CN106059705A - Relay physical layer safe transmission method - Google Patents

Abstract

The invention provides a relay physical level safe transmission method and belongs to the field of communication. According to the invention, a relay node is, while under a full duplex working mode, is configured with a plurality of antennas, and adopts decode-and-forward protocol. The method designs a pre-coded array to be an artificial noise, and allows the relay node to transmit a hybrid signal of an expectation signal and the artificial noise to an object node at each time slot. A bugging node can bug the hybrid signal. The object node can eliminate the designed artificial noise from the hybrid signal so as to decode the expectation signal, while the bugging node fails to do the decoding. According to the invention, the method also derives and acquires the safety rate of the system, acquires the optimal power allocation polity between the expectation signal and the artificial noise. According to the invention, the method can effectively deal with bugging behavior and explicitly increase safety property of the system.

Description

A kind of relaying safe transmission method of physical layer

Technical field

The present invention relates to the communications field, relay safe transmission method of physical layer particularly to one.

Background technology

Safety of physical layer application in wireless communications causes considerable concern recently, because from information-theoretical viewpoint On see, it is not necessary to upper layer data encryption be also possible to prevent to be ravesdropping.Along with safety of physical layer research is goed deep into, its research Model is the most complicated, and wherein the safety of physical layer of relay system is one of focus in recent years.

Cooperating relay communication is one of effective ways improving safe rate.Cooperating relay (cooperative Relaying, CR) can strengthen the security performance from source node to destination node, and the interference (cooperative that cooperates Jamming, CJ) it is that eavesdropping node is disturbed by the mode by man made noise (artificial noise, AN).And be correlated with The scheme of mixing CR and CJ that document proposes, and the power of the relaying between CR and CJ is distributed maximize safe rate and [join Examine file 1:C.Wang, H.-M.Wang, and X.-G.Xia, " Hybrid opportunistic relaying and jamming with power allocation for secure cooperative networks,”IEEE Trans.Wireless Commun.,vol.14,no.2,pp.589-605,Feb.2015].But, most of existing researchs Focus on half-duplex relaying (half duplex relay, HDR), seldom touch full duplex relaying (full duplex Relay, FDR), the especially situation of multiple antennas full duplex relaying.

Traditional equipment general work in a half-duplex mode, and full duplex technology have become as 5G communication key technology One of.Half-duplex refers under same time slot, the function that equipment can only be transmitted or receive, and full duplex is then with for the moment Under gap, the function that equipment can be transmitted simultaneously and receive, thus the twice of almost half-duplex spectrum efficiency.FDR overcomes HDR The defect of intrinsic spectrum efficiency, but there is relatively strong self-interference signal in full duplex transmission.Existing relevant technology shows When self-interference is suppressed significantly, FDR has more preferable security performance than HDR.

Summary of the invention

The present invention, in order to overcome the deficiencies in the prior art, relays for full duplex, it is provided that a kind of relaying safety of physical layer Transmission method, can be described as Full Duplex Hybrid Relaying-and-Jamming, referred to as FDHRJ with English.This The relaying work of inventive method in a full-duplex mode, is equipped with many antennas, uses decoding to forward (decode-and- Forward, DF) agreement.

A kind of relaying safe transmission method of physical layer of the present invention, the relay system applied includes a single antenna source Node S, a M root antenna relay node R, a single antenna eavesdropping node E, and single antenna destination node D.M is big In the integer of 1.If the experience block decline of all channels, i.e. channel transfer characteristic keep constant on fixing character cycle, and phase The most independent average of obeying is 0, and variance is the multiple Gauss distribution of 1.Source node S to via node R, source node S to eavesdropping node E, Via node R is respectively to destination node D, via node R to the channel transfer matrix of eavesdropping node E WhereinRepresent complex field.(·)^H,(·)^T,(·)^*,(·)^-1Represent respectively The conjugate transpose of matrix, transposition, conjugation, take inverse operation.| | and | | | | represent absolute value of a complex number and the model of matrix respectively Number.The determinant of det () representing matrix.E () represents the mathematic expectaion of stochastic variable.[x]⁺=max (x, 0).Concrete step Suddenly include:

Step 1, in t, single antenna source node S sends desired signal x to via node R_s(t)。

Step 2, in t, via node R receives the desired signal x sent from source node S in the t-1 moment_s(t-1), Simultaneously as via node uses the agreement of decoding forwarding and is operated in full-duplex mode, the letter that therefore via node can forward Number be

