CN104904288A - Method and apparatus for cognitive radio networks - Google Patents

Abstract

Methods and apparatuses for sensing and power allocation for cognitive radio (CR) networks are provided. The method comprises: receiving a sequence of signals from a primary node with multiple primary transmit powers; sensing a status of the primary node based on the sequence of signals; recognizing at least one feature of the primary node based on the status of the primary node and the sequence of signals; and determining at least one transmission parameter for a secondary node based on the at least one feature.

Description

For the method and apparatus of cognitive radio networks

Technical field

Embodiments of the invention relate in general to the communication technology.More particularly, embodiments of the invention relate to for the method for cognitive radio networks, device, network node and computer program.

Background technology

This part is introduced and can be helped to promote the aspect to better understanding of the present invention.Therefore, the statement of this part should to be understood to by reading in this angle to about what be prior art or what be not admitting of prior art.

Cognitive radio (CR) is by the potential technology be familiar with for being for improving the frequency spectrum scarcity problem in spectrum utilization and solution next generation wireless communication.If the frequency spectrum licensing to primary user (PU) is not utilized by this PU or to the interference of this PU lower than given level, then the secondary user's (SU) in CR network is allowed to access this frequency spectrum.

Current, exist for three kinds of CR network main frequency spectrum access methods: i) eclipsed form (Underlay) or so-called frequency spectrum share scheme, as long as wherein the service quality (QoS) of PU is protected, then SU is allowed to coexist with this PU, ii) opportunistic frequency spectrum access, wherein only when PU is detected the free time, SU can access main band; And iii) the above two combination, also namely based on the frequency spectrum share of perception, wherein SU first perceived spectral to determine the state (movable/idle) of this PU, and then select its through-put power based on judgement.

The existing spectrum sensing scheme observed based on this locality of SU can be divided into matched filter, energy measuring, cyclo-stationary detection, Wavelet Detection and covariance detection.A kind of improving one's methods for the frequency spectrum perception in CR has been suggested during the name that Tao Cui, Feifei Gao and Arumugam Nallanathan publishes in the IEEE TRANSACTIONS ON VEHICULAR TECHNOLOGY in May, 2011 is called " Optimization of Cooperative Spectrum Sensing in Cognitive Radio ".In advised method, multiple SU is designed to cooperation mutually and, with the problem solving concealed terminal and boundary effect, result in collaborative spectrum sensing.For keeping the QoS of PU-T reflector (PU-Tx) and protecting it not by harmful interference; and the non-linear and long term power budget of the power amplifier at SU place; have employed the constraint of the peak value/average transmission power at SU place and the peak value/average interference power at PU place; wherein show that optimal power allocation maximizes to make secondary accessible speed, this secondary accessible speed is used for the various combination of the power constraint caused due to actual needs.

Summary of the invention

Inventor notices in most of work on hand, and important hypothesis is main through-put power is unmodifiable and constant, and sensor model is conventional binary system hypothesis test model.But, in Modern Communication System, due to different channels signal to noise ratio (SNR), use adaptive tracking control to provide constant speed.

Therefore, the technical scheme be provided for the CR perception and/or power division with multiple main through-put power will be expected in the art.

One or more in order to what better solve in above misgivings, in a first aspect of the present invention, provide a kind of method for cognitive radio (CR) network.The method comprises: from the host node Received signal strength sequence with multiple main through-put power; Based on the state of this this host node of burst perception; Based on the state of this host node and at least one feature of this burst identification host node; And at least one transformation parameter of secondary nodes is determined based on this at least one feature.

In certain embodiments, this at least one feature comprise main through-put power and modulation and encoding scheme (MCS) at least one.

In certain embodiments, the step of perception can comprise: the cumlative energy calculating this burst; And by cumlative energy is compared with predefined threshold value the existence judging this host node.

In certain embodiments, the step of identification can comprise: define the multiple subspaces corresponding with the plurality of main through-put power; And by the energy comparison of this cumlative energy and the plurality of subspace being estimated this host node is using which the main through-put power in the plurality of main through-put power.

In certain embodiments, determining step can comprise: by using at least one predetermined criterion, based on this, at least one is characterized as this secondary nodes distribution secondary transmission power.It is one or more that this at least one predetermined criterion can comprise in the following: the average accessible speed of this secondary nodes is maximized; Average transmission power constraint under predefined power; And the average interference power under the maximum interference of this host node is being retrained.

In a second aspect of the present invention, provide a kind of device to implement the various embodiments of the method for a first aspect of the present invention.Especially, the device of cognitive radio (CR) network is provided for.This device comprises: receiving element, is arranged to the host node Received signal strength sequence from having multiple main through-put power; Perception unit, is arranged to the state based on this this host node of burst perception; Recognition unit, is arranged at least one feature of state based on this host node and this this host node of burst identification; And determining unit, be arranged at least one transformation parameter determining secondary nodes based on this at least one feature.

In a third aspect of the present invention, provide a kind of secondary nodes, it at least one memory comprising at least one processor and comprise computer program code.This memory and computer program code are configured to make this device perform the embodiment of the method for a first aspect of the present invention.

In a fourth aspect of the present invention, provide a kind of computer program, it comprises at least one computer readable memory medium it storing computer readable program code part.This computer readable program code part comprises the code instructions of the embodiment of the method for performing a first aspect of the present invention.

The specific embodiment of the theme described in this manual can be implemented so that one or more in realizing advantage below.

Adopt the specific embodiment of the technology described in this specification, the scheme that proposed utilizes the scene of multiple main through-put power transmission for wherein PU.The scheme proposed is suitable for actual conditions better.

