CN108322277B - Frequency spectrum sensing method based on inverse eigenvalue of covariance matrix - Google Patents
Frequency spectrum sensing method based on inverse eigenvalue of covariance matrix Download PDFInfo
- Publication number
- CN108322277B CN108322277B CN201810292741.XA CN201810292741A CN108322277B CN 108322277 B CN108322277 B CN 108322277B CN 201810292741 A CN201810292741 A CN 201810292741A CN 108322277 B CN108322277 B CN 108322277B
- Authority
- CN
- China
- Prior art keywords
- eigenvalue
- sampling
- covariance matrix
- inverse
- time slot
- 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.)
- Active
Links
Images
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
Abstract
The invention discloses a frequency spectrum sensing method based on inverse eigenvalues of a covariance matrix, which is characterized in that the covariance matrix is calculated according to sampling signals obtained by simultaneously sampling signals received by all receiving antennas in the current sensing time slot; then acquiring a maximum eigenvalue, a next maximum eigenvalue, a minimum eigenvalue and a next minimum eigenvalue of the covariance matrix; then, acquiring a first inverse eigenvalue of the covariance matrix according to the maximum eigenvalue and the minimum eigenvalue; acquiring a second inverse eigenvalue of the covariance matrix according to the second largest eigenvalue and the second smallest eigenvalue; then constructing a test statistic according to the first inverse characteristic value and the second inverse characteristic value; calculating a decision threshold according to the target false alarm probability; finally, whether an authorized user signal exists in a current sensing time slot is judged by comparing the test statistic with the judgment threshold so as to realize spectrum sensing; the method has the advantages that prior information of the signal and noise power of the authorized user does not need to be known, and the spectrum sensing performance is good.
Description
Technical Field
The invention relates to a spectrum sensing method in a cognitive radio system, in particular to a spectrum sensing method based on inverse eigenvalues of a covariance matrix.
Background
The rapid development of diversification of mobile communication services greatly enriches and facilitates people's work and life, but the demand for the number of wireless devices and mobile data traffic correspondingly increases explosively, which causes a problem of shortage of spectrum resources. In recent years, the problem of shortage of spectrum resources has been gradually emerging and will become increasingly serious in the foreseeable future. However, this is not caused by insufficient physical spectrum resources, but because many spectrum resources cannot be fully utilized by the existing fixed spectrum allocation strategy, the spectrum utilization rate is greatly reduced. Therefore, improving spectrum utilization becomes a key to solving this problem. To address this problem, doctor Mitola proposed cognitive radio technology. The cognitive radio technology means that a wireless device can interact with a communication environment and change self transmission parameters according to an interaction result, so that potential idle frequency spectrum is flexibly utilized in a dynamic and self-adaptive mode. In order to avoid interference to authorized users, the cognitive radio technology needs to be able to accurately and quickly find a free spectrum, and implement robust spectrum sensing. Therefore, spectrum sensing becomes one of the key technologies of cognitive radio.
At present, the property of eigenvalue of covariance matrix of received signals has been widely applied in cognitive network, and eigenvalue-based spectrum sensing methods are also emerging, such as: maximum eigenvalue detection, maximum-minimum eigenvalue detection, maximum-arithmetic mean eigenvalue detection, etc. The maximum characteristic value detection method needs to know the noise power to set a judgment threshold, and when the noise power is unknown, the spectrum sensing performance of the method is poor; although other spectrum sensing methods based on characteristic values have stronger robustness in resisting noise interference, the spectrum sensing performance needs to be improved.
Disclosure of Invention
The technical problem to be solved by the invention is to provide a spectrum sensing method based on inverse eigenvalues of a covariance matrix, which does not need to know prior information of authorized user signals and noise power, has good spectrum sensing performance and has stronger robustness in the aspect of resisting noise interference.
