CN104703246A - Quick and slow link adaptive fusion based relay selection method - Google Patents

Abstract

The invention discloses a quick and slow link adaptive fusion based relay selection method. The method comprises safe decoding-and-forwarding collaborative network relay selection and ordinary collaborative network relay selection. The method is characterized in that a channel state change frequency division method is put forward, a quick or slow link adaptive mode is selected according to the method, and accordingly, the system performance is improved, and the system complexity is reduced; (2) considering that reality system users link adaptive transmission generally, the method can be applied to reality systems and has realistic feasibility; (3) the secrecy capacity and whether the switching frequency between interference nodes is used are considered in security collaboration network relay selection, the relay selection is reasonable and meaningful by establishing relay selection in ordinary collaborative network on the basis of determining towards actual channels.

Description

Based on the relay selection method that fast, slow speed link self adaptation merge

Technical field

The present invention relates to mobile communication technology field, be specifically related to a kind of relay selection method merged based on fast, slow speed link self adaptation.

Background technology

In security cooperation network, due to the character of the wireless medium of broadcast, all users in system Covering domain can eavesdrop source information.Performance and the protection communication link of information privacy in communication are not eavesdropped by possible listener-in, become more and more important.Information privacy in conventional wireless network is the field of higher-layer protocols, and access to your password security scheme.But the high-rise information security realized is in the face of the challenge of ever-increasing potential risk.So in order to improve the security performance of mobile communication system further, people start to pay close attention to physical layer information safety problem.Hurry up, slow speed link adaptive mode may be used to security cooperation network and common collaborative network, but in the relay selection of two kinds of networks, all do not consider impact that is fast, slow speed link adaptive technique at present.

Summary of the invention

The invention provides a kind of relay selection method merged based on fast, slow speed link self adaptation, the method is respectively for security cooperation network and common collaborative network, on the basis that fast, slow speed link self adaptation merge, respective relay selection method is provided respectively, have effectively achieved the relay selection of the cooperation communication system adopting link circuit self-adapting transmission means.Feature of the present invention is: (1) proposes a kind of method distinguishing channel state variations frequency, channel conditions change is more frequent, can be considered that channel conditions is more undesirable, quick or the slow speed link adaptive mode according to the method choice for use, effectively achieve the fusion of two class link circuit self-adaptings in practical communication scene, thus improve the performance of system to a certain extent and reduce the complexity of system.(2) relay selection method is in the past different from, relay selection method that the present invention carries considers link circuit self-adapting transmission means, in view of reality system generally adopts link circuit self-adapting to transmit, therefore institute of the present invention extracting method can be used for reality system, has realistic feasibility.(3) taken into account secrecy capacity and whether used the switching frequency between interfering nodes in security cooperation network trunk is selected, relay selection is based upon on the judgement basis to actual channel in common collaborative network, make relay selection more reasonable, there is realistic meaning.

Set forth technical scheme of the present invention below.

Based on the relay selection method that fast, slow speed link self adaptation merge, described method comprises the safe relay selection of decoding forward collaboration network and the relay selection of common collaborative network.The relay selection of described safe decoding forward collaboration network, comprises the steps:

(1) choose two via nodes, a via node R is used for normal transmission data, and another node J is as interfering nodes.

(2) under the prerequisite of interfering nodes set non-NULL, if use interfering nodes, then:

$C_S(i_Y^{*},J^{*})=\arg \underset{J\∈J_{\mathrm{set}}}{\underset{i\∈C}{\underset{i\≠J}{\max }}}\left\{C_S(i,J)\right\}=\arg \underset{J\∈J_{\mathrm{set}}}{\underset{i\∈C}{\underset{i\≠J}{\max }}}\left\{\frac{1}{2}{\log }_2\left(\frac{1+\frac{{\mathrm{\γ}}_{i,D}^W}{1+{\mathrm{\γ}}_{J,D}^W}}{1+\frac{1+{\mathrm{\γ}}_{i,E}^W}{{\mathrm{\γ}}_{J,E}^W}}\right)\right\};$

If do not use interfering nodes, then:

$C_S\left(i_N^{*}\right)=\arg \underset{i\∈C_d}{\max }\left\{C_S\left(i\right)\right\}=\arg \underset{i\∈C_d}{\max }\left\{\frac{1}{2}{\log }_2\left(\frac{1+{\mathrm{\γ}}_{i,D}^W}{1+{\mathrm{\γ}}_{i,E}^W}\right)\right\};$

