CN109831265A - A kind of wideband signal spectrum cognitive method and system based on airspace filter - Google Patents
A kind of wideband signal spectrum cognitive method and system based on airspace filter Download PDFInfo
- Publication number
- CN109831265A CN109831265A CN201910069398.7A CN201910069398A CN109831265A CN 109831265 A CN109831265 A CN 109831265A CN 201910069398 A CN201910069398 A CN 201910069398A CN 109831265 A CN109831265 A CN 109831265A
- Authority
- CN
- China
- Prior art keywords
- signal
- frequency
- array
- covariance matrix
- frequency point
- 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
Landscapes
- Radar Systems Or Details Thereof (AREA)
Abstract
The present invention relates to a kind of wideband signal spectrum cognitive method and system based on airspace filter, this method comprises: the time domain of construction array broadband signal receives model and frequency domain receives model；Determine do not occur the maximum array element spacing of false spectral peak；According to the maximum array element spacing, it is combined TCT algorithm, broadband signal is transformed to by focusing frequency point by construction focussing matrix, fusion narrowband WMUSIC algorithm carries out the signal detection of spatial spectrum analysis；New variable is constructed to each frequency point using the denoising covariance matrix in TCT algorithm, the accuracy for carrying out angle to the signal in conjunction with narrowband WMUSIC algorithm is estimated.The present invention breaches the limitation that traditional array element spacing takes half-wavelength, enhances angular resolution；Reduce calculation amount；Without estimating that information source number can obtain the spatial spectrum of array received signal；Improve the accuracy of signal incidence angle.
Description
Technical field
The invention belongs to cognitive radios and wideband signal spectrum to perceive field, more particularly to a kind of based on airspace filter
Wideband signal spectrum cognitive method and system.
Background technique
With WLAN (WLAN), wireless MAN (WCAN), wireless personal area network (WPAN) and 4G technology at
Ripe development, radio communication service develop towards broad spectrum, and the surge of terminal user's number results in the in short supply of frequency spectrum resource.It is quiet
The mode of state distribution frequency spectrum resource has been unable to meet demand, and emerging dynamic allocation frequency spectrum resource premise seeks to correctly feel
Know whether frequency spectrum has signal occupied, thus finds suitable frequency spectrum sensing method to have the utilization rate for improving frequency spectrum resource
Important meaning.
Existing broader frequency spectrum aware scheme carries out adopting for signal using information theory criterion or compressive sensing theory mostly
Then the frequency spectrum of broadband signal is divided into multiple continuous sub-bands by sample processing, signal reconstruction, traverse in conjunction with energy detection algorithm
Each sub-band, finds free sub-band.
Problem of the existing technology is:
The applicable signal of existing broadband cognition technology includes the ultra wide band of discontinuous narrow band signal composition on multiple frequency domains
Range signal, substantially each sub-band belong to narrow band signal range.Secondly, for meet relative bandwidth and center carrier frequence it
Than for 0.1 to 0.2 general broadband signal, frequency spectrum cannot be divided into multiple continuous son frequencies with sparse sub-band
Band also just can not carry out frequency spectrum perception using existing broadband cognition technology.Then, multiple sub-bands are divided for broadband range,
The each sub-band range taken is different, also will affect the accuracy of perception.Furthermore existing broadband cognition technology is fundamentally based on
Frequency division multiplexing, with application of the cognitive radio in Future cellular networks, wide-band spectrum resources may be no longer sparse, existing width
Limitation with cognition technology is also just more obvious.
Summary of the invention
In order to solve the above-mentioned problems in the prior art, the present invention provides a kind of, and the broadband based on airspace filter is believed
Number frequency spectrum sensing method and system.The technical problem to be solved in the present invention is achieved through the following technical solutions:
A kind of wideband signal spectrum cognitive method based on airspace filter, comprising the following steps:
The time domain for constructing array broadband signal receives model and frequency domain receives model；
Determine do not occur the maximum array element spacing of false spectral peak；
According to the maximum array element spacing, it is combined TCT algorithm, is transformed to broadband signal by construction focussing matrix
Frequency point is focused, fusion narrowband WMUSIC algorithm carries out the signal detection of spatial spectrum analysis；
New variable is constructed to each frequency point using the denoising covariance matrix in TCT algorithm, is calculated in conjunction with narrowband WMUSIC
The accuracy that method carries out angle to the signal is estimated.
A kind of wideband signal spectrum cognitive method based on airspace filter the following steps are included:
S100, model, the time-domain signal that will be received are received to the time domain of broadband signal building array antenna broadband signal
By FFT transform to frequency domain, the frequency domain for obtaining array broadband signal receives model；
S200, the selection for carrying out maximum array element spacing analyze the value size of array element spacing and angular resolution and right
The signal space answered composes the relationship of false spectral peak number；
S300, frequency domain received data is obtained using the frequency domain reception model of the array broadband signal, acquired at each frequency point
Array output covariance matrix R (fj), the denoising covariance matrix P (f of each frequency pointj)；
S400, angle direction vector theta is introduced, in conjunction with the maximum array element spacing, acquires the signal covariance of each frequency point
Matrix Rs(fj), focus frequency point at covariance matrix Rs(f0), and focus the denoising covariance matrix P (f on frequency point0)；
S500, the denoising covariance matrix R (f to each frequency point of the variablej) and focus frequency point on denoising covariance
Matrix P (f0) improve, construct new variable Q (fj), Q (fj)=P (fj)PH(f0)P(f0)PH(fj), (j=J)；
S600, using TCT algorithm, obtain the ratio between the maxima and minima of spatial spectrum in conjunction with narrowband WMUSIC algorithm, it is right
Than decision threshold γ, existing signal is detected；
S700, the new variable is introducedIn conjunction with narrowband WMUSIC algorithm, Eigenvalues Decomposition is carried out, at weighting
Reason constructs new spatial noise U, determines that the direction vector under focusing frequency, the angular range of this direction vector of traversal search obtain
Broadband incident angle.
In one embodiment of the invention, the step S100 includes:
It S110, is M for antenna number, incoming signal number is the communication system of K, and the time domain of construction array broadband signal connects
Receive model；Wherein, K < M；
S120, model progress frequency-domain transform is received to the time domain of the array broadband signal, obtain array broadband signal
Frequency domain receives model X (f)=A (f, θ) S (f)+N (f)；Wherein, X (f) is array received signal frequency domain data；A (f, θ) is to become
Broadband signal after changing is oriented to matrix；S (f) is broadband signal frequency domain data；N (f) is noise frequency domain data.
