CN103684626A - Multi user cooperative frequency spectrum sensing data fusion method and device - Google Patents

Abstract

The invention discloses a multi user cooperative frequency spectrum sensing data fusion method and device. The method comprises the steps that a data fusion center receives local frequency spectrum sensing data from a number of cognitive radio users; the local frequency spectrum sensing data comprise local frequency spectrum sensing statistics of the cognitive radio users; the local frequency spectrum sensing statistics are based on generalized likelihood ratio and are the logarithm of the radio of arithmetic mean to geometric mean, wherein the arithmetic mean and the geometric mean are the characteristic values of a sample covariance matrix of signals received by the cognitive radio users; and the data fusion center carries out data fusion on the local frequency spectrum sensing statistics of a number of cognitive radio users according to the local frequency spectrum sensing data. According to the invention, the basis is the generalized likelihood ratio, thus the application performance is not affected by noise uncertainty; and whether the impact of the noise uncertainty is considered or not, in the situation of large reliability difference of local frequency spectrum sensing, a better frequency spectrum sensing performance can be acquired compared with energy detection based on equal gain combination.

Description

The data fusion method of multi-user collaborative frequency spectrum perception and device

Technical field

The present invention relates to the communications field, in particular to a kind of data fusion method and device of multi-user collaborative frequency spectrum perception.

Background technology

Cognitive radio (Cognitive Radio, referred to as CR) technology puts forward in order to solve the rare problem of frequency spectrum resource facing at present, and frequency spectrum perception algorithm is one of key technology of cognitive radio.For fear of cognitive radio system, authorized user is produced to harmful interference, require frequency spectrum perception algorithm can under low signal-to-noise ratio, detect reliably the signal of authorized user.

Due to the fading characteristic of wireless channel, only rely on alone family frequency spectrum perception algorithm to detect primary user (Primary User, referred to as PU) to authorizing the service condition of frequency spectrum, reliability is not high.Research shows, multi-user collaborative frequency spectrum perception can obviously improve authorizing the perceptual performance of frequency spectrum.

Existing multi-user collaborative frequency spectrum perception data fusion scheme mostly is the multi-user collaborative frequency spectrum perception based on energy detection algorithm (Energy Detection, referred to as ED).The cooperation spectrum realized perception data integration program based on energy detection algorithm has following two kinds: a kind of is equal gain combining (Equal Gain Combination, referred to as EGC), another kind of for selecting maximum normalized energy (Maximum Normalized Energy, referred to as MNE) to merge.

But the performance of multi-user collaborative energy measuring is very easily subject to the impact of noise power estimation error, noise is uncertain.

Summary of the invention

The invention provides a kind of data fusion method and device of multi-user collaborative frequency spectrum perception, at least to solve in correlation technique, multi-user collaborative frequency spectrum perception realizes based on energy detection algorithm, has noise uncertain problem.

According to an aspect of the present invention, a kind of data fusion method of multi-user collaborative frequency spectrum perception is provided, comprise: data fusion center receives from a plurality of cognitive radio users (Secondary User, referred to as SU) local frequency spectrum perception data, wherein, local frequency spectrum perception data comprise: the local frequency spectrum perception statistic of cognitive radio users, local frequency spectrum perception statistic, based on Generalized Likelihood Ratio, is the arithmetic average of characteristic value of sample covariance matrix and the logarithm of the ratio of geometric average that cognitive radio users receives signal; Data fusion center is carried out data fusion according to the local frequency spectrum perception data of a plurality of cognitive radio users to the local frequency spectrum perception statistic of a plurality of cognitive radio users.

Preferably, before data fusion center receives the local frequency spectrum perception data from a plurality of cognitive radio users, said method also comprises: a plurality of cognitive radio users are carried out local frequency spectrum perception in perception time slot, calculates local frequency spectrum perception statistic.

Preferably, calculating local frequency spectrum perception statistic comprises: cognitive radio users is sampled to the signal receiving; Cognitive radio users is calculated local sample covariance matrix according to sampled result; Cognitive radio users is carried out Eigenvalues Decomposition to local sample covariance matrix, obtains local sample covariance matrix characteristic of correspondence value; The logarithm of the arithmetic average of cognitive radio users computation of characteristic values and the ratio of geometric average, as the local frequency spectrum perception statistic of cognitive radio users.

Preferably, in a plurality of cognitive radio users, in perception time slot, carry out local frequency spectrum perception, after calculating local frequency spectrum perception statistic, said method also comprises: a plurality of cognitive radio users report local frequency spectrum perception statistic separately to data fusion center by up channel.

Preferably, data fusion center is carried out data fusion according to the local frequency spectrum perception data of a plurality of cognitive radio users to the local frequency spectrum perception statistic of a plurality of cognitive radio users and is comprised: data fusion center distributes corresponding weighted factor to respectively the local frequency spectrum perception statistic of a plurality of cognitive radio users, wherein, weighted factor is according to the frequency spectrum perception capability distribution of each cognitive radio users; The linear weighted function of local frequency spectrum perception statistic is calculated at data fusion center according to weighted factor, as overall frequency spectrum perception statistic.

Preferably, the local frequency spectrum perception data that data fusion center receives also comprise: the antenna number of cognitive radio users, signal sampling sample number, weighted factor is the product of antenna number and signal sampling sample number.

Preferably, after data fusion center is carried out data fusion according to the local frequency spectrum perception data of a plurality of cognitive radio users to the local frequency spectrum perception statistic of a plurality of cognitive radio users, said method also comprises: data fusion center is according to overall frequency spectrum perception statistic and the decision threshold obtaining in advance, whether idlely adjudicates current frequency spectrum resource.

Preferably, data fusion center is according to overall frequency spectrum perception statistic and the decision threshold obtaining in advance, and whether the free time comprises to adjudicate current frequency spectrum resource: the size of the more overall frequency spectrum perception statistic in data fusion center and decision threshold; If overall frequency spectrum perception statistic is more than or equal to decision threshold, data fusion center is adjudicated current frequency spectrum resource and is taken by primary user; If overall frequency spectrum perception statistic is less than decision threshold, it is idle that current frequency spectrum resource is adjudicated at data fusion center.

Preferably, before data fusion center receives the local frequency spectrum perception data from a plurality of cognitive radio users, said method also comprises: data fusion center notifies a plurality of cognitive radio users in its coverage to participate in frequency spectrum perception by control channel.

According to a further aspect in the invention, a kind of data fusion device of multi-user collaborative frequency spectrum perception is provided, be applied to data fusion center, comprise: receiver module, for receiving the local frequency spectrum perception data from a plurality of cognitive radio users, wherein, local frequency spectrum perception data comprise: the local frequency spectrum perception statistic of cognitive radio users, local frequency spectrum perception statistic, based on Generalized Likelihood Ratio, is the arithmetic average of characteristic value of sample covariance matrix and the logarithm of the ratio of geometric average that cognitive radio users receives signal; Data fusion module, for carrying out data fusion according to the local frequency spectrum perception data of a plurality of cognitive radio users to the local frequency spectrum perception statistic of a plurality of cognitive radio users.