Wherein: candidate relay node when using interfering nodes and interfering nodes respectively; it is candidate relay node when not using interfering nodes; represent Instant SNR or the average signal-to-noise ratio of link between node i to destination node D; represent Instant SNR or the average signal-to-noise ratio of link between node i to eavesdropping node E; represent Instant SNR or the average signal-to-noise ratio of link between interfering nodes j to destination node D; represent Instant SNR or the average signal-to-noise ratio of link between interfering nodes j to eavesdropping node E; I belongs to set of relay nodes C _d, C _dfor can decoding success forward from the set of relay nodes of source node signal; Interfering nodes J belongs to interfering nodes set J _set, and J ≠ i, interfering nodes set J _setby forming as the via node of interfering nodes.

(3) if then use interfering nodes, and choose and J ^*respectively as via node and interfering nodes; If then do not use interfering nodes, and choose as via node.

The relay selection of described common collaborative network, comprises the steps:

(1) the i-th (i=1,2 are obtained by estimation ... N _d) individual via node to the Instant SNR sample of destination node D link is ask for the average of this link Instant SNR sample respectively variance the degree of bias and kurtosis

(2) suppose that this link channel type is φ, estimate channel parameter, the Instant SNR probability density function of all kinds channel wherein, N _i,D, for describing parameter during different channels type; Obtain supposing that channel type is the average of the channel Instant SNR of φ according to Instant SNR probability density function variance the degree of bias and kurtosis

(3) calculate:

${\mathrm{\ϵ}}_{\mathrm{\φ}}=\frac{{[V_1^X-v_{\mathrm{\φ},1}^Y]}^2}{v_2^X}+\frac{1}{2}{[1-\frac{v_{\mathrm{\φ},2}^Y}{v_2^X}]}^2+\frac{1}{6}{(v_3^X-v_{\mathrm{\φ},3}^Y)}^2+\frac{1}{24}{(v_4^X-v_{\mathrm{\φ},4}^Y)}^2;$

Minimum ε _φcorresponding hypothesis channel type is the actual channel type of this link, and determines channel parameter N _i,D, with

(4) outage probability during this link employing slow speed link self adaptation:

This link adopts outage probability during rapid link self adaptation:

Wherein: M is the progression of M-QAM modulation, gauss-Q function, represent the maximum integer being less than or equal to x; H represents bit number, k represents dimension, l represents the number of Gamma component, l=1 ..., N _i,D.

(5) basis select via node; Wherein: R ^*for selected via node; When link adopts rapid link self adaptation, when link adopts slow speed link self adaptation, $P_{\mathrm{out}}^Z\left({\mathrm{\γ}}_{i,D}^W\right)=P_{\mathrm{out}}^{\mathrm{SAM}}\left({\mathrm{\γ}}_{i,D}^S\right).$

Embodiment

Each link adopts fast or Adaptive Transmission mode at a slow speed, and rapid link self adaptation is according to Instant SNR determination modulation system, and slow speed link self adaptation is according to average signal-to-noise ratio determination modulation system.According to judging channel, frequent variations choice for use is fast or slow speed link adaptive technique.In view of the close ties of translational speed and channel variation frequency, when destination node translational speed is υ _d≤ υ ₀time, judge that this link channel conditions is non-frequent variations, be suitable for rapid link adaptive technique; Work as υ ₀< υ _d≤ υ ₁time, judge that this link channel conditions is common variation, link circuit self-adapting mode corresponding when making channel security capacity maximum choose the prerequisite only changing link circuit self-adapting mode in security cooperation network under, link circuit self-adapting mode corresponding when making outage probability minimum choose the prerequisite only changing link circuit self-adapting mode in common collaborative network under; Work as υ _d> υ ₁time, judge that this link channel conditions is frequent variations, adopt slow speed link adaptive technique.Wherein: υ _dfor node motion speed; υ ₀be the first thresholding speed, 10km/h≤υ ₀≤ 30km/h; υ ₁be the second thresholding speed, 40km/h≤υ ₁≤ 60km/h.With represent Instant SNR (W=F) or the average signal-to-noise ratio (W=S) of link (being designated as i → j link) between node i to node j.

