CN109495197A - A kind of adaptive wideband cooperation compression frequency spectrum sensing method - Google Patents
A kind of adaptive wideband cooperation compression frequency spectrum sensing method Download PDFInfo
- Publication number
- CN109495197A CN109495197A CN201811590953.2A CN201811590953A CN109495197A CN 109495197 A CN109495197 A CN 109495197A CN 201811590953 A CN201811590953 A CN 201811590953A CN 109495197 A CN109495197 A CN 109495197A
- Authority
- CN
- China
- Prior art keywords
- signal vector
- reconstruction
- frequency spectrum
- cognitive
- obtains
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04B—TRANSMISSION
- H04B17/00—Monitoring; Testing
- H04B17/30—Monitoring; Testing of propagation channels
- H04B17/382—Monitoring; Testing of propagation channels for resource allocation, admission control or handover
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W16/00—Network planning, e.g. coverage or traffic planning tools; Network deployment, e.g. resource partitioning or cells structures
- H04W16/14—Spectrum sharing arrangements between different networks
Abstract
The invention discloses a kind of adaptive wideband cooperations to compress frequency spectrum sensing method, comprising: S1, obtains the compression sampling that single cognitive user carries out M random measurement to the time-domain signal received, obtains M low-dimensional observation signal vector；S2, signal vector reconstruct is carried out to low-dimensional observation signal vector according to OSAMP algorithm, obtains reconstruction signal vector；S3, the reconstructed error for calculating reconstruction signal vector, and judge whether the reconstructed error is greater than preset error threshold, when the judgment result is yes, L random measurement is increased for the cognitive user and executes step S1；Otherwise, step S4 is executed；S4, step S1 to step S3 is repeated, until the reconstruction signal vector of all cognitive users is respectively less than error threshold；S5, the reconstruction signal vector of all cognitive users is merged, obtains reconstruction result；S6, cascading judgement is carried out to reconstruction result, court verdict is sent to all cognitive users in a broadcast manner.
Description
Technical field
The present invention relates to radio communication technology fields more particularly to a kind of adaptive wideband cooperation to compress frequency spectrum perception side
Method.
Background technique
In recent years, the rapid growth of wireless communication led to the wireless service needs on authorized frequency bands and unlicensed band all
Sharp increase.However, current fixed frequency spectrum allocation strategy makes frequency spectrum service efficiency very low.In order to solve frequency spectrum resource shortage
The not high problem with the availability of frequency spectrum, there has been proposed the concepts of cognitive radio (CR).CR technology can effectively utilize not
Band occupancy, and then improve the availability of frequency spectrum.In CR system, one of most important task of each cognitive user (CU) is exactly frequency
Spectrum perception, i.e., radio frequency environment is detected, with judge whether to have on interested channel or frequency range primary user (PU) into
Row communication.The main purpose of frequency spectrum perception is to efficiently identify unused frequency range or physical channel, i.e. frequency spectrum cavity-pocket, for
CR is used, to improve the handling capacity of entire cognitive system and service quality and protect the communication of PU interference-free.
Frequency spectrum perception is that cognitive radio (Cognitive Radio, CR) technology is able to the premise and basis applied, with
Traffic rate and bandwidth demand it is growing, broader frequency spectrum cognition technology receives more and more attention.However, traditional width
It is limited with frequency spectrum sensing method by nyquist sampling theorem, it is desirable that radio-frequency front-end has very high sampling rate, causes hard
Part is realized difficult.
Summary of the invention
Technical problems based on background technology, the invention proposes a kind of adaptive wideband cooperations to compress frequency spectrum perception
Method；
Frequency spectrum sensing method is compressed in a kind of adaptive wideband cooperation proposed by the present invention, comprising:
S1, the compression sampling that a cognitive user carries out M random measurement to the time-domain signal received is obtained, obtains M
A low-dimensional observation signal vector；
S2, signal vector reconstruct is carried out to low-dimensional observation signal vector according to OSAMP algorithm, obtains reconstruction signal vector；
S3, the reconstructed error for calculating reconstruction signal vector, and judge whether the reconstructed error is greater than preset error door
Limit increases L random measurement for the cognitive user and executes step S1 when the judgment result is yes；Otherwise, step is executed
S4；
S4, step S1 to step S3 is repeated, until the reconstruction signal vector of all cognitive users is respectively less than error threshold；
S5, the reconstruction signal vector of all cognitive users is merged, obtains reconstruction result；
S6, cascading judgement is carried out to reconstruction result, court verdict is sent to all cognitive users in a broadcast manner.
