classification results to the test MME tagsorts using the grader, utilize
The classification results calculate detection probability.
Specifically, eigenmatrix is imported grader by sort module 404, obtains classification results, and utilize classification results meter
Detection probability is calculated, that is, calculates detection performance index.
At the same time it can also further calculate classification results, its false-alarm probability is calculated, to know that primary user is not present
When, probability existing for primary user is thought in system mistake.
By above scheme, a kind of frequency spectrum sensing device based on K mean algorithms provided in an embodiment of the present invention,
MME feature calculation modules 401 calculate the MME features of training signal, and then classifier calculated module 402 is obtained using K mean algorithms
To grader, sort module 404 is again classified the test MME features importing grader of test signal after training grader
Obtain classification results.Combined using MME features with K mean algorithms, improve the detection performance of frequency spectrum perception.
The embodiment of the present invention provides a kind of frequency spectrum sensing device specifically based on K mean algorithms, is different from above-mentioned implementation
Example, the embodiment of the present invention have made specific restriction and explanation, other modules to MME feature calculation modules 401 in above-described embodiment
Content is roughly the same with above-described embodiment, and particular content may be referred to above-described embodiment, and here is omitted.MME feature calculations
Module 401 specifically includes：
Training signal collecting unit, for gathering the first signal characteristic and secondary signal feature, first signal characteristic
With the secondary signal feature composing training signal；Wherein described first signal characteristic is secondary user's collection in the presence of primary user
The signal characteristic arrived, when the secondary signal is characterized as that the primary user is not present, signal characteristic that secondary user collects；
Specifically, in training, training signal collecting unit utilizes secondary user (SU) first in the presence of primary user (PU)
Signal data is gathered, then when primary user is not present, gathers signal data using secondary user, both signal datas constitute
Training signal.Training signal can be expressed asWherein si(n) represent
Primary user (PU) signal, wi(n) it is 0 to represent average, variance σ2White Gaussian noise signal.H0It is expressed as signal to be not present, H1Table
It is shown as signal presence；M is the number of time user, and the sampling number of each user is N.
First perceives matrix calculation unit, multiple in a detecting period section for being calculated using the training signal
The first of secondary user perceives matrix；
Specifically, Xi=[xi(1)xi(2)....xi(N) i-th of SU sampling matrix] is represented, then in a detecting period
The perception matrix of M time users can be expressed as in section：
Grouped element, matrix is perceived for the described first perception matrix to be divided into the second of the first predetermined number；
Specifically, grouped element is by user grouping, the perception matrix after being grouped.Packet number is set according to actual conditions
It is fixed.Assuming that there is 9 user collaborations to perceive, sampled point 1000 is default to be divided into 3 groups；The data of so this 9 user's collections
It is exactly the matrix X of 9 rows 1000 row,It is divided into three groups：
Group one：
Group two：
Group three：
MME feature calculation units, for each second MME features for perceiving matrix to be calculated.
Specifically, MME feature calculation units obtain each second association side for perceiving matrix using each second perception matrix
Poor matrix, so as to obtain the MME features of every group of userWherein it is λmaxThe eigenvalue of maximum of covariance matrix, λminFor
Minimal eigenvalue.
The embodiment of the present invention provides a kind of frequency spectrum sensing device specifically based on K mean algorithms, is different from above-mentioned implementation
Example, the embodiment of the present invention has been made specific restriction to classifier calculated module 402 in above-described embodiment and illustrated, in other modules
Appearance is roughly the same with above-described embodiment, and particular content may be referred to above-described embodiment, and here is omitted.Classifier calculated module
402 specifically include：
Class center determining unit, it is random to determine that second is default for being concentrated in the data object being made up of the MME features
Class central point of the MME feature objects of number as the second predetermined number, form the initial classes of the second predetermined number；Wherein, often
Individual class includes class central point and non-class central point；
Specifically, MME features constitute a set of data objects, and class center determining unit is concentrated in data object select at random
K object is selected as class central point, so as to obtain K class, there is a central point in class, and remaining point is non-central point.