(1) safe decoding forward collaboration network trunk is selected

All users in security cooperation network can eavesdrop source information, when information source S (Source) sends data to destination node D (Destination), can be ravesdropping node E (Eavesdropper) eavesdropping.Choose two via nodes, a via node R (Relay) is used for normal transmission data, and another node J (Jammer) is as interfering nodes, and the signal sent from J is interference signal, plays the effect of interference.Because interference is always not favourable to system, so in order to overcome this restriction, proposition system can switch at use interfering nodes and not using between interfering nodes.

represent Instant SNR or the average signal-to-noise ratio of link i → D, represent Instant SNR or the average signal-to-noise ratio of link i → E, represent Instant SNR or the average signal-to-noise ratio of link J → D, represent Instant SNR or the average signal-to-noise ratio of link J → E.Wherein i belongs to set of relay nodes C _d, C _dfor can decoding success forwarding from the set of relay nodes of source node signal.J belongs to interfering nodes set J _set, and J ≠ i, interfering nodes set J _setby forming as the via node of interfering nodes.When time, represent that " just disturbing " (J is to interference of E) is greater than " negative interference " (J is to interference of D), when state also continues for some time t (t > t ₀) time, this node can as interfering nodes; Otherwise this node not can be used as interfering nodes.As interfering nodes set J _setwhen changing, relay selection need be re-started.

(1) under the prerequisite of interfering nodes set non-NULL, if use interfering nodes:

$C_S(i_Y^{*},J^{*})=\arg \underset{J\&Element;J_{\mathrm{set}}}{\underset{i\&Element;C}{\underset{i\&NotEqual;J}{\max }}}\left\{C_S(i,J)\right\}=\arg \underset{J\&Element;J_{\mathrm{set}}}{\underset{i\&Element;C}{\underset{i\&NotEqual;J}{\max }}}\left\{{\left[\frac{1}{2}{\log }_2\left(\frac{1+\frac{{\mathrm{\&gamma;}}_{i,D}^W}{1+{\mathrm{\&gamma;}}_{J,D}^W}}{1+\frac{1+{\mathrm{\&gamma;}}_{i,E}^W}{{\mathrm{\&gamma;}}_{J,E}^W}}\right)\right]}^{+}\right\}---\left(1\right)$

candidate relay node when using interfering nodes and interfering nodes.Wherein

Due to the uncertainty of eavesdropping node E, and the signal to noise ratio of not all link all can be estimated to obtain, so eavesdropping node can be divided into determine that position eavesdropping node and undefined position eavesdrop node two kinds.For determining position eavesdropping node, with can estimate to obtain; For undefined position eavesdropping node, ${\mathrm{\&gamma;}}_{i,E}^W=E\left|{\mathrm{\&gamma;}}_{i,E}^W\right|=P^{\left(R\right)}{\mathrm{\&sigma;}}^2,{\mathrm{\&gamma;}}_{J,E}^W=E\left|{\mathrm{\&gamma;}}_{J,E}^W\right|=P^{\left(J\right)}{\mathrm{\&sigma;}}^2,$ Wherein P ^(R), P ^(J)be respectively R → E, the through-put power of J → E link, σ ²for white Gaussian noise variance; with still can be obtained by estimation.

(2) if do not use interfering nodes:

$C_S\left(i_N^{*}\right)=\arg \underset{i\&Element;C_d}{\max }\left\{C_S\left(i\right)\right\}=\arg \underset{i\&Element;C_d}{\max }\left\{{\left[\frac{1}{2}{\log }_2\left(\frac{1+{\mathrm{\&gamma;}}_{i,D}^W}{1+{\mathrm{\&gamma;}}_{i,E}^W}\right)\right]}^{+}\right\}---\left(2\right)$

it is candidate relay node when not using interfering nodes.For determining position eavesdropping node, can estimate to obtain; For undefined position eavesdropping node, still can be obtained by estimation.

Relay selection algorithm is as follows:

If then use interfering nodes, choose and J ^*respectively as via node and interfering nodes; If then do not use interference joint, choose as via node.

(2) common collaborative network relay selection.

