CN103974284B - A kind of broader frequency spectrum cognitive method based on partial reconfiguration - Google Patents
A kind of broader frequency spectrum cognitive method based on partial reconfiguration Download PDFInfo
- Publication number
- CN103974284B CN103974284B CN201410127331.1A CN201410127331A CN103974284B CN 103974284 B CN103974284 B CN 103974284B CN 201410127331 A CN201410127331 A CN 201410127331A CN 103974284 B CN103974284 B CN 103974284B
- Authority
- CN
- China
- Prior art keywords
- frequency spectrum
- iterations
- partial reconfiguration
- energy
- broader frequency
- 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.)
- Expired - Fee Related
Links
- 238000000034 method Methods 0.000 title claims abstract description 49
- 238000001228 spectrum Methods 0.000 title claims abstract description 40
- 230000001149 cognitive effect Effects 0.000 title claims abstract description 25
- 238000001514 detection method Methods 0.000 claims abstract description 25
- 238000007906 compression Methods 0.000 claims description 13
- 230000006835 compression Effects 0.000 claims description 13
- 238000004364 calculation method Methods 0.000 claims description 11
- 239000011159 matrix material Substances 0.000 claims description 6
- 239000000463 material Substances 0.000 claims description 5
- 230000004927 fusion Effects 0.000 claims description 4
- 230000008859 change Effects 0.000 abstract description 9
- 238000005070 sampling Methods 0.000 description 13
- 238000004422 calculation algorithm Methods 0.000 description 12
- 238000005516 engineering process Methods 0.000 description 6
- 230000008447 perception Effects 0.000 description 5
- 230000006870 function Effects 0.000 description 4
- 230000019771 cognition Effects 0.000 description 3
- 238000004891 communication Methods 0.000 description 3
- 230000000694 effects Effects 0.000 description 3
- 230000008569 process Effects 0.000 description 3
- 238000006243 chemical reaction Methods 0.000 description 2
- 238000012545 processing Methods 0.000 description 2
- 230000009467 reduction Effects 0.000 description 2
- 238000011160 research Methods 0.000 description 2
- 238000005303 weighing Methods 0.000 description 2
- 230000009471 action Effects 0.000 description 1
- 230000003044 adaptive effect Effects 0.000 description 1
- 238000013459 approach Methods 0.000 description 1
- 210000001367 artery Anatomy 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 238000001914 filtration Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000012804 iterative process Methods 0.000 description 1
- 239000000155 melt Substances 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 230000002035 prolonged effect Effects 0.000 description 1
- 230000000452 restraining effect Effects 0.000 description 1
- 238000004088 simulation Methods 0.000 description 1
- 230000003595 spectral effect Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
- 210000003462 vein Anatomy 0.000 description 1
Classifications
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
- Y02D30/00—Reducing energy consumption in communication networks
- Y02D30/70—Reducing energy consumption in communication networks in wireless communication networks
Landscapes
- Radar Systems Or Details Thereof (AREA)
- Mobile Radio Communication Systems (AREA)
Abstract
The invention provides a kind of broader frequency spectrum cognitive method based on partial reconfiguration, the thought of partial reconfiguration is applied to based on minimumThe interior point method that norm is solved.The present invention by successive ignition computing, progressively adjusts according to sampled value, is different from traditional reconstructing method without adjustment.The iterations used is not fixed, is changed with the energy of subchannel after reconstruct.If energy is big, iterations is reduced；Energy is small, increases iterations.The present invention can be used in that degree of rarefication is unknown, even in the broader frequency spectrum scene of degree of rarefication change, and reduce operation time in the case where ensureing accuracy of detection, improve the real-time of detection.
Description
Technical field
The present invention relates to wireless communication field, specifically a kind of broader frequency spectrum cognitive method based on partial reconfiguration.