The technical scheme adopted by the invention for solving the technical problems is as follows: a frequency spectrum sensing method based on inverse eigenvalue of covariance matrix is characterized in that the main processing procedures of the frequency spectrum sensing method are as follows: calculating a covariance matrix according to sampling signals obtained by simultaneously sampling signals received by all receiving antennas in a current sensing time slot; then, carrying out characteristic decomposition on the covariance matrix to obtain a maximum eigenvalue, a next maximum eigenvalue, a minimum eigenvalue and a next minimum eigenvalue of the covariance matrix; then, according to the maximum eigenvalue and the minimum eigenvalue of the covariance matrix, a first inverse eigenvalue of the covariance matrix is obtained; obtaining a second inverse eigenvalue of the covariance matrix according to the second largest eigenvalue and the second smallest eigenvalue of the covariance matrix; then constructing test statistics according to the first inverse eigenvalue and the second inverse eigenvalue of the covariance matrix; calculating a decision threshold according to the target false alarm probability; and finally, judging whether an authorized user signal exists in the current sensing time slot or not by comparing the test statistic with the judgment threshold to realize spectrum sensing.
The frequency spectrum sensing method based on the inverse eigenvalue of the covariance matrix comprises the following steps:
the method comprises the following steps: setting M receiving antennas at a receiving end, setting the number of sampling points in each sensing time slot to be N, and simultaneously sampling signals received by all the receiving antennas in each sensing time slot; wherein M >1, N > 1;
step two: marking a sampling signal obtained by sampling signals received by all receiving antennas simultaneously in a current sensing time slot as Xcur，XcurExpressed in vector form as: xcur＝[x(1),x(2),…,x(N)]，x(1)＝[x1(1),x2(1),…,xM(1)]T，x(2)＝[x1(2),x2(2),…,xM(2)]T，x(N)＝[x1(N),x2(N),…,xM(N)]T(ii) a Wherein the symbol "[ alpha ],")]"is a vector representation symbol, x (1), x (2), …, x (N) correspondingly represents that signals received by all receiving antennas are simultaneously sampled in a current sensing time slot, the signals are formed by all obtained 1 st sampling point signals in sequence, the signals are formed by all obtained 2 nd sampling point signals in sequence, …, and the signals are formed by all obtained N nd sampling point signals in sequence, the dimensions of x (1), x (2) and x (N) are M x 1, and x (1), x (2) and x (N) are M x 11(1),x2(1),…,xM(1) Correspondingly represents the 1 st sampling point signal obtained by sampling the signal received by the 1 st receiving antenna in the current sensing time slot, the 1 st sampling point signal obtained by sampling the signal received by the 2 nd receiving antenna, …, the 1 st sampling point signal obtained by sampling the signal received by the Mth receiving antenna, x1(2),x2(2),…,xM(2) Correspondingly represents the 2 nd sampling point signal obtained by sampling the signal received by the 1 st receiving antenna in the current sensing time slot, the 2 nd sampling point signal obtained by sampling the signal received by the 2 nd receiving antenna, …, the 2 nd sampling point signal obtained by sampling the signal received by the Mth receiving antenna, x1(N),x2(N),…,xM(N) corresponding representationAn nth sampling point signal obtained by sampling a signal received by a 1 st receiving antenna in a current sensing time slot, an nth sampling point signal obtained by sampling a signal received by a 2 nd receiving antenna, …, an nth sampling point signal obtained by sampling a signal received by an mth receiving antenna, [ x ] x1(1),x2(1),…,xM(1)]TIs [ x ]1(1),x2(1),…,xM(1)]Is transposed, [ x ]1(2),x2(2),…,xM(2)]TIs [ x ]1(2),x2(2),…,xM(2)]Is transposed, [ x ]1(N),x2(N),…,xM(N)]TIs [ x ]1(N),x2(N),…,xM(N)]Transposing;
step three: according to XcurCalculate the covariance matrix, note
Then to
Performing characteristic decomposition to obtain
The maximum eigenvalue, the next maximum eigenvalue, the minimum eigenvalue and the next minimum eigenvalue of (A) are correspondingly recorded as (lambda)1、λ2、λ3And λ4(ii) a Then according to lambda1And λ3Calculating
Is noted as v1，
And according to λ2And λ4Calculating
Second inverse eigenvalue of (v)2，
Wherein, Xcur HIs XcurThe conjugate transpose of (1);
step four: according to v1And v2Constructing test statistic, and recording as T, T ═ v1+v2(ii) a And calculating a decision threshold, and recording as gamma, gamma is 2I-1(1-Pf) (ii) a Wherein, I-1() Inverse function, P, representing a regularized incomplete beta functionfRepresenting a target false alarm probability;
step five: the spectrum sensing is realized by comparing the magnitude of T with the magnitude of gamma, if T < gamma, the authorized user signal is judged to exist in the current sensing time slot; if T ≧ gamma, it is determined that there is no authorized user signal within the current one of the sensing timeslots.