$x_r\left(t\right)=\sqrt{\mathrm{\α}}{w}_1{x}_s(t-1)+\sqrt{1-\mathrm{\α}}{w}_{2}z\left(t\right)$

Wherein, α ∈ (0,1) is the power allocation factor between desired signal and man made noise；w₂= H_⊥For the man made noise of design, wherein H_⊥For matrix h_rdThe column vector of the orthogonal basis of kernel, can obtain h_rdw₂=0

Z (t)=[z₁(t),z₂(t),...,z_M-1(t)]^TIt is system noise, wherein z_m(t), m=1,2 ..., M-1 is only The vertical multiple Gaussian random variable with distribution, average is zero, and variance isNoise section w₂z(t) The spatial isotropy launched, does not affects destination node D.x_s(t-1) the most separate with z (t).

The signal y that via node R can receive_r(t) be:

$y_r\left(t\right)=\sqrt{P_s}{h}_{sr}{x}_s\left(t\right)+\sqrt{P_r}{h}_{rr}{x}_r\left(t\right)+n_r\left(t\right)$

Wherein,Represent the via node R remaining self-interference after self-interference technology for eliminating, if h_rrIt is Independent Rayleigh distribution variable.P_sIt is the transmitting power of source node.P_rIt is the transmitting power of via node.n_rT () is t relaying Additive noise at node, in matrix, each element is independent identically distributed multiple Gaussian random variable, and average is zero, and variance is

Step 3, in t, destination node D receives signal y_d(t) be:

$y_d\left(t\right)=\sqrt{P_r}{h}_{rd}{x}_r\left(t\right)+n_d\left(t\right)=\sqrt{{\mathrm{\αP}}_r}{h}_{rd}{w}_1{x}_s(t-1)+n_d\left(t\right)$

Wherein, n_dT () is the additive noise at t destination node, in matrix, each element is independent identically distributed Multiple Gaussian random variable, average is zero, and variance is

Step 4, in t, eavesdropping node E receives signal y_e(t) be:

$\begin{array}{c}{y}_e\left(t\right)=\sqrt{P_s}{h}_{se}{x}_s\left(t\right)+\sqrt{P_r}{h}_{re}{x}_r\left(t\right)+n_e\left(t\right)\\ =\sqrt{P_s}{h}_{se}{x}_s\left(t\right)+\sqrt{{\mathrm{\αP}}_r}{h}_{re}{w}_1{x}_s(t-1)+\sqrt{(1-\mathrm{\α})P_r}{h}_{re}{w}_{2}z\left(t\right)+n_e\left(t\right)\end{array}$

Wherein, n_eT () is the additive noise at t eavesdropping node, in matrix, each element is independent identically distributed Multiple Gaussian random variable, average is zero, and variance is

Step 5, it is thus achieved that the achievable rate R of source node S to via node R_srFor:

$R_{sr}={\log }_2(1+P_s{h}_{sr}^H{\left(I+P_r{h}_{rr}^H{h}_{rr}\right)}^{-1}{h}_{sr})$

Obtain the via node R achievable rate R to destination node D_rdFor:

R_rd=log₂(1+αP_r||h_rdw₁||²)

Obtain the achievable rate R of tapping channel_eFor:

$R_e\≈{\log }_2(1+\frac{P_s|h_{se}{|}^2+{\mathrm{\αP}}_r|\left|h_{re}{w}_1\right|{|}^2}{\frac{1-\mathrm{\α}}{M-1}{P}_r\left|\right|h_{re}{w}_2|{|}^2+1})$

Obtain security of system speed R_sFor:

$R_s={\[{\log }_2(1+\min (P_s{h}_{sr}^H{\left(I+P_r{h}_{rr}^H{h}_{rr}\right)}^{-1}{h}_{sr},{\mathrm{\αP}}_r\left|\right|h_{rd}{w}_1|{|}^2\left)\right)-{\log }_2(1+\frac{P_s|h_{se}{|}^2+{\mathrm{\αP}}_r|\left|h_{re}{w}_1\right|{|}^2}{\frac{1-\mathrm{\α}}{M-1}{P}_r\left|\right|h_{re}{w}_2|{|}^2+1})\]}^{+}$

Step 6, utilizes two way classification, and the optimal power allocation obtaining security of system speed is

${\mathrm{\α}}_{opt}=\frac{P_s{h}_{sr}^H{(I+P_r{h}_{rr}^H{h}_{rr})}^{-1}{h}_{sr}}{P_r\left|\right|h_{rd}{w}_1|{|}^2}$

Wherein I is unit matrix, according to α_optCarry out the power distribution between desired signal and man made noise.