Preferably, step S2 is specifically included:
S21, initialization reconstruction signal X0=0, supported collection size I=M, remaining r0=Y, supported collection F0=[], the number of iterations
N=1；
S22, supported collection size I=I/2, pre- selected works S are calculatedn=Max (| Φ*rn-1|, I), Candidate Set Cn=Fn-1∪Sn,
Final supported collectionIt is remaining
S23,When, step S22 is executed, untilWhen, it holds
Row step S24；
S24, general | | r | |2With | | rn-1||2It is compared, | | r | |2> | | rn-1||2When, execute step S26；| | r |
|2=| | rn-1||2, enable I=I+1；| | r | |2< | | rn-1||2When, enable Fn=F, rn=r, n=n+1；
S25, judgement | | r | |2Whether opt is greater than | | Y | |2, if so, executing step S22；Otherwise, step S26 is executed；
S26, output reconstruction signal vectorWherein, Y indicates that low-dimensional observation signal vector, M indicate that low-dimensional is seen
Surveying signal vector number is, Φ indicates that preset observing matrix, * indicate conjugate transposition,Indicate generalized inverse, Sn、Cn、Fn、rnPoint
Not Biao Shi pre- selected works, Candidate Set, final supported collection and observation surplus in nth iteration, ΦFIt indicates to be located at supported collection in Φ
The submatrix of column composition corresponding to subscript in F,It indicates to be located at subscript collection CnIn subscript corresponding to atom composition,
Opt indicates preset stopping iterative parameter.
Preferably, step S3, the reconstructed error for calculating reconstruction signal vector, specifically includes:
Reconstructed error is
Preferably, step S5 is specifically included:
Reconstruction resultJ indicates cognitive user quantity,Indicate the reconstruct letter of jth position cognitive user
Number vector, j=1,2 ..., J.
The present invention can determine that successfully reconstruct is former in the case where spectrum sparse degree priori knowledge lacks by proceeding measurement
Best random measurement number required for beginning signal, while rationally utilizing system resource, avoiding unnecessary sampling overhead,
Ensure the broader frequency spectrum perceptual performance of system.The algorithm estimates excessive criterion according to degree of rarefication, right in supported collection selection course
Degree of rarefication carries out pre-estimation, while searching for optimal supported collection on the basis of degree of rarefication pre-estimation, optimizes the item of iteration stopping
Part, the whole flow process of OSAMP algorithm and the prior information for not containing any signal degree of rarefication, are self-adapting for signal degree of rarefication
Algorithm, and optimal supported collection is adaptively searched for according to concrete signal, be conducive to improve reconstruction accuracy, give based on OSAMP
Frequency spectrum sensing method is compressed in broadband under more cognitive user collaboration scenarios of algorithm, and the emulation under MATLAB communication simulation environment is real
Verify the validity and accuracy that this method is illustrated.
Detailed description of the invention
Fig. 1 is the flow diagram that frequency spectrum sensing method is compressed in a kind of adaptive wideband cooperation proposed by the present invention；
Fig. 2 is the first ROC curve figure in the embodiment of the present invention；
Fig. 3 is the second ROC curve figure in the embodiment of the present invention；
Perceptual performance changes schematic diagram with SNR in Fig. 4 embodiment of the present invention；
Average operating time is with sparse than changing schematic diagram in Fig. 5 inventive embodiments.
Specific embodiment
Referring to Fig.1, frequency spectrum sensing method is compressed in a kind of adaptive wideband cooperation proposed by the present invention, comprising:
Step S1 obtains the compression sampling that a cognitive user carries out M random measurement to the time-domain signal received, obtains
To M low-dimensional observation signal vector.