Preferably, above-mentioned data fusion module comprises: allocation units, for distributing corresponding weighted factor to respectively the local frequency spectrum perception statistic of a plurality of cognitive radio users, wherein, weighted factor is according to the frequency spectrum perception capability distribution of each cognitive radio users; Computing unit, for calculate the linear weighted function of local frequency spectrum perception statistic according to weighted factor, as overall frequency spectrum perception statistic.

Preferably, the local frequency spectrum perception data that data fusion center receives also comprise: the antenna number of cognitive radio users, signal sampling sample number, weighted factor is the product of antenna number and signal sampling sample number.

By the present invention, the local frequency spectrum perception statistic based on Generalized Likelihood Ratio that a plurality of cognitive radio users of data fusion center reception participation frequency spectrum perception report, and it is carried out to data fusion.Due to based on Generalized Likelihood Ratio, application performance is not subject to the probabilistic impact of noise, has solved the probabilistic problem of noise; And, no matter whether consider the probabilistic impact of noise, in the situation that the local frequency spectrum perception reliability difference of cognitive radio users is larger, also can obtain more excellent frequency spectrum perception performance than the energy measuring based on equal gain combining.

Accompanying drawing explanation

Accompanying drawing described herein is used to provide a further understanding of the present invention, forms the application's a part, and schematic description and description of the present invention is used for explaining the present invention, does not form inappropriate limitation of the present invention.In the accompanying drawings:

Fig. 1 is according to the flow chart of the data fusion method of the multi-user collaborative frequency spectrum perception of the embodiment of the present invention;

Fig. 2 is the flow chart of the data fusion method of multi-user collaborative frequency spectrum perception according to the preferred embodiment of the invention;

Fig. 3 is the schematic diagram that calculates according to the preferred embodiment of the invention local frequency spectrum perception statistic;

Fig. 4 be multi-user collaborative frequency spectrum perception according to the preferred embodiment of the invention data fusion method realize schematic diagram;

Fig. 5 is the simulation result comparison diagram of the data fusion method in typical according to the preferred embodiment of the invention heterogeneous network situation;

Fig. 6 is according to the structured flowchart of the data fusion device of the multi-user collaborative frequency spectrum perception of the embodiment of the present invention;

Fig. 7 is the structured flowchart one of the data fusion device of multi-user collaborative frequency spectrum perception according to the preferred embodiment of the invention;

Fig. 8 is the structured flowchart two of the data fusion device of multi-user collaborative frequency spectrum perception according to the preferred embodiment of the invention.

Embodiment

It should be noted that, in the situation that not conflicting, embodiment and the feature in embodiment in the application can combine mutually.Describe below with reference to the accompanying drawings and in conjunction with the embodiments the present invention in detail.

The embodiment of the present invention provides a kind of data fusion method of multi-user collaborative frequency spectrum perception, and Fig. 1 is according to the flow chart of the data fusion method of the multi-user collaborative frequency spectrum perception of the embodiment of the present invention, as shown in Figure 1, comprises that following step S102 is to step S104.

Step S102, data fusion center receives the local frequency spectrum perception data from a plurality of cognitive radio users, wherein, local frequency spectrum perception data comprise: the local frequency spectrum perception statistic of cognitive radio users, local frequency spectrum perception statistic, based on Generalized Likelihood Ratio, is the arithmetic average of characteristic value of sample covariance matrix and the logarithm of the ratio of geometric average that cognitive radio users receives signal.

Step S104, data fusion center is carried out data fusion according to the local frequency spectrum perception data of a plurality of cognitive radio users to the local frequency spectrum perception statistic of a plurality of cognitive radio users.

In correlation technique, multi-user collaborative frequency spectrum perception realizes based on energy detection algorithm, exists noise uncertain.In the embodiment of the present invention, data fusion center receives a plurality of cognitive radio users (the Secondary User that participates in frequency spectrum perception, referred to as SU) report based on Generalized Likelihood Ratio (Generalized Likelihood Ratio Test, referred to as GLRT) local frequency spectrum perception statistic, and it is carried out to data fusion.Due to based on Generalized Likelihood Ratio, application performance is not subject to the probabilistic impact of noise, has solved the probabilistic problem of noise; And, no matter whether consider the probabilistic impact of noise, in the situation that the local frequency spectrum perception reliability difference of cognitive radio users is larger, also can obtain more excellent frequency spectrum perception performance than the energy measuring based on equal gain combining.

Because the frequency spectrum perception algorithm based on Generalized Likelihood Ratio is a kind of total blindness's detection algorithm, cognitive radio users is without any need for prior information, is not also subject to that noise is probabilistic to be affected simultaneously.Therefore, the multi-user collaborative perception based on Generalized Likelihood Ratio will have better application prospect.

Before step S102, a plurality of cognitive radio users that participate in frequency spectrum perception are carried out local frequency spectrum perception in perception time slot, calculate local frequency spectrum perception statistic separately, can realize by following steps: cognitive radio users is sampled to the signal receiving; Cognitive radio users is calculated local sample covariance matrix according to sampled result; Cognitive radio users is carried out Eigenvalues Decomposition to local sample covariance matrix, obtains local sample covariance matrix characteristic of correspondence value; The logarithm of the arithmetic average of cognitive radio users computation of characteristic values and the ratio of geometric average, as the local frequency spectrum perception statistic of cognitive radio users.In this preferred embodiment, calculate the local frequency spectrum perception statistic based on Generalized Likelihood Ratio, thereby can not be subject to that noise is probabilistic to be affected.

After a plurality of cognitive radio users that participate in frequency spectrum perception are calculated local frequency spectrum perception statistic, said method also comprises: a plurality of cognitive radio users report local frequency spectrum perception statistic separately to data fusion center by up channel.

Step S104 comprises: data fusion center distributes corresponding weighted factor to respectively the local frequency spectrum perception statistic of a plurality of cognitive radio users, and wherein, weighted factor is according to the frequency spectrum perception capability distribution of each cognitive radio users; The linear weighted function of local frequency spectrum perception statistic is calculated at data fusion center according to weighted factor, as overall frequency spectrum perception statistic.In this preferred embodiment, according to the corresponding weighted factor of frequency spectrum perception capability distribution of cognitive radio users, the result that data fusion obtains is more reasonable.

Preferably, the local frequency spectrum perception data that data fusion center receives also comprise: the antenna number of cognitive radio users, signal sampling sample number, weighted factor is the product of antenna number and signal sampling sample number.

Preferably, after step S104, said method also comprises: data fusion center is according to overall frequency spectrum perception statistic and the decision threshold obtaining in advance, whether idlely adjudicates current frequency spectrum resource.Thereby whether all cognitive radio users that can adjudicate in the coverage of data fusion center can utilize current frequency spectrum resource.