When read in conjunction with the accompanying drawings, according to the following description of specific embodiment, also will understand other Characteristics and advantages of embodiments of the invention, accompanying drawing illustrates the principle of embodiments of the invention by example.

Accompanying drawing explanation

Above or other aspects, the characteristic sum benefit of various embodiment of the present invention will become more completely obviously by example from following detailed description and accompanying drawing, in the accompanying drawings:

Fig. 1 illustrates the example system model for cognitive radio networks;

Fig. 2 diagram is according to an embodiment of the invention for the flow chart of the method for cognitive radio networks;

Fig. 3 is the schematic block diagram of the device 300 that can be configured to put into practice example embodiment of the present invention; And

Fig. 4 is suitable for the schematic block diagram putting into practice the network node used in example embodiment of the present invention.

Same reference numerals and name instruction identical element in various accompanying drawing.

Embodiment

Hereinafter, with reference to illustrative embodiment, principle of the present invention and spirit are described.It should be understood that all these embodiments are only presented for making those skilled in the art better understand and putting into practice the present invention further, instead of for limiting the scope of the invention.Such as, as the part of an embodiment and the feature illustrating or describe can make together with another embodiment for producing other embodiment.In order to clear, all features of actual execution mode are not described in this manual.What will of course be appreciated that is in the exploitation of these practical embodiments, should be understood that a lot of judgement specific to execution mode is to realize the specific objective of developer, such as meet the constraint that system is relevant and business is relevant, this constraint can change from an embodiment to another embodiment.And, be understood that such development effort may be complicated with consuming time, but be only born routine work for benefiting from those of ordinary skill in the art of the present disclosure.

Referring now to accompanying drawing, theme of the present disclosure is described.Only schematically describe various structure, system and equipment in the accompanying drawings for illustrative purposes, and so that make description do not well known to a person skilled in the art details cover.But, accompanying drawing is included to the illustrated examples of description and the theme disclosed in explanation.Word used herein and phrase should be understood and interpreted to has the implication consistent with the understanding of various equivalent modifications to those words and phrase.The consistent use of term or phrase herein is not intended to the specifically defined of this term implicit or phrase, is namely different from the definition of common and usual implication understood by one of ordinary skill in the art.The implication be intended to that there is particular meaning with regard to term or phrase, being namely different from technical staff's understanding, like this specifically defined will come to be proposed expressly in the description in a defined manner, and the mode of this definition directly or is clearly provided for the specifically defined of this term or phrase.

Fig. 1 illustrates the example system model for cognitive radio networks.As shown in fig. 1, CR network is considered to have a pair main reflector (PU-Tx) and receiver (PU-Rx) and under given interference constraints, shares a pair secondary emitters (SU-Tx) and the receiver (SU-Rx) of frequency spectrum with main band.Make γ ₁, γ ₂, h and g represents the transient channel power gain from PU-Tx to SU-Tx, the transient channel power gain from PU-Tx to SU-Rx, the transient channel power gain from SU-Tx to PU-Rx and the transient channel power gain from SU-Tx to SU-Rx respectively.Channel gain is assumed that ergodic, stable and is known at SU place.In practice, SU can be provided with the channel gain of PU, or SU can detect channel gain by the training symbol of monitoring PU.

Fig. 2 diagram is according to the flow chart of a kind of method for cognitive radio networks of the embodiment of the present invention.The method of Fig. 2 can be performed at the secondary nodes of cognitive radio networks (i.e. secondary user's) place.

The method starts from step S201 and proceeds to step S202, and wherein SU is from PU Received signal strength sequence.In considered scene, PU can transmit with multiple main through-put power.

Then, in step S203 place, SU is based on the state of burst perception PU.The state of PU can be presence or absence.In one embodiment, perception is performed based on energy measuring.Those skilled in the art are to be understood that the state that other technologies can be used to carry out perception PU.

In step S204 place, SU identifies at least one feature of PU based on the state of the PU of institute's perception and burst.This at least one feature can comprise main through-put power and modulation and encoding scheme (MCS) at least one.Certainly, if need also to identify other features relevant with the transformation parameter of PU.

Then, in step S205 place, SU can determine himself at least one transformation parameter based on this at least one feature of PU.In one embodiment, if transformation parameter is through-put power, then this determines that can comprise by using at least one predetermined criterion to come based on this at least one feature (such as main through-put power) of PU is SU distribution secondary transmission power.It is one or more that this at least one predetermined criterion can comprise in the following: the average accessible speed of SU is maximized; Average transmission power constraint under predefined power; And the average interference power under the maximum interference of host node is being retrained.

Finally, SU can use determined (multiple) transformation parameter to be used for communication, and the method terminates in step S206 place.

Frequency spectrum perception and feature identification will be described below in detail.Draw optimum perception rule based on likelihood ratio (likelihood ratio), this shows this rule and is equal to energy measuring rule.Exemplarily, this feature is main through-put power.Theoretical based on optimal detection, optimum main through-put power is estimated and total detection probability is derived.Then, for wherein PU with the situation of multiple main through-put power transmission, propose multiple power levels allocation strategy, and especially, a secondary transmission power depends on a main through-put power.