Background technology
In cognitive radio (CR, Cognitive Radio) there is provided reliable communication and efficiently using radio-frequency spectrum this two
Individual target determines the importance for quickly and accurately carrying out frequency spectrum perception.But it is required when being perceived to broader frequency spectrum
High sampling rate and magnanimity sampled-data processing ability all proposes stern challenge to the hardware of frequency spectrum perception.In reality,
The number of sub-bands that primary user takes on broader frequency spectrum is far smaller than sum, meets openness.Then people associate naturally
Compression sampling technology (CS, Compressed Sampling, also referred to as compressed sensing, Compressive Sensing).It
As long as showing that signal is compressible or is sparse in some transform domain, so that it may with one and the conversion incoherent observation square of base
Battle array projects to the high dimensional signal obtained by conversion on one lower dimensional space, then just can be a small amount of from these by optimized algorithm
Projection in original signal is reconstructed with high probability, handled finally according to the signal after reconstruct, make detection judgement.Compression is adopted
Sample technology can substantially reduce requirement of the equipment to sample rate, wherein it is a step crucial during compression sampling is studied to reconstruct, if
If rapidly and efficiently original signal can not being reconstructed according to sampled value, then compression sampling is theoretical fixed for nyquist sampling
The superiority of rule can not be highlighted Chu Lai yet.
At present, the research emphasis of compression sampling be the acquisition of low speed observation sequence and the high probability of signal waveform, it is high-precision
Degree reconstruct.Wherein commonly use restructing algorithm and be largely divided into three major types：Based on l1The convex optimized algorithm of Norm minimum, based on l0Norm is most
Small greedy algorithm and combinational algorithm.The reconstruction accuracy that these methods have is high, and what is had is applied to the big occasion of data volume, but altogether
Same shortcoming has two：Amount of calculation is huge and compression ratio is influenceed by original signal degree of rarefication.In addition, existing based on compression sampling
Cognitive radio wideband frequency spectrum cognition technology assumes that broader frequency spectrum degree of rarefication is known.But in CR scenes, due to master
Between user (PU, Primary User) and cognitive user (or secondary user's, SU, Second User) can not direct communication or
Information exchange, and the spectrum occupancy of primary user is dynamic change, so this openness prior information of broader frequency spectrum
It is not readily available.Existing document proposes degree of rarefication algorithm for estimating：First quickly estimate actual degree of rarefication with fraction sampled value,
Computing is carried out again after total hits is adjusted according to estimate.This method does not need the priori of degree of rarefication, but to carry out volume
Outer sampling.Another conservative method is to obtain the upper bound of degree of rarefication according to the prolonged statistical process of cognitive user, with this
Determine compression ratio.However, this often causes number of samples excessive, the amount of calculation of signal reconstruction after further increasing.
On the other hand, Perfect Reconstruction original signal it is many processing application in it is not necessary to.Many times, we are
Wish the uninterested information from useful information is extracted in observation sequence or subsequent treatment is filtered out.Now, it is intended to pass through observation
It is apparently not best selection that sequence, which reconstructs and solves signal extraction and filtering problem again after all signals,.For example, recognizing
Know in radio system, whether cognitive user is only concerned channel occupancy, without concern for the concrete condition of channel, therefore, not always
It is to need Accurate Reconstruction to go out primary signal.Either minimize l1The restructing algorithm or greedy algorithm of norm, they pass through
Successive ignition, is progressively adjusted, and finally gives optimal solution, is a process approached.Base follows the trail of Method of Noise (BPDN, Basis
Pursuit De-noising) it is classical based on l1Algorithm of the reconstruct of Norm minimum by noise polluted signal.Existing scholar
This method is improved, but focuses on the precision for improving BPDN algorithms or expands it in other sparse noise (such as arteries and veins
Rush noise) performance under bad border.How amount of calculation is reduced on the premise of required precision is met, in addition it is also necessary to further research.
The content of the invention
Amount of calculation is huge in existing broader frequency spectrum cognition technology and compression ratio is sparse by original signal in order to solve by the present invention
The problem of degree influence, it was unknown to be used in degree of rarefication, very there is provided a kind of broader frequency spectrum cognitive method based on partial reconfiguration
To be degree of rarefication change broader frequency spectrum scene in, and ensure accuracy of detection in the case of reduce operation time, improve
The real-time of detection.