Preferably, data fusion center is according to overall frequency spectrum perception statistic and the decision threshold obtaining in advance, and whether the free time can realize by following steps to adjudicate current frequency spectrum resource: the size of the more overall frequency spectrum perception statistic in data fusion center and decision threshold; If overall frequency spectrum perception statistic is more than or equal to decision threshold, data fusion center is adjudicated current frequency spectrum resource and is taken by primary user PU, that is, all cognitive radio users in the coverage of data fusion center can not be utilized current frequency spectrum resource; If overall frequency spectrum perception statistic is less than decision threshold, it is idle that current frequency spectrum resource is adjudicated at data fusion center, that is, all cognitive radio users in the coverage of data fusion center can be utilized current frequency spectrum resource.

Before step S102, said method also comprises: data fusion center notifies a plurality of cognitive radio users in its coverage to participate in frequency spectrum perception by control channel.In this preferred embodiment, by data fusion center, notify a plurality of cognitive radio users in its coverage to participate in frequency spectrum perception, only need the cognitive radio users of notified participation frequency spectrum perception to calculate local frequency spectrum perception statistic, can avoid unnecessary operation.

Fig. 2 is the flow chart of the data fusion method of multi-user collaborative frequency spectrum perception according to the preferred embodiment of the invention, and in this preferred embodiment, the data fusion method of above-mentioned multi-user collaborative frequency spectrum perception can realize by following steps:

Step S202, data fusion center (Fusion Center, referred to as FC) informs that SU participates in frequency spectrum perception.

Step S204, SU calculates local frequency spectrum perception statistic.

Step S206, SU reports local frequency spectrum perception data to FC.

Step S208, FC calculates overall frequency spectrum perception statistic.

Step S210, frequency spectrum state is authorized in FC judgement.

Step S212, the current mandate frequency spectrum of SU state is informed in FC broadcast.

By above description, can be found out, the embodiment of the present invention provides the optimal data fusion method of the multi-user collaborative frequency spectrum perception based on Generalized Likelihood Ratio in heterogeneous network, comprise that local cognitive radio users generates the data of reporting to data fusion center, and data merging is carried out at data fusion center.Particularly, frequency spectrum perception is in the cycle, and each participates in the cognitive radio users node of frequency spectrum perception and samples by local data, correspondingly calculates the local frequency spectrum perception statistic based on Generalized Likelihood Ratio.Then, cognitive radio users node will send data fusion node to based on the local frequency spectrum perception statistic of Generalized Likelihood Ratio, local antenna number and signal sampling sample number by control channel.Finally, the data that data fusion node reports according to cognitive radio users, for each cognitive radio users reported data is distributed the different linear weighted function factors, the local frequency spectrum perception statistic of linear combining, and make the court verdict of this frequency spectrum perception.

To the preferred embodiment shown in Fig. 2 be described in detail below.

Step 1, sends frequency spectrum perception signaling

Data fusion center notifies K the cognitive radio users (SU) in its coverage to participate in frequency spectrum perception by control channel (Cognitive Pilot Channel, referred to as CPC).

Step 2, calculates local frequency spectrum perception statistic

The cognitive radio users of notified participation frequency spectrum perception (SU) k=1 in step 1 ..., K carries out local frequency spectrum perception in perception time slot, calculates local frequency spectrum perception statistic T _k, in this step, calculate local frequency spectrum perception statistic T _kprocess as shown in Figure 3, be described in detail below.

Each cognitive radio users (SU) that participates in frequency spectrum perception is sampled to received signal, and for example, k cognitive radio users (SU) passed through M constantly at n _kroot antenna is sampled to received signal, obtains M _k* 1 sample of signal y _k(n);

If N _kfor participating in the signal sampling sample number of k the cognitive radio users (SU) of frequency spectrum perception, the local sample covariance matrix that k cognitive radio users (SU) calculated is

wherein, H represents conjugate transpose;

K the cognitive radio users (SU) that participates in frequency spectrum perception carried out Eigenvalues Decomposition to local sample covariance matrix, obtains local sample covariance matrix characteristic of correspondence value λ _{k, 1}...,

The characteristic value of the local sample covariance matrix that k the cognitive radio users (SU) of participation frequency spectrum perception obtains according to above-mentioned steps, logarithm (the LogarithmArithmetic-to-Geometric of the arithmetic average of computation of characteristic values and the ratio of geometric average, referred to as LAGM), form the local frequency spectrum perception statistic T of k cognitive radio users (SU) _k:

$T_k=\ln \left\{\frac{\frac{1}{M_k}\underset{i=1}{\overset{M_k}{\mathrm{\Σ}}}{\mathrm{\λ}}_{k,i}}{{\left(\underset{i=1}{\overset{M_k}{\mathrm{\Π}}}{\mathrm{\λ}}_{k,i}\right)}^{\frac{1}{M_k}}}\right\}.$

Step 3: report local frequency spectrum perception data

Each cognitive radio users (SU) that participates in frequency spectrum perception reports local frequency spectrum perception statistic, number of antennas and signal sampling sample number separately to data fusion center by up channel.As, k cognitive radio users (SU) will report local frequency spectrum perception statistic T _k, number of antennas M _kwith signal sampling number of samples N _k.

Step 4: frequency spectrum perception data fusion

Data fusion is carried out to the local frequency spectrum perception statistic of cognitive radio users (SU) in data fusion center, according to Generalized Likelihood Ratio etection theory, the in the situation that of fixing false alarm probability, for detection probability is maximized, the corresponding weighted factor ω of local frequency spectrum perception statistic that data fusion central dispense reports to Different Cognitive radio subscriber (SU) _k, obtain overall frequency spectrum perception statistic T _gLRT:

it is the linear weighted function that overall frequency spectrum perception statistic is local frequency spectrum perception statistic.

The weighted factor of distributing to each local frequency spectrum perception statistic is the product of local cognitive radio users (SU) number of antennas and signal sampling number of samples, is the local frequency spectrum perception statistic that the cognitive radio users of different spectral perception reports and distributes different weighted factors.For example, to k cognitive radio users (SU), the merging weighted factor of its correspondence is the product of its antenna number and sample number, that is: ω _k=M _kn _k.

Step 5: judgement

Data fusion center is by the overall frequency spectrum perception statistic T obtaining _gLRTwith the decision threshold γ comparison setting in advance.

When detecting overall frequency spectrum perception statistic, be more than or equal to decision threshold, i.e. T _gLRTduring>=γ, adjudicate current frequency spectrum resource and taken by primary user, all cognitive radio users (SU) in this data fusion center coverage can not be utilized this frequency spectrum resource;

When detecting overall frequency spectrum perception statistic, be less than decision threshold, i.e. T _gLRTduring < γ, adjudicate the current frequency spectrum resource free time, all cognitive radio users (SU) in this data fusion center coverage can be utilized this frequency spectrum resource.