In one embodiment of the invention, the step S200 includes:
The quantitative relationship of S210, analysis array element spacing and false spectral peak；
S220, illustrate the beam angle with the relationship of array aperture and array resolution, half-power from the angle of linear array
The beam angle of point is BW0.5≈θc/ (D/ λ)=θc/ ((M-1) * d/ λ), wherein θcIt is the definite value in current environment；D is array element
Spacing；M is number of antennas；D is array aperture；In narrow band signal, λ is carrier wavelength；In broadband signal, λ is focusing frequency
Locate corresponding wavelength；
Wherein, S210 includes: again
S211, as d/ λ < 1/2, without falseness spectral peak occur；
S212, as 1/2 < d/ λ < 1, if 1/2* λ < d < λ/(1+sin (θinput)), then occur without false spectral peak, if λ/(1+
sin(θinput)) < d < λ, then there is false spectral peak to occur；
S212, as 1 < d/ λ, there is false spectral peak to occur；
In one embodiment of the invention, the step S300 includes:
S310, TCT algorithm is introduced, transformation is focused using the relationship between each frequency point noise free data；Assuming that focusing
Transformation matrix is T (fj), and meet T (fj)A(fj,θ)S(fj)=A (f0,θ)S(f0)；
S320, obtain the array broadband signal frequency domain receive model after, if by observing time T0Interior array
It receives data and is divided into L field, every time is Td, leaf transformation in J point discrete Fourier, array broadband then are carried out to observation data
The frequency domain of signal receives model and becomes: Xl(fj)=A (fj,θ)Sl(fj)+Nl(fj)；
In formula, Xl(fj)、Sl(fj)、Nl(fj) it is respectively l segment data in frequency fj(l=1,2 ..., L；J=1,
2 ..., J) at reception data, signal and noise DFT transform；
S330, by the Xl(fj) acquire each frequency point fjThe frequency domain received data covariance matrix R (f at placej)；
S340, by formulaAcquire the denoising covariance matrix P (f of each frequency pointj), choose frequency point fj
Data covariance matrix minimal eigenvalue as noise power；Wherein σ is noise power；I is unit matrix.
In one embodiment of the invention, the step S400 includes:
S410, determine that focusing frequency, the focusing frequency are 1.1 times of broadband signal centre frequency；
S420, the angle pre-estimation process for omitting TCT algorithm, with angle direction vector θ=[θ1,θ2,...,θk] replace
Angular range is carried out the division at equal intervals of k point by pre-estimation process from [- 60 ° ,+60 °]；
S430, angle direction vector is introduced, acquires the signal covariance matrix at each frequency point；Calculation formula is as follows:
Rs(fj)=[AH(fj,θ)A(fj,θ)]-1AH(fj,θ)P(fj)A(fj,θ)[AH(fj,θ)A(fj,θ)]-1；
S440, according to formulaCalculate focusing frequency f0The signal covariance matrix R at places(f0)；
S450, according to formula P (f0)=A (f0,θ)Rs(f0)AH(f0, θ), calculate focusing frequency f0The denoising covariance at place
Matrix P (f0)。
In one embodiment of the invention, the step S600 includes:
S610, the denoising covariance matrix P (f to each frequency pointj) and the denoising covariance square focused on frequency point
Battle array P (f0) feature decomposition is carried out, obtain Uj,θAnd U0,θRespectively P (fj) and P (f0) feature matrix；Further gathered
Burnt matrix Tθ(fj)=U0,θUj,θ H；
S620, pass through the focussing matrix Tθ(fj) by the frequency domain data covariance matrix focusing transform of each frequency point to frequency
f0Place, frequency f at this time0The data covariance matrix at place is
S630, to the frequency f0Data covariance matrix R (the f at place0) special using the WMUSIC algorithm progress of narrow band signal
Value indicative is decomposed, and corresponding eigenvalue λ is obtainedi, (i=1,2 ..., M), corresponding characteristic vector space is U=[u1,u2,...,
uM]；
S640, processing is weighted to characteristic vector space with obtained characteristic value, defines new characteristic vector space
S650, determine that direction vector under focusing frequency, the angular range of this direction vector of traversal search obtain broadband letter
Number spatial spectrum, using signal whether there is or not and the system of the ratio between spatial spectrum peak value difference, construction maximum value spectrum peak and minimum spectrum peak
Metering adjudicates existing signal compared with decision threshold γ.
In one embodiment of the invention, the step S700 includes:
S710, the new variable formula using construction: Q (fj)=P (fj)PH(f0)P(f0)PH(fj), (j=J) will be described
New variable Q (fj) instead of the frequency f in step S620 and step S6300Data covariance matrix R (the f at place0)；Carry out feature
Value is decomposed, and corresponding eigenvalue λ is obtainedQ i, (i=1,2 ..., M), corresponding characteristic vector space is UQ=[uQ 1,uQ 2,...,
uQ M]；
S720, processing is weighted to characteristic vector space with obtained characteristic value, defines new characteristic vector space
S730, determine that direction vector under focusing frequency, the angular range of this direction vector of traversal search obtain broadband letter
Number the corresponding incident angle of peak value.
A kind of wideband signal spectrum sensory perceptual system based on airspace filter, for executing any of the above-described kind based on airspace filter
Wideband signal spectrum cognitive method the step of.
Compared with prior art, beneficial effects of the present invention:
The present invention carries out the frequency spectrum perception of general broadband signal using array antenna model, for the frequency domain of broadband signal
It is not required with the presence or absence of sparse sub-band.When unauthorized user on airspace perceives authorized user and occupies frequency range, pass through tune
Whole angle can share same frequency range simultaneously；The present invention analyzes the relationship of array element spacing and angular resolution and space spectral peak, dashes forward
The limitation that traditional array element spacing takes carrier wave half-wavelength has been broken, ensure that under the premise of there is not false spectral peak, angle point is improved
Resolution, although can estimate the incident angle of signal in turn in the presence of judgement has signal using TCT algorithm, it estimates angle
The accuracy of degree is not high enough, and the present invention carries out the estimation of angle by constructing new variable, so that angle estimation accuracy is more
It is high.It can make the non-interfering communication of multi-user using the directional beam technologies of existing maturation, greatly improve frequency spectrum resource
Utilization rate.
Detailed description of the invention
Fig. 1 is a kind of process of wideband signal spectrum cognitive method based on airspace filter provided in an embodiment of the present invention
Figure；
Fig. 2 is the process of another wideband signal spectrum cognitive method based on airspace filter provided in an embodiment of the present invention
Figure；
Fig. 3 is the process of another wideband signal spectrum cognitive method based on airspace filter provided in an embodiment of the present invention
Figure；
Fig. 4 is that array element spacing provided in an embodiment of the present invention and false spectral peak number and angular resolution result are presented.
Fig. 5 is the detection probability figure of different signal-to-noise ratio provided in an embodiment of the present invention.
Fig. 6 is the procedural block diagram of weighting MUSIC method provided in an embodiment of the present invention.
Fig. 7 is the selection process and thresholding for the threshold value that perception stage provided in an embodiment of the present invention carries out signal detection
It is worth the proof diagram of correctness.
Fig. 8 is the detection probability result of variations figure under Parameters variation provided in an embodiment of the present invention.
Fig. 9 is the angle estimation value accurate performance comparison diagram before and after introducing new variables provided in an embodiment of the present invention.
Specific embodiment
Further detailed description is done to the present invention combined with specific embodiments below, but embodiments of the present invention are not limited to
This.