The present invention comprises the following steps：
1) cognitive user is sampled to frequency spectrum and low speed sample sequence is transferred into fusion center；
2) fusion center carries out Perfect Reconstruction to original signal with interior point method, and record iterations is T, updates iterations
For T=T-1；
3) energy of each sub-channels is calculated, maximum is designated as Jmax, minimum is designated as Jmin, define threshold value
Gmma=0.5 (Jmax-Jmin)；
4) partial reconfiguration is carried out with interior point method；
5) volume of the energy value of each channel after reconstruct, channel number of the record more than threshold value Gmma and subsignal is calculated
Number, feed back to cognitive user；
6) iterations is updated according to the energy situation of subchannel, if the energy that there is subchannel is more than 2Gmma, or existed
The difference of two sub-channels energy values is less than 1.5Gmma, then in next detection cycle, T=T+1, otherwise T=T-1；
7) repeat step 4) to step 6)；
Described comprises the following steps with interior point method to original signal progress Perfect Reconstruction：
1) error originated from input tolerance ε and maximum iteration Tmax；
2) according to target frequency bands Spectrum compression detection model y=Φ x, initialization data, if current iteration number of times T=0, x
=0, u=1 ∈ RN, one is defined than larger number, is designated as λ, it is desirable to and λ >=| | 2 ΦTy||∞, wherein Φ is M × N-dimensional observation square
Battle array, y is the sequence that observation is obtained, and u is convergence error；
3) T=T+1 is made, is passed throughCalculating the direction of search, (Δ x, wherein Δ u), H are Hessian matrix, and g is
The current gradient of (x, u)；
4) material calculation s, s for vector λ/| 2 ΦT(Φ x-y) | } in least member；
5) (x, u)=(x, u)+s (Δ x, Δ u) are updated;
6) convergence error u is calculated；
7) repeat step 3) to step 6) until meeting fault tolerance ε, i.e. min | | x | |1S.t. | | y- Φ x | | ＜ ε or
Reach greatest iteration number TmaxWhen stop.
Described comprises the following steps with interior point method progress partial reconfiguration：
1) T=T+1 is made, is passed throughCalculating the direction of search, (Δ x, wherein Δ u), H are Hessian matrix, and g is
The current gradient of (x, u)；
2) material calculation s, s for vector λ/| 2 ΦT(Φx- y) | } in least member；
3) (x, u)=(x, u)+s (Δ x, Δ u) are updated;
4) convergence error u is calculated；
5) repeat step 1) to step 4), iteration forces to stop when running to the iterations T after updating.
Beneficial effect of the present invention is：The thought of partial reconfiguration is applied to be based on minimum l1The interior point method that norm is solved,
Propose a kind of frequency spectrum sensing method based on partial reconfiguration.The present invention, by successive ignition computing, is progressively adjusted according to sampled value
It is whole, it is different from traditional reconstructing method without adjustment.The iterations used is not fixed, with subchannel after reconstruct
Energy and change.If energy is big, iterations is reduced；Energy is small, increases iterations.The present invention can be used in
Degree of rarefication is unknown, even in the broader frequency spectrum scene of degree of rarefication change, and is reduced in the case where ensureing accuracy of detection
Operation time, improve the real-time of detection.
Brief description of the drawings
Fig. 1 (a) is that a CR broader frequency spectrum perceives the original observation sequence figure of scene.
Fig. 1 (b) is that a CR broader frequency spectrum perceives scene with the sequence pattern after 25 Perfect Reconstructions of interior point method iteration.
Fig. 1 (c) is that a CR broader frequency spectrum perceives the second part reconstruct figure of scene iteration 12.
Fig. 1 (d) is that a CR broader frequency spectrum perceives the second part reconstruct figure of scene iteration 6.
Fig. 2 applies to the change of iterations in Fig. 1 scenes with detection cycle for the present invention.
Fig. 3 is comparison of the present invention with Perfect Reconstruction detection time again in 100 detection cycles.
Embodiment
The invention will be further described below in conjunction with the accompanying drawings.
The core concept of the present invention is：By compression sampling technology and the perception phase that broader frequency spectrum is directed in cognitive radio
With reference to.Using the method for partial reconfiguration to based on minimum l1The interior point method that norm is solved is improved.Make its iterations with
The energy of subchannel after reconstruct and change.The present invention can be applied not only in the constant broader frequency spectrum environment of degree of rarefication, dilute
Good Detection results can also be obtained in the case of dredging degree time-varying.On the premise of precision is ensured, the present invention can reduce fortune
Evaluation time, improves detection real-time.