It should be noted that, decision threshold γ is relevant with the false alarm probability of cognitive radio users (SU) number, each local cognitive radio users (SU) number of antennas, signal sampling sample number and the system requirements of participation frequency spectrum perception.Those skilled in the art can obtain circular by existing open source literature, thereby obtain the value of decision threshold γ.Because the computational methods of decision threshold γ belong to prior art, and non-summary of the invention of the present invention.Therefore, the present invention no longer launches to describe for the circular of decision threshold γ.Can by DATA REASONING, build the decision threshold table under typical application scenarios in advance in actual applications, during frequency spectrum perception, data fusion node (FC) can be tabled look-up and be obtained corresponding decision threshold.

Fig. 4 be multi-user collaborative frequency spectrum perception according to the preferred embodiment of the invention data fusion method realize schematic diagram, wherein,

with

respectively null hypothesis and the alternative hypothesis in hypothesis testing, wherein null hypothesis is that the signal that cognitive radio users receives is white Gaussian noise, and alternative hypothesis is that signal that cognitive radio users receives is that primary user after channel fading transmits and white Gaussian noise sum.

expression proposes null hypothesis and alternative hypothesis according to actual conditions.For being elaborated above implementation procedure, repeat no more herein.

In order to make technical scheme of the present invention and implementation method clearer, below in conjunction with preferred embodiment, its implementation procedure is described in detail, in following preferred embodiment one to four, cognitive radio users is all SU(Secondary User).

Preferred embodiment one

Step 1, sends frequency spectrum perception signaling.

In this step, data fusion center notifies 4 cognitive radio users in its coverage to participate in frequency spectrum perception (being K=4) by control channel, and it is M that the number of antennas of each cognitive radio users is 4( ₁=4, M ₂=4, M ₃=4, M ₄=4), local signal sample number corresponding to each cognitive radio users is respectively 2160,1080,540,270(is N ₁=2160, N ₂=1080, N ₃=540, N ₄=270).

Step 2, calculates local frequency spectrum perception statistic.

In this step, 4 cognitive radio users that participate in frequency spectrum perception are carried out local frequency spectrum perception in perception time slot, calculate local frequency spectrum perception statistic T separately ₁, T ₂, T ₃, T ₄.

First, 4 cognitive radio users that participate in frequency spectrum perception are sampled to received signal, obtain 4 sample of signal vector y that cognitive radio users is corresponding ₁(1) ..., y ₁(2160), y ₂(1) ..., y ₂(1080), y ₃(1) ..., y ₃(540), y ₄(1) ..., y ₄(270);

Secondly, each cognitive radio users that participates in frequency spectrum perception is according to the local sample covariance matrix of signal sampling sample calculation ${\hat{R}}_{y,1}=\frac{1}{2160}\underset{n=1}{\overset{2160}{\mathrm{\&Sigma;}}}{y}_1\left(n\right)y_1^H\left(n\right),$ ${\hat{R}}_{y,2}=\frac{1}{1080}\underset{n=1}{\overset{1080}{\mathrm{\&Sigma;}}}{y}_2\left(n\right)y_2^H\left(n\right),$ ${\hat{R}}_{y,3}=\frac{1}{540}\underset{n=1}{\overset{540}{\mathrm{\&Sigma;}}}{y}_3\left(n\right)y_3^H\left(n\right)$ With ${\hat{R}}_{y,4}=\frac{1}{270}\underset{n=1}{\overset{270}{\mathrm{\&Sigma;}}}{y}_4\left(n\right)y_4^H\left(n\right);$

Again, each cognitive radio users that participates in frequency spectrum perception is carried out Eigenvalues Decomposition to sample covariance matrix, obtains 4 cognitive radio users sample covariance matrix characteristic of correspondence value λ separately _1,1..., λ _{isosorbide-5-Nitrae}, λ _2,1..., λ _2,4, λ _3,1..., λ _3,4and λ _4,1..., λ _4,4;

Finally, calculate the local frequency spectrum perception statistic T of 4 local cognitive radio users based on characteristic value ₁, T ₂, T ₃and T ₄:

$T_1=\ln \left\{\frac{\frac{1}{4}\underset{i=1}{\overset{4}{\mathrm{\&Sigma;}}}{\mathrm{\&lambda;}}_{1,i}}{{\left(\underset{i=1}{\overset{4}{\mathrm{\&Pi;}}}{\mathrm{\&lambda;}}_{1,i}\right)}^{1/4}}\right\}$

$T_2=\ln \left\{\frac{\frac{1}{4}\underset{i=1}{\overset{4}{\mathrm{\&Sigma;}}}{\mathrm{\&lambda;}}_{2,i}}{{\left(\underset{i=1}{\overset{4}{\mathrm{\&Pi;}}}{\mathrm{\&lambda;}}_{2,i}\right)}^{1/4}}\right\}$

$T_3=\ln \left\{\frac{\frac{1}{4}\underset{i=1}{\overset{4}{\mathrm{\&Sigma;}}}{\mathrm{\&lambda;}}_{3,i}}{{\left(\underset{i=1}{\overset{4}{\mathrm{\&Pi;}}}{\mathrm{\&lambda;}}_{3,i}\right)}^{1/4}}\right\}$

$T_4=\ln \left\{\frac{\frac{1}{4}\underset{i=1}{\overset{4}{\mathrm{\&Sigma;}}}{\mathrm{\&lambda;}}_{4,i}}{{\left(\underset{i=1}{\overset{4}{\mathrm{\&Pi;}}}{\mathrm{\&lambda;}}_{4,i}\right)}^{1/4}}\right\}$

Step 3: report local frequency spectrum perception data.

In this step, 4 cognitive radio users that participate in frequency spectrum perception by up channel by the local frequency spectrum perception statistic T calculating in each comfortable previous step ₁, T ₂, T ₃and T ₄, number of antennas 4, and signal sampling number of samples 2160,1080,540 and 270 sends data fusion center to.

Step 4: frequency spectrum perception data fusion.

In this step, it is 8640,4320,2160 and 1080 that data fusion center calculation obtains 4 the linear weighted function factors corresponding to cognitive radio users, then the local frequency spectrum perception statistic T to 4 cognitive radio users ₁, T ₂, T ₃and T ₄carry out linear weighted function merging, obtain overall frequency spectrum perception statistic T _cLRT=8640T ₁+ 4320T ₂+ 2160T ₃+ 1080T ₄.

Step 5: judgement.

In this step, data fusion center is by the overall frequency spectrum perception statistic T obtaining _gLRTwith decision threshold γ comparison, in this preferred embodiment, decision threshold γ=37.28;

If overall frequency spectrum perception statistic is more than or equal to decision threshold, i.e. T _gLRT>=γ, adjudicates current frequency spectrum resource and is taken by primary user, and all cognitive radio users in this data fusion center coverage can not be utilized this frequency spectrum resource;

If overall frequency spectrum perception statistic is less than decision threshold, i.e. T _gLRT< γ, adjudicates the current frequency spectrum resource free time, and all cognitive radio users in this data fusion center coverage can be utilized this frequency spectrum resource.

Preferred embodiment two

Step 1, sends frequency spectrum perception signaling.