In concrete scheme, from process flow, compressive sensing theory model can be divided into rarefaction representation, compression is surveyed
Amount and three links of signal reconstruction.Real signal X is tieed up for a N × 1, orthogonal base vectors can be tieed up with N × 1Linear group
Closing indicates are as follows:Wherein, Ψ={ ψ1,ψ2,…,ψNIt is the orthogonal basic matrix of N × N-dimensional, s is X in matrix Ψ
On N × 1 tie up projection coefficient vector.If the nonzero element number K of coefficient vector s meets K < < N, claiming s is sparse on Ψ
, degree of rarefication K, Ψ are known as rarefaction representation matrix, sparse signal X projected on a M × N-dimensional observing matrix Φ,
The observation for completing sparse signal obtains the observation signal Y of the dimension of M × 1, the matrix representation forms of observation signal are as follows: Y=Φ X, due to
Dimension of the dimension well below original signal X of M < < N, observation signal Y, that is, realize complete while sampling the signal it is dilute
The compression for dredging signal, by formulaIn substitution formula Y=Φ X, Y=Φ X=Φ Ψ s=Θ s is obtained, wherein Θ
=Φ Ψ is M × N-dimensional matrix, referred to as projection matrix, and since observation dimension M is much smaller than signal dimension N, signal reconstruction faces
The problem of solving the underdetermined system of equations, using the sparsity of signal, compressive sensing theory can turn underdetermined system of equations Solve problems
It turns to l0Norm or l1The solution of norm minimum problem, in general, when other conditions are fixed, original signal is more sparse, at
Random measurement number required for function reconstructs is fewer, and in the broader frequency spectrum perception scene of cognitive radio, primary user is for frequency
The occupancy situation of spectrum determines the degree of rarefication of frequency spectrum, and the utilization rate of frequency spectrum resource is lower, and broader frequency spectrum is more sparse, due to cognition
There is usually no direct communications between radio system and authorized master user system, and primary user is for the use state of frequency spectrum
The lasting variation with time and location, this learn cognitive radio system often can not in advance wait perceive in practical applications
The practical degree of rarefication of broader frequency spectrum also can not accurately determine successfully to reconstruct random measurement number required for original signal, for this
One problem, the solution generallyd use are observational learning of the cognitive radio system by a period of time, obtain statistical significance
On degree of rarefication, the practical degree of rarefication of broader frequency spectrum signal is denoted as K by the present inventionr, and its priori in statistical significance is dilute
Thin degree is denoted as Ks.In the existing broader frequency spectrum perception algorithm based on compressive sensing theory, K is generally assumed thats=Kr, i.e., with system
Meter degree of rarefication replaces practical degree of rarefication, and determines required random measurement number M with this.It is equidistant theoretical according to being limited, when
Utilize l1When norm optimization's method restores original signal, M needs to meet M >=C μ2(Φ,Ψ)·KsLogN, wherein μ (Φ,
It is Ψ) coherence factor of observing matrix Φ and rarefaction representation matrix Ψ, constant C > 0；
In actual compression perception, following rule of thumb M >=4K is also commonly useds, the practical degree of rarefication K of signalrIt is dilute with counting
Dredge degree KsIt is general different, work as Ks< KrWhen, according to KsDetermining compression pendulous frequency MsIt is less than successfully needed for reconstruction signal
Pendulous frequency Mr, incite somebody to action so that cognitive user can not accurately restore source signal with higher probability, and cause broader frequency spectrum perceptibility
It can decline；Work as Ks> KrWhen, according to KsDetermining compression pendulous frequency MsPendulous frequency M needed for being greater than successfully reconstruction signalr,
The sampling wasting of resources will be caused so that sampling rate is excessively high.
Centralized broadband CR network system includes isolated data channel by J cognitive user (CU) and 1 cognitive base station
And control channel, each CUj(j=1,2 ..., J) to the time-domain signal receivedThe compression for carrying out M random measurement is adopted
Sample.
Step S2, according to OSAMP algorithm to low-dimensional observation signal vector carry out signal vector reconstruct, obtain reconstruction signal to
Amount.