Embodiment 1:
Referring to Figure 1, Fig. 1 is a kind of wideband signal spectrum perception side based on airspace filter provided in an embodiment of the present invention
The flow chart of method.A kind of wideband signal spectrum cognitive method based on airspace filter of the present embodiment, comprising:
The time domain for constructing array broadband signal receives model and frequency domain receives model；
Determine do not occur the maximum array element spacing of false spectral peak；
According to the maximum array element spacing, it is combined TCT algorithm, is transformed to broadband signal by construction focussing matrix
Frequency point is focused, fusion narrowband WMUSIC algorithm carries out the signal detection of spatial spectrum analysis；
New variable is constructed to each frequency point using the denoising covariance matrix in TCT algorithm, is calculated in conjunction with narrowband WMUSIC
The accuracy that method carries out angle to the signal is estimated.
It should be noted that broadband signal according to the present invention refer to meet relative bandwidth and center carrier frequence ratio between
0.1 to 0.2 general broadband signal.
The difficulty and meaning of the technical problems to be solved by the invention are:
Existing broadband cognitive method is directed to the ultra-broadband signal for having sparse range on frequency domain, or by multiple frequencies
The ultra wide band range signal of discontinuous narrow band signal composition on domain.With application of the cognitive radio in Future cellular networks,
Wide-band spectrum resources may be no longer sparse.And general broadband signal is different from ultra-broadband signal, frequency spectrum may be not present sparse
Range, can not carry out division, the traversal of sub-band on frequency domain, therefore cannot carry out general broadband with existing broadband cognitive method
The sampling of signal, restructuring procedure.
Existing broadband cognitive method is based on frequency division multiplexing, if a certain frequency range has primary user's signal occupancy, other are recognized
Signal just can not access this frequency range simultaneously, and frequency spectrum perception is mainly exactly to improve the utilization rate of frequency spectrum, therefore how to access simultaneously same
One frequency range is to be particularly important.And the method in airspace then can be very good to solve the problems, such as while access with frequency, how carry out
The wideband signal spectrum perception in airspace is difficult point place.
For above-mentioned difficult point, the present invention selects array antenna to establish broadband signal Array Model and carries out frequency spectrum perception.?
It during this, is handled by the way that general broadband signal to be placed under Array Model, general broadband is carried out by improved TCT algorithm
The frequency spectrum perception of signal.Perception broadband signal there are while, also accurately obtain the incident orientation information of broadband signal, nothing
It need to be concerned about whether frequency spectrum has sparse range sub-band, and break through traditional frequency division multiplexing dimension, utilize maturation from airspace dimension
Directional beam technologies make it possible to the different angle simultaneously in same frequency range and are multiplexed, frequency spectrum resource is significantly increased
Utilization rate.
Embodiment 2:
Fig. 2 is referred to, Fig. 2 is another wideband signal spectrum perception based on airspace filter provided in an embodiment of the present invention
The flow chart of method.A kind of wideband signal spectrum cognitive method based on airspace filter of the present embodiment, comprising:
S100, model, the time-domain signal that will be received are received to the time domain of broadband signal building array antenna broadband signal
By FFT transform to frequency domain, the frequency domain for obtaining array broadband signal receives model；It specifically includes:
It S110, is M for antenna number, incoming signal number is the communication system of K, and the time domain of construction array broadband signal connects
Receive model；Wherein, K < M；
S120, model progress frequency-domain transform is received to the time domain of the array broadband signal, obtain array broadband signal
Frequency domain receives model X (f)=A (f, θ) S (f)+N (f)；Wherein, X (f) is array received signal frequency domain data；A (f, θ) is to become
Broadband signal after changing is oriented to matrix；S (f) is broadband signal frequency domain data；N (f) is noise frequency domain data.
S200, the selection for carrying out maximum array element spacing analyze the value size of array element spacing and angular resolution and right
The signal space answered composes the relationship of false spectral peak number；It specifically includes:
The quantitative relationship of S210, analysis array element spacing and false spectral peak；
S220, illustrate the beam angle with the relationship of array aperture and array resolution, half-power from the angle of linear array
The beam angle of point is BW0.5≈θc/ (D/ λ)=θc/ ((M-1) * d/ λ), wherein θcIt is the definite value in current environment；D is array element
Spacing；M is number of antennas；D is array aperture；In narrow band signal, λ is carrier wavelength；In broadband signal, λ is focusing frequency
Locate corresponding wavelength；
Wherein, S210 includes: again
S211, as d/ λ < 1/2, without falseness spectral peak occur；
S212, as 1/2 < d/ λ < 1, if 1/2* λ < d < λ/(1+sin (θinput)), then occur without false spectral peak, if λ/(1+
sin(θinput)) < d < λ, then there is false spectral peak to occur；
S212, as 1 < d/ λ, there is false spectral peak to occur；
S300, frequency domain received data is obtained using the frequency domain reception model of the array broadband signal, acquired at each frequency point
Array output covariance matrix R (fj), the denoising covariance matrix P (f of each frequency pointj)；It specifically includes:
S310, TCT algorithm is introduced, transformation is focused using the relationship between each frequency point noise free data；Assuming that focusing
Transformation matrix is T (fj), and meet T (fj)A(fj,θ)S(fj)=A (f0,θ)S(f0)；
S320, obtain the array broadband signal frequency domain receive model after, if by observing time T0Interior array
It receives data and is divided into L field, every time is Td, leaf transformation in J point discrete Fourier, array broadband then are carried out to observation data
The frequency domain of signal receives model and becomes: Xl(fj)=A (fj,θ)Sl(fj)+Nl(fj)；
In formula, Xl(fj)、Sl(fj)、Nl(fj) it is respectively l segment data in frequency fj(l=1,2 ..., L；J=1,
2 ..., J) at reception data, signal and noise DFT transform；
S330, by the Xl(fj) acquire each frequency point fjThe frequency domain received data covariance matrix R (f at placej)；
S340, by formulaAcquire the denoising covariance matrix P (f of each frequency pointj), choose frequency point fj
Data covariance matrix minimal eigenvalue as noise power；Wherein σ is noise power；I is unit matrix.
S400, angle direction vector theta is introduced, in conjunction with the maximum array element spacing, acquires the signal covariance of each frequency point
Matrix Rs(fj), focus frequency point at covariance matrix Rs(f0), and focus the denoising covariance matrix P (f on frequency point0)；Specifically
Include:
S410, determine that focusing frequency, the focusing frequency are 1.1 times of broadband signal centre frequency；
S420, the angle pre-estimation process for omitting TCT algorithm, with angle direction vector θ=[θ1,θ2,...,θk] replace
Angular range is carried out the division at equal intervals of k point by pre-estimation process from [- 60 ° ,+60 °]；
S430, angle direction vector is introduced, acquires the signal covariance matrix at each frequency point；Calculation formula is as follows:
Rs(fj)=[AH(fj,θ)A(fj,θ)]-1AH(fj,θ)P(fj)A(fj,θ)[AH(fj,θ)A(fj,θ)]-1；
S440, according to formulaCalculate focusing frequency f0The signal covariance matrix R at places(f0)；
S450, according to formula P (f0)=A (f0,θ)Rs(f0)AH(f0, θ), calculate focusing frequency f0The denoising covariance at place
Matrix P (f0)。
S500, the denoising covariance matrix R (f to each frequency point of the variablej) and focus frequency point on denoising covariance
Matrix P (f0) improve, construct new variable Q (fj), Q (fj)=P (fj)PH(f0)P(f0)PH(fj), (j=J).
