CN103281143B - Selecting type cooperation spectrum sensing method based on double-threshold energy detection - Google Patents

Abstract

The invention discloses a selecting type cooperation spectrum sensing method based on double-threshold energy detection. The selecting type cooperation spectrum sensing method comprises the following steps: (1) in a master user signal detection stage, adopting a double-threshold energy detection method by various cognitive users in a cognitive radio system to carry out local spectrum sensing, if an energy value exceeds two thresholds, carrying out local judgment 'H0 or H1'by the cognitive users, and if the energy value is between the two thresholds, reserving the energy value as a primary energy detection value by the cognitive users; (2) in an initial detection result reporting stage, evenly distributing a master user frequency band by the cognitive users, adopting a selecting type strategy by the cognitive users to report initial detection results respectively to a fusion center, preventing cognitive detection results with unreliable parameters from being introduced, meanwhile, saving traditional exclusive control channel resources, adopting the equal gain criterion by the fusion center to carry out master user judgment on the received primary energy detection value, allowing the judgment result to be equivalent to a node strategy by the fusion center, and then adopting the 'or' criterion by the fusion center to make the final judgment whether a master user exists. The simulation results show that according to the selecting type cooperation spectrum sensing method based on the double-threshold energy detection, and on the premise that ROC properties are not lost, the exclusive control channel resources are effectively saved, and meanwhile higher detection efficiency can be obtained.

Description

A kind of selecting type cooperative frequency spectrum sensing method based on double threshold energy measuring

Technical field

The invention belongs to wireless communication technology field, relate to a kind of frequency spectrum sensing method, particularly a kind of selecting type cooperative frequency spectrum sensing method based on double threshold energy measuring.

Background technology

Along with the sustainable growth of radio communication service, the demand of wireless communication system to frequency spectrum resource constantly increases, thus makes radio spectrum resources become more and more rare.But radio spectrum resources is a kind of non-renewable resources, can only recycle frequency spectrum resource, therefore studies the utilance how improving frequency spectrum resource, could alleviate the difficult problem that frequency spectrum resource is rare.Cognitive radio technology is as a kind of emerging intelligent wireless communication technology, breach the policy of traditional fixed allocation frequency spectrum, by Real-Time Monitoring target frequency bands, under the prerequisite that any interference is not caused to primary user, allow cognitive user " to wait for an opportunity " access temporarily not by the idle frequency range that primary user uses, effectively raise the availability of frequency spectrum.If once find that primary user reuses this frequency range, cognitive user should exit this frequency range in time, normal and reliable to ensure that primary user communicates.

Owing to there is the problems such as path loss, shadow effect, multipath effect and concealed terminal in wireless environment, the detection probability that traditional single node is detected reduces, and even cannot detect the existence of primary user, thus causes interference to primary user's communication.Therefore collaboration frequency spectrum detects and is paid close attention to widely as the technology that can improve frequency spectrum detection.Document " the collaborative spectrum sensing Part I in cognitive radio: two users' network " (Cooperative spectrum sensing incognitive radio, Part I:Two user networks, G.Ganesan and Y.G.Li, IEEE Transactions on Wireless Communications, 2007, 6 (6): 2204-2213) point out in traditional collaborative spectrum sensing scheme, general all exist proprietary control channel between hypothesis cognitive user and fusion center, but this needs extra radio spectrum resources, and need to carry out dynamic management to proprietary control channel resource, add the complexity that system realizes.

Summary of the invention

For the problems referred to above, the present invention proposes a kind of selecting type cooperative frequency spectrum sensing method based on double threshold energy measuring for cognitive radio networks, this method is difficult to determine the problems such as unique decision threshold, multipath fading and shadow fading in avoiding collaboration frequency spectrum to detect, based on double threshold energy measuring, cognitive user mean allocation primary user frequency range, the time slot be assigned to adopts selecting type strategy report its initial examination and measurement result to fusion center; First fusion center adjudicates primary energy detected value according to equal gain combining criterion, this court verdict is equivalent to a node decision-making, then (is pointed to fusion center by "or" criterion and other decision nodes and report initial examination and measurement result H ₀or H ₁cognitive user) make the conclusive judgement whether primary user exist, improve the detection probability of system.