Compared with the prior art, the invention has the advantages that:
1) the method utilizes the inverse eigenvalue of the covariance matrix of the received signals to construct the test statistic, the inverse eigenvalue can fully reflect the characteristics of the covariance matrix of the received signals, and the method has better spectrum sensing performance than a maximum-minimum eigenvalue detection method and a maximum-arithmetic mean eigenvalue detection method.
2) The method of the invention does not need prior information of authorized user signals and noise power when calculating the decision threshold, so the method of the invention is a blind spectrum detection method.
3) The method has stronger robustness in the aspect of resisting noise interference.
Drawings
FIG. 1 is a block flow diagram of the method of the present invention;
FIG. 2 shows the false alarm probability P of the targetfWhen the value is equal to 0.1, curve comparison graphs of the missed detection probability changing along with the signal-to-noise ratio are obtained by using the method of the invention and the existing maximum and minimum eigenvalue detection method, the existing maximum and minimum eigenvalue mean-square ratio detection method and the existing maximum-arithmetic mean eigenvalue detection method respectively.
Detailed Description
The invention is described in further detail below with reference to the accompanying examples.
Research shows that the inverse characteristic value of the covariance matrix of the received signals can also be used for realizing spectrum detection, the robustness in resisting noise interference is stronger, and the detection performance is superior to that of other spectrum sensing methods based on the characteristic value. However, few people use this property to realize spectrum sensing, and therefore, the present application proposes a spectrum sensing method based on inverse eigenvalues of covariance matrix.
The invention provides a frequency spectrum sensing method based on inverse eigenvalues of covariance matrix, the flow chart of which is shown in figure 1, and the main processing procedures are as follows: calculating a covariance matrix according to sampling signals obtained by simultaneously sampling signals received by all receiving antennas in a current sensing time slot; then, carrying out characteristic decomposition on the covariance matrix to obtain a maximum eigenvalue, a next maximum eigenvalue, a minimum eigenvalue and a next minimum eigenvalue of the covariance matrix; then, according to the maximum eigenvalue and the minimum eigenvalue of the covariance matrix, a first inverse eigenvalue of the covariance matrix is obtained; obtaining a second inverse eigenvalue of the covariance matrix according to the second largest eigenvalue and the second smallest eigenvalue of the covariance matrix; then constructing test statistics according to the first inverse eigenvalue and the second inverse eigenvalue of the covariance matrix; calculating a decision threshold according to the target false alarm probability; and finally, judging whether an authorized user signal exists in the current sensing time slot or not by comparing the test statistic with the judgment threshold to realize spectrum sensing.
The frequency spectrum sensing method specifically comprises the following steps:
the method comprises the following steps: setting M receiving antennas at a receiving end, setting the number of sampling points in each sensing time slot to be N, and simultaneously sampling signals received by all the receiving antennas in each sensing time slot; where M >1, M is 8 and N >1 in this example, and N is 100 in this example.