S600, using TCT algorithm, obtain the ratio between the maxima and minima of spatial spectrum in conjunction with narrowband WMUSIC algorithm, it is right
Than decision threshold γ, existing signal is detected；It specifically includes:
S610, the denoising covariance matrix P (f to each frequency pointj) and the denoising covariance square focused on frequency point
Battle array P (f0) feature decomposition is carried out, obtain Uj,θAnd U0,θRespectively P (fj) and P (f0) feature matrix；Further gathered
Burnt matrix Tθ(fj)=U0,θUj,θ H；
S620, pass through the focussing matrix Tθ(fj) by the frequency domain data covariance matrix focusing transform of each frequency point to frequency
f0Place, frequency f at this time0The data covariance matrix at place is
S630, to the frequency f0Data covariance matrix R (the f at place0) special using the WMUSIC algorithm progress of narrow band signal
Value indicative is decomposed, and corresponding eigenvalue λ is obtainedi, (i=1,2 ..., M), corresponding characteristic vector space is U=[u1,u2,...,
uM]；
S640, processing is weighted to characteristic vector space with obtained characteristic value, defines new characteristic vector space
S650, determine that direction vector under focusing frequency, the angular range of this direction vector of traversal search obtain broadband letter
Number spatial spectrum, using signal whether there is or not and the system of the ratio between spatial spectrum peak value difference, construction maximum value spectrum peak and minimum spectrum peak
Metering adjudicates existing signal compared with decision threshold γ.
S700, the new variable Q (f is introducedj), in conjunction with narrowband WMUSIC algorithm, carry out Eigenvalues Decomposition, weighting processing
New spatial noise U is constructed, determines that the direction vector under focusing frequency, the angular range of this direction vector of traversal search obtain width
Band incident angle；It specifically includes:
S710, the new variable formula using construction: Q (fj)=P (fj)PH(f0)P(f0)PH(fj), (j=J) will be described
New variable Q (fj) instead of the frequency f in step S620 and step S6300Data covariance matrix R (the f at place0)；Carry out feature
Value is decomposed, and corresponding eigenvalue λ is obtainedQ i, (i=1,2 ..., M), corresponding characteristic vector space is UQ=[uQ 1,uQ 2,...,
uQ M]；
S720, processing is weighted to characteristic vector space with obtained characteristic value, defines new characteristic vector space
S730, determine that direction vector under focusing frequency, the angular range of this direction vector of traversal search obtain broadband letter
Number the corresponding incident angle of peak value.
The present invention carries out the frequency spectrum perception of general broadband signal using array antenna model, for the frequency domain of broadband signal
It is not required with the presence or absence of sparse sub-band.When unauthorized user on airspace perceives authorized user and occupies frequency range, pass through tune
Whole angle can share same frequency range simultaneously；The present invention analyzes the relationship of array element spacing and angular resolution and space spectral peak, dashes forward
The limitation that traditional array element spacing takes carrier wave half-wavelength has been broken, ensure that under the premise of there is not false spectral peak, angle point is improved
Resolution, although can estimate the incident angle of signal in turn in the presence of judgement has signal using TCT algorithm, it estimates angle
The accuracy of degree is not high enough, and the present invention carries out the estimation of angle by constructing new variable, so that angle estimation accuracy is more
It is high.It can make the non-interfering communication of multi-user using the directional beam technologies of existing maturation, greatly improve frequency spectrum resource
Utilization rate.
Embodiment 3:
On the basis of the above embodiments, the present embodiment is to a kind of wideband signal spectrum based on airspace filter of the invention
Cognitive method is described in further details.
Fig. 3 is referred to, Fig. 3 is that the present invention implements another the broadband signal cognitive method stream based on airspace filter provided
Cheng Tu.As shown in figure 3, the broadband signal cognitive method based on airspace filter of this implementation, comprising:
Array broadband model construction；
The maximum array element spacing for not occurring false spectral peak determines；
Signal detection process；
The accurate estimation procedure of angle.
In terms of array broadband model construction, model construction is carried out using to broadband time-domain signal FFT transform；Between maximum array element
Away from aspect is determined, to improve the resolution ratio of angle；In terms of signal detection process, with the WMUSIC algorithm of TCT algorithm combination narrowband
The ratio between minimax spectrum peak is obtained, with decision threshold multilevel iudge signal；In terms of the accurate estimation procedure of angle；It is calculated using TCT
The structure's variable new variables of method, obtains more accurate angle estimation.
The present invention is built upon under Array Model into broadband signal row frequency spectrum perception, and parameter setting procedure is refined as following several
A step:
A. the parameter setting of array antenna model.Antenna number is 16, and array element spacing is set as
Fig. 4 is referred to, Fig. 4 is array element spacing provided in an embodiment of the present invention and false spectral peak number and angular resolution
As a result it presents.Fig. 4 analyzes the number and array element spacing and angular resolution that array element spacing takes the corresponding false spectral peak of different value
Relationship.Taking array element spacing is half-wavelength, maximum without being greater than half-wave long value, integer wavelength, signal incidence angle under false spectral peak
Spending adjustment is respectively to be compared for 30 ° and for the second time 60 ° for the first time.Furthermore array element spacing size and angular resolution are also illustrated
Relationship, it is half-wavelength and maximum without being greater than half-wave long value under false spectral peak that we, which take array element spacing, between array element of the invention
Take discern signal similar in two angles the maximum half-wave long value time-division that is greater than without under false spectral peak, and array element spacing takes half-wavelength
Then it can not be identified.
Specifically, " number that analysis array element spacing takes the corresponding false spectral peak of different value " can be with detailed table in step A
It is shown as:
The incident angle of setting signal is 30 °, and for convenience of analyzing, the fixed narrow band signal of selecting frequency sets noise
Than keeping other parameters constant, array element spacing successively takes equal to half-wavelength, is equal to 60 ° of corresponding maximum array element spacing, is equal to whole
Number wavelength, successively observes the number of the spacing spectrum peak of incoming signal；
The incident angle of setting signal is 60 °, and for convenience of analyzing, the fixed narrow band signal of selecting frequency sets noise
Than keeping other parameters constant, array element spacing successively takes equal to half-wavelength, is equal to 60 ° of corresponding maximum array element spacing, is equal to whole
Number wavelength, successively observes the number of the spacing spectrum peak of incoming signal；
Specifically, " relationship of analysis array element spacing and angular resolution " detailed can indicate in step A are as follows:
The angle for setting two incoming signals is respectively 20 ° and 25 °, for convenience of analyzing, the fixed narrowband letter of selecting frequency
Number, it sets signal-to-noise ratio, keeps other parameters constant, array element spacing takes half carrier wavelength for the first time, takes set for the second timeCan the signal space spectrum under the conditions of two kinds of observation, distinguish signal similar in two angles.