Advantages of the present invention with have the active effect that the inventive method devises pre-coding matrix conduct at via node Human disturbance, relays in a full-duplex mode and sends desired signal and Human disturbance at each time slot simultaneously, and destination node is permissible Decode desired signal, and eliminate the adverse effect of Human disturbance, and eavesdrop node and cannot accomplish this point.The inventive method is also The safe rate of derivation acquisition system, obtains the optimal power allocation strategy between desired signal and man made noise.Use this Inventive method can be effective against eavesdropping behavior, is obviously improved the security performance of system.

Accompanying drawing explanation

Fig. 1 is the model schematic of the relay system safety of physical layer of the embodiment of the present invention；

Fig. 2 is to work as P_s/σ²=20dB, P_r/σ²During=30dB, FDHRJ uses optimal power allocation, and FDHRJ uses average merit Rate distribution and the security of system speed schematic diagram changed with antenna number without three kinds of schemes of man made noise；

Fig. 3 is to work as P_s/σ²=15dB, γ_se=γ_reDuring=-25dB, FDHRJ uses optimal power allocation, and FDHRJ uses Average power allocation and the security of system speed schematic diagram changed with relay power without three kinds of schemes of man made noise.

Detailed description of the invention

Describe embodiments of the present invention in detail below in conjunction with drawings and Examples, whereby how the present invention is applied Technological means solves technical problem, and the process that realizes reaching relevant art effect can fully understand and implement according to this.This Shen Please each feature in embodiment and embodiment, can be combined with each other under not colliding premise, the technical scheme formed All within protection scope of the present invention.

Fig. 1 is the illustraton of model of the relay system safety of physical layer of the embodiment of the present invention.Described relay system includes one Single antenna source node S, a M root antenna relay node R, a single antenna eavesdropping node E, single antenna destination node D；M For the integer more than 1.If the experience block decline of all channels, i.e. channel transfer characteristic keep constant on fixing character cycle, And separate obedience average is 0, variance is the multiple Gauss distribution of 1.Source node S is to via node R, and source node S is to eavesdropping joint Point E, via node R to destination node D, via node R are expressed as to the channel transfer matrix of eavesdropping node EWhereinRepresent complex field.(·)^H,(·)^T,(·)^*, (·)^-1The conjugate transpose of representing matrix respectively, transposition, conjugation, take inverse operation.| | and | | | | represent the absolute of plural number respectively Value and norm of matrix.The determinant of det () representing matrix.E () represents the mathematic expectaion of stochastic variable.

The relaying safe transmission method of physical layer of the present invention includes that step 1 is to step 6.Illustrate below each to realize step.

Step 1, in t, single antenna source node S sends desired signal x to via node R_sT (), now eavesdrops node This signal can also be intercepted.

Step 2, in t, multiple antennas via node R can receive the expectation letter sent from source node S in the t-1 moment Number x_s(t-1), simultaneously as via node uses the agreement of decoding forwarding and is operated in full-duplex mode, therefore via node exists The signal that t can forward is x_rT (), is expressed as:

$x_r\left(t\right)=\sqrt{\mathrm{\α}}{w}_1{x}_s(t-1)+\sqrt{1-\mathrm{\α}}{w}_{2}z\left(t\right)---\left(1\right)$

Wherein, α ∈ (0,1) is the power allocation factor between desired signal and man made noise；w₂= H_⊥The man made noise designed for the present invention, wherein H_⊥For matrix h_rdThe column vector of the orthogonal basis of kernel, can obtain h_rdw₂= 0.Z (t)=[z₁(t),z₂(t),...,z_M-1(t)]^TIt is system noise, wherein z_m(t), m=1,2 ..., M-1 is independent with dividing The multiple Gaussian random variable of cloth, average is zero, and variance isI_M-1Unit square for M-1 dimension Battle array.Noise section w₂The spatial isotropy that z (t) launches, does not affects destination node D.x_s(t-1) simultaneously mutual with z (t) Independent.The signal y that via node R can receive_rT (), is expressed as:

$y_r\left(t\right)=\sqrt{P_s}{h}_{sr}{x}_s\left(t\right)+\sqrt{P_r}{h}_{rr}{x}_r\left(t\right)+n_r\left(t\right)---\left(2\right)$