In this step, data fusion center notifies 4 cognitive radio users in its coverage to participate in frequency spectrum perception (being K=4) by control channel, and the number of antennas of each cognitive radio users is respectively 8,6,4,2(is M ₁=8, M ₂=6, M ₃=4, M ₄=2), to be 270(be N to local signal sample number corresponding to each cognitive radio users ₁=270, N ₂=270, N ₃=270, N ₄=270).

Step 2, calculates local frequency spectrum perception statistic.

In this step, 4 cognitive radio users that participate in frequency spectrum perception are carried out local frequency spectrum perception in perception time slot, calculate local frequency spectrum perception statistic T separately ₁, T ₂, T ₃, T ₄.

First, 4 cognitive radio users that participate in frequency spectrum perception are sampled to received signal, obtain 4 sample of signal vector y that cognitive radio users is corresponding ₁(1) ..., y ₁(270), y ₂(1) ..., y ₂(270), y ₃(1) ..., y ₃(270), y ₄(1) ..., y ₄(270);

Secondly, each cognitive radio users that participates in frequency spectrum perception is according to the local sample covariance matrix of signal sampling sample calculation ${\hat{R}}_{y,1}=\frac{1}{270}\underset{n=1}{\overset{270}{\mathrm{\&Sigma;}}}{y}_1\left(n\right)y_1^H\left(n\right),$ ${\hat{R}}_{y,2}=\frac{1}{270}\underset{n=1}{\overset{270}{\mathrm{\&Sigma;}}}{y}_2\left(n\right)y_2^H\left(n\right),$ ${\hat{R}}_{y,3}=\frac{1}{270}\underset{n=1}{\overset{270}{\mathrm{\&Sigma;}}}{y}_3\left(n\right)y_3^H\left(n\right)$ With ${\hat{R}}_{y,4}=\frac{1}{270}\underset{n=1}{\overset{270}{\mathrm{\&Sigma;}}}{y}_4\left(n\right)y_4^H\left(n\right);$

Again, each cognitive radio users that participates in frequency spectrum perception is carried out Eigenvalues Decomposition to sample covariance matrix, obtains 4 cognitive radio users sample covariance matrix characteristic of correspondence value λ separately _1,1..., λ _1,8, λ _2,1..., λ _2,6, λ _3,1..., λ _3,4and λ _4,1..., λ _4,2;

Finally, calculate the local frequency spectrum perception statistic T of 4 local cognitive radio users based on characteristic value ₁, T ₂, T ₃and T ₄:

$T_1=\ln \left\{\frac{\frac{1}{8}\underset{i=1}{\overset{8}{\mathrm{\&Sigma;}}}{\mathrm{\&lambda;}}_{1,i}}{{\left(\underset{i=1}{\overset{8}{\mathrm{\&Pi;}}}{\mathrm{\&lambda;}}_{1,i}\right)}^{1/8}}\right\}$

$T_2=\ln \left\{\frac{\frac{1}{6}\underset{i=1}{\overset{6}{\mathrm{\&Sigma;}}}{\mathrm{\&lambda;}}_{2,i}}{{\left(\underset{i=1}{\overset{6}{\mathrm{\&Pi;}}}{\mathrm{\&lambda;}}_{2,i}\right)}^{1/6}}\right\}$

$T_3=\ln \left\{\frac{\frac{1}{4}\underset{i=1}{\overset{4}{\mathrm{\&Sigma;}}}{\mathrm{\&lambda;}}_{3,i}}{{\left(\underset{i=1}{\overset{4}{\mathrm{\&Pi;}}}{\mathrm{\&lambda;}}_{3,i}\right)}^{1/4}}\right\}$

$T_4=\ln \left\{\frac{\frac{1}{2}\underset{i=1}{\overset{2}{\mathrm{\&Sigma;}}}{\mathrm{\&lambda;}}_{4,i}}{{\left(\underset{i=1}{\overset{2}{\mathrm{\&Pi;}}}{\mathrm{\&lambda;}}_{4,i}\right)}^{1/2}}\right\}$

Step 3: report local frequency spectrum perception data;

In this step, 4 cognitive radio users that participate in frequency spectrum perception by up channel by the local frequency spectrum perception statistic T calculating in each comfortable previous step ₁, T ₂, T ₃and T ₄, number of antennas 8,6,4 and 2, and signal sampling number of samples 270 sends data fusion center to.

Step 4: frequency spectrum perception data fusion.

In this step, it is 2160,1620,1080 and 540 that data fusion center calculation obtains 4 the linear weighted function factors corresponding to cognitive radio users, then the local frequency spectrum perception statistic T to 4 cognitive radio users ₁, T ₂, T ₃and T ₄carry out linear weighted function merging, obtain overall frequency spectrum perception statistic T _gLRT=2160T ₁+ 1620T ₂+ 1080T ₃+ 540T ₄.

Step 5: judgement.

In this step, data fusion center is by the overall frequency spectrum perception statistic T obtaining _gLRTwith decision threshold γ comparison, in this preferred embodiment, decision threshold γ=68.44;

If overall frequency spectrum perception statistic is more than or equal to decision threshold, i.e. T _gLRT>=γ, adjudicates current frequency spectrum resource and is taken by primary user, and all cognitive radio users in this data fusion center coverage can not be utilized this frequency spectrum resource;

If overall frequency spectrum perception statistic is less than decision threshold, i.e. T _gLRT< γ, adjudicates the current frequency spectrum resource free time, and all cognitive radio users in this data fusion center coverage can be utilized this frequency spectrum resource.

Preferred embodiment three

Step 1, sends frequency spectrum perception signaling.

In this step, data fusion center notifies 4 cognitive radio users in its coverage to participate in frequency spectrum perception (being K=4) by control channel, and it is M that the number of antennas of each cognitive radio users is 4( ₁=4, M ₂=4, M ₃=4, M ₄=4), local signal sample number corresponding to each cognitive radio users is N for being 270( ₁=270, N ₂=270, N ₃=270, N ₄=270).

Step 2, calculates local frequency spectrum perception statistic.

In this step, 4 cognitive radio users that participate in frequency spectrum perception are carried out local frequency spectrum perception in perception time slot, calculate local frequency spectrum perception statistic T separately ₁, T ₂, T ₃, T ₄.