B. incoming signal chooses 1, and incident angle is 10 degree, and broadband signal is that meet relative bandwidth and carrier frequency ratio be 0.2
General broadband signal, sampling number choose 1280.
C. signal-to-noise ratio variation range is set as -20dB to 0dB, and each signal-to-noise ratio carries out 1000 Monte-Carlo Simulations,
To observe the inspection rate variation under the conditions of different signal-to-noise ratio.
Fig. 5 is referred to, Fig. 5 is the detection probability figure of different signal-to-noise ratio provided in an embodiment of the present invention.Fig. 5 illustrates this hair
The effect that the bright detection probability under different signal-to-noise ratio changes with signal-to-noise ratio.Furthermore in order to illustrate signal detection proposed by the present invention
Perceived effect, construction new variables is used to carry out the effect of signal detection to compare in the present invention.The base that the present invention uses
The detection probability for being used to detect signal based on new variables is apparently higher than in the signal-detection probability of focussing matrix.Thus also illustrate
The detection of signal is carried out in the method for perception stage focussing matrix.
The present invention detects the general broadband signal under Array Model, which detailed can indicate are as follows:
A. broadband signal array antenna received arrived carries out FFT transform, transforms from the time domain to frequency domain, before transformation, obtains
To signal S be a M*N matrix, M be array-element antenna number, N be time-domain sampling count.Assuming that the sampled point of FFT transform is
Nf, broadband signal X after converting are the matrix of a M*Nf in frequency domain
B. the building that frequency domain correlation matrix is carried out to the matrix of obtained M*Nf, carries out on each frequency point of broadband signal
The transformation of X*X ' obtains the array output covariance matrix on each frequency point, i.e. R (fj), the denoising covariance matrix of each frequency pointIt can be seen that first it is to be understood that noise power, and it is general there are two types of modes, one is frequency point fjData association
The small characteristic value of variance matrix is averaged, and the method needs to know number of source, and frequency spectrum perception is not aware that information source number, otherwise
Also just without necessity existing for perceptual signal.Another method is that noise power takes frequency point fjMinimal eigenvalue, the method
The estimation of information source number is avoided well, so we obtain noise power using second method, denoising is then calculated
Covariance matrix.
C. array element spacing is made a concrete analysis of in " parameter setting ", it is determined that maximum array element spacing also determines that width
Band signal is oriented to the form of matrix, but to obtain the signal covariance matrix R of each frequency points(fj), it needs to carry out angle pre-
Estimation, angle pre-estimation will increase additional calculation amount.In order to avoid angle pre-estimation, angle direction arrow is introduced herein
Amount, is embodied as:
- 60 ° to 60 ° of angular interval is divided at equal intervals, obtains a series of constant angles of difference, these angles
Value covers the true incident angle of broadband signal, and the guiding of fixed broadband signal is combined by introducing angle direction vector
Matrix carries out the solution of each frequency point signal covariance matrix, then average, acquires the signal covariance square focused on frequency point
Battle array.
D. step C has acquired the signal covariance matrix focused on frequency point, but does not provide the choosing method for focusing frequency point,
Herein, focusing frequency is generally chosen for 1.1 times of broadband signal centre frequency, must focus frequency point then in conjunction with guiding Matrix Calculating
On denoising covariance matrix.
E. step B and step D has respectively obtained each frequency point and has focused the denoising covariance matrix on frequency point, by right
The feature decomposition of matrix respectively obtains corresponding feature matrix, constructs focussing matrix, and then obtains focusing on frequency point
Data covariance matrix.
F. it by step E, obtains to be carried out with narrowband estimation method after focusing the data covariance matrix on frequency point
Estimation of Spatial Spectrum, because can not know information source number, a kind of MUSIC method for choosing weighting herein carries out Estimation of Spatial Spectrum,
The MUSIC method flow of weighting is as shown in fig. 6, specific as follows in detail:
Feature decomposition is carried out to the focusing frequency point data covariance matrix that step E is obtained, conventional MUSIC method is in feature point
Solution needs to sort after obtaining characteristic value, and needs to estimate that number of source to carry out the division of noise subspace, herein, carries out
After Eigenvalues Decomposition, does not need to carry out Sources number estimation, directly corresponding feature vector is weighted with characteristic value.It does so
The reason of be analyzed as follows:
The corresponding characteristic value very little of noise subspace and approximately equal, therefore can use the consistent amplification of these characteristic values
Noise feature vector；The corresponding characteristic value of signal subspace is larger, therefore can use the consistent diminution signal of these characteristic values
Feature vector.By above-mentioned processing, the sequence of characteristic value and the division of noise subspace are avoided, not by Eigenvalue Bounds mould
The influence of paste problem.New characteristic vector space can largely retain subspace with approximate noise subspace
Orthogonality.Therefore it redefines and receives signal characteristic vector space are as follows:
Wherein α is weighting coefficient, and α > 0
The selection of weighted value, is analyzed as follows:
α crosses conference and inhibits spectral peak, owes estimation to be formed, α is too small, can not effectively inhibit signal subspace component.Compromise
Processing, and consider computation complexity, choose α=1.
E. angle is then searched for,Traversal obtains the spectral peak of spatial spectrum, finds spectral peak maximum
The ratio between minimum value is made comparisons with threshold value, the presence of decision signal.And the choosing method of threshold value is as follows in detail:
False-alarm probability is to judge probability existing for signal by accident in the absence of signal.False-alarm probability is set as 0.1, i.e. P now
(T>γ|H0)=0.1, H0Indicate that signal is not present.It is obtained by probability theory knowledge, F (X)=P (X<γ)=1-P (X>γ), in H0It is false
Set, by the ratio between spectral peak maximin T replace X, obtain F (T)=P (T < γ | H0)=1-P (T > γ | H0)=1-Pf=0.9,
It is decision threshold value that the cdfplot curve drawn, which corresponds to abscissa when ordinate of orthogonal axes is 0.9,.It has been determined that false-alarm probability is
After threshold value under 0.1, it is added without signal, by the variation of noise intensity, whether observes practical false-alarm probability at this time 0.1
Below.Fig. 7 is referred to, Fig. 7 is the selection process for the threshold value that perception stage provided in an embodiment of the present invention carries out signal detection
And the proof diagram of threshold value correctness, Fig. 7 illustrate the verifying of the selection process and thresholding selected value of threshold value.
The present invention, by the variation of parameter, observes Parameters variation to the shadow of detection probability after obtaining the simulation result of Fig. 5
It rings:
By taking 1 signal as an example, keep sampling number constant, array antenna number is different, observes detection probability with signal-to-noise ratio
Variation；Keep array antenna number constant, sampling number is different, and observation detection probability keeps sampling number not with the variation of signal-to-noise ratio
Become, array antenna number is constant, by taking 2 incoming signals as an example, when different angle difference, observation detection probability with signal-to-noise ratio variation；
Fig. 8 is referred to, Fig. 8 is the detection probability result of variations figure under Parameters variation provided in an embodiment of the present invention, and Fig. 8 gives specifically
Analyze result.