By estimating to obtain the i-th (i=1,2 ... N _d) individual via node to the Instant SNR sample of destination node D link is

The average of this link Instant SNR sample:

$v_1^X=\frac{{\mathrm{\&gamma;}}_{i,D}^{\left(1\right)}+{\mathrm{\&gamma;}}_{i,D}^{\left(2\right)}+...{\mathrm{\&gamma;}}_{i,D}^{\left(n\right)}}{n}---\left(3\right)$

The variance of this link Instant SNR sample is:

$v_2^X=\frac{\underset{k=1}{\overset{n}{\mathrm{\&Sigma;}}}{[{\mathrm{\&gamma;}}_{i,D}^{\left(k\right)}-v_1^X]}^2}{n-1}---\left(4\right)$

The degree of bias of this link Instant SNR sample is:

$v_3^X=\frac{1}{(n-1){\left(v_2^X\right)}^{\frac{3}{2}}}\underset{k=1}{\overset{n}{\mathrm{\&Sigma;}}}{({\mathrm{\&gamma;}}_{i,D}^{\left(k\right)}-v_1^X)}^3---\left(5\right)$

The kurtosis of this link Instant SNR sample is:

$v_4^X=\frac{1}{(n-1){\left(v_2^X\right)}^2}\underset{k=1}{\overset{n}{\mathrm{\&Sigma;}}}{({\mathrm{\&gamma;}}_{i,D}^{\left(k\right)}-v_1^X)}^4---\left(6\right)$

Suppose that channel type is φ, estimate channel parameter, method of estimation has implementation method, repeats no more.Mixing gamma distribution is adopted to describe the Instant SNR probability density function of various channel $f_{\mathrm{\&phi;}}\left({\mathrm{\&gamma;}}_{i,D}^F\right)=\underset{l=1}{\overset{N_{i,D}}{\mathrm{\&Sigma;}}}{\mathrm{\&alpha;}}_{i,D}^l{\left({\mathrm{\&gamma;}}_{i,D}^F\right)}^{{\mathrm{\&beta;}}_{i,D}^l-1}{e}^{-{\mathrm{\&xi;}}_{i,D}^l{\mathrm{\&gamma;}}_{i,D}^F},$ Wherein, N _i,D, for parameter.When this distribution describes different channels, parameter has different expression formulas.Such as, when channel is Rayleigh channel, N _i,D=1, wherein for i → D link average signal-to-noise ratio; When channel is Nakagami-m channel, N _{i, D}=1, ${\mathrm{\&alpha;}}_{i,D}^l=\frac{m^m}{\mathrm{\&Gamma;}\left(m\right){\left({\mathrm{\&gamma;}}_{i,D}^S\right)}^m},{\mathrm{\&beta;}}_{i,D}^l=m,{\mathrm{\&xi;}}_{i,D}^l=\frac{m}{{\mathrm{\&gamma;}}_{i,D}^S},$ Wherein m is the channel parameter can estimating to obtain, for i → D link average signal-to-noise ratio.

The Instant SNR average supposing channel model can be obtained according to instant channel signal to noise ratio probability density function and the Instant SNR variance of hypothesis channel model the Instant SNR degree of bias $v_{\mathrm{\&phi;},3}^Y=E{\left[\frac{{\mathrm{\&gamma;}}_{i,D}^F-v_{\mathrm{\&phi;},1}^Y}{\sqrt{v_{\mathrm{\&phi;},2}^Y}}\right]}^3$ With Instant SNR kurtosis $v_{\mathrm{\&phi;},4}^Y=E{\left[\frac{{\mathrm{\&gamma;}}_{i,D}^F-v_{\mathrm{\&phi;},1}^Y}{\sqrt{v_{\mathrm{\&phi;},2}^Y}}\right]}^4.$ Wherein φ is respectively different channel types, as Rayleigh channel, and K channel, K _gwhen channel, η-μ channel, Nakagami channel etc., calculate corresponding respectively with then calculate:

${\mathrm{\&epsiv;}}_{\mathrm{\&phi;}}=\frac{{[V_1^X-v_{\mathrm{\&phi;},1}^Y]}^2}{v_2^X}+\frac{1}{2}{[1-\frac{v_{\mathrm{\&phi;},2}^Y}{v_2^X}]}^2+\frac{1}{6}{(v_3^X-v_{\mathrm{\&phi;},3}^Y)}^2+\frac{1}{24}{(v_4^X-v_{\mathrm{\&phi;},4}^Y)}^2---\left(7\right)$

ε _φless, illustrate that the statistical property of Instant SNR sample is more identical with the statistical property of hypothesis channel φ, so minimum ε _φcorresponding hypothesis channel is the actual channel of this link.N _i,D, parameter adopts the parameter of this channel model.