In cognition wireless network, the task of frequency spectrum perception is to realize that dynamic spectrum connects on the premise of primary user is protected
Enter, cognitive user needs to be detected within each sampling period to judge to whether there is primary user in the frequency range.In order to reduce
The data acquisition and storage of sensing node, each cognitive user are direct to broadband analog signal using simulation/transcriber
Carry out acquisition of information.The present invention discussed in the case of not exclusively reconstruct original signal, with the energy of subchannel after partial reconfiguration with
Threshold value compares, and thus two kinds of hypothesis are tested, judgement is made.Target frequency bands Spectrum compression detection model can be represented
For：Y=Φ x.
Corresponding to primary user the absence and presence of two kinds of situations, two kinds are assumed as follows respectively：
H0:X=ω
H1:X=ω+s
Wherein s=[s (0) ..., s (N-1)]TWith ω=[ω (0) ..., ω (N-1)]TBe respectively in some channel with
The sampling value sequence of machine sparse signal and noise, and it is separate between s and ω.M is number of sampling points, and Φ is M × N-dimensional observation square
Battle array.
When Φ meets limited iso-distance constraint limitation, s can be reconstructed.Its solve model be：
min||x||1S.t. | | y- Φ x | | ＜ ε
Or writing unconstrained problem：
min||y-Φx||+λ||x||1
Base is referred to as to the method that above-mentioned two expression formula is solved and follows the trail of Method of Noise (BPDN).Parameter ε or λ effect are to use
To control the openness balance between allowable error, the table for ensureing to cause signal while signal errors is minimum as far as possible is made every effort to
Show most sparse.Compared with other restructing algorithms, BPDN algorithms are relatively good to the denoising effect of white Gaussian noise.But, BPDN
The complexity of method is very high.It requires that number of samples M meets M ＞ cK, c ≈ log2(N/K+1), time complexity is O (N3), K is
Degree of rarefication.The present invention scene unknown applied to degree of rarefication, M takes the upper bound.
Seen from the above description its essence is a principle of optimality by BPDN.Notice it is equally that the optimization of belt restraining is asked
Inscribe, the canonical form of linear programming (LP, Linear Program) is：
mincTX s.t.Ax=b
Wherein, variable x ∈ Rm, cTX is object function, and Ax=b is equality constraint.Obviously, expression formula min | | x1s.t.||
Y- Φ x | | ＜ ε and expression formula mincTX s.t.Ax=b can be changed mutually.Therefore the method for solution linear programming can be for solution
Certainly BPDN problems.
Simplex method is the universal method for solving linear programming.Its theoretical foundation is, if the feasible zone of linear programming problem
In the presence of (contradiction is not present between constraints), then feasible zone is vector space RNIn convex set.Feasible solution corresponding to summit
As basic feasible solution.The optimal value of linear programming problem is if it is present optimal solution must reach in certain apex of the convex set.
The thinking of simplex method is：A basic feasible solution is first found out, the target function value of the apex is calculated, sees whether be optimal
Solution；If it is not, comparing the target function value of adjacent vertex around, go on that smaller summit, and repeat this process.
In simplex method, adjacent vertex is searched only for, the amount of calculation very little of iteration is often walked, but needing to access many summits can just find most
Excellent solution.It may need to access all non-optimal summits in the worst cases.In order to reduce iterative steps, many scholars propose interior
Point method, its substitution method be it is mobile along " shortest path " inside feasible zone, i.e., next iteration when along making object function
Value obtains the most fast direction of minimum value.
The method according to the invention, first with interior point method, carries out weighing completely according to sample sequence to the signal of former channel
Structure, specific implementation step has：
Input：Fault tolerance ε, maximum iteration Tmax；
Initialization：Current iteration number of times T=0, x=0, u=1 ∈ RN, one is defined than larger number, is designated as λ, it is desirable to λ
≥||2ΦTy||∞；
Restructing operation：
Make T=T+1；
Pass through
Calculating the direction of search, (Δ x, wherein Δ u), H are Hessian matrix, and g is the gradient of current (x, u)；
Material calculation s, s for vector λ/| 2 ΦT(Φ x-y) | } in least member；
Update：(x, u)=(x, u)+s (Δ x, Δ u)；
Convergence error u is calculated,
Restructing operation is repeated until meeting error requirements or reaching that greatest iteration number then stops and output result, is recorded simultaneously
Lower iterations T used.