After the completion of the present invention detects the general broadband signal under Array Model, estimate from the available angle of spatial spectrum
Evaluation, but there is deviations for angle estimation value and real angle value.The accuracy of angle estimated value is improved, it is improved
As a result as shown in figure 9, the process detailed can indicate are as follows:
A. the step B when carrying out signal detection and step D has obtained each frequency point and has focused the denoising association side on frequency point
Poor matrix P (fj),P(f0), construct new variable Q (fj)=P (fj)PH(f0)P(f0)PH(fj), (j=J)；
What B. the step E when carrying out signal detection was obtained is the data covariance matrix focused at frequency point, is used herein
The new variable Q (f of constructionj) replace, and then the angle carried out in the step F and step G of " when carrying out signal detection " is searched
Rope traversal；
C. the angle estimation obtained with focussing matrix is presented on the angle estimation Comparative result analysis obtained with new variables
In Fig. 9.
Fig. 9 illustrates the present invention and is detecting signal, and then carries out the effect of angle estimation.In addition, in order to illustrate this hair
The new variables of bright proposition carries out the accuracy of angle estimation, compares by the present invention and with the scheme of focussing matrix.This hair
It is bright it is middle the method for angle estimation is carried out when incidence angle is 15 degree using new variables, the angle value of estimation closer to real angle,
And the accurate sexual deviation of angle value that the method for using focussing matrix is estimated is larger.
The present invention by establish array antenna broadband signal time-frequency domain receive model, analyze array element spacing without falseness
Spectral peak maximum value selection range breaches the limitation that traditional array element spacing takes half-wavelength, enhances angular resolution；Pass through TCT
Algorithm introduces angle direction vector and replaces angle pre-estimation process, reduces calculation amount.In conjunction with narrowband WMUSIC algorithm, construction
New noise subspace makes without estimating that information source number can obtain the spatial spectrum of array received signal, by spectral peak maximin
The ratio between compared with threshold value with the presence or absence of decision signal；And it is further introduced into new variable combination narrowband WMUSIC algorithm and has improved
The accuracy of signal incidence angle.
Embodiment 4:
The present embodiment is a kind of wideband signal spectrum sensory perceptual system based on airspace filter, and the system is for executing above-mentioned reality
The method for applying example, have the present invention is based on the wideband signal spectrum of airspace filter perceive possessed by beneficial effect, herein no longer
It repeats.
The above content is a further detailed description of the present invention in conjunction with specific preferred embodiments, and it cannot be said that
Specific implementation of the invention is only limited to these instructions.For those of ordinary skill in the art to which the present invention belongs, exist
Under the premise of not departing from present inventive concept, a number of simple deductions or replacements can also be made, all shall be regarded as belonging to of the invention
Protection scope.
Claims (9)
1. a kind of wideband signal spectrum cognitive method based on airspace filter characterized by comprising
The time domain for constructing array broadband signal receives model and frequency domain receives model；
Determine do not occur the maximum array element spacing of false spectral peak；
According to the maximum array element spacing, it is combined TCT algorithm, broadband signal is transformed to by focusing by construction focussing matrix
Frequency point, fusion narrowband WMUSIC algorithm carry out the signal detection of spatial spectrum analysis；
New variable is constructed to each frequency point using the denoising covariance matrix in TCT algorithm, in conjunction with narrowband WMUSIC algorithm pair
The signal carries out the accuracy estimation of angle.
2. a kind of wideband signal spectrum cognitive method based on airspace filter, which comprises the following steps:
S100, model is received to the time domain of broadband signal building array antenna broadband signal, the time-domain signal received is passed through
For FFT transform to frequency domain, the frequency domain for obtaining array broadband signal receives model；
S200, the selection for carrying out maximum array element spacing analyze the value size of array element spacing and angular resolution and corresponding
Signal space composes the relationship of false spectral peak number；
S300, frequency domain received data is obtained using the frequency domain reception model of the array broadband signal, acquires the battle array at each frequency point
Column output covariance matrix R (fj), the denoising covariance matrix P (f of each frequency pointj)；
S400, angle direction vector theta is introduced, in conjunction with the maximum array element spacing, acquires the signal covariance matrix R of each frequency points
(fj), focus frequency point at covariance matrix Rs(f0), and focus the denoising covariance matrix P (f on frequency point0)；
S500, the denoising covariance matrix R (f to each frequency point of the variablej) and focus frequency point on denoising covariance matrix P
(f0) improve, construct new variable Q (fj), Q (fj)=P (fj)PH(f0)P(f0)PH(fj), (j=J)；
S600, using TCT algorithm, obtain the ratio between the maxima and minima of spatial spectrum in conjunction with narrowband WMUSIC algorithm, comparison is sentenced
Certainly thresholding γ detects existing signal；
S700, the new variable Q (f is introduced to the signalj), in conjunction with narrowband WMUSIC algorithm, Eigenvalues Decomposition is carried out, is added
Power processing constructs new spatial noise U, determines the direction vector under focusing frequency, the angular range of this direction vector of traversal search,
Obtain broadband incident angle.
3. the wideband signal spectrum cognitive method according to claim 2 based on airspace filter, which is characterized in that the step
Suddenly S100 includes:
It S110, is M for antenna number, incoming signal number is the communication system of K, and the time domain of construction array broadband signal receives mould
Type；Wherein, K < M；
S120, model progress frequency-domain transform is received to the time domain of the array broadband signal, obtain the frequency domain of array broadband signal
Receive model X (f)=A (f, θ) S (f)+N (f)；Wherein, X (f) is array received signal frequency domain data；A (f, θ) is after converting
Broadband signal be oriented to matrix；S (f) is broadband signal frequency domain data；N (f) is noise frequency domain data.
4. the wideband signal spectrum cognitive method according to claim 2 based on airspace filter, which is characterized in that the step
Suddenly S200 includes:
The quantitative relationship of S210, analysis array element spacing and false spectral peak；
S220, from the angle of linear array come illustrate beam angle with the relationship of array aperture and array resolution, half power points
Beam angle is BW0.5≈θc/ (D/ λ)=θc/ ((M-1) * d/ λ), wherein θcIt is the definite value in current environment；D is between array element
Away from；M is number of antennas；D is array aperture；In narrow band signal, λ is carrier wavelength；In broadband signal, λ is at focusing frequency
Corresponding wavelength；
Wherein, S210 includes: again
S211, as d/ λ < 1/2, without falseness spectral peak occur；
S212, as 1/2 < d/ λ < 1, if 1/2* λ < d < λ/(1+sin (θinput)), then occur without false spectral peak, if λ/(1+
sin(θinput)) < d < λ, then there is false spectral peak to occur；
S212, as 1 < d/ λ, there is false spectral peak to occur.