The signal being in a jth sample place reception in step S202 is modeled as:

$r_j=\left\{\begin{array}{cc}{n}_j,& H_0,\\ \sqrt{P_{p,i}}\sqrt{\mathrm{\γ}1}{s}_j+n_j,& H_1,i=1,...,N,\end{array}\right.---\left(1\right)$

Wherein H ₀and H ₁represent the hypothesis that PU-Tx does not exist or exists respectively; P _p,i, i=1 ... N meets 0<P _p,i<P _{p, i+1}, main discrete transmissions power; s _jthe jth symbol from PU-Tx transmission, s _jbe assumed to be the Gaussian Profile in accordance with having zero-mean and unit variance, i.e. s _j~ N (0,1); And n _jsuppose for all situations in accordance with N (0, N ₀) additive noise.Suppose s _jand n _jseparate.

In one embodiment, in step S203 place, perception can comprise the cumlative energy of the burst that calculating receives.The detection statistics y of the accumulative reception energy of sample is used to be written as

$y=\underset{j=1}{\overset{M}{\mathrm{\&Sigma;}}}{\left|r_j\right|}^2---\left(2\right)$

Wherein M is the total quantity of the sample received at SU-Tx place in a frame.Therefore at H ₀, P _p,iand H ₁the probability density function (pdf) of the y under condition is given respectively:

$f\left(y\right|H_0)=\frac{y^{\frac{M}{2}-1}{e}^{-\frac{y}{{2N}_0}}}{r\left(\frac{M}{2}\right){\left({2N}_o\right)}^{\frac{M}{2}}}---\left(3\right)$

$f\left(y\right|P_{p,i})=\frac{V^{\frac{M}{2}-1}{e}^{-\frac{y}{{2N}_D+{2r}_j{P}_{p,i}}}}{\mathrm{\&Gamma;}\left(\frac{M}{2}\right){(2M_o+2{\mathrm{\&gamma;}}_1{P}_{p,i})}^{\frac{M}{2}}}---\left(4\right)$

$f\left(y\right|H_1)={\mathrm{\&Sigma;}}_{i=1}^{N}f\left(y\right|P_{p,i})\frac{\mathrm{Pr}\left(P_{p,i}\right)}{\mathrm{Pr}\left(H_1\right)}---\left(5\right)$

Wherein Γ (.) represents gamma function, Pr (P _p,i) represent that PU is with power P _p,ithe prior probability of transmission, this probability meets pr (H ₀) and Pr (H ₁) be respectively the idle and busy probability of PU.

The optimum detector of likelihood ratio test is used to be written as:

$L\left(y\right)=\frac{f\left(H_1\right|y)}{f\left(H_0\right|y)}=\underset{i=1}{\overset{N}{\mathrm{\&Sigma;}}}{\left(\frac{N_0}{N_0+P_{p,i}{\mathrm{\&gamma;}}_1}\right)}^{\frac{M}{2}}\exp \left(\frac{y{P}_{p,i}{\mathrm{\&gamma;}}_1}{{2N}_0(N_0+P_{p,i}{\mathrm{\&gamma;}}_1)}\right)\frac{\mathrm{Pr}\left(P_{p,i}\right)}{\mathrm{Pr}\left(H_0\right)}---\left(6\right)$

Therefore, the perception of step S203 may further include by the energy of accumulation is compared with predefined threshold value the existence judging PU.

Due to the strictly increasing function that L (y) is y, wherein η is the hard decision rule of decision threshold be equivalent to wherein θ is new threshold value of equal value.Therefore optimum detector is energy detector.

Then, the probability that false-alarm (false alarm) and mistake (false) detect can be calculated as

$P_f\left(\mathrm{\&theta;}\right)=\mathrm{Pr}\left({\hat{H}}_0\right|H_0)={\&Integral;}_0^{+\&infin;}f\left(y\right|H_0)\mathrm{dy}=\frac{\mathrm{\&gamma;}(\frac{M}{2},\frac{0}{{2N}_0})}{\mathrm{\&Gamma;}\left(\frac{M}{2}\right)}---\left(7\right)$

With

$P_d\left(\mathrm{\&theta;}\right)=\mathrm{Pr}\left({\hat{H}}_1\right|H_1)=\underset{i=1}{\overset{N}{\mathrm{\&Sigma;}}}{\&Integral;}_0^{+\&infin;}f\left(y\right|P_{p,i})\mathrm{dy}\frac{\mathrm{Pr}\left(P_{p,i}\right)}{\mathrm{Pr}\left(H_1\right)}=\underset{i=1}{\overset{N}{\mathrm{\&Sigma;}}}\frac{\mathrm{\&gamma;}(\frac{M}{2},\frac{0}{{2N}_0+{2P}_{p,i}{\mathrm{\&gamma;}}_1})}{\mathrm{\&Gamma;}\left(\frac{M}{2}\right)}\frac{\mathrm{Pr}\left(P_{p,i}\right)}{\mathrm{Pr}\left(H_1\right)}---\left(8\right)$

Wherein γ (.) is the imperfect gamma function of low order.

A vital task of frequency spectrum perception detects the state of PU (idle or busy, namely not exist or exist).But, when multiple main through-put power, estimate that the main through-put power of through-put power and the main transmission of protection that can be used to judge SU is also significant.

Therefore, this at least one feature is in an embodiment of the main through-put power of PU wherein, the identification of step S204 may further include the definition multiple subspaces corresponding with multiple main through-put power, and by the energy comparison of cumlative energy and multiple subspace being estimated PU is using which main through-put power of multiple main through-put power.