Target class forms unit, for calculating the non-class central point in each initial classes to the distance of each class central point,
The class each non-class central point being assigned to where the nearest class central point of each non-class central point of distance, forms target class；
Specifically, in K mean cluster algorithm, target class, which forms unit, to be needed to distinguish the non-class central point in all classes
It is assigned in the class where the class central point nearest apart from it, target class is formed after being assigned.
Judging unit, calculate target class in all MME feature objects average value, judge the average value whether with it is described
The class central point of target class is identical, if so, then stopping, determining grader；If it is not, then the class central point of the target class is reduced
For the non-class central point of the target class, the class central point using the average value as the target class, the target class is made
For initial classes, continue to call the target class to form unit.
Specifically, the average value of all MME feature objects in target class is calculated, judging unit is judged in current goal class
Whether average value is exactly current class center, if it is not, then need class central point being reduced to non-class central point, and will be average
Value is then back to S302 and continues to calculate distance and distribute to form new target class, until in target class as new class central point
Average value when being exactly its central point, stop circulation and determine grader.
In order to improve classifying quality, so as to further improve the accuracy of frequency spectrum perception, the embodiment of the present invention provides a kind of
Frequency spectrum sensing device specifically based on K mean algorithms, based on above-described embodiment, the embodiment of the present invention also includes：
Decomposing module, after the described first perception matrix is divided into the second perception matrix of predetermined number, by described in
The signal vector of second perception matrix is decomposed into I, Q two parts, obtains the 3rd perception matrix；
The then MME feature calculation units, are specifically used for：
The each 3rd MME features for perceiving matrix are calculated.
Specifically, it is assumed that the signal vector of the sampling of i-th of SU user is Xi=[xi(1)xi(2)...xi(N)].Decompose
Module is by XiI and Q two parts are resolved into, expression formula is as follows：
Wherein fcIt is to carry frequency, fsIt is sample frequency.So the perception matrix of the group one after being decomposed in above-described embodiment can
To be expressed as：
Group two, group three are also equally decomposed, after decomposition
Perception matrix obtain new covariance matrix, so as to obtain MME features.It ensure that reception signal is deposited in new covariance matrix
Time, the maximum of signal correlation between space or phase related (I and Q component) utilize, therefore obtained after decomposing
MME characteristic values can more reflect the specifying information of current demand signal, improve classifying quality, so as to improve the performance of detection.
Each embodiment is described by the way of progressive in this specification, what each embodiment stressed be and other
The difference of embodiment, between each embodiment identical similar portion mutually referring to.
The foregoing description of the disclosed embodiments, professional and technical personnel in the field are enable to realize or using the present invention.
A variety of modifications to these embodiments will be apparent for those skilled in the art, as defined herein
General Principle can be realized in other embodiments without departing from the spirit or scope of the present invention.Therefore, it is of the invention
The embodiments shown herein is not intended to be limited to, and is to fit to and principles disclosed herein and features of novelty phase one
The most wide scope caused.
Claims (10)
- A kind of 1. frequency spectrum sensing method based on K mean algorithms, it is characterised in that including：Training signal is obtained, the MME features of the training signal are calculated；Grader is obtained using K mean cluster algorithm and the MME features；Test signal is obtained, the test MME features of the test signal are calculated；Classification results are obtained to the test MME tagsorts using the grader, is calculated and detected using the classification results Probability.
- 2. frequency spectrum sensing method according to claim 1, it is characterised in that the acquisition training signal, institute is calculated The MME features of training signal are stated, including：Gather the first signal characteristic and secondary signal feature, first signal characteristic and the secondary signal feature composing training Signal；Wherein described first signal characteristic is the signal characteristic that in the presence of primary user, secondary user collects, and the secondary signal is special Levy when being not present for the primary user, the signal characteristic that secondary user collects；Multiple secondary users in a detecting period section are calculated using the training signal first perceives matrix；Described first perception matrix is divided into the second of the first predetermined number and perceives matrix；The each second MME features for perceiving matrix are calculated.