This step specifically includes:
S21, initialization reconstruction signal X0=0, supported collection size I=M, remaining r0=Y, supported collection F0=[], the number of iterations
N=1；
S22, supported collection size I=I/2, pre- selected works S are calculatedn=Max (| Φ*rn-1|, I), Candidate Set Cn=Fn-1∪Sn,
Final supported collectionIt is remaining
S23,When, step S22 is executed, untilWhen, it holds
Row step S24；
S24, general | | r | |2With | | rn-1||2It is compared, | | r | |2> | | rn-1||2When, execute step S26；| | r |
|2=| | rn-1||2, enable I=I+1；| | r | |2< | | rn-1||2When, enable Fn=F, rn=r, n=n+1；
S25, judgement | | r | |2Whether opt is greater than | | Y | |2, if so, executing step S22；Otherwise, step S26 is executed；
S26, output reconstruction signal vectorWherein, Y indicates that low-dimensional observation signal vector, M indicate that low-dimensional is seen
Surveying signal vector number is, Φ indicates that preset observing matrix, * indicate conjugate transposition,Indicate generalized inverse, Sn、Cn、Fn、rnPoint
Not Biao Shi pre- selected works, Candidate Set, final supported collection and observation surplus in nth iteration, ΦFIt indicates to be located at supported collection in Φ
The submatrix of column composition corresponding to subscript in F,It indicates to be located at subscript collection CnIn subscript corresponding to atom composition,
Opt indicates preset stopping iterative parameter.
In concrete scheme, algorithm introduces step information, and segmentation carries out, and gradually expands supported collection, and the selection of step-length t is difficult
To meet the requirement of reconstruction accuracy and reconstructed velocity simultaneously, if selecting smaller, a large amount of operation time is consumed；If select compared with
Greatly, then signal reconstruction precision can be reduced, for this purpose, OSAMP algorithm estimates excessive criterion according to the degree of rarefication proposed in document [18],
Pre-estimation is carried out to degree of rarefication；
If it is (2K, δ that Φ, which meets parameter,2K) RIP property, ifThen estimate degree of rarefication k >
K, wherein Λ0The set constituted is indexed for k maximum value, | Λ0|=k,It indicates with Λ0For respectively column are constituted in the Φ of index
Submatrix,For its transposed matrix；
In OSAMP algorithm, estimation degree of rarefication k=M/2 is first set, obtains corresponding pre- selected works, Candidate Set and supported collection, if full
FootK=k/2 is then taken, makes to estimate degree of rarefication and gradually halves, its quick approaching to reality can be made sparse
Degree；Otherwise, supported collection size is increased with I=I+1, stops iterated conditional until meeting；
For SAMP algorithm, it is being unsatisfactory for stop condition | | r | |2≤opt·||Y||2In the case of, when remaining norm | | r |
|2When increase, section transformation is carried out, while increasing the size of supported collection, the selection for stopping iterative parameter opt is particularly important.If opt
It chooses larger, then will affect reconstruction accuracy；It is on the contrary then supported collection can be increased always, when the size of supported collection is increased to much larger than true
When real degree of rarefication, then a large amount of wrong atoms can be introduced to supported collection, reduce reconstruction accuracy again instead.It is often selected in SAMP algorithm
A fixed lesser stopping iterative parameter opt is selected, is desirably to obtain more preferably quality reconstruction, but due to by original signal and dilute
The influence of matrix is dredged, stop condition can not be met by often extending to supported collection close to observation, may be introduced instead more
Wrong atom；
OSAMP algorithm searches for optimal supported collection on the basis of degree of rarefication pre-estimation, when remaining norm | | r | |2Than previous
When secondary big, illustrate to have been introduced into wrong atom, then it is assumed that the supported collection obtained at this time in the current constant situation of supported collection size
Be it is optimal, exit circulation；When | | r | |2With it is preceding primary equal when, then explanation has searched current supported collection size
Rope has arrived relatively optimal supported collection, increases the size of supported collection, further decreases | | r | |2。
Step S3, calculates the reconstructed error of reconstruction signal vector, and judges whether the reconstructed error is greater than preset mistake
Poor thresholding increases L random measurement for the cognitive user and executes step S1 when the judgment result is yes；Otherwise, step is executed
Rapid S4.