This is that one typically supposes test problem more, and decision rule is

If $f\left(P_{p,i}\right|y)>f\left(P_{p,k}\right|y),\&ForAll;k\&NotEqual;i---\left(9\right)$

Set up, then judge it is P _p,i.Will (4) substitute into (9), we obtain for P _p,idecision space be

$S\left(P_{p,i}\right)=\left\{y\right|\underset{k\&Element;(1,i)}{\max }d(k,i)<y<\underset{k\&Element;(i,N)}{\min }d(k,i)\}---\left(10\right)$

Wherein

$d(k,i)=\frac{2(N_0+P_{p,i}{\mathrm{\&gamma;}}_1)(N_0+P_{p,k}{\mathrm{\&gamma;}}_1)}{P_{p,k}{\mathrm{\&gamma;}}_1-P_{p,i}{\mathrm{\&gamma;}}_1}\ln \left(\frac{\mathrm{Pr}\left(P_{p,i}\right)}{\mathrm{Pr}\left(P_{p,k}\right)}\frac{{(N_0+P_{p,k}{\mathrm{\&gamma;}}_1)}^{\frac{M}{2}}}{{(N_0+P_{p,i}{\mathrm{\&gamma;}}_1)}^{\frac{M}{2}}}\right)---\left(11\right)$

Note, in (10), if Pr is (P _p,i) very little, then can compare larger, and for Pr (P _p,i) decision space become sky.Make λ _i, i=1 ... K+1 is breakpoint (break point), and β _i, i=1 ... K is at interval [λ _i, λ _i+1) in correspondence power estimate.Obviously, we have K≤N.Definition λ ₁=0 and λ _k+1=+∞, then the estimation of main through-put power can be written as if y ∈ is [λ _i, λ _i+1), i=1 ..., K (12)

Problem changes how to determine optimum λ into _iand β _i.Following lemma is used to solve this problem.

Lemma 1: for any constant y ₁and y ₂if, y ₁<y ₂, i<k and f (P _{p, i}| y ₁) < f (P _{p, k}| y ₁), then we have f (P _{p, i}| y ₂) < f (P _{p, k}| y ₂).

Prove: first, Wo Menyou

$\begin{array}{c}f\left(P_{p,i}\right|y_2)=\frac{f\left(y_2\right|P_{p,i})\mathrm{Pr}\left(P_{p,i}\right)}{f\left(y_2\right)}=\frac{y^{\frac{M}{2}-1}{e}^{-\frac{y}{{2N}_0+{2\mathrm{\&gamma;}}_1{P}_{p,i}}}}{\mathrm{\&Gamma;}\left(\frac{M}{2}\right){({2N}_0+{2\mathrm{\&gamma;}}_1{P}_{p,i})}^{\frac{M}{2}}}\frac{\mathrm{Pr}\left(P_{p,i}\right)}{f\left(y_2\right)}\\ =f\left(P_{p,i}\right|y_1)\frac{f\left(y_1\right)}{f\left(y_2\right)}{e}^{-\frac{y_2-y_1}{{2N}_0+{3P}_{p,i}{\mathrm{\&gamma;}}_1}},\end{array}---\left(13\right)$

With $f\left(P_{p,k}\right|y_2)=f\left(P_{p,k}\right|y_1)\frac{f\left(y_1\right)}{f\left(y_2\right)}{e}^{-\frac{y_2-y_1}{{2N}_0+2P_{p,k}{\mathrm{\&gamma;}}_1}}.$ Due to i<k, therefore P _p,i<P _p,kset up.With f (P _{p, i}| y ₁) < f (P _{p, k}| y ₁) together, according to (13), we obtain f (P _{p, i}| y ₂) < f (P _{p, k}| y ₂) set up.

Infer from (11), d (k, i) represents f (P _{p, k}| y=f (P _{p, i}| some y y).Together with lemma 1, we can obtain λ as follows _iand β _ioptimal solution.First, at a y=0 place, calculate f (P _{p, k}| 0), k ∈ [1, N], select by the maximum represented, and set β ₁=P _p,i.As i=N, iteration stopping.The details of this algorithm is as table 1.

Table 1

Especially, for situation, first we have

$\frac{\&PartialD;d(k,i)}{{\&PartialD;P}_{p,k}}=\frac{M(N_0+P_{p,i}{\mathrm{\&gamma;}}_1)}{P_{p,k}-P_{p,i}}[1-\frac{N_0+P_{p,i}{\mathrm{\&gamma;}}_1}{P_{p,k}{\mathrm{\&gamma;}}_1-P_{p,i}{\mathrm{\&gamma;}}_1}\ln (1+\frac{P_{p,k}{\mathrm{\&gamma;}}_1-P_{p,i}{\mathrm{\&gamma;}}_1}{N_0+P_{p,i}{\mathrm{\&gamma;}}_1})].---\left(14\right)$

We can easily prove, set up, and set up.Therefore, if k>i, then set up, and d (k, i) is P _p,kincreasing function, otherwise d (k, i) is P _p,kdecreasing function.Then for such situation, (10) can be written as further

S(P _p，i)＝{y|d(i-1，i)＜y＜d(i+1，i)} (15)

And λ _j=d (j-1, i), β _j=P _{p, j}.

The performance that main through-put power is estimated is analyzed as follows.First we have

Then total detection probability can be written as

${\mathrm{Pr}}_e=\underset{i=1}{\overset{N}{\mathrm{\&Sigma;}}}\mathrm{Pr}\left({\hat{P}}_{p,i}\right|P_{p,i})\mathrm{Pr}\left(P_{p,i}\right)---\left(17\right)$

Have estimated the main through-put power of PU in step S205 place, when namely identifying the feature of the through-put power of PU, SU can determine its transformation parameter based on the feature identified of PU.

In the power allocation scheme of routine, for the situation of constant main through-put power, SU-Tx adjusts its through-put power by based on the judgement done during perception time slot.When PU be perceived as do not exist time, SU-Tx will with higher power transmission, otherwise, with lower-wattage transmission to reduce the interference caused by PU.