- 3. frequency spectrum sensing method according to claim 2, it is characterised in that special using K mean cluster algorithm and the MME Grader is obtained, including：S301, concentrated in the data object being made up of the MME features, the random MME feature objects for determining the second predetermined number As the class central point of the second predetermined number, the initial classes of the second predetermined number of formation；Wherein, each class includes class central point With non-class central point；S302, the non-class central point in each initial classes is calculated to the distance of each class central point, will each non-class central point point The class being fitted on where the nearest class central point of each non-class central point of distance, forms target class；S303, calculate target class in all MME feature objects average value, judge the average value whether with the target class Class central point is identical, if so, then stopping, determining grader；If it is not, the class central point of the target class is then reduced to the mesh Mark the non-class central point of class, the class central point using the average value as the target class, using the target class as initial classes, Return to S302.
- 4. frequency spectrum sensing method according to claim 2, it is characterised in that the described first perception matrix is divided into default After several second perceives matrix, in addition to：The signal vector of described second perception matrix is decomposed into I, Q two parts, obtains the 3rd perception matrix；The then MME features that each second perception matrix is calculated, including：The each 3rd MME features for perceiving matrix are calculated.
- 5. frequency spectrum sensing method as claimed in any of claims 1 to 4, it is characterised in that described to calculate detection generally After rate, in addition to：Calculate false-alarm probability.
- A kind of 6. frequency spectrum sensing device based on K mean algorithms, it is characterised in that including：MME feature calculation modules, for obtaining training signal, the MME features of the training signal are calculated；Classifier calculated module, for obtaining grader using K mean cluster algorithm and the MME features；MME feature calculation modules are tested, for obtaining test signal, the test MME features of the test signal are calculated；Sort module, for obtaining classification results to the test MME tagsorts using the grader, utilize the classification As a result detection probability is calculated.
- 7. frequency spectrum sensing device according to claim 6, it is characterised in that the MME feature calculation modules, including：Training signal collecting unit, for gathering the first signal characteristic and secondary signal feature, first signal characteristic and institute State secondary signal feature composing training signal；Wherein described first signal characteristic be primary user in the presence of, what secondary user collected Signal characteristic, when the secondary signal is characterized as that the primary user is not present, signal characteristic that secondary user collects；First perceives matrix calculation unit, for multiple use in a detecting period section to be calculated using the training signal The first of family perceives matrix；Grouped element, matrix is perceived for the described first perception matrix to be divided into the second of the first predetermined number；MME feature calculation units, for each second MME features for perceiving matrix to be calculated.
- 8. frequency spectrum sensing device according to claim 7, it is characterised in that the classifier calculated module, including：Class center determining unit, it is random to determine the second predetermined number for being concentrated in the data object being made up of the MME features Class central point of the MME feature objects as the second predetermined number, form the initial classes of the second predetermined number；Wherein, each class Include class central point and non-class central point；Target class forms unit, will be every for calculating the non-class central point in each initial classes to the distance of each class central point Individual non-class central point is assigned to the class where the nearest class central point of each non-class central point of distance, forms target class；Judging unit, calculate target class in all MME feature objects average value, judge the average value whether with the target The class central point of class is identical, if so, then stopping, determining grader；If it is not, the class central point of the target class is then reduced to institute The non-class central point of target class is stated, the class central point using the average value as the target class, using the target class as just Beginning class, continue to call the target class to form unit.
- 9. frequency spectrum sensing device according to claim 7, it is characterised in that also include：Decomposing module, after the described first perception matrix is divided into the second perception matrix of predetermined number, by described second The signal vector for perceiving matrix is decomposed into I, Q two parts, obtains the 3rd perception matrix；The then MME feature calculation units, are specifically used for：The each 3rd MME features for perceiving matrix are calculated.