The reconstructed error that this step calculates reconstruction signal vector specifically includes: reconstructed error is
In concrete scheme, the reconstructed error of reconstruction signal vector is calculated, and it is pre- to judge whether reconstruction signal vector is greater than
If error threshold, if reconstructed error be greater than error threshold, then it is assumed that M random measurement is not enough to accurate reconstruction original signal,
It is required that cognitive user CUjL random measurement is carried out, additionally again to improve the accuracy of reconstruct.
Step S4 repeats step S1 to step S3, until the reconstruction signal vector of all cognitive users is respectively less than error door
Limit.
Step S5 merges the reconstruction signal vector of all cognitive users, obtains reconstruction result.
This step specifically includes: reconstruction resultJ indicates cognitive user quantity,Indicate jth position
The reconstruction signal vector of cognitive user, j=1,2 ..., J.
Step S6 carries out cascading judgement to reconstruction result, and court verdict is sent to all cognitions in a broadcast manner and is used
Family.
Embodiment:
Cognitive radio networks include J=20 cognitive user, average signal-to-noise ratio SNR=0dB.In compressed sensing, signal
Length N=128, each cognitive user initial measurement number M0=32, the increased step-length L=8 of pendulous frequency.
Compression sampling match tracing (Compressive Sampling Matching PurCUit, CoSaMP) is chosen to calculate
K is set forth in the reference of method and SAMP as this paper algorithm performance, Fig. 2 and Fig. 3r=Ks=16 and Kr=32, Ks=16 two kinds
In the case of receiver performance characteristics (ROC) curve for perceiving of broader frequency spectrum carried out based on algorithms of different.
It can see referring to Fig. 2 and Fig. 3, when priori counts degree of rarefication KsWith practical degree of rarefication KrWhen equal, rule of thumb method
Then take Ms=4Ks=64, three kinds of algorithms can work normally in the case where identified compression ratio M/N=50%, obtain preferable
Detection performance.Work as Ks< KrWhen, CoSaMP algorithm counts degree of rarefication K according to priorisIdentified compression ratio M/N=50% makes
Measurement result needed for it can not obtain accurate reconstruction signal, causes perceptual performance to be greatly reduced；And mentioned method can be with herein
With the practical degree of rarefication K of signalrIncrease and be adaptively adjusted random measurement number, improve compression ratio, it is accurate that acquisition is enough
The random measurement of reconstruction signal is as a result, so as in the case where spectrum sparse degree prior information is inaccurate or lacks, still
Keep preferable perceptual performance.
At " constant false alarm ", the frequency spectrum perception performance of various methods takes K with the situation of change of Signal to Noise Ratio (SNR) for examinationr
=32, Ks=16, false-alarm probability Pf=1%, Fig. 4 give simulation result.
Referring to Fig. 4, it can be seen that the broader frequency spectrum cognitive method based on CoSaMP algorithm perceives under Low SNR
Performance is poor, and with the increase of SNR, perceptual performance is also mutually due for promotion.Mentioned method is under the conditions of compared with low signal-to-noise ratio herein
Also it can guarantee preferable perceptual performance, in SNR >=2dB situation, detection probability can achieve 90% or more.
Finally, examining or check algorithms of different average operating time with sparse than situation of change, emulate 500 times, Fig. 5 gives sparse
The comparative situation of algorithms of different in the case of being 0~0.5 than K/N variation range.
Referring to Fig. 5, it can be seen that compared to traditional SAMP algorithm, the average operating time of OSAMP algorithm has obvious excellent
Gesture is conducive to quickly approach practical degree of rarefication because of the degree of rarefication pre-estimation that the OSAMP algorithm first stage carries out, and saves big
Measure the time.And CoSaMP algorithm in advance according to statistics degree of rarefication determine random measurement number, will not adaptive change, it is relatively average
Runing time is less, but relatively poor from its reconstruction accuracy known to analysis above and simulation result.
The foregoing is only a preferred embodiment of the present invention, but scope of protection of the present invention is not limited thereto,
Anyone skilled in the art in the technical scope disclosed by the present invention, according to the technique and scheme of the present invention and its
Inventive concept is subject to equivalent substitution or change, should be covered by the protection scope of the present invention.