Wherein,Represent the via node R remaining self-interference after self-interference technology for eliminating, it is assumed that it It is through effectively suppression, so h_rrIt is considered independent Rayleigh distribution variable.P_sIt is the transmitting power of source node.P_rIn being Continue the transmitting power of node.n_rT () is the additive noise at via node, in matrix, each element is independent identically distributed multiple Gaussian random variable, average is zero, and variance is

Step 3, in t, destination node receives signal y_dT () is expressed as:

$y_d\left(t\right)=\sqrt{P_r}{h}_{rd}{x}_r\left(t\right)+n_d\left(t\right)=\sqrt{{\mathrm{\αP}}_r}{h}_{rd}{w}_1{x}_s(t-1)+n_d\left(t\right)---\left(3\right)$

Wherein, n_dT () is the additive noise at t destination node, matrix n_dT in (), each element is independent with dividing The multiple Gaussian random variable of cloth, average is zero, and variance is

Step 4, in t, eavesdropping node receives signal y_e(t) be:

$\begin{array}{c}{y}_e\left(t\right)=\sqrt{P_s}{h}_{se}{x}_s\left(t\right)+\sqrt{P_r}{h}_{re}{x}_r\left(t\right)+n_e\left(t\right)\\ =\sqrt{P_s}{h}_{se}{x}_s\left(t\right)+\sqrt{{\mathrm{\αP}}_r}{h}_{re}{w}_1{x}_s(t-1)+\sqrt{(1-\mathrm{\α})P_r}{h}_{re}{w}_{2}z\left(t\right)+n_e\left(t\right)\end{array}---\left(4\right)$

Wherein, n_eT () is the additive noise at t eavesdropping node E, noise matrix n_eT each element in () is only The vertical multiple Gaussian random variable with distribution, average is zero, and variance is

Step 5, is derived by the source node S achievable rate R to via node R_sr, via node R to destination node D can Reach speed R_rd, the achievable rate R of tapping channel_eIt is respectively as follows:

$R_{sr}={\log }_2(1+P_s{h}_{sr}^H{\left(I+P_r{h}_{rr}^H{h}_{rr}\right)}^{-1}{h}_{sr})---\left(5\right)$

R_rd=log₂(1+αP_r||h_rdw₁||²) (6)

$R_e\≈{\log }_2(1+\frac{P_s|h_{se}{|}^2+{\mathrm{\αP}}_r|\left|h_{re}{w}_1\right|{|}^2}{\frac{1-\mathrm{\α}}{M-1}{P}_r\left|\right|h_{re}{w}_2|{|}^2+1})---\left(7\right)$

Security of system speed R_sFor:

$R_s={\[{\log }_2(1+\min (P_s{h}_{sr}^H{\left(I+P_r{h}_{rr}^H{h}_{rr}\right)}^{-1}{h}_{sr},{\mathrm{\αP}}_r\left|\right|h_{rd}{w}_1|{|}^2\left)\right)-{\log }_2(1+\frac{P_s|h_{se}{|}^2+{\mathrm{\αP}}_r|\left|h_{re}{w}_1\right|{|}^2}{\frac{1-\mathrm{\α}}{M-1}{P}_r\left|\right|h_{re}{w}_2|{|}^2+1})\]}^{+}---\left(8\right)$

Wherein, [x]⁺=max (x, 0), represents the higher value in taking both 0 and x.

Step 6, utilizes two way classification, obtains optimal power allocation α of security of system speed_optFor:

${\mathrm{\α}}_{opt}=\frac{P_s{h}_{sr}^H{(I+P_r{h}_{rr}^H{h}_{rr})}^{-1}{h}_{sr}}{P_r\left|\right|h_{rd}{w}_1|{|}^2}---\left(9\right)$

Wherein, I is unit matrix.According to the α o obtained_ptCarry out the power distribution between desired signal and man made noise, i.e. α=α is substituted in formula (1)_optObtain forward signal x_r(t)。

Fig. 2 and Fig. 3 is the simulation result of the inventive method.Assume that channel gain is γ_sr=| | h_sr||²,γ_rd=| | h_rd| |²,γ_se=| h_se|²,γ_re=| | h_re||²,γ_rr=| | h_rr||²,γ_se=γ_re=0dB, γ_rr=-40dB, if noise power

Fig. 2 is the result that safe rate changes with relay antenna number.Wherein FDHRJ opt represents and uses optimal power allocation FDHRJ strategy, FDHRJ opt the most currently preferred relaying safe transmission method of physical layer；FDHRJ ave represents to use and puts down All FDHRJ strategies of power distribution, No AN represents and does not use FDHRJ strategy.On the whole, safe rate is along with relay antenna Number increase and increase.Work as γ_se=-25dB and γ_re=-15dB, the safe rate increasing degree of three kinds of contrast schemes and γ_se =γ_reThe situation of=-15dB is similar to, but works as γ_se=-15dB and γ_re=-25dB, FDHRJ opt has the fastest safety The increasing degree of speed, FDHRJ ave is also good than the performance of No AN scheme.In sum, FDHRJ side proposed by the invention Case is an advantage over other two kinds contrast schemes.