- 10. the frequency spectrum sensing device according to any one in claim 6 to 9, it is characterised in that also include：False-alarm probability computing module, after calculating detection probability, calculate false-alarm probability.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710706974.5A CN107395301B (en) | 2017-08-17 | 2017-08-17 | Spectrum sensing method and device based on K-means algorithm |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710706974.5A CN107395301B (en) | 2017-08-17 | 2017-08-17 | Spectrum sensing method and device based on K-means algorithm |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN107395301A true CN107395301A (en) | 2017-11-24 |
| CN107395301B CN107395301B (en) | 2020-09-11 |
Family
ID=60353274
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710706974.5A Expired - Fee Related CN107395301B (en) | 2017-08-17 | 2017-08-17 | Spectrum sensing method and device based on K-means algorithm |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107395301B (en) |
Cited By (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108462544A (en) * | 2018-03-27 | 2018-08-28 | 广东工业大学 | A kind of frequency spectrum sensing method and device |
| CN108494509A (en) * | 2018-03-09 | 2018-09-04 | 广东工业大学 | A kind of cooperative frequency spectrum sensing method and device |
| CN108566254A (en) * | 2018-03-27 | 2018-09-21 | 广东工业大学 | A kind of frequency spectrum sensing method and device |
| CN108712222A (en) * | 2018-05-17 | 2018-10-26 | 广东工业大学 | A kind of cooperative frequency spectrum sensing method and relevant apparatus |
| CN108736992A (en) * | 2018-05-17 | 2018-11-02 | 广东工业大学 | A kind of cooperative frequency spectrum sensing method and relevant apparatus |
| CN108768563A (en) * | 2018-05-17 | 2018-11-06 | 广东工业大学 | A kind of cooperative frequency spectrum sensing method and relevant apparatus |
| CN109004997A (en) * | 2018-08-09 | 2018-12-14 | 广东工业大学 | A kind of frequency spectrum sensing method, device and computer readable storage medium |
| CN109309538A (en) * | 2018-08-28 | 2019-02-05 | 广东工业大学 | A kind of frequency spectrum sensing method, device, equipment, system and storage medium |
| CN109743121A (en) * | 2019-03-01 | 2019-05-10 | 广东工业大学 | A kind of channel occupation status detection method, device, electronic equipment and storage medium |
| CN112003662A (en) * | 2020-06-24 | 2020-11-27 | 常州工学院 | Cooperative spectrum sensing method and device based on dimensionality reduction and clustering in cognitive network |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20080293353A1 (en) * | 2007-03-08 | 2008-11-27 | Mody Apurva N | Cognitive radio methodology, physical layer policies and machine learning |
| CN103166723A (en) * | 2013-03-18 | 2013-06-19 | 广东工业大学 | Cognitive radio sensor network spectrum sensing method based on extensible correlation function |
| CN103326797A (en) * | 2013-06-21 | 2013-09-25 | 上海交通大学 | Cooperation type frequency spectrum sensing method in cognitive network |
| CN104135327A (en) * | 2014-07-10 | 2014-11-05 | 上海大学 | Spectrum sensing method based on support vector machine |
-
2017
- 2017-08-17 CN CN201710706974.5A patent/CN107395301B/en not_active Expired - Fee Related
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20080293353A1 (en) * | 2007-03-08 | 2008-11-27 | Mody Apurva N | Cognitive radio methodology, physical layer policies and machine learning |
| CN103166723A (en) * | 2013-03-18 | 2013-06-19 | 广东工业大学 | Cognitive radio sensor network spectrum sensing method based on extensible correlation function |
| CN103326797A (en) * | 2013-06-21 | 2013-09-25 | 上海交通大学 | Cooperation type frequency spectrum sensing method in cognitive network |
| CN104135327A (en) * | 2014-07-10 | 2014-11-05 | 上海大学 | Spectrum sensing method based on support vector machine |
Non-Patent Citations (4)
| Title |
|---|
| HAOZHOU XUE 等: "A machine learning based spectrum-sensing algorithm using sample covariance matrix", 《2015 10TH INTERNATIONAL CONFERENCE ON COMMUNICATIONS AND NETWORKING IN CHINA (CHINACOM)》 * |
| VAIBHAV KUMAR 等: "K-mean Clustering based Cooperative Spectrum Sensing in Generalized k-μ Fading Channels", 《2016 TWENTY SECOND NATIONAL CONFERENCE ON COMMUNICATION (NCC)》 * |
| YINGQI LU, 等: "Machine learning techniques with probability vector for cooperative spectrum sensing in cognitive radio networks", 《2016 IEEE WIRELESS COMMUNICATIONS AND NETWORKING CONFERENCE》 * |
| 王永华 等: "无线认知传感器网络的研究", 《计算机科学》 * |