Step two: will be received simultaneously for all receiving antennas in the current one sensing time slotIs sampled to obtain a sampled signal denoted as Xcur，XcurExpressed in vector form as: xcur＝[x(1),x(2),…,x(N)]，x(1)＝[x1(1),x2(1),…,xM(1)]T，x(2)＝[x1(2),x2(2),…,xM(2)]T，x(N)＝[x1(N),x2(N),…,xM(N)]T(ii) a Wherein the symbol "[ alpha ],")]"is a vector representation symbol, x (1), x (2), …, x (N) correspondingly represents that signals received by all receiving antennas are simultaneously sampled in a current sensing time slot, the signals are formed by all obtained 1 st sampling point signals in sequence, the signals are formed by all obtained 2 nd sampling point signals in sequence, …, and the signals are formed by all obtained N nd sampling point signals in sequence, the dimensions of x (1), x (2) and x (N) are M x 1, and x (1), x (2) and x (N) are M x 11(1),x2(1),…,xM(1) Correspondingly represents the 1 st sampling point signal obtained by sampling the signal received by the 1 st receiving antenna in the current sensing time slot, the 1 st sampling point signal obtained by sampling the signal received by the 2 nd receiving antenna, …, the 1 st sampling point signal obtained by sampling the signal received by the Mth receiving antenna, x1(2),x2(2),…,xM(2) Correspondingly represents the 2 nd sampling point signal obtained by sampling the signal received by the 1 st receiving antenna in the current sensing time slot, the 2 nd sampling point signal obtained by sampling the signal received by the 2 nd receiving antenna, …, the 2 nd sampling point signal obtained by sampling the signal received by the Mth receiving antenna, x1(N),x2(N),…,xM(N) represents the Nth sampling point signal obtained by sampling the signal received by the 1 st receiving antenna in the current sensing time slot, the Nth sampling point signal obtained by sampling the signal received by the 2 nd receiving antenna, …, and the Nth sampling point signal obtained by sampling the signal received by the Mth receiving antenna, [ x ] x1(1),x2(1),…,xM(1)]TIs [ x ]1(1),x2(1),…,xM(1)]Is transposed, [ x ]1(2),x2(2),…,xM(2)]TIs [ x ]1(2),x2(2),…,xM(2)]Is transposed, [ x ]1(N),x2(N),…,xM(N)]TIs [ x ]1(N),x2(N),…,xM(N)]The transposing of (1).
Step three: according to XcurCalculate the covariance matrix, note
Then to
Performing characteristic decomposition to obtain
The maximum eigenvalue, the next maximum eigenvalue, the minimum eigenvalue and the next minimum eigenvalue of (A) are correspondingly recorded as (lambda)1、λ2、λ3And λ4(ii) a Then according to lambda1And λ3Calculating
Is noted as v1，
And according to λ2And λ4Calculating
Second inverse eigenvalue of (v)2，
Wherein, Xcur HIs XcurThe conjugate transpose of (c).
Step four: according to v1And v2Constructing test statistic, and recording as T, T ═ v1+v2(ii) a And calculating a decision threshold, and recording as gamma, gamma is 2I-1(1-Pf) (ii) a Wherein, I-1() Inverse function, P, representing a regularized incomplete beta functionfIndicating target false alarm probability, e.g. taking Pf＝0.1。
Step five: the spectrum sensing is realized by comparing the magnitude of T with the magnitude of gamma, if T < gamma, the authorized user signal is judged to exist in the current sensing time slot; if T ≧ gamma, it is determined that there is no authorized user signal within the current one of the sensing timeslots.
The feasibility and effectiveness of the method of the invention is further illustrated by the following simulations.