For achieving the above object, the present invention is based on the method for the selecting type collaborative spectrum sensing without the need to proprietary control channel of double threshold energy measuring, it is characterized in that, comprise the following steps: a kind of selecting type cooperative frequency spectrum sensing method based on double threshold energy measuring, it is characterized in that, comprise following steps:

(1), primary user's phase detected signal:

(1a), in cognitive radio system each cognitive user carries out local frequency spectrum perception to primary user's signal, obtains the primary user energy detection value Y perceived _i;

(1b) primary user's energy value Y that each cognitive user detects, is judged _iwith two threshold value λ ₁and λ ₂relation, as energy detection value Y _idrop on two threshold value λ ₁and λ ₂outside, directly carry out this locality judgement " H ₀or H ₁", H ₀represent that primary user does not exist, H ₁represent that primary user exists;

If (1c) energy detection value Y _idrop on two threshold value λ ₁and λ ₂between, directly do not carry out this locality judgement, cognitive user needs to retain primary energy detected value Y _i;

(2) the initial examination and measurement report the test stage

(2a), each cognitive user mean allocation primary user frequency range, cognitive user on the sub-slots be assigned to, by respective test results report to fusion center;

(2b) selecting type strategy is adopted to report initial examination and measurement result to fusion center; exist when cognitive user does not perceive primary user at primary user's phase detected signal; then send an index signal through coding to fusion center, otherwise do not send any signal to avoid producing interference with primary user to fusion center;

(2c), fusion center is to the primary energy detected value Y received _imake primary user's judgement " H ₀or H ₁", court verdict is equivalent to a node decision-making, then the testing result of reporting to fusion center in joint step (2b) (comprises H ₀or H ₁) make the conclusive judgement whether primary user exist.

Further technical scheme comprises:

2 threshold value λ varied in size are set in step (1) ₁and λ ₂, λ ₁< λ ₂, as energy detection value Y _i> λ ₂time, by testing result D _ibe judged to 1, namely there is H in primary user's signal ₁; As energy detection value Y _i< λ ₁time, by testing result D _ibe judged to 0, namely there is not H in primary user's signal ₀; When energy detection value is positioned at λ ₁< Y _i< λ ₂time interval, then retain primary energy detected value Y _i; Testing result D _ibe expressed as:

$D_i=\left\{\begin{array}{cc}0,& 0<Y_i<{\mathrm{\&lambda;}}_1\\ Y_i,& {\mathrm{\&lambda;}}_1<Y_i<{\mathrm{\&lambda;}}_2\\ 1,& Y_i>{\mathrm{\&lambda;}}_2\end{array}---\left(1\right)\right.$

In step (2), cognitive user CU _isignal beta is sent to fusion center at the i-th sub-channels of primary user's frequency range _i(k), the corresponding Received signal strength of fusion center is expressed as:

$y_c^i(k,2)=\sqrt{p_s}{h}_{\mathrm{ic}}(k,2){\mathrm{\&beta;}}_i(k,2)+\sqrt{p_p}{h}_{\mathrm{pc}}\left(k\right)\mathrm{\&alpha;}(k,2)+n_c^i(k,2)---\left(2\right)$

Wherein k represents the time slot that cognitive user is assigned to, p _prepresent the transmitting power of primary user, p _srepresent the transmitting power of cognitive user; h _ic(k) and h _pck () represents CU respectively _ito fusion center and primary user to the wireless channel gain of fusion center; (k, 2) represent that zero-mean and variance are N ₀additive white Gaussian noise;

${\mathrm{\&beta;}}_i\left(k\right)=\left\{\begin{array}{cc}{D}_i,& H_i(k,1)=H_0\\ 0,& H_i(k,1)=H_1\end{array}---\left(3\right)\right.$

$\mathrm{\&alpha;}(k,2)=\left\{\begin{array}{cc}0,& H\left(k\right)=H_0\\ s(k,2),& H\left(k\right)=H_1\end{array}---\left(4\right)\right.$

Wherein s (k, 2) represents primary user's transmitting in the initial examination and measurement report the test stage, according to formula (2) fusion center decoding β _i(k);