Interior point method usually requires to consider all feasible directions in every step iteration in order to find best moving direction.Change
Yan Zhi, compared to simplex method, interior point method exchanges the reduction of iterations for amount of calculation larger in each iteration.And reality exists
In frequency spectrum perception, just it can correctly detect whether primary user deposits by the feasible solution obtained in an iterative process.Figure of description 1
One CR broader frequency spectrum of simple hypothesis perceives scene, comprising 50 non-overlapping subchannels, and each channel is with 10 number tables
Show, 3 sub-channels of selection are occupied and spectral magnitude weighing apparatus is 1.Signal to noise ratio is 10dB.Fig. 1 (a) is original observation sequence, i.e. y,
Fig. 1 (b) is to use the sequence after interior point method Perfect Reconstruction, and iterations is 25, Fig. 1 (c) and Fig. 1 (d) is partial reconfiguration
As a result.Wherein Fig. 1 (c) figures iterations is that 12, Fig. 1 (d) is 6.Obviously, the reconstruction signal obtained after complete iteration is than calibrated
Really, with the reduction of iterations, the effect of reconstruct is also worse and worse.But this does not imply that the accurate of testing result can be influenceed
Property.In fact, in latter two figures, can still identify which channel by naked eyes occupied.It is primary even partial reconfiguration
Its energy of the subchannel of family occupancy is also far longer than the energy of vacant channels.
T=T-1 is updated, and calculates the energy of each sub-channels, select maximum is designated as Jmax, minimum is designated as Jmin.It is fixed
Adopted threshold value Gmma=0.5 (Jmax-Jmin), the threshold value will be employed in the detection after.Record is more than threshold value Gmma
Channel number and subsignal numbering, feed back to cognitive user；Think there is primary user in these channels, cognitive user can not
Take；
In second detection cycle, cognitive user is detected to frequency spectrum and is transferred to the result after projection again melts
Conjunction center.Middle heart action interior point method carries out partial reconfiguration but forces to stop when iterations reaches the T after updating.Calculate weight
The energy value of each channel, is compared and is judged with Gmma after structure；
It is still in the cycle, analyzes the energy of subchannel.If the energy that there is subchannel is more than 2Gmma, T=is updated
T+1；If the difference in the presence of two sub-channels energy values is less than 1.5Gmma, same T=T+1；Otherwise T=T-1.
Detection cycle afterwards is operated as second detection cycle, is carried out in summary according to following step.
1st, the iterations updated according to a upper detection cycle carries out partial reconfiguration；
2nd, each sub-channel energy is calculated according to the result after partial reconfiguration；
3rd, compared and entered a judgement with threshold value；
4th, iterations is updated.
According to the description of the invention, and Figure of description 2,3, those skilled in the art should be not difficult to find out, this hair
It is bright to be applied in the detection of the broader frequency spectrum of cognitive radio.According to channel circumstance and required precision, adaptive adjustment changes
Generation number.The present invention can not only be used in that degree of rarefication is unknown, even in the broader frequency spectrum scene of degree of rarefication change, Er Qie
Ensure to reduce operation time in the case of accuracy of detection, improve the real-time of detection.With the extensive scope of application.
Concrete application approach of the present invention is a lot, and described above is only the preferred embodiment of the present invention, it is noted that for
For those skilled in the art, under the premise without departing from the principles of the invention, some improvement can also be made, this
A little improve also should be regarded as protection scope of the present invention.The content not being described in detail in the present patent application book belongs to this area specialty
Prior art known to technical staff.