PU is with in the embodiments of the invention of multiple main through-put power transmission wherein, and for the given estimation of the main through-put power of PU, SU is by specific for use one secondary transmission power.In other words, multistage power division is proposed.Especially, if PU is perceived as there is not H ₀, then SU-Tx will with power P _{s, 0}transmit, PU is perceived as with through-put power P else if _p,jexist, then SU-Tx will with power P _s,jtransmit.Secondary transmission power can be distributed based on the main through-put power of estimated/identification by using at least one predetermined criterion.

This at least one predetermined criterion can comprise the one or more of the following: the average accessible speed of SU is maximized; Average secondary through-put power constraint under predefined power; And the average interference power under the maximum interference of PU is being retrained.Use description to the example determining optimum secondary transmission power below.

Due to the restriction of frequency spectrum perception technology, PU may be falsely detected or false-alarm may occur.Note, the non-existent situation of PU equals PU and transmits with power 0.Definition P _{p, 0}=0 is the non-existent situation of PU.Then, the instantaneous transmission speed of SU is provided by following formula

$R_{i,j}={\log }_2(1+\frac{g{P}_{\mathrm{\&beta;},j}}{N_0+P_{p,i}{\mathrm{\&gamma;}}_2})---\left(18\right)$

Wherein first index (index) is meant to real state, and second index is meant to result of determination, and i=0 ..., N, j=0 ... N.Subscript 0 is meant to the non-existent state of PU, therefore P _{p, 0}=0 and Pr (P _{p, 0})=Pr (H ₀).

Therefore, the average accessible speed of SU can be modeled as

$R=\underset{i=0}{\overset{N}{\mathrm{\&Sigma;}}}\underset{j=0}{\overset{N}{\mathrm{\&Sigma;}}}\mathrm{Pr}\left(P_{p,i}\right)R_{i,j}\mathrm{Pr}\left({\hat{P}}_{p,j}\right|P_{p,i})---\left(19\right)$

At parameter P _avunder average secondary through-put power constraint can be written as

$\underset{i=0}{\overset{N}{\mathrm{\&Sigma;}}}\underset{j=0}{\overset{N}{\mathrm{\&Sigma;}}}\mathrm{Pr}\left(P_{p,i}\right)P_{s,j}\mathrm{Pr}\left({\hat{P}}_{p,j}\right|P_{p,i})\&le;P_{\mathrm{av}}---\left(20\right)$

At H ₁lower SU produces interference to PU, and wherein main through-put power is P _p,i, i=1 ..., N and SU are with P _s,j, j=0 ..., any power delivery in N set.Therefore at maximum interference I _avunder average interference power constraint be modeled as

$\underset{i=0}{\overset{N}{\mathrm{\&Sigma;}}}\underset{j=0}{\overset{N}{\mathrm{\&Sigma;}}}\mathrm{hPr}\left(P_{p,i}\right)P_{s,j}\mathrm{Pr}\left({\hat{P}}_{p,j}\right|P_{p,i})\&le;I_{\mathrm{av}}---\left(21\right)$

Under constraint discussed above, the maximized optimization problem of accessible transmission rate of SU can be formulated as

$\begin{array}{cc}\underset{P_{p,j}}{\max }& R\end{array}$

$s.t.\left(20\right),\left(21\right),P_{s,j}\&GreaterEqual;0,\&ForAll;j---\left(22\right)$

Following lemma will be used.

Lemma 2: problem (22) is relative to through-put power P under the constraint of (20) and (21) _s,jit is convexity.

Prove: this proves usual, because

$\frac{{\&PartialD;}^{2}R}{{\&PartialD;}^2{P}_{a,j}}={\mathrm{\&Sigma;}}_{i=0}^N\frac{\mathrm{Pr}\left(P_{p,i}\right){\log }_2\left(e\right)\mathrm{Pr}\left({\hat{P}}_{p,i}\right|P_{p,i})}{{(P_{s,j}+(N_0+P_{p,i}{\mathrm{\&gamma;}}_1)/g)}^2}>0,\frac{{\&PartialD;}^{2}R}{\&PartialD;P_{s,j}\&PartialD;P_{s,j}}=0,\&ForAll;i\&NotEqual;j.$ And retraining (20) and (21) is all P _s,jlinear function, therefore problem (22) is about P _s,jit is convexity.

First, we with reference to the problem (22) under constraint (20) and (21), can build Lagrangian L (P _s,j, α, μ) be

$\begin{array}{c}L(P_{s,j},\mathrm{\&alpha;},\mathrm{\&mu;},)=\underset{i=0}{\overset{N}{\mathrm{\&Sigma;}}}\underset{j=0}{\overset{N}{\mathrm{\&Sigma;}}}\mathrm{Pr}\left(P_{p,i}\right)R_{i,j}\mathrm{Pr}\left({\hat{P}}_{p,j}\right|P_{p,i})+\mathrm{\&alpha;}(P_{\mathrm{av}}-\underset{i=0}{\overset{N}{\mathrm{\&Sigma;}}}\underset{j=0}{\overset{N}{\mathrm{\&Sigma;}}}\mathrm{Pr}\left(P_{p,i}\right)P_{s,j}\mathrm{Pr}\left({\hat{P}}_{p,j}\right|P_{p,i}))\\ +\mathrm{\&mu;}(I_{\mathrm{av}}-\underset{i=1}{\overset{N}{\mathrm{\&Sigma;}}}\underset{j=0}{\overset{N}{\mathrm{\&Sigma;}}}\mathrm{hPr}\left(P_{p,i}\right)P_{s,j}\mathrm{Pr}\left({\hat{P}}_{p,j}\right|P_{p,j}))\end{array}---\left(23\right),$

Wherein α, μ >=0 is the dual variable corresponding to (20) and (21).