Based on the result that channel judges, obtain this link and adopt as follows respectively with outage probability during rapid link self adaptation at a slow speed:

Wherein: M is the progression of M-QAM modulation; it is Gauss-Q function; represent the maximum integer being less than or equal to x; H represents bit number, k represents dimension, l represents the number of Gamma component, l=1 ..., N _i,D.

In sum, relay selection algorithm is as follows:

$R^{*}=\arg \min \left\{P_{\mathrm{out}}^Z\left({\mathrm{\&gamma;}}_{i,D}^W\right)\right\}---\left(8\right)$

R ^*for selected via node, when link adopts rapid link self adaptation, when link adopts slow speed link self adaptation,

Provide embodiment below in order to effect of the present invention to be described.

(1) relay selection of security cooperation network.

Have 5 via nodes in base station coverage field can receive and forwarding information, be respectively R ₁, R ₂, R ₃, R ₄, R ₅, destination node is D, and eavesdropping node is E.System is known eavesdropping node location, and each link signal to noise ratio is estimated to obtain by adaptive technique.

Get v ₀=10km/h, v ₁=60km/h.The screen v of destination node D _d=80km/h, eavesdropping node E is stationary node, i.e. v _e=0km/h.Due to v _d>v ₁, each via node is frequent to the link channel change of destination node D, with slow adaptive technique (SAM); Due to v _e<v ₀, each via node changes infrequently, with fast adaptive technique (FAM) to the link channel of interfering nodes E.

Estimate the signal to noise ratio obtaining i → j link (dB) as follows:

R ₁→ D link:

R ₂→ D link:

R ₃→ D link:

R ₄→ D link:

R ₅→ D link:

R ₁→ E link:

R ₂→ E link:

R ₃→ E link:

R ₄→ E link:

R ₅→ E link:

Get t ₀=0.1s.

For R ₁node, due to then R ₁can not as interfering nodes.

For R ₂node, due to and the time t of state continuance ₂>t ₀, then R ₂can as interfering nodes.

For R ₃node, due to and the time t of state continuance ₃>t ₀, then R ₃can as interfering nodes.

For R ₄node, due to then R ₄can not as interfering nodes.

For R ₅node, due to and the time t of state continuance ₅>t ₀, then R ₅can as interfering nodes.

From the above, can be J as the set of interfering nodes _set∈ { R ₂, R ₃.

Interfering nodes set is nonempty set, when using interfering nodes, and will bring C into _s(R, J), secrecy capacity is as follows:

C _S(R ₁,R ₂)＝0.3925；

C _S(R ₁,R ₃)＝0.3942；

C _S(R ₂,R ₃)＝0；

C _S(R ₃,R ₂)＝0.3070

C _S(R ₄,R ₂)＝0.3390；

C _S(R ₄,R ₃)＝0.3552；

C _S(R ₅,R ₂)＝0.02.3；

C _S(R ₅,R ₃)＝0；

Then $(R^{*},J^{*})=\arg \underset{R\&NotEqual;J}{\max }\left\{C_S(R,J)\right\}=(R_1,R_3)=0.3942,$ Namely R is chosen respectively ₁, R ₃for candidate relay node during use interfering nodes and candidate's interfering nodes.

When not using interfering nodes, will bring C into _s(R), secrecy capacity is as follows:

C _S(R ₁)＝0.4037；

C _S(R ₂)＝0；

C _S(R ₃)＝0；

C _S(R ₄)＝0.4312；

C _S(R ₅)＝0；

Then namely R is chosen ₄for not using candidate relay node during interfering nodes.

Due to max (C _s(R ₁, R ₃), C _s(R ₄))=C _s(R ₄), then do not use interfering nodes, choose R ₄for via node.

(2) relay selection of common collaborative network.

Have 2 via nodes in base station coverage field can receive and forwarding information, be respectively R ₁, R ₂.System is known eavesdropping node location, and each link signal to noise ratio is estimated to obtain by adaptive technique.