Claims (3)
1. a kind of broader frequency spectrum cognitive method based on partial reconfiguration, it is characterised in that comprise the following steps：
1) cognitive user is sampled to frequency spectrum and low speed sample sequence is transferred into fusion center；
2) fusion center carries out Perfect Reconstruction to original signal with interior point method, and record iterations is T, and renewal iterations is T
=T-1；
3) energy of each sub-channels is calculated, maximum is designated as Jmax, minimum is designated as Jmin, define threshold value Gmma=0.5
(Jmax-Jmin)；
4) partial reconfiguration is carried out with interior point method；
5) numbering of the energy value of each channel after reconstruct, channel number of the record more than threshold value Gmma and subsignal is calculated,
Feed back to cognitive user；
6) iterations is updated according to the energy situation of subchannel, if the energy that there is subchannel is more than 2Gmma, or there are two
The difference of sub-channel energy value is less than 1.5Gmma, then in next detection cycle, T=T+1, otherwise T=T-1；
7) repeat step 4) to step 6).
2. the broader frequency spectrum cognitive method according to claim 1 based on partial reconfiguration, it is characterised in that described utilization
Interior point method carries out Perfect Reconstruction to original signal and comprised the following steps：
1) error originated from input tolerance ε and maximum iteration Tmax；
2) according to target frequency bands Spectrum compression detection model y=Φ x, initialization data, if current iteration number of times T=0, x=0,One is defined than larger number, λ is designated as, it is desirable to λ >=| | 2 ΦTy||∞, wherein Φ is M × N-dimensional observing matrix, y
It is the sequence that observation is obtained, u is convergence error；
3) T=T+1 is made, is passed throughCalculate the direction of search (Δ x, wherein Δ u), H are Hessian matrix, g be it is current (x,
U) gradient；
4) material calculation s, s for vector λ/| 2 ΦT(Φ x-y) | } in least member；
5) (x, u)=(x, u)+s (Δ x, Δ u) are updated；
6) convergence error u is calculated；
7) repeat step 3) to step 6) until meeting fault tolerance ε, i.e. min | | x | |1S.t. | | y- Φ x | | ＜ ε reach
Greatest iteration number TmaxWhen stop.
3. the broader frequency spectrum cognitive method according to claim 1 based on partial reconfiguration, it is characterised in that：Described utilization
Interior point method carries out partial reconfiguration and comprised the following steps：
1) T=T+1 is made, is passed throughCalculate the direction of search (Δ x, wherein Δ u), H are Hessian matrix, g be it is current (x,
U) gradient；
2) material calculation s, s for vector λ/| 2 ΦT(Φ x-y) | } in least member；
3) (x, u)=(x, u)+s (Δ x, Δ u) are updated；
4) convergence error u is calculated；
5) repeat step 1) to step 4), iteration forces to stop when running to the iterations T after updating.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410127331.1A CN103974284B (en) | 2014-03-31 | 2014-03-31 | A kind of broader frequency spectrum cognitive method based on partial reconfiguration |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410127331.1A CN103974284B (en) | 2014-03-31 | 2014-03-31 | A kind of broader frequency spectrum cognitive method based on partial reconfiguration |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN103974284A CN103974284A (en) | 2014-08-06 |
| CN103974284B true CN103974284B (en) | 2017-10-31 |
Family
ID=51243217
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201410127331.1A Expired - Fee Related CN103974284B (en) | 2014-03-31 | 2014-03-31 | A kind of broader frequency spectrum cognitive method based on partial reconfiguration |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN103974284B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106301627B (en) * | 2015-06-01 | 2018-11-27 | 中国科学院上海微系统与信息技术研究所 | Distributed collaborative frequency spectrum sensing method in a kind of cognitive self-organizing network |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101640541A (en) * | 2009-09-04 | 2010-02-03 | 西安电子科技大学 | Reconstruction method of sparse signal |
| WO2011121443A2 (en) * | 2010-03-31 | 2011-10-06 | Alcatel Lucent | Shared cooperative spectrum sensing method, sensing nodes and fusion center in cognitive radio networks |
| CN103036574A (en) * | 2012-12-13 | 2013-04-10 | 南开大学 | Self-check sparseness self-adaption matching pursuit arithmetic based on compressive sensing |