5. the wideband signal spectrum cognitive method according to claim 2 based on airspace filter, which is characterized in that the step
Suddenly S300 includes:
S310, TCT algorithm is introduced, transformation is focused using the relationship between each frequency point noise free data；Assuming that focusing transform
Matrix is T (fj), and meet T (fj)A(fj,θ)S(fj)=A (f0,θ)S(f0)；
S320, obtain the array broadband signal frequency domain receive model after, if by observing time T0Interior array received number
According to L field is divided into, every time is Td, leaf transformation in J point discrete Fourier then is carried out to observation data, array broadband signal
Frequency domain receives model and becomes: Xl(fj)=A (fj,θ)Sl(fj)+Nl(fj)；
In formula, Xl(fj)、Sl(fj)、Nl(fj) it is respectively l segment data in frequency fj(l=1,2 ..., L；J=1,2 ..., J)
The DFT transform of the reception data at place, signal and noise；
S330, by the Xl(fj) acquire each frequency point fjThe frequency domain received data covariance matrix R (f at placej)；
S340, by formulaAcquire the denoising covariance matrix P (f of each frequency pointj), choose frequency point fjNumber
According to covariance matrix minimal eigenvalue as noise power；Wherein σ is noise power；I is unit matrix.
6. the wideband signal spectrum cognitive method according to claim 2 based on airspace filter, which is characterized in that the step
Suddenly S400 includes:
S410, determine that focusing frequency, the focusing frequency are 1.1 times of broadband signal centre frequency；
S420, the angle pre-estimation process for omitting TCT algorithm, with angle direction vector θ=[θ1,θ2,...,θk] replace estimating
Angular range is carried out the division at equal intervals of k point by meter process from [- 60 ° ,+60 °]；
S430, angle direction vector is introduced, acquires the signal covariance matrix at each frequency point；Calculation formula is as follows:
Rs(fj)=[AH(fj,θ)A(fj,θ)]-1AH(fj,θ)P(fj)A(fj,θ)[AH(fj,θ)A(fj,θ)]-1；
S440, according to formulaCalculate focusing frequency f0The signal covariance matrix R at places(f0)；
S450, according to formula P (f0)=A (f0,θ)Rs(f0)AH(f0, θ), calculate focusing frequency f0The denoising covariance matrix P at place
(f0)。
7. the wideband signal spectrum cognitive method according to claim 2 based on airspace filter, which is characterized in that the step
Suddenly S600 includes:
S610, the denoising covariance matrix P (f to each frequency pointj) and the denoising covariance matrix P focused on frequency point
(f0) feature decomposition is carried out, obtain Uj,θAnd U0,θRespectively P (fj) and P (f0) feature matrix；It further obtains focusing square
Battle array Tθ(fj)=U0,θUj,θ H；
S620, pass through the focussing matrix Tθ(fj) by the frequency domain data covariance matrix focusing transform of each frequency point to frequency f0Place,
Frequency f at this time0The data covariance matrix at place is
S630, to the frequency f0Data covariance matrix R (the f at place0) using the WMUSIC algorithm progress characteristic value of narrow band signal
It decomposes, obtains corresponding eigenvalue λi, (i=1,2 ..., M), corresponding characteristic vector space is U=[u1,u2,...,uM]；
S640, processing is weighted to characteristic vector space with obtained characteristic value, defines new characteristic vector space
S650, determine that direction vector under focusing frequency, the angular range of this direction vector of traversal search obtain broadband signal
Spatial spectrum, using signal whether there is or not and the statistic of the ratio between spatial spectrum peak value difference, construction maximum value spectrum peak and minimum spectrum peak,
Compared with decision threshold γ, existing signal is adjudicated.
8. the wideband signal spectrum cognitive method according to claim 7 based on airspace filter, which is characterized in that the step
Suddenly S700 includes:
S710, the new variable formula using construction: Q (fj)=P (fj)PH(f0)P(f0)PH(fj), (j=J) will be described new
Variable Q (fj) instead of the frequency f in step S620 and step S6300Data covariance matrix R (the f at place0)；Carry out characteristic value point
Solution, obtains corresponding eigenvalue λQ i, (i=1,2 ..., M), corresponding characteristic vector space is UQ=[uQ 1,uQ 2,...,uQ M]；
S720, processing is weighted to characteristic vector space with obtained characteristic value, defines new characteristic vector space
S730, determine that direction vector under focusing frequency, the angular range of this direction vector of traversal search obtain broadband signal
The corresponding incident angle of peak value.
9. a kind of wideband signal spectrum sensory perceptual system based on airspace filter, which is characterized in that for execute as claim 1~
The step of 8 described in any item wideband signal spectrum cognitive methods based on airspace filter.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910069398.7A CN109831265B (en) | 2019-01-24 | 2019-01-24 | Broadband signal spectrum sensing method and system based on spatial filtering |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910069398.7A CN109831265B (en) | 2019-01-24 | 2019-01-24 | Broadband signal spectrum sensing method and system based on spatial filtering |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109831265A true CN109831265A (en) | 2019-05-31 |
| CN109831265B CN109831265B (en) | 2020-06-12 |
Family
ID=66862480
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910069398.7A Active CN109831265B (en) | 2019-01-24 | 2019-01-24 | Broadband signal spectrum sensing method and system based on spatial filtering |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109831265B (en) |
Cited By (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110133574A (en) * | 2019-07-02 | 2019-08-16 | 华南理工大学 | Utilize the one-dimensional DOA estimation method of the secondary virtual extended of multiple-frequency signal |
| CN112034418A (en) * | 2020-09-15 | 2020-12-04 | 北京中电慧声科技有限公司 | Beam scanning method based on frequency domain Bark sub-band and sound source orientation device |
| CN112272066A (en) * | 2020-09-15 | 2021-01-26 | 中国民用航空飞行学院 | Frequency spectrum data cleaning method used in airport terminal area very high frequency communication |
| CN112363109A (en) * | 2020-11-12 | 2021-02-12 | 西安科技大学 | Weighted direction of arrival estimation method based on broadband strong and weak adjacent signals |
| CN112965026A (en) * | 2021-02-08 | 2021-06-15 | 华南理工大学 | DOA array element spacing setting method with prior positioning angle range |
| CN113030842A (en) * | 2021-03-05 | 2021-06-25 | 电子科技大学 | Angle super-resolution DOA estimation method based on broadband signals |
| CN113111470A (en) * | 2021-04-13 | 2021-07-13 | 中国民用航空飞行学院 | Arraying method for identifying number features of targets of low-altitude small unmanned aerial vehicle |
| CN113625222A (en) * | 2021-08-27 | 2021-11-09 | 东南大学 | Broadband signal high-resolution direction estimation method based on time delay difference norm weighting |
| CN113687305A (en) * | 2021-07-26 | 2021-11-23 | 浙江大华技术股份有限公司 | Method, device and equipment for positioning sound source azimuth and computer readable storage medium |
| CN114397010A (en) * | 2021-12-29 | 2022-04-26 | 南京中科声势智能科技有限公司 | Transient signal acoustic imaging method based on wavelet decomposition |