Get v ₀=10km/h, v ₁=60km/h.The screen v of destination node D _d=80km/h, each via node is frequent variations to the link channel complexity of destination node D, estimates link signal to noise ratio with SAM.Each via node obtains n sample to the signal to noise ratio of destination node, composition stochastic variable (dB) as follows:

(1) R ₁→ D link:

R ₁the average of → D link signal to noise ratio sample, variance, the degree of bias and kurtosis as follows:

$v_1^{X_1}=984.5834;$

$v_2^{X_1}=902580;$

$v_3^{X_1}=1.0098;$

$v_4^{X_1}=2.8231.$

Rayleigh channel average, variance, the degree of bias and kurtosis as follows:

Then ε _rayleigh=0.0107.

Estimate obtain Rice factor n=8, Rice channel average, variance, the degree of bias and kurtosis as follows:

Then ε _{lai Si}=0.8659.

Then ε _rayleigh< ε _{lai Si}, channel is Rayleigh channel, so N=1, outage probability $P_{\mathrm{out},R1}^{\mathrm{SAM}}=0.0054.$

(2) R ₂→ D link:

R ₂the average of → D link signal to noise ratio sample, variance, the degree of bias and kurtosis as follows:

$v_1^{X_2}=1142;$

$v_2^{X_2}=1213000;$

$v_3^{X_2}=1.0882;$

$v_4^{X_2}=2.8966.$

Rayleigh channel average, variance, the degree of bias and kurtosis as follows:

Then ε _rayleigh=0.0369.

Estimate obtain Rice factor n=8, Rice channel average, variance, the degree of bias and kurtosis as follows:

Then ε _{lai Si}=1.0199.

Then ε _rayleigh< ε _{lai Si}, channel model adopts Rayleigh channel model.Get N=1, outage probability $P_{\mathrm{out},R2}^{\mathrm{SAM}}=0.0047.$

then $R^{*}=\arg \min \left\{P_{\mathrm{out}}^Z\left({\mathrm{\&gamma;}}_{i,D}^W\right)\right\}=R_2,$ Namely R is selected ₂as relaying.

Claims (4)

1., based on the relay selection method that fast, slow speed link self adaptation merge, described method comprises:

The relay selection of safe decoding forward collaboration network;

The relay selection of common collaborative network;

The relay selection of described safe decoding forward collaboration network, comprises the steps:

(1) choose two via nodes, a via node R is used for normal transmission data, and another node J is as interfering nodes;

(2) under the prerequisite of interfering nodes set non-NULL, if use interfering nodes, then:

$C_S(i_Y^{*},J^{*})=\arg \underset{\underset{\underset{J\&Element;J_{\mathrm{set}}}{i\&Element;C_d}}{i\&NotEqual;J}}{\max }\left\{C_S(i,J)\right\}=\arg \underset{\underset{\underset{J\&Element;J_{\mathrm{set}}}{i\&Element;C_d}}{i\&NotEqual;J}}{\max }\left\{\frac{1}{2}{\log }_2\left(\frac{1+\frac{{\mathrm{\&gamma;}}_{i,D}^W}{1+{\mathrm{\&gamma;}}_{J,D}^W}}{1+\frac{1+{\mathrm{\&gamma;}}_{i,E}^W}{{\mathrm{\&gamma;}}_{J,E}^W}}\right)\right\};$

If do not use interfering nodes, then:

$C_S\left(i_N^{*}\right)=\arg \underset{i\&Element;C_d}{\max }\left\{C_S\left(i\right)\right\}=\arg \underset{i\&Element;C_d}{\max }\left\{\frac{1}{2}{\log }_2\left(\frac{1+{\mathrm{\&gamma;}}_{i,D}^W}{1+{\mathrm{\&gamma;}}_{i,E}^W}\right)\right\}$

Wherein: j ^*candidate relay node when using interfering nodes and interfering nodes respectively; it is candidate relay node when not using interfering nodes; represent Instant SNR or the average signal-to-noise ratio of link between node i to destination node D; represent Instant SNR or the average signal-to-noise ratio of link between node i to eavesdropping node E; represent Instant SNR or the average signal-to-noise ratio of link between interfering nodes j to destination node D; represent Instant SNR or the average signal-to-noise ratio of link between interfering nodes j to eavesdropping node E; I belongs to set of relay nodes C _d, C _dfor can decoding success forward from the set of relay nodes of source node signal; Interfering nodes J belongs to interfering nodes set J _set, and J ≠ i, interfering nodes set J _setby forming as the via node of interfering nodes;

(3) if $\max (C_S(i_Y^{*},J^{*}),C_S\left(i_N^{*}\right))=C_S(i_Y^{*},J^{*}),$ Then use interfering nodes, and choose and J ^*respectively as via node and interfering nodes; If then do not use interfering nodes, and choose as via node.