| CN103138859A (en) * | 2013-02-25 | 2013-06-05 | 东华大学 | Cognition wireless broadband frequency spectrum compressed sensing method based on backtracking and centralized type cooperation |
| CN103684472A (en) * | 2013-12-29 | 2014-03-26 | 哈尔滨工业大学 | Reconfiguration method of adaptive signal of 1-Bit sparse level based on compression perception |
-
2014
- 2014-03-31 CN CN201410127331.1A patent/CN103974284B/en not_active Expired - Fee Related
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101640541A (en) * | 2009-09-04 | 2010-02-03 | 西安电子科技大学 | Reconstruction method of sparse signal |
| WO2011121443A2 (en) * | 2010-03-31 | 2011-10-06 | Alcatel Lucent | Shared cooperative spectrum sensing method, sensing nodes and fusion center in cognitive radio networks |
| CN103036574A (en) * | 2012-12-13 | 2013-04-10 | 南开大学 | Self-check sparseness self-adaption matching pursuit arithmetic based on compressive sensing |
| CN103138859A (en) * | 2013-02-25 | 2013-06-05 | 东华大学 | Cognition wireless broadband frequency spectrum compressed sensing method based on backtracking and centralized type cooperation |
| CN103684472A (en) * | 2013-12-29 | 2014-03-26 | 哈尔滨工业大学 | Reconfiguration method of adaptive signal of 1-Bit sparse level based on compression perception |
Non-Patent Citations (3)
| Title |
|---|
| An Interior-Point Method for Large-Scale l1-Regularized Least Squares;Seung-Jean Kim等;《IEEE JOURNAL OF SELECTED TOPICS IN SIGNAL PROCESSING》;20071231;全文 * |
| 一种基于基追踪压缩感知信号重构的改进算法;芮国胜等;《电子测量技术》;20100430;全文 * |
| 认知无线网络中基于非重构序贯压缩的随机信号检测算法与分析;涂思怡等;《信号处理》;20140228;全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN103974284A (en) | 2014-08-06 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN105654049B (en) | The method and device of facial expression recognition | |
| KR102387020B1 (en) | Hypernetwork training method and device, electronic device and storage medium | |
| CN110824450A (en) | Radar target HRRP robust identification method in noise environment | |
| CN103295198B (en) | Based on redundant dictionary and the sparse non-convex compressed sensing image reconstructing method of structure | |
| McCoy et al. | Convexity in source separation: Models, geometry, and algorithms | |
| CN112512069B (en) | Network intelligent optimization method and device based on channel beam pattern | |
| CN110879351B (en) | Fault diagnosis method for non-linear analog circuit based on RCCA-SVM | |
| CN112381667B (en) | Distribution network electrical topology identification method based on deep learning | |
| Yang et al. | One-dimensional deep attention convolution network (ODACN) for signals classification | |
| CN103138859B (en) | Cognition wireless broadband frequency spectrum compressed sensing method based on backtracking and centralized type cooperation | |
| CN112468203B (en) | Low-rank CSI feedback method, storage medium and equipment for deep iterative neural network | |
| CN109815849A (en) | Chaotic signal Denoising Algorithm based on singular value decomposition | |
| CN107547088A (en) | Enhanced self-adapted segmentation orthogonal matching pursuit method based on compressed sensing | |
| CN108225332B (en) | Indoor positioning fingerprint map dimension reduction method based on supervision | |
| CN103974284B (en) | A kind of broader frequency spectrum cognitive method based on partial reconfiguration | |
| Han et al. | Communication emitter individual identification via 3D‐Hilbert energy spectrum‐based multiscale segmentation features | |
| Subekti et al. | Spectrum sensing for cognitive radio using deep autoencoder neural network and SVM | |
| CN118158607B (en) | UWB base station ranging precision improving method, device and equipment | |
| CN108107421A (en) | A kind of interior distance measuring method and device | |
| Aghabiglou et al. | The R2D2 deep neural network series paradigm for fast precision imaging in radio astronomy | |
| CN114757224A (en) | Specific radiation source identification method based on continuous learning and combined feature extraction | |
| Koziel | Shape-preserving response prediction for microwave circuit modeling | |
| CN116754213A (en) | Electric gate valve fault diagnosis method, device and equipment based on strong noise background | |
| CN103577877A (en) | Ship motion prediction method based on time-frequency analysis and BP neural network | |
| CN111083632A (en) | Ultra-wideband indoor positioning method based on support vector machine |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20171031 Termination date: 20190331 |
|
| CF01 | Termination of patent right due to non-payment of annual fee |