Then, we can build Lagrangian double optimization problem and are

$\underset{\mathrm{\&alpha;}\&GreaterEqual;0,\mathrm{\&mu;}\&GreaterEqual;0}{\min }g(\mathrm{\&alpha;},\mathrm{\&mu;})\stackrel{\mathrm{\&Delta;}}{=}\underset{P_{s,j\&GreaterEqual;0}}{\sup }L(P_{s,j},\mathrm{\&alpha;},\mathrm{\&mu;})---\left(24\right)$

Adopt lemma 2, we the optimal value of inference problems (24) can equal problem (22).Therefore we can solve double optimization problem (24) instead of solve (22).From (24), we must obtain L (P _s,j, α, μ) supremum.For finding optimum P _s,j, we get L (P _{s, j}, α, μ) and relative to P _{s, j}derivative, it can obtain and be

$\begin{array}{c}\frac{\&PartialD;L(P_{a,j},\mathrm{\&alpha;},\mathrm{\&mu;})}{{\&PartialD;P}_{a,j}}=\underset{i=0}{\overset{N}{\mathrm{\&Sigma;}}}\frac{\mathrm{Pr}\left(P_{p,i}\right){\log }_2\left(e\right)\mathrm{Pr}\left({\hat{P}}_{p,j}\right|P_{p,i})}{P_{s,j}+(N_0+P_{p,i}{\mathrm{\&gamma;}}_2)/g}-\mathrm{\&alpha;}\underset{i=0}{\overset{N}{\mathrm{\&Sigma;}}}\mathrm{Pr}\left(P_{p,i}\right)\mathrm{Pr}\left({\hat{P}}_{p,j}\right|P_{p,i})\\ -\mathrm{\&mu;}\underset{i=1}{\overset{N}{\mathrm{\&Sigma;}}}\mathrm{hPr}\left(P_{p,i}\right)\mathrm{Pr}\left({\hat{P}}_{p,j}\right|P_{p,i}).\end{array}---\left(25\right)$

From (25), we can see p _s,jdecreasing function.If therefore have unique meet if namely

${\mathrm{\&Sigma;}}_{i=0}^N\frac{\mathrm{Pr}\left(P_{p,i}\right){\log }_2\left(e\right)\mathrm{Pr}\left(P_{p,j}\right|P_{p,i})}{(N_0+P_{p,i}{\mathrm{\&gamma;}}_2)/g}\&GreaterEqual;\mathrm{\&alpha;}{\mathrm{\&Sigma;}}_{i=0}^N\mathrm{Pr}\left(P_{p,i}\right)\mathrm{Pr}\left({\hat{P}}_{p,j}\right|P_{p,i})+\mathrm{\&mu;}{\mathrm{\&Sigma;}}_{i=1}^N\mathrm{hPr}\left(P_{p,i}\right)\mathrm{Pr}\left({\hat{P}}_{p,j}\right|P_{p,i})$ , then for given α and μ, optimal power allocation is otherwise P _{s, j}=0.

Then need to find the optimal value of Lagrange's multiplier α and μ to obtain optimum power distribution strategies P _s,j.Here use the method based on subgradient to find optimal solution, wherein subgradient is provided by following proposition, such as, and ellipsoid method and Newton method.

Proposition 1: the subgradient of Lagrangian bifunction g (α, μ) is [C, D], wherein

$C=P_{\mathrm{av}}-{\mathrm{\&Sigma;}}_{i=0}^N{\mathrm{\&Sigma;}}_{j=0}^N\mathrm{Pr}\left(P_{p,i}\right){\hat{P}}_{s,j}\mathrm{Pr}\left({\hat{P}}_{p,j}\right|P_{p,i})$ And

$C=I_{\mathrm{av}}-{\mathrm{\&Sigma;}}_{i=1}^N{\mathrm{\&Sigma;}}_{j=1}^N\mathrm{hPr}\left(P_{p,i}\right){\hat{P}}_{s,j}\mathrm{Pr}\left({\hat{P}}_{p,j}\right|P_{p,i})$

it is the optimal power allocation for fixing α and μ.

Prove: will with be expressed as the feasible value of bifunction g (α, μ), and it is corresponding optimal power allocation.Prove that proposition 1 equals to prove $g(\widetilde{\mathrm{\&alpha;}},\widetilde{\mathrm{\&mu;}})\&GreaterEqual;g(\mathrm{\&alpha;},\mathrm{\&mu;})+\left(\right[\widetilde{\mathrm{\&alpha;}},\widetilde{\mathrm{\&mu;}}]-[\mathrm{\&alpha;},\mathrm{\&mu;}){[C,D]}^T$ For arbitrarily with set up.Therefore, Wo Menyou