According to Shannon channel coding theorem, as cognitive user CU _iwith fusion center produce interrupt time, fusion center cannot Received signal strength or successfully decode in respective sub-channel, and now fusion center thinks cognitive user CU _ido not send signal, acquiescence CU _iinitial examination and measurement result H _i(k, 2)=H ₁or Y _i; When fusion center successful Received signal strength decoding in respective sub-channel, so fusion center thinks CU _ihave sent signal, i.e. fusion center acquiescence CU _iinitial examination and measurement result be H _i(k, 2)=H ₀; Fusion center receive from cognitive user CU _iinitial examination and measurement result be expressed as:

Wherein Θ (k, 2)=1 represents cognitive user CU _iinterrupt between fusion center, wherein μ represents time bandwidth product, γ _srepresent cognitive user transmitting power, γ _prepresent primary user's transmitting power; Θ (k, 2)=0 represents cognitive user CU _ido not interrupt between fusion center;

Generation interruption is expressed as:

$\frac{1-a}{M}{\log }_2(1+\frac{{\left|h_{\mathrm{ic}}\left(k\right)\right|}^2{\&CenterDot;\mathrm{\&gamma;}}_s{\left|{\mathrm{\&beta;}}_i\left(k\right)\right|}^2}{{\left|h_{\mathrm{pc}}\left(k\right)\right|}^2{\&CenterDot;\mathrm{\&gamma;}}_p{\left|\mathrm{\&alpha;}(k,2)\right|}^2+1})<\frac{1}{\mathrm{\&mu;}}---\left(6\right)$

Wherein a is the input overhead time, and M is cognitive user quantity.

In step 2 (c), fusion center adopts equal gain combining criterion to the primary energy detected value Y received _imake primary user's judgement " H ₀or H ₁", more comprehensively make by the testing result of reporting to fusion center in "or" criterion and step (2b) conclusive judgement whether primary user exist.

Sensing results corresponding to conclusive judgement is expressed as:

(7) wherein y refers to primary user's court verdict that fusion center employing equal gain combining criterion is made the primary energy detected value received, h _i(k, 2) process for adopting logical "or" criterion the result obtained to the testing result that this locality is adjudicated.

Beneficial effect of the present invention: the present invention adopts collaborative spectrum sensing technology, avoid because there is the problems such as path loss, shadow effect, multipath effect and concealed terminal in wireless environment, the detection probability that traditional single node is detected reduces, even cannot detect the existence of primary user, thus interference is caused to primary user's communication.Due in traditional collaborative spectrum sensing scheme, general all exist proprietary control channel between hypothesis cognitive user and fusion center, but this needs extra radio spectrum resources, and need to carry out dynamic management to proprietary control channel resource, add the complexity that system realizes, the present invention casts aside traditional proprietary control channel, by mean allocation not by frequency range that primary user uses, each cognitive user on the time slot be assigned to by the data transmission scheme of selecting type, initial examination and measurement result is reported to fusion center, avoiding communicates to primary user produces interference, effectively raise detection probability, and improve the availability of frequency spectrum.

Accompanying drawing explanation

Fig. 1 is collaborative spectrum sensing system model block diagram of the present invention;

Fig. 2 is double threshold energy measuring decision diagram;

Fig. 3 is under different detection probability, and traditional scheme compares performance chart with the false alarm probability of improvement project;

Fig. 4 is that under different primary user's input overhead time, false alarm probability is with the change curve of detection probability;

Fig. 5 is the flow chart of the selecting type cooperative frequency spectrum sensing method based on double threshold energy measuring of the present invention.

Specific embodiments

Below in conjunction with accompanying drawing, principle of the present invention and embodiment are described in further detail:

Figure 1 shows that collaborative spectrum sensing system model figure of the present invention, collaborative spectrum sensing system comprises a primary user, a M cognitive user and an information fusion center.