First, 4 cognitive radio users that participate in frequency spectrum perception are sampled to received signal, obtain 4 sample of signal vector y that cognitive radio users is corresponding ₁(1) ..., y ₁(270), y ₂(1) ..., y ₂(270), y ₃(1) ..., y ₃(270), y ₄(1) ..., y ₄(270);

Secondly, each cognitive radio users that participates in frequency spectrum perception is according to the local sample covariance matrix of signal sampling sample calculation ${\hat{R}}_{y,1}=\frac{1}{270}\underset{n=1}{\overset{270}{\mathrm{\&Sigma;}}}{y}_1\left(n\right)y_1^H\left(n\right),$ ${\hat{R}}_{y,2}=\frac{1}{270}\underset{n=1}{\overset{270}{\mathrm{\&Sigma;}}}{y}_2\left(n\right)y_2^H\left(n\right),$ ${\hat{R}}_{y,3}=\frac{1}{270}\underset{n=1}{\overset{270}{\mathrm{\&Sigma;}}}{y}_3\left(n\right)y_3^H\left(n\right)$ With ${\hat{R}}_{y,4}=\frac{1}{270}\underset{n=1}{\overset{270}{\mathrm{\&Sigma;}}}{y}_4\left(n\right)y_4^H\left(n\right);$

Again, each cognitive radio users that participates in frequency spectrum perception is carried out Eigenvalues Decomposition to sample covariance matrix, obtains 4 cognitive radio users sample covariance matrix characteristic of correspondence value λ separately _1,1..., λ _{isosorbide-5-Nitrae}, λ _2,1..., λ _2,4, λ _3,1..., λ _3,4and λ _4,1..., λ _4,4;

Finally, calculate the local frequency spectrum perception statistic T of 4 local cognitive radio users based on characteristic value ₁, T ₂, T ₃and T ₄:

$T_1=\ln \left\{\frac{\frac{1}{4}\underset{i=1}{\overset{4}{\mathrm{\&Sigma;}}}{\mathrm{\&lambda;}}_{1,i}}{{\left(\underset{i=1}{\overset{4}{\mathrm{\&Pi;}}}{\mathrm{\&lambda;}}_{1,i}\right)}^{1/4}}\right\}$

$T_2=\ln \left\{\frac{\frac{1}{4}\underset{i=1}{\overset{4}{\mathrm{\&Sigma;}}}{\mathrm{\&lambda;}}_{2,i}}{{\left(\underset{i=1}{\overset{4}{\mathrm{\&Pi;}}}{\mathrm{\&lambda;}}_{2,i}\right)}^{1/4}}\right\}$

$T_3=\ln \left\{\frac{\frac{1}{4}\underset{i=1}{\overset{4}{\mathrm{\&Sigma;}}}{\mathrm{\&lambda;}}_{3,i}}{{\left(\underset{i=1}{\overset{4}{\mathrm{\&Pi;}}}{\mathrm{\&lambda;}}_{3,i}\right)}^{1/4}}\right\}$

$T_4=\ln \left\{\frac{\frac{1}{4}\underset{i=1}{\overset{4}{\mathrm{\&Sigma;}}}{\mathrm{\&lambda;}}_{4,i}}{{\left(\underset{i=1}{\overset{4}{\mathrm{\&Pi;}}}{\mathrm{\&lambda;}}_{4,i}\right)}^{1/4}}\right\}$

Step 3: report local frequency spectrum perception data.

In this step, 4 cognitive radio users that participate in frequency spectrum perception by up channel by the local frequency spectrum perception statistic T calculating in each comfortable previous step ₁, T ₂, T ₃and T ₄, number of antennas 4, and signal sampling number of samples 270 sends data fusion center to.

Step 4: frequency spectrum perception data fusion.

In this step, data fusion center calculation obtains 4 the linear weighted function factors corresponding to cognitive radio users and is 1080, then the local frequency spectrum perception statistic T to 4 cognitive radio users ₁, T ₂, T ₃and T ₄carry out linear weighted function merging, obtain overall frequency spectrum perception statistic T _gLRT=1080T ₁+ 1080T ₂+ 1080T ₃+ 1080T ₄.

Step 5: judgement.

In this step, data fusion center is by the overall frequency spectrum perception statistic T obtaining _gLRTwith decision threshold γ comparison, in this preferred embodiment, decision threshold γ=37.38;

If overall frequency spectrum perception statistic is more than or equal to decision threshold, i.e. T _gLRT>=γ, adjudicates current frequency spectrum resource and is taken by primary user, and all cognitive radio users in this data fusion center coverage can not be utilized this frequency spectrum resource;

If overall frequency spectrum perception statistic is less than decision threshold, i.e. T _cLRT< γ, adjudicates the current frequency spectrum resource free time, and all cognitive radio users in this data fusion center coverage can be utilized this frequency spectrum resource.

Preferred embodiment four

Step 1, sends frequency spectrum perception signaling.

In this step, data fusion center notifies 4 cognitive radio users in its coverage to participate in frequency spectrum perception (being K=4) by control channel, and the number of antennas of each cognitive radio users is respectively 8,6,4,2(is M ₁=8, M ₂=6, M ₃=4, M ₄=2), local signal sample number corresponding to each cognitive radio users is respectively 2160,1080,540,270(is N ₁=2160, N ₂=1080, N ₃=540, N ₄=270).

Step 2, calculates local frequency spectrum perception statistic.

In this step, 4 cognitive radio users that participate in frequency spectrum perception are carried out local frequency spectrum perception in perception time slot, calculate local frequency spectrum perception statistic T separately ₁, T ₂, T ₃, T ₄.

First, 4 cognitive radio users that participate in frequency spectrum perception are sampled to received signal, obtain 4 sample of signal vector y that cognitive radio users is corresponding ₁(1) ..., y ₁(2160), y ₂(1) ..., y ₂(1080), y ₃(1) ..., y ₃(540), y ₄(1) ..., y ₄(270);

Secondly, each cognitive radio users that participates in frequency spectrum perception is according to the local sample covariance matrix of signal sampling sample calculation ${\hat{R}}_{y,1}=\frac{1}{2160}\underset{n=1}{\overset{2160}{\mathrm{\&Sigma;}}}{y}_1\left(n\right)y_1^H\left(n\right),$ ${\hat{R}}_{y,2}=\frac{1}{1080}\underset{n=1}{\overset{1080}{\mathrm{\&Sigma;}}}{y}_2\left(n\right)y_2^H\left(n\right),$ ${\hat{R}}_{y,3}=\frac{1}{540}\underset{n=1}{\overset{540}{\mathrm{\&Sigma;}}}{y}_3\left(n\right)y_3^H\left(n\right)$ With ${\hat{R}}_{y,4}=\frac{1}{270}\underset{n=1}{\overset{270}{\mathrm{\&Sigma;}}}{y}_4\left(n\right)y_4^H\left(n\right);$

Again, each cognitive radio users that participates in frequency spectrum perception is carried out Eigenvalues Decomposition to sample covariance matrix, obtains 4 cognitive radio users sample covariance matrix characteristic of correspondence value λ separately _1,1..., λ _1,8, λ _2,1..., λ _2,6, λ _3,1..., λ _3,4and λ _4,1..., λ _4,2;

Finally, calculate the local frequency spectrum perception statistic T of 4 local cognitive radio users based on characteristic value ₁, T ₂, T ₃and T ₄:

$T_1=\ln \left\{\frac{\frac{1}{8}\underset{i=1}{\overset{8}{\mathrm{\&Sigma;}}}{\mathrm{\&lambda;}}_{1,i}}{{\left(\underset{i=1}{\overset{8}{\mathrm{\&Pi;}}}{\mathrm{\&lambda;}}_{1,i}\right)}^{1/8}}\right\}$