The safe rate of Fig. 3 display system first increases with total transmitting power of relaying and subtracts afterwards.Now it is configured to M=4, γ_se= γ_re=-25dB.Owing to the channel speed of trunk channel is to take channel speed less in S → R and R → D, certain at channel In the case of, work as P_rTime less, the signal to noise ratio of the R → D signal to noise ratio less than S → R, now trunk channel speed is by P_rDetermine, can be with P_r Increase and increase；Work as P_rIncrease to and P_sTime close, the speed of trunk channel depends on this link of S → R.Due to relaying The existence of self-interference channel, can cause trunk channel speed with P_rIncrease and reduce；For tapping channel, along with P_r's Increase, both can increase the useful signal that eavesdropping node is received, also can increase its interference signal received, the most not Can be with P_rIncrease and produce large change.As it is shown on figure 3, in the case of channel status is stable, work as P_rTime less, FDHRJ exists Use α_optTime consistent with the effect without man made noise's scheme, but work as P_rAfter increasing to about 11dB, use α_opt's FDHRJ scheme to be much better than other two kinds of contrast schemes.To sum up analyzing and understand, the safe rate of system can always launch merit with relaying The increase of rate and first increase and subtract afterwards, when without man made noise, eavesdropping node will not receive the interference signal of middle secondary, at this moment Its tapping channel speed can be along with P_rIncrease have significantly increase, therefore in FIG. 3, it is seen that this curve of No AN exists Second half section can be more faster than what other two curves declined.

In sum, the relay system safety of physical layer mode that the embodiment of the present invention provides, use the present invention to propose FDHRJ scheme, at via node, devise pre-coding matrix as Human disturbance, relay in a full-duplex mode each Time slot sends desired signal and Human disturbance simultaneously, and destination node can decode desired signal, and eliminates Human disturbance not Profit impact, and eavesdrop node and cannot accomplish this point.The most under the model, it is thus achieved that the safe rate of system, have studied the phase Hope the optimal power allocation strategy between signal and man made noise, use proposed scheme can be obviously improved the safety of system Energy.

Embodiment of above is merely to illustrate the present invention, and not limitation of the present invention, common about technical field Technical staff, without departing from the spirit and scope of the present invention, it is also possible to make a variety of changes and modification, therefore own The technical scheme of equivalent falls within scope of the invention, and the scope of patent protection of the present invention should be defined by the claims.

Claims (2)

1. a relaying safe transmission method of physical layer, it is characterised in that relay system model used includes a single antenna Source node S, the via node R of a M root antenna, a single antenna eavesdropping node E and single antenna destination node D；M is Integer more than 1；Via node uses the agreement of decoding forwarding and is operated in full-duplex mode；The experience block decline of all channels, Channel transfer characteristic keeps constant on fixing character cycle, and separate obedience average is 0, and variance is the multiple Gauss of 1 Distribution；If source node S is to via node R, source node S to eavesdropping node E, via node R to destination node D and relaying joint Point R is respectively to the channel transfer matrix of eavesdropping node E Table Give instructions in reply number field；

Described transmission method comprises the steps:

Step 1, in t, single antenna source node S sends desired signal x to via node R_s(t)；

Step 2, in t, via node R receives the signal x sent from source node S in the t-1 moment_s(t-1), relaying joint The signal x that point forwards_r(t) be:

$x_r\left(t\right)=\sqrt{\mathrm{\α}}{w}_1{x}_s(t-1)+\sqrt{1-\mathrm{\α}}{w}_{2}z\left(t\right)$

Wherein, α is the power allocation factor between desired signal and man made noise, α ∈ (0,1)；w₁、w₂For manually making an uproar of design Sound,w₂=H_⊥, H_⊥For matrix h_rdThe column vector of the orthogonal basis of kernel, h_rdw₂=0；Z (t) is system noise Sound, z (t)=[z₁(t),z₂(t),...,z_M-1(t)]^T, x_s(t-1) separate with z (t)；Noise section w₂The sky that z (t) launches Between isotropism, do not affect destination node D；(·)^H、(·)^TThe conjugate transpose of representing matrix, transposition respectively, | | | | table Show norm of matrix；