Based on a selecting type cooperative frequency spectrum sensing method for double threshold energy measuring, comprise the steps:

1, primary user's phase detected signal

In the first stage of time slot k, i.e. primary user's phase detected signal.What primary user's phase detected signal adopted is double threshold energy detection method, by arranging 2 threshold value λ varied in size ₁and λ ₂(λ ₁< λ ₂) dividing Received signal strength energy value (Fig. 2 is double threshold energy measuring judgement block diagram), M cognitive user carries out frequency spectrum perception to primary user's signal alone, obtains energy detection value Y _i, i=1,2 ... M.As energy detection value Y _i> λ ₂time, by testing result D _ibe judged to 1, namely there is H in primary user's signal ₁, as energy detection value Y _i< λ ₁time, by testing result D _ibe judged to 0, namely there is not H in primary user's signal ₀.When detected energy value drops on λ ₁< Y _i< λ ₂time interval, then cognitive user retains primary energy detected value Y _i.

Then testing result D _ican be expressed as:

$D_i=\left\{\begin{array}{cc}0,& 0<Y_i<{\mathrm{\&lambda;}}_1\\ Y_i,& {\mathrm{\&lambda;}}_1<Y_i<{\mathrm{\&lambda;}}_2\\ 1,& Y_i>{\mathrm{\&lambda;}}_2\end{array}---\left(1\right)\right.$

2, the initial examination and measurement report the test stage

In the second stage of time slot k, i.e. the initial examination and measurement report the test stage.Each cognitive user CU _imean allocation primary user frequency range, each cognitive user CU _ion the sub-slots be assigned to, by respective test results report to fusion center.Each cognitive user CU _iselecting type strategy is adopted to report initial examination and measurement result, if cognitive user CU to fusion center _i(local court verdict is H not perceive primary user in the primary user detection stage ₀cognitive user), then to fusion center send one through coding after index signal; Otherwise do not send any signal to avoid producing interference with primary user to fusion center, the testing result being now reported to fusion center is defaulted as Y _ior H ₁, this testing result is not be reported to fusion center by the transmission of index signal, but showed by this form of signal interruption.

Cognitive user CU _isignal beta is sent to fusion center at the i-th sub-channels of primary user's frequency range _i(k), the corresponding Received signal strength of fusion center can be expressed as:

$y_c^i(k,2)=\sqrt{p_s}{h}_{\mathrm{ic}}(k,2){\mathrm{\&beta;}}_i(k,2)+\sqrt{p_p}{h}_{\mathrm{pc}}\left(k\right)\mathrm{\&alpha;}(k,2)+n_c^i(k,2)---\left(2\right)$

Wherein p _prepresent the transmitting power of primary user, p _srepresent the transmitting power of cognitive user; h _ic(k) and h _pck () represents CU respectively _ito fusion center and primary user to the wireless channel gain of fusion center; (k, 2) represent that zero-mean and variance are N ₀additive white Gaussian noise.

${\mathrm{\&beta;}}_i\left(k\right)=\left\{\begin{array}{cc}{D}_i,& H_i(k,1)=H_0\\ 0,& H_i(k,1)=H_1\end{array}\right.---\left(3\right)$

$\mathrm{\&alpha;}(k,2)=\left\{\begin{array}{cc}0,& H\left(k\right)=H_0\\ s(k,2),& H\left(k\right)=H_1\end{array}---\left(4\right)\right.$

Wherein s (k, 2) represents primary user's transmitting in time slot k second stage, according to formula (2) fusion center decoding β _i(k).

According to Shannon channel coding theorem, if channel capacity is lower than the rate of information throughput, adopt which kind of decoder all correctly can not recover primary signal, and be called and interrupt.As cognitive user CU _iwith fusion center produce interrupt time, fusion center cannot Received signal strength or successfully decode in respective sub-channel, and now fusion center thinks cognitive user CU _ido not send signal, acquiescence CU _iinitial examination and measurement result H _i(k, 2)=H ₁or Y _i, namely cognitive user initial examination and measurement result for primary user exist or detected value be original energy value.If fusion center successful Received signal strength decoding in respective sub-channel, so fusion center then thinks CU _ihave sent signal, i.e. fusion center acquiescence CU _iinitial examination and measurement result be H _i(k, 2)=H ₀.Therefore, fusion center receives from cognitive user CU _iinitial examination and measurement result can be expressed as:

Wherein Θ (k, 2)=1 represents cognitive user CU _iinterrupt between fusion center, wherein μ represents time bandwidth product, γ _srepresent cognitive user transmitting power and γ _prepresent primary user's transmitting power; Θ (k, 2)=0 expression cognitive user CU _ido not interrupt between fusion center.Generation interruption is expressed as:

$\frac{1-a}{M}{\log }_2(1+\frac{{\left|h_{\mathrm{ic}}\left(k\right)\right|}^2{\&CenterDot;\mathrm{\&gamma;}}_s{\left|{\mathrm{\&beta;}}_i\left(k\right)\right|}^2}{{\left|h_{\mathrm{pc}}\left(k\right)\right|}^2{\&CenterDot;\mathrm{\&gamma;}}_p{\left|\mathrm{\&alpha;}(k,2)\right|}^2+1})<\frac{1}{\mathrm{\&mu;}}---\left(6\right)$

Wherein a is the input overhead time, and M is cognitive user quantity.

Fusion center first by retain in step (1) and the primary energy detected value Y received by fusion center _icarry out gain such as grade (EGC) and merge fusion, make primary user " H ₀or H ₁" judgement, this court verdict is equivalent to a node decision-making, then (for the cognitive user sending index signal, testing result is H according to "or" criterion and other decision nodes ₀; For not sending index signal and not retaining the cognitive user of primary energy detected value Yi, testing result is H ₁) make final sensing results, namely make the conclusive judgement whether primary user exists.

Consider logical "or" criterion, sensing results can be expressed as: (7)

Under rayleigh fading channel, in cognitive radio system, the initial examination and measurement probability of cognitive user, false alarm probability, false dismissal probability are expressed as follows respectively:

${\mathrm{Pd}}_i\left(\mathrm{\&mu;}\right)=P\{Y_i>{\mathrm{\&lambda;}}_2|H_1\}$

$={\&Integral;}_x{Q}_{\mathrm{\&mu;}}(\sqrt{2\stackrel{\&OverBar;}{\mathrm{\&gamma;}}},\sqrt{{\mathrm{\&lambda;}}_2})\&CenterDot;f_{\mathrm{\&gamma;}}\left(x\right)\mathrm{dx}---\left(8\right)$

$=e^{-\frac{{\mathrm{\&lambda;}}_2}{2}}\underset{n=0}{\overset{\mathrm{\&mu;}-2}{\mathrm{\&Sigma;}}}\frac{1}{n!}{\left(\frac{{\mathrm{\&lambda;}}_2}{2}\right)}^n+{\left(\frac{1+\stackrel{\&OverBar;}{\mathrm{\&gamma;}}}{\stackrel{\&OverBar;}{\mathrm{\&gamma;}}}\right)}^{\mathrm{\&mu;}-1}\left[e^{-\frac{{\mathrm{\&lambda;}}_2}{2(1+\stackrel{\&OverBar;}{\mathrm{\&gamma;}})}}\right]-e^{-\frac{{\mathrm{\&lambda;}}_2}{2}}\underset{n=0}{\overset{\mathrm{\&mu;}-2}{\mathrm{\&Sigma;}}}\frac{1}{n!}{\left(\frac{{\mathrm{\&lambda;}}_2\stackrel{\&OverBar;}{\mathrm{\&gamma;}}}{2(1+\stackrel{\&OverBar;}{\mathrm{\&gamma;}})}\right)}^n]$

${\mathrm{Pm}}_i=1-{\mathrm{\&Delta;}}_{1,j}-{\mathrm{Pd}}_i---\left(9\right)$

${\mathrm{Pf}}_i=P\{Y_i>{\mathrm{\&lambda;}}_2|H_o\}=\frac{\mathrm{\&Gamma;}(\mathrm{\&mu;},\frac{{\mathrm{\&lambda;}}_2}{2})}{\mathrm{\&Gamma;}\left(\mathrm{\&mu;}\right)}---\left(10\right)$

Wherein Q _urepresent broad sense horse khoum (Marcum) function; Γ (μ) and Γ (μ, λ ₂/ 2) complete and imperfect Gamma function is represented; represent average signal-to-noise ratio, f _γthe probability-distribution function of signal to noise ratio under (x) expression fading profiles.