In the simulation, the receiving end is set to have 8 receiving antennas, the number of sampling points in each sensing time slot is set to 100, and the power of white gaussian noise
FIG. 2 shows the false alarm probability P of the targetfWhen the value is equal to 0.1, curve comparison graphs of the missed detection probability changing along with the signal-to-noise ratio are obtained by using the method of the invention and the existing maximum and minimum eigenvalue detection method, the existing maximum and minimum eigenvalue mean-square ratio detection method and the existing maximum-arithmetic mean eigenvalue detection method respectively. As can be seen from fig. 2, the probability of missed detection obtained by using the method of the present invention is lower than the probability of missed detection obtained by using the existing three feature value-based detection methods, so that it can be shown that the spectrum sensing performance of the method of the present invention is better than that of the existing feature value-based detection method.
Claims (1)
1. A frequency spectrum sensing method based on inverse eigenvalue of covariance matrix is characterized in that the main processing procedures of the frequency spectrum sensing method are as follows: calculating a covariance matrix according to sampling signals obtained by simultaneously sampling signals received by all receiving antennas in a current sensing time slot; then, carrying out characteristic decomposition on the covariance matrix to obtain a maximum eigenvalue, a next maximum eigenvalue, a minimum eigenvalue and a next minimum eigenvalue of the covariance matrix; then, according to the maximum eigenvalue and the minimum eigenvalue of the covariance matrix, a first inverse eigenvalue of the covariance matrix is obtained; obtaining a second inverse eigenvalue of the covariance matrix according to the second largest eigenvalue and the second smallest eigenvalue of the covariance matrix; then constructing test statistics according to the first inverse eigenvalue and the second inverse eigenvalue of the covariance matrix; calculating a decision threshold according to the target false alarm probability; finally, whether an authorized user signal exists in a current sensing time slot is judged by comparing the test statistic with the judgment threshold so as to realize spectrum sensing;
the frequency spectrum sensing method comprises the following steps:
the method comprises the following steps: setting M receiving antennas at a receiving end, setting the number of sampling points in each sensing time slot to be N, and simultaneously sampling signals received by all the receiving antennas in each sensing time slot; wherein M is more than 1, and N is more than 1;
step two: marking a sampling signal obtained by sampling signals received by all receiving antennas simultaneously in a current sensing time slot as Xcur，XcurExpressed in vector form as: xcur＝[x(1),x(2),…,x(N)]，x(1)＝[x1(1),x2(1),…,xM(1)]T，x(2)＝[x1(2),x2(2),…,xM(2)]T，x(N)＝[x1(N),x2(N),…,xM(N)]T(ii) a Wherein the symbol "[ alpha ],")]"is a vector representation symbol, x (1), x (2), …, x (N) correspondingly represents that signals received by all receiving antennas are simultaneously sampled in a current sensing time slot, the signals are formed by all obtained 1 st sampling point signals in sequence, the signals are formed by all obtained 2 nd sampling point signals in sequence, …, and the signals are formed by all obtained N nd sampling point signals in sequence, the dimensions of x (1), x (2) and x (N) are M x 1, and x (1), x (2) and x (N) are M x 11(1),x2(1),…,xM(1) Correspondingly represents the 1 st sampling point signal obtained by sampling the signal received by the 1 st receiving antenna in the current sensing time slot, the 1 st sampling point signal obtained by sampling the signal received by the 2 nd receiving antenna, …, the 1 st sampling point signal obtained by sampling the signal received by the Mth receiving antenna, x1(2),x2(2),…,xM(2) The correspondence is expressed inSampling the signal received by the 1 st receiving antenna in the previous