$\begin{array}{c}g(\widetilde{\mathrm{\&alpha;}},\widetilde{\mathrm{\&mu;}})=\underset{P_{a,j}}{\sup }L(P_{a,j},\widetilde{\mathrm{\&alpha;}},\widetilde{\mathrm{\&mu;}})=L({\stackrel{\&OverBar;}{P}}_{s,j},\widetilde{\mathrm{\&alpha;}},\widetilde{\mathrm{\&mu;}})\\ \&GreaterEqual;L({\stackrel{\&OverBar;}{P}}_{s,j},\widetilde{\mathrm{\&alpha;}},\widetilde{\mathrm{\&mu;}})\\ =L({\stackrel{\&OverBar;}{P}}_{s,j},\mathrm{\&alpha;},\mathrm{\&mu;})+\left(\right[\widetilde{\mathrm{\&alpha;}},\widetilde{\mathrm{\&mu;}}]-[\mathrm{\&alpha;},\mathrm{\&mu;}\left]\right){[C,D]}^T\\ =g(\mathrm{\&alpha;},\mathrm{\&mu;})+\left(\right[\widetilde{\mathrm{\&alpha;}},\widetilde{\mathrm{\&mu;}}]-[\mathrm{\&alpha;},\mathrm{\&mu;}){[C,D]}^T.\end{array}---\left(26\right)$

Determined (multiple) transformation parameter (such as secondary transmission power) of SU, then SU can communicate with determined (multiple) transformation parameter.

Simulation result display is as maximum interference I _avtime low, the Performance Ratio conventional scheme of the scheme proposed better.But, as maximum interference I _avtime large, the performance of the scheme proposed is better than conventional scheme significantly.Simulation result is also shown in real system, and first SU should detect main signal and identify (multiple) feature of PU, then distributes owing to the different through-put power of institute's recognition feature or due to modulation and encoding scheme (MCS).By this, SU can protect main transmission and improve its accessible speed.It will be appreciated by those skilled in the art that proposed scheme can be extended to wherein SU and can detect the feature of PU and determine that the transformation parameter of himself is to protect the QoS of PU and to improve other situations of its throughput.Such as, SU can detect the frequency of operation of PU, modulation and encoding scheme and/or main through-put power, and then determines the transformation parameter of himself.The scheme proposed can be used in the environment of wherein multiple coexistence of communication systems, to reduce the interference in user.Such as, Femto cell can be detected LTE system and determine its suitable transformation parameter.

Fig. 3 can be configured to put into practice the schematic block diagram according to the device 300 of the exemplary embodiment of the embodiment of the present invention.This device 300 can be incorporated in the secondary nodes in CR network, and is configured to perform the method as with reference to the exemplary embodiment of the present invention illustrated in figure 2.

As shown in Figure 3, device 300 can comprise receiving element 310, perception unit 320, recognition unit 330 and determining unit 340.

Receiving element 310 is configured to from PU Received signal strength sequence.Such as, receiving element 310 can receive M symbol in the frame with perception time slot.In considered scene, PU can with multiple main through-put power transmission.

Perception unit 320 is configured the state carrying out perception PU based on received burst.The state of PU can be presence or absence.In one embodiment, perception is performed based on energy measuring.In such embodiments, perception unit 320 comprises computing unit 321 and identifying unit 322 further.Computing unit 321 is configured to the cumlative energy calculating received signal sequence, as shown in formula (2).Identifying unit 322 is configured to by accumulated energy is compared with predefined threshold value the existence judging PU, described by reference formula (6).

Recognition unit 330 is configured at least one feature identifying PU based on the state of the PU of institute's perception and burst.This at least one feature can comprise main through-put power and at least one in modulation and encoding scheme (MCS).Certainly, such as, if need can identify the further feature relevant with the transformation parameter of PU, the frequency of operation of PU, the communication protocol etc. of PU.

The feature of PU is that in an embodiment of main through-put power, recognition unit 330 may further include definition unit 330 and estimation unit 332 wherein.Definition unit 331 is configured to define the multiple subspaces corresponding with multiple main through-put power, described by formula (10).Estimation unit 332 is configured to by the energy comparison of cumlative energy and multiple subspace being estimated PU is using which main through-put power of multiple main through-put power.

Determining unit 340 is configured at least one transformation parameter determining SU based at least one feature of PU.In one embodiment, this at least one transformation parameter is the secondary transmission power of SU.Then, determining unit 340 may further include allocation units 341, and it is configured to by using at least one predetermined criterion to come for SU distributes secondary transmission power based on the main through-put power of estimated PU.

For the given estimation of the main through-put power of PU, a specific secondary transmission power will be assigned to SU.

It is one or more that this at least one predetermined criterion can comprise in the following: the average accessible speed of SU is maximized; Average secondary through-put power constraint under predefined power; And the average interference power under the maximum interference of host node is being retrained.

Then, SU can use determined transformation parameter for communication.

It should be understood that the unit be included in device 300 is arranged to and put into practice exemplary embodiment of the present invention.Therefore, be also applicable to the unit of device 300 and Qi Nei about the operation described above of Fig. 2 and feature, and omit their detailed description here.

Fig. 4 illustrates the simplified block diagram be suitable at the network node 400 (such as secondary nodes) put into practice in the CR network used in exemplary embodiment of the present invention.

As shown in Figure 4, network node 400 comprises data processor (DP) 401, is coupled to the memory (MEM) 402 of DP 401 and is coupled to suitable RF reflector TX and the receiver RX 404 of DP 401.MEM 402 storage program (PROG) 403.TX/RX 404 is for the two-way wireless communication with other network nodes.Such as, TX/RX 404 receiving element 310 that can implement Fig. 3 is in order to from host node (i.e. PU) Received signal strength sequence.

PROG 403 is assumed that and comprises program command, and when being performed by the DP 401 be associated, program command makes network node 400 can operate, as what discuss together with the method shown in Fig. 2 according to exemplary embodiment of the present invention herein.Such as, PROG 403 and DP 401 can implement perception unit 320, recognition unit 330 and determining unit 340 in order to perform corresponding function.

Embodiments of the invention can be implemented or implement by hardware or by the combination of software and hardware by the computer software performed by the DP 401 of network node 400.