The signal y that via node R receives in t_r(t) be:

$y_r\left(t\right)=\sqrt{P_s}{h}_{sr}{x}_s\left(t\right)+\sqrt{P_r}{h}_{rr}{x}_r\left(t\right)+n_r\left(t\right)$

Wherein,Represent the via node R remaining self-interference after self-interference technology for eliminating, if h_rrIt is independent Rayleigh distributed variable；P_sIt is the transmitting power of source node S, P_rIt is the transmitting power of via node R；n_rT () is t relaying joint Additive noise at Dian, noise matrix n_rT in (), each element is independent identically distributed multiple Gaussian random variable, average is Zero, variance is

Step 3, in t, destination node D receives signal y_d(t) be:

$y_d\left(t\right)=\sqrt{P_r}{h}_{rd}{x}_r\left(t\right)+n_d\left(t\right)=\sqrt{{\mathrm{\αP}}_r}{h}_{rd}{w}_1{x}_s(t-1)+n_d\left(t\right)$

Wherein, n_dT () is the additive noise at t destination node, matrix n_dT in (), each element is independent identically distributed Multiple Gaussian random variable；

Step 4, in t, eavesdropping node E receives signal y_e(t) be:

$\begin{array}{c}{y}_e\left(t\right)=\sqrt{P_s}{h}_{se}{x}_s\left(t\right)+\sqrt{P_r}{h}_{re}{x}_r\left(t\right)+n_e\left(t\right)\\ =\sqrt{P_s}{h}_{se}{x}_s\left(t\right)+\sqrt{{\mathrm{\αP}}_r}{h}_{re}{w}_1{x}_s\left(t-1\right)+\sqrt{(1-\mathrm{\α})P_r}{h}_{re}{w}_{2}z\left(t\right)+n_e\left(t\right)\end{array}$

Wherein, n_eT () is the additive noise at t eavesdropping node E, noise matrix n_eT each element in () is independent same The multiple Gaussian random variable of distribution；

Step 5, it is thus achieved that the achievable rate R of source node S to via node R_srFor:

$R_{sr}={\log }_2(1+P_s{h}_{sr}^H{\left(I+P_r{h}_{rr}^H{h}_{rr}\right)}^{-1}{h}_{sr})$

Wherein, ()^-1Representing matrix take inverse operation；

Obtain the via node R achievable rate R to destination node D_rdFor:

R_rd=log₂(1+αP_r||h_rdw₁||²)

Obtain the achievable rate R of tapping channel_eFor:

$R_e\≈{\log }_2(1+\frac{P_s|h_{se}{|}^2+{\mathrm{\αP}}_r|\left|h_{re}{w}_1\right|{|}^2}{\frac{1-\mathrm{\α}}{M-1}{P}_r\left|\right|h_{re}{w}_2|{|}^2+1})$

Wherein, | | represent and seek absolute value of a complex number；

Obtain security of system speed R_sFor:

$R_s={\[{\log }_2\left(1+\min \left(P_s{h}_{sr}^H{\left(I+P_r{h}_{rr}^H{h}_{rr}\right)}^{-1}{h}_{sr},{\mathrm{\αP}}_r\left|\right|h_{rd}{w}_1|{|}^2\right)\right)-{\log }_2\left(1+\frac{P_s|h_{se}{|}^2+{\mathrm{\αP}}_r|\left|h_{re}{w}_1\right|{|}^2}{\frac{1-\mathrm{\α}}{M-1}{P}_r\left|\right|h_{re}{w}_2|{|}^2+1}\right)\]}^{+}$

Step 6, utilizes two way classification, obtains optimal power allocation α of security of system speed_optFor:

${\mathrm{\α}}_{opt}=\frac{P_s{h}_{sr}^H{(I+P_r{h}_{rr}^H{h}_{rr})}^{-1}{h}_{sr}}{P_r\left|\right|h_{rd}{w}_1|{|}^2}$

Wherein I is unit matrix, according to α_optCarry out the power distribution between desired signal and man made noise.

A kind of relaying safe transmission method of physical layer the most according to claim 1, it is characterised in that described system noise Element z in z (t)_m(t), m=1,2 ..., M-1, is independent identically distributed multiple Gaussian random variable, and its average is zero, variance ForE () represents the mathematic expectaion of stochastic variable, I_M-1Unit matrix for M-1 dimension.