According to formula (6), (7), (8), can show that when adopting logical "or" criterion, the detection probability of collaborative spectrum sensing when fusion center without the need to proprietary control channel is:

${\mathrm{Pd}}_{\mathrm{or}}^{\mathrm{proposed}}=1-\underset{K=0}{\overset{M-1}{\mathrm{\&Sigma;}}}\left(\begin{array}{c}M\\ K\end{array}\right)\underset{i=1}{\overset{K}{\mathrm{\&Pi;}}}\mathrm{Pm}\&CenterDot;\underset{j=1}{\overset{M-K}{\mathrm{\&Pi;}}}{\mathrm{\&Delta;}}_{1,j}\&CenterDot;\{1-{\mathrm{Pd}}_j[(M-K)\&CenterDot;\mathrm{\&mu;}\left]\right\}-\underset{i=1}{\overset{M}{\mathrm{\&Pi;}}}\mathrm{Pm}$

（11）

In formula

detection limit Y when it represents that primary user exists _idrop on threshold value λ ₁and λ ₂between probability. with represent the wireless channel gain h of cognitive user CUi to fusion center respectively _ick average that () obeys and primary user are to the wireless channel gain h of fusion center _pck average that () obeys.

According to formula (6), (7), (10), can show that when adopting logical "or" criterion, the false alarm probability of collaborative spectrum sensing when fusion center without the need to proprietary control channel is:

${\mathrm{Pf}}_{\mathrm{or}}^{\mathrm{proposed}}=1-\underset{K=0}{\overset{M-1}{\mathrm{\&Sigma;}}}\left(\begin{array}{c}M\\ K\end{array}\right)\underset{i=1}{\overset{K}{\mathrm{\&Pi;}}}\mathrm{Pn}\&CenterDot;\underset{j=1}{\overset{M-K}{\mathrm{\&Pi;}}}{\mathrm{\&Delta;}}_{0,j}\&CenterDot;\{1-\frac{\mathrm{\&Gamma;}\left\{\right[(M-K)\&CenterDot;\mathrm{\&mu;}],\frac{\mathrm{\&lambda;}}{2}\}}{\mathrm{\&Gamma;}[(M-K)\&CenterDot;\mathrm{\&mu;}]}\}-\underset{i=1}{\overset{M}{\mathrm{\&Pi;}}}\mathrm{Pn}$

（13）

In formula

Δ _{0, j}=P{ λ ₁< Y _i< λ ₂| H ₀represent primary user detection limit Y when not existing _idrop on threshold value λ ₁and λ ₂between probability.

The false alarm probability of the method for the selecting type collaborative spectrum sensing without the need to proprietary control channel based on double threshold energy measuring that the present invention proposes compares can consult Fig. 3 with the false alarm probability of traditional double threshold energy measuring method.As can be seen from Figure 3, the false alarm probability of the collaborative spectrum sensing scheme proposed in low detection probability region is obviously greater than traditional collaborative spectrum sensing scheme, this is because unaffected in order to ensure primary user's telecommunication service quality, make the initial examination and measurement probability P d of cognitive user _icertain threshold condition (as the formula (15)) must be met;

${\mathrm{Pd}}_i\&GreaterEqual;1-\frac{{\mathrm{\&gamma;}}_p\&CenterDot;{\mathrm{\&sigma;}}_{\mathrm{pd}}^2\&CenterDot;M[1-{(1-\mathrm{Pout})}^{1/M}]}{{\mathrm{\&gamma;}}_s\&CenterDot;{\mathrm{\&sigma;}}_{\mathrm{id}}^2\&CenterDot;{(1-\mathrm{Pout})}^{1/M}\&CenterDot;(2^{R_p}-1)}---\left(15\right)$

Pout represents given primary user's communication interruption probability thresholding, R _prepresent primary user's message transmission rate.

In high detection probability region, the collaborative spectrum sensing scheme of false alarm probability almost with traditional of the collaborative spectrum sensing scheme proposed is the same, in actual cognitive radio system, such as IEEE radio area network standard specifies that the detection probability of frequency spectrum perception must be greater than 0.9, from this angle scheme in this paper under the prerequisite of not losing ROC performance, effectively save proprietary reporting channel resource.