Cited By (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108494509A (en) * | 2018-03-09 | 2018-09-04 | 广东工业大学 | A kind of cooperative frequency spectrum sensing method and device |
| CN108462544A (en) * | 2018-03-27 | 2018-08-28 | 广东工业大学 | A kind of frequency spectrum sensing method and device |
| CN108566254A (en) * | 2018-03-27 | 2018-09-21 | 广东工业大学 | A kind of frequency spectrum sensing method and device |
| CN108462544B (en) * | 2018-03-27 | 2021-09-17 | 广东工业大学 | Spectrum sensing method and device |
| CN108712222A (en) * | 2018-05-17 | 2018-10-26 | 广东工业大学 | A kind of cooperative frequency spectrum sensing method and relevant apparatus |
| CN108736992A (en) * | 2018-05-17 | 2018-11-02 | 广东工业大学 | A kind of cooperative frequency spectrum sensing method and relevant apparatus |
| CN108768563A (en) * | 2018-05-17 | 2018-11-06 | 广东工业大学 | A kind of cooperative frequency spectrum sensing method and relevant apparatus |
| CN109004997A (en) * | 2018-08-09 | 2018-12-14 | 广东工业大学 | A kind of frequency spectrum sensing method, device and computer readable storage medium |
| CN109309538A (en) * | 2018-08-28 | 2019-02-05 | 广东工业大学 | A kind of frequency spectrum sensing method, device, equipment, system and storage medium |
| CN109743121A (en) * | 2019-03-01 | 2019-05-10 | 广东工业大学 | A kind of channel occupation status detection method, device, electronic equipment and storage medium |
| CN112003662A (en) * | 2020-06-24 | 2020-11-27 | 常州工学院 | Cooperative spectrum sensing method and device based on dimensionality reduction and clustering in cognitive network |
Also Published As
| Publication number | Publication date |
|---|---|
| CN107395301B (en) | 2020-09-11 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107395301A (en) | A kind of frequency spectrum sensing method based on K mean algorithms | |
| CN104158611B (en) | Wireless signal Interference Detection system and method based on spectrum analysis | |
| CN107360577A (en) | A kind of frequency spectrum sensing method and device based on machine learning | |
| CN102404058B (en) | Electromagnetic wave discrimination device, electromagnetic wave discrimination method, and electromagnetic wave discrimination program | |
| Costa et al. | Constrained spanning tree algorithms for irregularly-shaped spatial clustering | |
| CN103914475B (en) | A kind of Forecasting Methodology, system and the device of video playing amount | |
| CN107733541A (en) | Method, apparatus, equipment and the computer-readable recording medium of frequency spectrum perception | |
| CN105849581A (en) | Detecting incorrectly placed access points | |
| CN107547154A (en) | A kind of method and device for establishing video traffic prediction model | |
| CN109039503A (en) | A kind of frequency spectrum sensing method, device, equipment and computer readable storage medium | |
| CN104039008B (en) | A kind of hybrid locating method | |
| CN104038944B (en) | Cognitive radio spectrum sensing method based on random matrix | |
| CN109565716A (en) | For controlling the control method and control system of the access point in wireless network | |
| CN110224771B (en) | Spectrum sensing method and device based on BP neural network and information geometry | |
| CN103281142A (en) | Energy detection method and device combining time domain double thresholds and frequency domain variable point number | |
| CN103167505A (en) | Method and system of neighborhood data channel configuration | |
| CN103326797A (en) | Cooperation type frequency spectrum sensing method in cognitive network | |
| CN107438264A (en) | Estate performance analysis method and device | |
| CN106408571B (en) | A kind of variable class remote sensing image segmentation method based on the selection of optimal fuzzy factor | |
| Koufos et al. | Distributed sensing in multiband cognitive networks | |
| CN106488554A (en) | A kind of fingerprint database method for building up and system | |
| CN103036626B (en) | Wireless communication method based on cognitive radio cooperation users and threshold testing combined selection | |
| CN113702703A (en) | Weak signal detection and identification method and system | |
| CN108768563A (en) | A kind of cooperative frequency spectrum sensing method and relevant apparatus | |
| CN110913407A (en) | Method and device for analyzing overlapping coverage |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20200911 |
|
| CF01 | Termination of patent right due to non-payment of annual fee |