MEM 402 can be any type being suitable for local technical environment, and any suitable data storage technology can be used to implement, such as the memory device of the based semiconductor of non-limiting example, magnetic storage apparatus and system, light storage device and system, read-only storage and removable memory.Although an only MEM shown in network node 400, can have several physically different memory cell in network node 400.DP 401 can be any type being suitable for local technical environment, and can comprise all-purpose computer, special-purpose computer, microprocessor, digital signal processor (DSP) and one or more based in the processor of polycaryon processor framework as non-limiting example.Network node 400 can have multiple processor, and be such as such as dedicated IC chip, it is driven in the clock synchronous with primary processor in time.

Below the block diagram of reference method and device and flow chart diagram describe exemplary embodiment of the present invention.Be understood that the combination of block diagram and the illustrated each frame of flow chart and block diagram and the illustrated frame of flow chart can be implemented by the various devices comprising computer program instructions respectively.These computer program instructions can be loaded into produce a kind of machine on all-purpose computer, special-purpose computer or other programmable data processing unit, so that the instruction performed on computer or other programmable data processing unit creates for implementing function specified in one or more flow chart box.

Aforesaid computer program instructions can be such as subroutine and/or function.Computer program in one embodiment of the invention comprises at least one computer readable memory medium, and this computer readable memory medium stores aforesaid computer program instructions.Computer readable memory medium can be the electronic memory device of such as CD or picture RAM (random access storage device) or ROM (read-only memory).

Although this specification comprises many particular implementation details; these should not be construed as the restriction in any execution mode scope; or to the restriction to content required for protection, and should as may specific to the description of the feature of the specific embodiment of particular implementation.Some feature in this manual described in the background of the embodiment be separated also can be implemented in combination in single embodiment.On the contrary, the various feature described in the background of single embodiment also can be implemented discretely in many embodiment: or be implemented with any suitable sub-portfolio.And; do although feature may be described to be combined into action with some above; and even so initially by so claimed; one or more features from combination required for protection can be deleted by from combination in some cases, and combination required for protection can the modification of directed sub-portfolio or sub-portfolio.

Also it should be noted that above-described embodiment is presented for describing instead of restriction the present invention, and should be understood that as readily understood by the skilled person when without departing from the spirit and scope of the present invention, can modifications and variations be taked.These modifications and variations be considered be in the present invention and claims scope in.Protection scope of the present invention is limited by claims.In addition, any Reference numeral in detail in the claims should not be construed as limitations on claims.Verb " comprises " and the existence of element outside those elements or step of stating in claim or step is not got rid of in its paradigmatic use.Indefinite article " one " before element or step does not get rid of the existence of multiple such element or step.

Claims (12)

1., for a method for cognitive radio (CR) network, comprising:

From the host node Received signal strength sequence with multiple main through-put power;

The state of host node described in perception is carried out based on described burst;

At least one feature of described host node is identified based on the described state of described host node and described burst; And

At least one transformation parameter of secondary nodes is determined based at least one feature described.

2. method according to claim 1, at least one feature wherein said comprise main through-put power and modulation and encoding scheme (MCS) at least one.

3. method according to claim 1 and 2, wherein saidly determine to comprise:

By use at least one predetermined criterion based on described at least one be characterized as described secondary nodes and distribute secondary transmission power.

4. method according to claim 3, it is one or more that at least one predetermined criterion wherein said comprises in the following:

The average accessible speed of described secondary nodes is maximized;

Average secondary through-put power constraint under predefined power; And

Average interference power constraint under the maximum interference to described host node.

5. the method according to any one in claim 1-4, wherein said identification comprises:

Define the multiple subspaces corresponding with described multiple main through-put power; And

By the energy comparison of the cumlative energy of described burst and described multiple subspace is estimated described host node is using which the main through-put power in described multiple main through-put power.

6. the method according to any one of claim 1-5, wherein said perception comprises:

Calculate the cumlative energy of described burst; And

By described cumlative energy is compared with predefined threshold value the existence judging described host node.

7., for a device for cognitive radio (CR) network, comprising:

Receiving element, is arranged to the host node Received signal strength sequence from having multiple main through-put power;

Perception unit, is arranged to the state carrying out host node described in perception based on described burst;

Recognition unit, is arranged at least one feature identifying described host node based on the described state of described host node and described burst; And

Determining unit, is arranged at least one transformation parameter determining secondary nodes based at least one feature described.

8. device according to claim 7, at least one feature wherein said comprise main through-put power and modulation and encoding scheme (MCS) at least one.

9. the device according to claim 7 or 8, wherein said determining unit comprises:

Allocation units, be arranged to by use at least one predetermined criterion based on described at least one be characterized as described secondary nodes and distribute secondary transmission power.

10. device according to claim 9, it is one or more that at least one predetermined criterion wherein said comprises in the following:

The average accessible speed of described secondary nodes is maximized;

Average secondary through-put power constraint under predefined power; And

Average interference power constraint under the maximum interference to described host node.

11. devices according to any one of claim 7-10, wherein said recognition unit comprises:

Definition unit, is arranged to multiple subspaces that definition is corresponding with described multiple main through-put power; And

Estimation unit, is arranged to by the energy comparison of the cumlative energy of described burst and described multiple subspace is estimated described host node is using which the main through-put power in described multiple main through-put power.

12. devices according to any one of claim 7-10, wherein said perception unit comprises:

Computing unit, is arranged to the cumlative energy calculating described burst; And

Identifying unit, is arranged to by described cumlative energy is compared with predefined threshold value the existence judging described host node.