sensing time slot to obtain the 2 nd sampling point signal, sampling the signal received by the 2 nd receiving antenna to obtain the 2 nd sampling point signal, …, sampling the signal received by the Mth receiving antenna to obtain the 2 nd sampling point signal, x1(N),x2(N),…,xM(N) represents the Nth sampling point signal obtained by sampling the signal received by the 1 st receiving antenna in the current sensing time slot, the Nth sampling point signal obtained by sampling the signal received by the 2 nd receiving antenna, …, and the Nth sampling point signal obtained by sampling the signal received by the Mth receiving antenna, [ x ] x1(1),x2(1),…,xM(1)]TIs [ x ]1(1),x2(1),…,xM(1)]Is transposed, [ x ]1(2),x2(2),…,xM(2)]TIs [ x ]1(2),x2(2),…,xM(2)]Is transposed, [ x ]1(N),x2(N),…,xM(N)]TIs [ x ]1(N),x2(N),…,xM(N)]Transposing;
step three: according to XcurCalculate the covariance matrix, note
Then to
Performing characteristic decomposition to obtain
The maximum eigenvalue, the next maximum eigenvalue, the minimum eigenvalue and the next minimum eigenvalue of (A) are correspondingly recorded as (lambda)1、λ2、λ3And λ4(ii) a Then according to lambda1And λ3Calculating
Is noted as v1，
And according to λ2And λ4Calculating
Second inverse eigenvalue of (v)2，
Wherein, Xcur HIs XcurThe conjugate transpose of (1);
step four: according to v1And v2Constructing test statistic, and recording as T, T ═ v1+v2(ii) a And calculating a decision threshold, and recording as gamma, gamma is 2I-1(1-Pf) (ii) a Wherein, I-1() Inverse function, P, representing a regularized incomplete beta functionfRepresenting a target false alarm probability;
step five: the spectrum sensing is realized by comparing the T with the gamma, if the T is less than the gamma, the authorized user signal is judged to exist in the current sensing time slot; if T ≧ gamma, it is determined that there is no authorized user signal within the current one of the sensing timeslots.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810292741.XA CN108322277B (en) | 2018-04-04 | 2018-04-04 | Frequency spectrum sensing method based on inverse eigenvalue of covariance matrix |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810292741.XA CN108322277B (en) | 2018-04-04 | 2018-04-04 | Frequency spectrum sensing method based on inverse eigenvalue of covariance matrix |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN108322277A CN108322277A (en) | 2018-07-24 |
| CN108322277B true CN108322277B (en) | 2021-01-12 |
Family
ID=62895949
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810292741.XA Active CN108322277B (en) | 2018-04-04 | 2018-04-04 | Frequency spectrum sensing method based on inverse eigenvalue of covariance matrix |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108322277B (en) |
Families Citing this family (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109194417A (en) * | 2018-09-12 | 2019-01-11 | 宁波大学 | A kind of covariance matrix frequency spectrum sensing method |
| CN109120361B (en) * | 2018-09-12 | 2021-05-14 | 宁波大学 | Covariance matrix frequency spectrum sensing method using past sensing time slot data |
| CN109286937B (en) * | 2018-09-12 | 2023-03-24 | 宁波大学 | Covariance matrix spectrum sensing method for estimating noise power by using small eigenvalue |
| CN109150342A (en) * | 2018-09-12 | 2019-01-04 | 宁波大学 | A kind of maximum eigenvalue frequency spectrum sensing method |
| CN111864347B (en) * | 2020-06-24 | 2021-04-13 | 宁波大学 | Polarization dynamic matching method of VICTS antenna |
| CN112615652B (en) * | 2020-12-17 | 2022-06-10 | 吉首大学 | Semi-blind spectrum sensing method based on eigenvalue test and suitable for high-order MIMO |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103297160A (en) * | 2013-05-27 | 2013-09-11 | 东南大学 | Spectrum sensing method and spectrum sensing device for goodness-of-fit test based on normalized eigenvalues |
| CN103684626A (en) * | 2012-09-20 | 2014-03-26 | 中兴通讯股份有限公司 | Multi user cooperative frequency spectrum sensing data fusion method and device |
| CN103973382A (en) * | 2014-05-19 | 2014-08-06 | 西安电子科技大学 | Frequency spectrum detecting method based on limited random matrix |