$T_2=\ln \left\{\frac{\frac{1}{6}\underset{i=1}{\overset{6}{\mathrm{\&Sigma;}}}{\mathrm{\&lambda;}}_{2,i}}{{\left(\underset{i=1}{\overset{6}{\mathrm{\&Pi;}}}{\mathrm{\&lambda;}}_{2,i}\right)}^{1/6}}\right\}$

$T_3=\ln \left\{\frac{\frac{1}{4}\underset{i=1}{\overset{4}{\mathrm{\&Sigma;}}}{\mathrm{\&lambda;}}_{3,i}}{{\left(\underset{i=1}{\overset{4}{\mathrm{\&Pi;}}}{\mathrm{\&lambda;}}_{3,i}\right)}^{1/4}}\right\}$

$T_4=\ln \left\{\frac{\frac{1}{2}\underset{i=1}{\overset{2}{\mathrm{\&Sigma;}}}{\mathrm{\&lambda;}}_{4,i}}{{\left(\underset{i=1}{\overset{2}{\mathrm{\&Pi;}}}{\mathrm{\&lambda;}}_{4,i}\right)}^{1/2}}\right\}$

Step 3: report local frequency spectrum perception data.

In this step, 4 cognitive radio users that participate in frequency spectrum perception by up channel by the local frequency spectrum perception statistic T calculating in each comfortable previous step ₁, T ₂, T ₃and T ₄, number of antennas 8,6,4 and 2, and signal sampling number of samples 2160,1080,540 and 270 sends data fusion center to.

Step 4: frequency spectrum perception data fusion.

In this step, it is 17280,6480,2160 and 540 that data fusion center calculation obtains 4 the linear weighted function factors corresponding to cognitive radio users, then the local frequency spectrum perception statistic T to 4 cognitive radio users ₁, T ₂, T ₃and T ₄carry out linear weighted function merging, obtain overall frequency spectrum perception statistic T _gLRT=17280T ₁+ 6480T ₂+ 2160T ₃+ 540T ₄.

Step 5: judgement.

In this step, data fusion center is by the overall frequency spectrum perception statistic T obtaining _gLRTwith decision threshold γ comparison, in this preferred embodiment, decision threshold γ=68.04;

If overall frequency spectrum perception statistic is more than or equal to decision threshold, i.e. T _gLRT>=γ, adjudicates current frequency spectrum resource and is taken by primary user, and all cognitive radio users in this data fusion center coverage can not be utilized this frequency spectrum resource;

If overall frequency spectrum perception statistic is less than decision threshold, i.e. T _gLRT< γ, adjudicates the current frequency spectrum resource free time, and all cognitive radio users in this data fusion center coverage can be utilized this frequency spectrum resource.

Method described in employing above preferred embodiment four is carried out emulation testing, when primary user's number of transmit antennas is 1, adopt biphase phase shift keying (Binary Phase Shift Keying, referred to as BPSK) modulation, bit rate is 270kb/s, 4 cognitive radio users reception antenna spatial coherences are under correlation of indices model, reception antenna coefficient correlation is 0.9, local cognitive radio users signal to noise ratio (dB) difference is 6, 2,-2,-6 o'clock, simulation result as shown in Figure 5, as can be seen from the figure, compare with the energy detection method of tradition based on equal gain combining, the method of the embodiment of the present invention has better detection performance, for example, in detection probability, it is 0.9 o'clock, the embodiment of the present invention do not have noise under uncertain etc. gain energy measuring has 4dB is to gain, have under 0.5dB noise uncertainty etc. gain energy measuring have the performance gain of 15dB.

It should be noted that, in the step shown in the flow chart of accompanying drawing, can in the computer system such as one group of computer executable instructions, carry out, and, although there is shown logical order in flow process, but in some cases, can carry out shown or described step with the order being different from herein.

The embodiment of the present invention also provides a kind of data fusion device of multi-user collaborative frequency spectrum perception, is applied to data fusion center, and this device can be for realizing the data fusion method of above-mentioned multi-user collaborative frequency spectrum perception.Fig. 6 is according to the structured flowchart of the data fusion device of the multi-user collaborative frequency spectrum perception of the embodiment of the present invention, and as shown in Figure 6, this device comprises receiver module 62 and data fusion module 64.Below its structure is described in detail.

Receiver module 62, for receiving the local frequency spectrum perception data from a plurality of cognitive radio users, wherein, local frequency spectrum perception data comprise: the local frequency spectrum perception statistic of cognitive radio users, local frequency spectrum perception statistic, based on Generalized Likelihood Ratio, is the arithmetic average of characteristic value of sample covariance matrix and the logarithm of the ratio of geometric average that cognitive radio users receives signal; Data fusion module 64, is connected to receiver module 62, for the local frequency spectrum perception statistic of a plurality of cognitive radio users being carried out to data fusion according to the local frequency spectrum perception data of a plurality of cognitive radio users.

As shown in Figure 7, above-mentioned data fusion module 64 comprises: allocation units 642, for distributing corresponding weighted factor to respectively the local frequency spectrum perception statistic of a plurality of cognitive radio users, wherein, weighted factor is according to the frequency spectrum perception capability distribution of each cognitive radio users; Computing unit 644, is connected to allocation units 642, for calculate the linear weighted function of local frequency spectrum perception statistic according to weighted factor, as overall frequency spectrum perception statistic.

Preferably, the local frequency spectrum perception data that data fusion center receives also comprise: the antenna number of cognitive radio users, signal sampling sample number, weighted factor is the product of antenna number and signal sampling sample number.

As shown in Figure 8, said apparatus also comprises: judging module 66, be connected to data fusion module 64, and for according to overall frequency spectrum perception statistic (being the result of data fusion) and the decision threshold obtaining in advance, whether idlely adjudicate current frequency spectrum resource.

Preferably, judging module 66 comprises: comparing unit, for the size of more overall frequency spectrum perception statistic and decision threshold; The first decision unit, in the situation that overall frequency spectrum perception statistic is more than or equal to decision threshold, adjudicates current frequency spectrum resource and is taken by primary user PU; The second decision unit, in the situation that overall frequency spectrum perception statistic is less than decision threshold, adjudicates the current frequency spectrum resource free time.

Preferably, said apparatus also comprises: notification module, participates in frequency spectrum perception for a plurality of cognitive radio users by control channel notification data fusion center coverage.

It should be noted that, the data fusion device of the multi-user collaborative frequency spectrum perception of describing in device embodiment is corresponding to above-mentioned embodiment of the method, and its concrete implementation procedure had been carried out detailed description in embodiment of the method, did not repeat them here.

One of ordinary skill in the art will appreciate that all or part of step that realizes above-described embodiment is to come the hardware that instruction is relevant to complete by program, described program can be stored in a kind of computer-readable recording medium, this program, when carrying out, comprises step of embodiment of the method one or a combination set of.

In addition, each functional unit in each examples of implementation of the present invention can adopt the form of hardware to realize, and also can adopt the form of software function module to realize.If described integrated module usings that the form of software function module realizes and during as production marketing independently or use, also can be stored in a computer read/write memory medium.