Claims (4)
1. frequency spectrum sensing method is compressed in a kind of adaptive wideband cooperation characterized by comprising
S1, the compression sampling that single cognitive user carries out M random measurement to the time-domain signal received is obtained, it is a low obtains M
Tie up observation signal vector；
S2, signal vector reconstruct is carried out to low-dimensional observation signal vector according to OSAMP algorithm, obtains reconstruction signal vector；
S3, the reconstructed error for calculating reconstruction signal vector, and judge whether the reconstructed error is greater than preset error threshold,
Judging result is to increase L random measurement when being for the cognitive user and execute step S1；Otherwise, step S4 is executed；
S4, step S1 to step S3 is repeated, until the reconstruction signal vector of all cognitive users is respectively less than error threshold；
S5, the reconstruction signal vector of all cognitive users is merged, obtains reconstruction result；
S6, cascading judgement is carried out to reconstruction result, court verdict is sent to all cognitive users in a broadcast manner.
2. frequency spectrum sensing method is compressed in adaptive wideband cooperation according to claim 1, which is characterized in that step S2, tool
Body includes:
S21, initialization reconstruction signal X0=0, supported collection size I=M, remaining r0=Y, supported collection F0=[], the number of iterations n=
1；
S22, supported collection size I=I/2, pre- selected works S are calculatedn=Max (| Φ*rn-1|, I), Candidate Set Cn=Fn-1∪Sn, final to prop up
Support collectionIt is remaining
S23,When, step S22 is executed, untilWhen, execute step
Rapid S24；
S24, general | | r | |2With | | rn-1||2It is compared, | | r | |2> | | rn-1||2When, execute step S26；| | r | |2=|
|rn-1||2, enable I=I+1；| | r | |2< | | rn-1||2When, enable Fn=F, rn=r, n=n+1；
S25, judgement | | r | |2Whether opt is greater than | | Y | |2, if so, executing step S22；Otherwise, step S26 is executed；
S26, output reconstruction signal vectorWherein, Y indicates that low-dimensional observation signal vector, M indicate low-dimensional observation letter
Number vector number is that Φ indicates that preset observing matrix, * indicate conjugate transposition,Indicate generalized inverse, Sn、Cn、Fn、rnTable respectively
Show pre- selected works, Candidate Set, final supported collection and the observation surplus in nth iteration, ΦFIt indicates to be located in supported collection F in Φ
Subscript corresponding to column composition submatrix,It indicates to be located at subscript collection CnIn subscript corresponding to atom composition, opt
Indicate preset stopping iterative parameter.
3. frequency spectrum sensing method is compressed in adaptive wideband cooperation according to claim 2, which is characterized in that step S3, institute
The reconstructed error for calculating reconstruction signal vector is stated, is specifically included:
Reconstructed error is
4. frequency spectrum sensing method is compressed in adaptive wideband cooperation according to claim 3, which is characterized in that step S5, tool
Body includes:
Reconstruction resultJ indicates cognitive user quantity,Indicate jth position cognitive user reconstruction signal to
Amount, j=1,2 ..., J.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811590953.2A CN109495197A (en) | 2018-12-21 | 2018-12-21 | A kind of adaptive wideband cooperation compression frequency spectrum sensing method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811590953.2A CN109495197A (en) | 2018-12-21 | 2018-12-21 | A kind of adaptive wideband cooperation compression frequency spectrum sensing method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109495197A true CN109495197A (en) | 2019-03-19 |
Family
ID=65711805
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811590953.2A Pending CN109495197A (en) | 2018-12-21 | 2018-12-21 | A kind of adaptive wideband cooperation compression frequency spectrum sensing method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109495197A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112737711A (en) * | 2020-12-24 | 2021-04-30 | 成都戎星科技有限公司 | Adaptive noise floor estimation method and broadband carrier detection method thereof |
| CN113708863A (en) * | 2021-09-10 | 2021-11-26 | 中国人民解放军63891部队 | Method and device for constructing spectrum sensing training data set |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102291341A (en) * | 2011-07-08 | 2011-12-21 | 南京邮电大学 | Broadband frequency spectrum perceiving method based on compression perception |