| US9379924B2 (en) * | 2013-12-20 | 2016-06-28 | King Fahd University Of Petroleum And Minerals | Cognitive radio spectrum sensing with improved edge detection of frequency bands |
-
2018
- 2018-04-04 CN CN201810292741.XA patent/CN108322277B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103684626A (en) * | 2012-09-20 | 2014-03-26 | 中兴通讯股份有限公司 | Multi user cooperative frequency spectrum sensing data fusion method and device |
| CN103297160A (en) * | 2013-05-27 | 2013-09-11 | 东南大学 | Spectrum sensing method and spectrum sensing device for goodness-of-fit test based on normalized eigenvalues |
| US9379924B2 (en) * | 2013-12-20 | 2016-06-28 | King Fahd University Of Petroleum And Minerals | Cognitive radio spectrum sensing with improved edge detection of frequency bands |
| CN103973382A (en) * | 2014-05-19 | 2014-08-06 | 西安电子科技大学 | Frequency spectrum detecting method based on limited random matrix |
Non-Patent Citations (2)
| Title |
|---|
| Antieigenvalues provideaboundonrealizedsignaltonoiseratio;Ravindra Khattree;《JOURNAL OF STATISTICAL PLANNING AND INFERENCE》;20101030;第140卷(第10期);摘要，第2节 * |
| 认知无线电中的本地频谱感知;陈娟;《中国优秀硕士学位论文全文数据库信息科技辑》;20170315(第03期);第3.2.1节，第3.2.3节，第4.2节 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN108322277A (en) | 2018-07-24 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108322277B (en) | Frequency spectrum sensing method based on inverse eigenvalue of covariance matrix | |
| Ureten et al. | Wireless security through RF fingerprinting | |
| CN107682103B (en) | Double-feature spectrum sensing method based on maximum feature value and principal feature vector | |
| CN103141067B (en) | Frequency band recognition methods, device | |
| CN110932807B (en) | Spectrum sensing method and system for MIMO (multiple input multiple output) system under non-Gaussian noise | |
| CN106713190B (en) | MIMO transmitting antenna number blind estimation calculation method based on random matrix theory and characteristic threshold estimation | |
| Li et al. | A two-step spectrum sensing scheme for cognitive radio networks | |
| Huynh-The et al. | MIMO-OFDM modulation classification using three-dimensional convolutional network | |
| CN110932806A (en) | Multi-antenna spectrum sensing method under alpha stable noise fading channel | |
| CN107770778B (en) | Blind cooperative spectrum sensing method based on soft fusion strategy | |
| WO2021012859A1 (en) | Spectrum sensing method based on symmetric peaks of cyclic autocorrelation function of modulation signal | |
| Zhang et al. | Riemannian mean shift-based data fusion scheme for multi-antenna cooperative spectrum sensing | |
| CN109219055B (en) | Main user duty ratio estimation method | |
| Jin et al. | Spectrum sensing using multiple large eigenvalues and its performance analysis | |
| CN108400826B (en) | Frequency spectrum sensing method based on circulation matrix eigenvalue | |
| CN108900268B (en) | Maximum eigenvalue frequency spectrum sensing method for estimating noise power by using small eigenvalue | |
| CN109600181B (en) | Spectrum sensing method for multiple antennas | |
| CN113556157B (en) | Method and system for estimating number of transmitting antennas of MIMO system under non-Gaussian interference | |
| CN109286937B (en) | Covariance matrix spectrum sensing method for estimating noise power by using small eigenvalue | |
| CN109067483B (en) | Maximum eigenvalue frequency spectrum sensing method using past sensing time slot data | |
| CN114727286A (en) | Non-threshold multi-attribute physical layer authentication method and related equipment | |
| CN112073135B (en) | Centralized spectrum sensing method, system, storage medium and communication equipment | |
| CN107196720B (en) | OFDM signal frequency spectrum sensing method under condition of time synchronization and no frequency offset | |
| CN107222870B (en) | Ofdm signal frequency spectrum sensing method asynchronous for the time and being not present under offset frequency situation | |
| CN107623553B (en) | Robust cooperative spectrum sensing method based on reliable evaluation standard |
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 |