2. method according to claim 1, the determination principle of wherein said interfering nodes: when being in state when continuing for some time t, t > t ₀, t ₀for time set point, this node can as interfering nodes; Otherwise this node not can be used as interfering nodes.

3. method according to claim 1, the relay selection of described common collaborative network, comprises the steps: that (1) obtains the i-th (i=1,2...N by estimation _d) individual via node to the Instant SNR sample of destination node D link is ask for the average of this link Instant SNR sample respectively variance the degree of bias and kurtosis

(2) suppose that this link channel type is φ, estimate channel parameter, the Instant SNR probability density function of all kinds channel $f_{\mathrm{\&phi;}}\left({\mathrm{\&gamma;}}_{i,D}^F\right)=\underset{l=1}{\overset{N_{i,D}}{\mathrm{\&Sigma;}}}{\mathrm{\&alpha;}}_{i,D}^l{\left({\mathrm{\&gamma;}}_{i,D}^F\right)}^{{\mathrm{\&beta;}}_{i,D}^l-1}{e}^{-{\mathrm{\&xi;}}_i^l,{\mathrm{\&gamma;}}_{i,D}^F},$ Wherein, N _i,D, for describing parameter during different channels type; Obtain supposing that channel type is the average of the channel Instant SNR of φ according to Instant SNR probability density function variance the degree of bias and kurtosis

(3) calculate:

${\mathrm{\&epsiv;}}_{\mathrm{\&phi;}}=\frac{{[v_1^X-v_{\mathrm{\&phi;},1}^Y]}^2}{v_2^X}+\frac{1}{2}{[1-\frac{v_{\mathrm{\&phi;},2}^Y}{v_2^X}]}^2+\frac{1}{6}{(v_3^X-v_{\mathrm{\&phi;},3}^Y)}^2+\frac{1}{24}{(v_4^X-v_{\mathrm{\&phi;},4}^Y)}^2;$

Minimum ε _φcorresponding hypothesis channel type is the actual channel type of this link, and determines channel parameter N _i,D, with

(4) outage probability during this link employing slow speed link self adaptation:

This link adopts outage probability during rapid link self adaptation:

Wherein: M is the progression of M-QAM modulation; it is Gauss-Q function; represent the maximum integer being less than or equal to x; H represents bit number, k represents dimension, l represents the number of Gamma component, l=1..., N _i,D;

(5) basis select via node; Wherein: R ^*for selected via node; When link adopts rapid link self adaptation, when link adopts slow speed link self adaptation, $P_{\mathrm{out}}^Z\left({\mathrm{\&gamma;}}_{i,D}^W\right)=P_{\mathrm{out}}^{\mathrm{SAM}}\left({\mathrm{\&gamma;}}_{i,D}^S\right).$

4. the method according to claim 1 or 2 or 3, described method also comprises choice for use fast or the principle of slow speed link adaptive technique, and described principle is as follows:

(1) υ is worked as _d≤ υ ₀time, judge that this link channel conditions is non-frequent variations, be suitable for rapid link adaptive technique;

(2) υ is worked as ₀< υ _d≤ υ ₁time, judge that this link channel conditions is common variation, link circuit self-adapting mode corresponding when making channel security capacity maximum choose the prerequisite only changing link circuit self-adapting mode in security cooperation network under; Link circuit self-adapting mode corresponding when making outage probability minimum choose the prerequisite only changing link circuit self-adapting mode in common collaborative network under;

(3) υ is worked as _d> υ ₁time, judge that this link channel conditions is frequent variations, adopt slow speed link adaptive technique; Wherein: υ _dfor node motion speed; υ ₀be the first thresholding speed, 10km/h≤υ ₀≤ 30km/h; υ ₁be the second thresholding speed, 40km/h≤υ ₁≤ 60km/h.