| CN103139828A (en) * | 2012-12-29 | 2013-06-05 | 东北大学 | Broadband spectrum sensing device and method |
-
2018
- 2018-12-21 CN CN201811590953.2A patent/CN109495197A/en active Pending
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102291341A (en) * | 2011-07-08 | 2011-12-21 | 南京邮电大学 | Broadband frequency spectrum perceiving method based on compression perception |
| CN103139828A (en) * | 2012-12-29 | 2013-06-05 | 东北大学 | Broadband spectrum sensing device and method |
Non-Patent Citations (3)
| Title |
|---|
| JIAO CHUANHAI 等: "Wideband compressed spectrum sensing based on modified sparsity adaptive matching pursuit algorithm", 《强激光与粒子》 * |
| 吕伟杰 等: "基于压缩感知的稀疏度自适应匹配追踪改进算法", 《控制与决策》 * |
| 吕伟杰 等: "基于压缩感知的自适应匹配追踪算法优化", 《系统工程与电子技术》 * |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112737711A (en) * | 2020-12-24 | 2021-04-30 | 成都戎星科技有限公司 | Adaptive noise floor estimation method and broadband carrier detection method thereof |
| CN113708863A (en) * | 2021-09-10 | 2021-11-26 | 中国人民解放军63891部队 | Method and device for constructing spectrum sensing training data set |
| CN113708863B (en) * | 2021-09-10 | 2023-08-01 | 中国人民解放军63891部队 | Method and device for constructing spectrum sensing training data set |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN102833020B (en) | Bayes compression broadband frequency spectrum detection method in cognitive radio network based on self-adaptive measurement | |
| CN102497337B (en) | Compressed sensing wireless communication channel estimation method based on sparsity self-adapting | |
| CN101630982B (en) | Broadband cognitive radio frequency spectrum detection method based on parallel compressed sensing | |
| CN101944961B (en) | Double threshold cooperative sensing method in cognitive wireless network | |
| CN105610525B (en) | A kind of broadband compression frequency spectrum sensing method based on sub-band match selection | |
| CN106131958A (en) | A kind of based on channel condition information with the indoor Passive Location of support vector machine | |
| CN109617850A (en) | OFDM condition of sparse channel estimation method based on self-adapting compressing perception | |
| CN102986269B (en) | A kind of method and apparatus estimating down channel quality | |
| CN104703216A (en) | Energy-efficient-based detecting method of multi-task Bayes compressed sensing broadband spectrum | |
| CN102984711A (en) | Multi-user collaborative spectrum sensing method based on single bit compression sensing technology | |
| WO2009038271A1 (en) | Method for automatic clustering, and method and apparatus for multipath clustering in wireless communication using the same | |
| CN105680963B (en) | A kind of distributed compression perceived spectral detection of energy efficiency priority and power distribution method | |
| CN103888145A (en) | Method for reconstructing signals | |
| CN106291452B (en) | A kind of direction of arrival estimation method based on enhanced greedy algorithm | |
| CN109495197A (en) | A kind of adaptive wideband cooperation compression frequency spectrum sensing method | |
| CN102915735B (en) | Noise-containing speech signal reconstruction method and noise-containing speech signal device based on compressed sensing | |
| CN106789766A (en) | Sparse OFDM channel estimation method based on homotopy method | |
| CN104780008A (en) | Broadband spectrum sensing method based on self-adaptive compressed sensing | |
| CN105553896A (en) | Non-uniform sampling and reconstruction method of broadband multi-frequency sparse signals | |
| KR101009827B1 (en) | Estimating apparatus and method of mobile station speed in a mobile communication system | |
| CN103326797B (en) | Cooperative frequency spectrum sensing method in cognition network | |
| CN101982953A (en) | Frequency domain multi-dimensional parameterized model of broadband wireless communication channel and modeling method | |
| CN103916969A (en) | Combined authorized user perception and link state estimation method and device | |
| CN110166383A (en) | One kind being based on tree-shaped random search pilot design method | |
| CN105610479B (en) | A kind of extensive MU-MIMO system channel estimation methods |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20190319 |
|
| RJ01 | Rejection of invention patent application after publication |