Under Fig. 4 gives different primary user's input overhead time, false alarm probability is with the change curve of detection probability.As can be seen from Figure 4 by reasonably increasing the detection time of primary user's signal, effectively can improve the initial examination and measurement performance of cognitive user, the false alarm probability of the present invention program also declines thereupon.

The contribution of method of the present invention is in sum:

1, collaborative spectrum sensing technology is adopted, avoid because there is the problems such as path loss, shadow effect, multipath effect and concealed terminal in wireless environment, the detection probability that traditional single node is detected reduces, and even cannot detect the existence of primary user, thus causes interference to primary user's communication.

2, adopt selecting type strategy to report testing result technology, avoid the insecure transfer of data of some testing results to fusion center, reduce the complexity of fusion center data processing, avoid the interference to primary user's communication simultaneously.

3, adopt without the need to proprietary control channel technology, reduce because supposing to there is proprietary control channel between cognitive user and fusion center in traditional collaborative spectrum sensing scheme, decrease extra radio spectrum resources, avoid and dynamic management is carried out to proprietary control channel resource, reduce the complexity that system realizes.

4, adopt double threshold energy detection technique, avoid existence in collaboration frequency spectrum detection and be difficult to determine the problems such as unique decision threshold, multipath fading and shadow fading.

Above-described embodiment does not limit the present invention in any form, the technical scheme that the mode that all employings are equal to replacement or equivalent transformation obtains, and all drops on protection scope of the present invention.

Claims (5)

1., based on a selecting type cooperative frequency spectrum sensing method for double threshold energy measuring, it is characterized in that, comprise following steps:

(1), primary user's phase detected signal:

(1a), in cognitive radio system each cognitive user carries out local frequency spectrum perception to primary user's signal, obtains the primary user energy detection value Y perceived _i;

(1b) primary user's energy value Y that each cognitive user detects, is judged _iwith two threshold value λ ₁and λ ₂relation, as energy detection value Y _idrop on two threshold value λ ₁and λ ₂outside, directly carry out this locality judgement " H ₀or H ₁", H ₀represent that primary user does not exist, H ₁represent that primary user exists;

If (1c) energy detection value Y _idrop on two threshold value λ ₁and λ ₂between, directly do not carry out this locality judgement, cognitive user needs to retain primary energy detected value Y _i;

(2) the initial examination and measurement report the test stage

(2a), each cognitive user mean allocation primary user frequency range, cognitive user on the sub-slots be assigned to, by respective test results report to fusion center;

(2b) adopt selecting type strategy to report initial examination and measurement result to fusion center, exist when cognitive user does not perceive primary user at primary user's phase detected signal, then send an index signal through coding to fusion center; Otherwise do not send any signal to avoid producing interference with primary user to fusion center, testing result is defaulted as Y _ior H ₁;

(2c), fusion center is to the primary energy detected value Y received _imake primary user's judgement " H ₀or H ₁", court verdict is equivalent to a node decision-making, then the testing result of reporting to fusion center in joint step (2b) makes the conclusive judgement whether primary user exist.

2. a kind of selecting type cooperative frequency spectrum sensing method based on double threshold energy measuring according to claim 1, is characterized in that, arrange 2 threshold value λ varied in size in step (1) ₁and λ ₂, λ ₁< λ ₂, as energy detection value Y _i> λ ₂time, by testing result D _ibe judged to 1, namely there is H in primary user's signal ₁; As energy detection value Y _i< λ ₁time, by testing result D _ibe judged to 0, namely there is not H in primary user's signal ₀; When energy detection value is positioned at λ ₁< Y _i< λ ₂time interval, then retain primary energy detected value Y _i;

Testing result D _ibe expressed as:

$D_i=\left\{\begin{array}{cc}0,& 0<Y_i<{\mathrm{\&lambda;}}_1\\ Y_i,& {\mathrm{\&lambda;}}_1<Y_i<{\mathrm{\&lambda;}}_2\\ 1,& Y_i>{\mathrm{\&lambda;}}_2\end{array}\right.---\left(1\right)$