The above-mentioned storage medium of mentioning can be read-only memory, disk or CD etc.

In sum, according to the abovementioned embodiments of the present invention, provide a kind of data fusion method and device of multi-user collaborative frequency spectrum perception.By the above embodiment of the present invention, the local frequency spectrum perception statistic based on Generalized Likelihood Ratio that a plurality of cognitive radio users of data fusion center reception participation frequency spectrum perception report, and it is carried out to data fusion.Due to based on Generalized Likelihood Ratio, application performance is not subject to the probabilistic impact of noise, has solved the probabilistic problem of noise; And, no matter whether consider the probabilistic impact of noise, in the situation that the local frequency spectrum perception reliability difference of cognitive radio users is larger, also can obtain more excellent frequency spectrum perception performance than the energy measuring based on equal gain combining, can in heterogeneous network, obtain than tradition based on adopting the energy measuring of equal gain combining better to detect performance.

Obviously, those skilled in the art should be understood that, above-mentioned each module of the present invention or each step can realize with general calculation element, they can concentrate on single calculation element, or be distributed on the network that a plurality of calculation elements form, alternatively, they can be realized with the executable program code of calculation element, thereby, they can be stored in storage device and be carried out by calculation element, or they are made into respectively to each integrated circuit modules, or a plurality of modules in them or step are made into single integrated circuit module to be realized.Like this, the present invention is not restricted to any specific hardware and software combination.

The foregoing is only the preferred embodiments of the present invention, be not limited to the present invention, for a person skilled in the art, the present invention can have various modifications and variations.Within the spirit and principles in the present invention all, any modification of doing, be equal to replacement, improvement etc., within all should being included in protection scope of the present invention.

Claims (12)

1. a data fusion method for multi-user collaborative frequency spectrum perception, is characterized in that comprising:

Data fusion center receives the local frequency spectrum perception data from a plurality of cognitive radio users, wherein, described local frequency spectrum perception data comprise: the local frequency spectrum perception statistic of described cognitive radio users, described local frequency spectrum perception statistic, based on Generalized Likelihood Ratio, is the arithmetic average of characteristic value of sample covariance matrix and the logarithm of the ratio of geometric average that described cognitive radio users receives signal;

Described data fusion center is carried out data fusion according to the local frequency spectrum perception data of described a plurality of cognitive radio users to the local frequency spectrum perception statistic of described a plurality of cognitive radio users.

2. method according to claim 1, is characterized in that, before data fusion center receives the local frequency spectrum perception data from a plurality of cognitive radio users, described method also comprises:

Described a plurality of cognitive radio users is carried out local frequency spectrum perception in perception time slot, calculates local frequency spectrum perception statistic.

3. method according to claim 2, is characterized in that, calculates local frequency spectrum perception statistic and comprises:

Described cognitive radio users is sampled to the signal receiving;

Described cognitive radio users is calculated local sample covariance matrix according to sampled result;

Described cognitive radio users is carried out Eigenvalues Decomposition to described local sample covariance matrix, obtains described local sample covariance matrix characteristic of correspondence value;

Described cognitive radio users is calculated the logarithm of the arithmetic average of described characteristic value and the ratio of geometric average, as the local frequency spectrum perception statistic of described cognitive radio users.

4. method according to claim 2, is characterized in that, in described a plurality of cognitive radio users, carries out local frequency spectrum perception in perception time slot, and after calculating local frequency spectrum perception statistic, described method also comprises:

Described a plurality of cognitive radio users reports local frequency spectrum perception statistic separately to described data fusion center by up channel.

5. method according to claim 1, is characterized in that, described data fusion center is carried out data fusion according to the local frequency spectrum perception data of described a plurality of cognitive radio users to the local frequency spectrum perception statistic of described a plurality of cognitive radio users and comprised:

Described data fusion center distributes corresponding weighted factor to respectively the local frequency spectrum perception statistic of described a plurality of cognitive radio users, and wherein, described weighted factor is according to the frequency spectrum perception capability distribution of described each cognitive radio users;

The linear weighted function of described local frequency spectrum perception statistic is calculated at described data fusion center according to described weighted factor, as overall frequency spectrum perception statistic.

6. method according to claim 5, it is characterized in that, the described local frequency spectrum perception data that described data fusion center receives also comprise: the antenna number of described cognitive radio users, signal sampling sample number, described weighted factor is the product of described antenna number and described signal sampling sample number.

7. method according to claim 5, it is characterized in that, after described data fusion center is carried out data fusion according to the local frequency spectrum perception data of described a plurality of cognitive radio users to the local frequency spectrum perception statistic of described a plurality of cognitive radio users, described method also comprises:

Described data fusion center is according to described overall frequency spectrum perception statistic and the decision threshold obtaining in advance, whether idlely adjudicates current frequency spectrum resource.

8. method according to claim 7, is characterized in that, described data fusion center is according to described overall frequency spectrum perception statistic and the decision threshold obtaining in advance, and whether the free time comprises to adjudicate current frequency spectrum resource:

The size of the more described overall frequency spectrum perception statistic in described data fusion center and described decision threshold;

If described overall frequency spectrum perception statistic is more than or equal to described decision threshold, the described data fusion center described current frequency spectrum resource of judgement is taken by primary user;

If described overall frequency spectrum perception statistic is less than described decision threshold, judgement described current frequency spectrum resource in described data fusion center is idle.

9. according to the method described in any one in claim 1 to 8, it is characterized in that, before data fusion center receives the local frequency spectrum perception data from a plurality of cognitive radio users, described method also comprises:

Described data fusion center notifies a plurality of cognitive radio users in its coverage to participate in frequency spectrum perception by control channel.

10. the data fusion device of a multi-user collaborative frequency spectrum perception, be applied to data fusion center, it is characterized in that comprising: receiver module, for receiving the local frequency spectrum perception data from a plurality of cognitive radio users, wherein, described local frequency spectrum perception data comprise: the local frequency spectrum perception statistic of described cognitive radio users, described local frequency spectrum perception statistic, based on Generalized Likelihood Ratio, is the arithmetic average of characteristic value of sample covariance matrix and the logarithm of the ratio of geometric average that described cognitive radio users receives signal;

Data fusion module, for carrying out data fusion according to the local frequency spectrum perception data of described a plurality of cognitive radio users to the local frequency spectrum perception statistic of described a plurality of cognitive radio users.

11. devices according to claim 10, is characterized in that, described data fusion module comprises:

Allocation units, for distributing corresponding weighted factor to respectively the local frequency spectrum perception statistic of described a plurality of cognitive radio users, wherein, described weighted factor is according to the frequency spectrum perception capability distribution of described each cognitive radio users;

Computing unit, for calculate the linear weighted function of described local frequency spectrum perception statistic according to described weighted factor, as overall frequency spectrum perception statistic.

12. devices according to claim 11, it is characterized in that, the described local frequency spectrum perception data that described data fusion center receives also comprise: the antenna number of described cognitive radio users, signal sampling sample number, described weighted factor is the product of described antenna number and described signal sampling sample number.

Images