| CN116626591A (en) * | 2023-05-26 | 2023-08-22 | 长安大学 | Broadband sound source positioning method, system, equipment and medium based on unmanned aerial vehicle harmonic focusing |
| WO2023193473A1 (en) * | 2022-04-06 | 2023-10-12 | 中兴通讯股份有限公司 | Spectrum sensing method, electronic device and storage medium |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20120313809A1 (en) * | 2011-06-10 | 2012-12-13 | Sony Corporation | Signal processing unit and method |
| CN103995221A (en) * | 2014-05-29 | 2014-08-20 | 西安交通大学 | Ultrasonic receiving array orientation method for transformer partial discharge positioning |
| CN104716982A (en) * | 2015-03-25 | 2015-06-17 | 北京北斗星通导航技术股份有限公司 | Robust anti-interference processing method and device of spread frequency system |
| CN107255793A (en) * | 2017-06-16 | 2017-10-17 | 中国电子科技集团公司第二十九研究所 | A kind of array df method and device for wideband OFDM signal of communication |
-
2019
- 2019-01-24 CN CN201910069398.7A patent/CN109831265B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20120313809A1 (en) * | 2011-06-10 | 2012-12-13 | Sony Corporation | Signal processing unit and method |
| CN103995221A (en) * | 2014-05-29 | 2014-08-20 | 西安交通大学 | Ultrasonic receiving array orientation method for transformer partial discharge positioning |
| CN104716982A (en) * | 2015-03-25 | 2015-06-17 | 北京北斗星通导航技术股份有限公司 | Robust anti-interference processing method and device of spread frequency system |
| CN107255793A (en) * | 2017-06-16 | 2017-10-17 | 中国电子科技集团公司第二十九研究所 | A kind of array df method and device for wideband OFDM signal of communication |
Non-Patent Citations (1)
| Title |
|---|
| 范辉: "基于MUSIC算法的空间谱测向方法研究", 《CNKI优秀硕士学位论文》 * |
Cited By (17)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110133574A (en) * | 2019-07-02 | 2019-08-16 | 华南理工大学 | Utilize the one-dimensional DOA estimation method of the secondary virtual extended of multiple-frequency signal |
| CN110133574B (en) * | 2019-07-02 | 2022-12-16 | 华南理工大学 | One-dimensional DOA estimation method utilizing secondary virtual expansion of multi-frequency signals |
| CN112272066B (en) * | 2020-09-15 | 2022-08-26 | 中国民用航空飞行学院 | Frequency spectrum data cleaning method used in airport terminal area very high frequency communication |
| CN112034418A (en) * | 2020-09-15 | 2020-12-04 | 北京中电慧声科技有限公司 | Beam scanning method based on frequency domain Bark sub-band and sound source orientation device |
| CN112272066A (en) * | 2020-09-15 | 2021-01-26 | 中国民用航空飞行学院 | Frequency spectrum data cleaning method used in airport terminal area very high frequency communication |
| CN112363109A (en) * | 2020-11-12 | 2021-02-12 | 西安科技大学 | Weighted direction of arrival estimation method based on broadband strong and weak adjacent signals |
| CN112965026A (en) * | 2021-02-08 | 2021-06-15 | 华南理工大学 | DOA array element spacing setting method with prior positioning angle range |
| CN112965026B (en) * | 2021-02-08 | 2023-07-18 | 华南理工大学 | DOA array element spacing setting method with priori positioning angle range |
| CN113030842A (en) * | 2021-03-05 | 2021-06-25 | 电子科技大学 | Angle super-resolution DOA estimation method based on broadband signals |
| CN113111470A (en) * | 2021-04-13 | 2021-07-13 | 中国民用航空飞行学院 | Arraying method for identifying number features of targets of low-altitude small unmanned aerial vehicle |
| CN113687305A (en) * | 2021-07-26 | 2021-11-23 | 浙江大华技术股份有限公司 | Method, device and equipment for positioning sound source azimuth and computer readable storage medium |
| CN113625222A (en) * | 2021-08-27 | 2021-11-09 | 东南大学 | Broadband signal high-resolution direction estimation method based on time delay difference norm weighting |
| CN113625222B (en) * | 2021-08-27 | 2022-10-28 | 东南大学 | Broadband signal high-resolution direction estimation method based on time delay difference norm weighting |
| CN114397010A (en) * | 2021-12-29 | 2022-04-26 | 南京中科声势智能科技有限公司 | Transient signal acoustic imaging method based on wavelet decomposition |
| WO2023193473A1 (en) * | 2022-04-06 | 2023-10-12 | 中兴通讯股份有限公司 | Spectrum sensing method, electronic device and storage medium |
| CN116626591A (en) * | 2023-05-26 | 2023-08-22 | 长安大学 | Broadband sound source positioning method, system, equipment and medium based on unmanned aerial vehicle harmonic focusing |
| CN116626591B (en) * | 2023-05-26 | 2024-02-20 | 长安大学 | Broadband sound source positioning method, system, equipment and medium based on unmanned aerial vehicle harmonic focusing |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109831265B (en) | 2020-06-12 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109831265A (en) | A kind of wideband signal spectrum cognitive method and system based on airspace filter | |
| Wang et al. | Multiantenna-assisted spectrum sensing for cognitive radio | |
| Yucek et al. | Spectrum characterization for opportunistic cognitive radio systems | |
| Eldemerdash et al. | Identification of cellular networks for intelligent radio measurements | |
| Liu et al. | Information theoretic criterion-based spectrum sensing for cognitive radio | |
| Zhang et al. | GLRT-based spectrum sensing with blindly learned feature under rank-1 assumption | |
| CN109889288B (en) | Continuous spectrum sensing method based on ensemble average value iterative detection | |
| Zhang | Theoretical performance and thresholds of the multitaper method for spectrum sensing | |
| Owayed et al. | Probabilities of detection and false alarm in multitaper based spectrum sensing for cognitive radio systems in AWGN | |
| Jin et al. | Spectrum sensing using multiple large eigenvalues and its performance analysis | |
| CN105429913B (en) | More level detections of feature based value and recognition methods | |
| WO2012092751A1 (en) | Method and system for neighboring cell interference detection | |
| Eldemerdash et al. | Fast and robust identification of GSM and LTE signals | |
| Qin et al. | Adaptive threshold for energy detector based on discrete wavelet packet transform | |
| Wang et al. | A multitaper spectrum based detector for cognitive radio | |
| Yawada et al. | Comparative study of spectrum sensing techniques base on techniques non-cooperative in cognitive radio networks | |
| De Vito et al. | Eigenvalue-based signal detector for an automatic modulation classifier | |
| Lu et al. | Novel spectrum sensing method based on the spatial spectrum for cognitive radio systems | |
| Diaz et al. | Cooperative spectrum sensing technique for identifying illegal FM broadcast radio stations using an energy detector and a peaks detector | |
| Rop et al. | Spectrum Sensing on High Density Cognitive Radio Vehicular Ad Hoc Network. | |
| Moghaddam et al. | A comparative study on the two popular cognitive radio spectrum sensing methods: Matched filter versus energy detector | |
| Jing et al. | Blind multiband spectrum sensing in cognitive radio network | |
| Zhang et al. | Spectrum sensing based on blindly learned signal feature | |
| Kumar et al. | Comprehensive analysis of cyclo-stationary feature detection technique for efficient spectrum usage: Future research and recent advantages | |
| Chen | A Weight Extreme Value Detection Method of Radio Frequency I/Q Data Component for Software Defined Radio Electromagnetic Spectrum Dynamic Access Strategy |
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 |