3. a kind of selecting type cooperative frequency spectrum sensing method based on double threshold energy measuring according to claim 1, is characterized in that, in step (2),

Cognitive user CU _isignal beta is sent to fusion center at the i-th sub-channels of primary user's frequency range _i(k), the corresponding Received signal strength of fusion center is expressed as:

$y_c^i(k,2)=\sqrt{p_s}{h}_{\mathrm{ic}}(k,2){\mathrm{\&beta;}}_i(k,2)+\sqrt{p_p}{h}_{\mathrm{pc}}\left(k\right)\mathrm{\&alpha;}(k,2)+n_c^i(k,2)---\left(2\right)$

Wherein k represents the time slot that cognitive user is assigned to, p _prepresent the transmitting power of primary user, p _srepresent the transmitting power of cognitive user; h _ic(k) and h _pck () represents CU respectively _ito fusion center and primary user to the wireless channel gain of fusion center; represent that zero-mean and variance are N ₀additive white Gaussian noise;

${\mathrm{\&beta;}}_i\left(k\right)=\left\{\begin{array}{cc}{D}_i,& H_i(k,1)=H_0\\ 0,& H_i(k,1)=H_1\end{array}\right.---\left(3\right)$

$\mathrm{\&alpha;}(k,2)=\left\{\begin{array}{cc}0,& H\left(k\right)=H_0\\ s(k,2),& H\left(k\right)=H_1\end{array}\right.---\left(4\right)$

Wherein s (k, 2) represents primary user's transmitting in the initial examination and measurement report the test stage, according to formula (2) fusion center decoding β _i(k);

According to Shannon channel coding theorem, as cognitive user CU _iwith fusion center produce interrupt time, fusion center cannot Received signal strength or successfully decode in respective sub-channel, and now fusion center thinks cognitive user CU _ido not send signal, acquiescence CU _iinitial examination and measurement result H _i(k, 2)=H ₁or Y _i; When fusion center successful Received signal strength decoding in respective sub-channel, so fusion center thinks CU _ihave sent signal, i.e. fusion center acquiescence CU _iinitial examination and measurement result be H _i(k, 2)=H ₀; Fusion center receive from cognitive user CU _iinitial examination and measurement result be expressed as:

Wherein represent cognitive user CU _iinterrupt between fusion center, Θ (k, 2)=0 represents cognitive user CU _ido not interrupt between fusion center;

Generation interruption is expressed as:

$\frac{1-a}{M}{\log }_2(1+\frac{{\left|h_{\mathrm{ic}}\left(k\right)\right|}^2\&CenterDot;{\mathrm{\&gamma;}}_s{\left|{\mathrm{\&beta;}}_i\left(k\right)\right|}^2}{{\left|h_{\mathrm{pc}}\left(k\right)\right|}^2\&CenterDot;{\mathrm{\&gamma;}}_p{\left|\mathrm{\&alpha;}(k,2)\right|}^2+1})<\frac{1}{\mathrm{\&mu;}}---\left(6\right)$

Wherein a is the input overhead time, and M is cognitive user quantity, and μ represents time bandwidth product, γ _srepresent cognitive user transmitting power, γ _prepresent primary user's transmitting power.

4. a kind of selecting type cooperative frequency spectrum sensing method based on double threshold energy measuring according to claim 1, is characterized in that, in step 2 (c), fusion center adopts equal gain combining criterion to the primary energy detected value Y received _imake primary user's judgement " H ₀or H ₁", more comprehensively make by the testing result of reporting to fusion center in "or" criterion and step (2b) conclusive judgement whether primary user exist.

5. a kind of selecting type cooperative frequency spectrum sensing method based on double threshold energy measuring according to claim 3 or 4 any one, is characterized in that, sensing results corresponding to conclusive judgement is expressed as:

$H_c\left(k\right)=\&CirclePlus;Y\underset{i=1}{\overset{K}{\&CirclePlus;}}{H}_i(k,2)---\left(7\right)$

Wherein y refers to primary user's court verdict that fusion center employing equal gain combining criterion is made the primary energy detected value received, the testing result that this locality is adjudicated is processed to the result obtained for adopting logical "or" criterion.