CN106413099A - Optimal power allocation method of full duplex collaborative relay network - Google Patents

Abstract

The invention discloses an optimal power allocation method of a full duplex collaborative relay network, in order to mainly solve the problems of limited actual application scenes and low reliability in the resource allocation of the full duplex collaborative relay network resulting from ignorance to direct links in the prior art. The technical scheme is as follows: 1) estimating channel parameters via a minimum mean square error; 2) acquiring received signals of a relay node and a destination node according to the channel parameters; 3) determining a signal to interference plus noise power ratio according to the received signals; 4) establishing a relation between volume and the signal to interference plus noise power ratio; 5) establishing a target function for maximizing the signal to interference plus noise power ratio through the relation between the volume and the signal to interference plus noise power ratio; 6) constructing a new target function only related to relay power according to the theory of maximizing the minimum value; and 7) carrying out derivation on the relay power by using the new target function to obtain the optimal power of the relay node and a source node. The optimal power allocation method disclosed by the invention can be used in networks with different transmission distances to improve the resource utilization rate and the system performance in the case of limited resources.

Description

The optimal power allocation method of full duplex cooperative relay network

Technical field

The invention belongs to communication technical field, it is related to optimal power allocation method, can be used for full duplex cooperative relay network.

Background technology

Relaying is by the reception of data signal and forwarding, to expand the distance of network transmission.In beginning one's study most The network that continues is in a half-duplex mode, and half-duplex relay cooperation communication system that can be traditional is sent and received using orthogonal time slot Signal, significant wastage radio spectrum resources, and full duplex relaying collaboration communication is doubled with its channel capacity and lifts frequency spectrum money Source utilization rate has obtained the extensive concern of academia.But full duplex relaying system has that intrinsic is exactly self-interference, by It is directly proportional in the power of self-interference and the power of relaying, so one rational power allocation scheme of research is conducive to lift system Performance.

In cooperation communication system, because cooperative diversity technique requires to assist it while transmission information between node Its node-node transmission information, will necessarily produce the consumption of the resources such as extra bandwidth, time slot and power, therefore with no collaboration communication phase , with greater need for taking reasonably optimized distribution to the power of each user and bandwidth, guarantee is with relatively low complexity for collaboration communication for ratio Degree and higher resource utilization go to obtain good systematic function.

In today of resource requirement growing tension, the research to resource allocation problem is continuing always.Current resource is divided Join mainly in the assignment problem of power, if limited system resources, how reasonably to carry out power distribution, preferably have Helping lift system performance just becomes the emphasis of research.In relay system, source node and via node adopt average power allocation Method, be easier realize, but due to there is the impact of system channel quality and time variation, should in this way can not Fully utilize system resource and reduce overhead it is therefore desirable to study optimum power distribution method.

The power distribution method of cooperation communication system has been carried out very many research at present, and great majority are in three sections In the system of point double bounce, ignore the impact of tie link simultaneously or directly tie link is analyzed as interference, in selection Continue and set up model under decoding retransmission protocol or under amplification forwarding agreement, the starting point of research generally comprises:The capacity of system, Error rate of system, outage probability, computation complexity and overhead etc., its method of salary distribution includes：

L) transmitting terminal power adopts independent power distribution, and this mode is all to be distributed whole transmission powers by transmitting terminal To source node, it is only applicable to small-sized short range network it is impossible to expand network transmission distance.

2) transmitting terminal power adopts constant power to distribute, and this mode is can be simply obtained very much necessarily by constant power distribution Power system capacity under channel status, but it is unable to effectively utilizes power resource.

3) transmitting terminal power adopts optimal power allocation, and this mode is by minimizing outage probability, maximum channel Capacity or minimize the bit error rate and set out consideration power distribution, the above two described can effectively utilizes power provide this method of salary distribution relatively Source, but because which is to ignore the optimum allocation side of tie link under single power limited or general power limited situation simultaneously Formula, and do not consider that in single power and the united situation of general power and different size of network, tie link is under systematic influence Best practice, thus results in limited by application scenarios it is impossible to meet in different size of real network for resource optimal allocation The demand of lifting network performance.

Content of the invention

Present invention aims to the deficiency of above-mentioned prior art, a kind of full duplex cooperative relay network is proposed Excellent power distribution method, to enter for different size of full duplex cooperative relay network lower limited by single power and general power joint Row optimal power allocation, the service quality of lifting communication and handling capacity, and the reliability of cooperation communication system.

For achieving the above object, technical scheme include as follows：

(1) adopt least mean-square error channel estimation methods, respectively estimate source node to via node channel parameter h_SR、 Via node is to the channel parameter h of destination node_RD, the channel parameter h of source node to destination node_SDWith via node loop letter Road parameter h_RR；

(2) source node is modulated to sending signal, and signal x (i) after modulation is sent to via node and purpose section Point, via node obtains receipt signal g (i), and destination node obtains receipt signal d (i)；

(3) via node recompiles after total receipt signal r (i) is demodulated again, and by coding after signal t I () is sent to destination node, destination node obtains final receipt signal：

Y (i)=d (i)+h_RDt(i)+n_D(i),

Wherein, n_DI () is the Gaussian noise of purpose node, obeying average is 0, and variance is σ_D ²Multiple Gauss distribution；

(4) set up the relational expression maximizing system channel capacity and system Signal to Interference plus Noise Ratio：

S.t.0 ＜ P_s+P_r≤P_t

0 ＜ P_s≤P_max, 0 ＜ P_r≤P_max,

Wherein, C is system channel capacity, and B is system bandwidth, P_sFor the transmission power of source node, P_rSending out for via node Penetrate power, P_tFor the general power higher limit of system, P_max=wP_tFor single power upper limit value, w is to control single power maximum restriction scope Coefficient, 0 ＜ w ＜ 1,0 ＜ P_s+P_r≤P_tConstraints limited by system total power, 0 ＜ P_s≤P_maxIt is subject to for source node power The constraints of limit, 0 ＜ P_r≤P_maxConstraints for via node power limited；γ is system Signal to Interference plus Noise Ratio, its value root Determine according to the network being used：

γ=min { γ in tie link is as the network of interference signal_SR,γ_RD,For via node pair The Signal to Interference plus Noise Ratio of source node,For the Signal to Interference plus Noise Ratio to via node for the purpose node, wherein, k_srFor source node in Continue the channel gain of node, k_rdFor the channel gain of via node to destination node, k_sdChannel for source node to destination node Gain and k_rrFor via node loop channel gain；

γ=min { γ in tie link is as the network of useful signal_SR,γ_SD+γ_RD,For relaying The Signal to Interference plus Noise Ratio to source node for the node, γ_SD=k_sdP_sFor the Signal to Interference plus Noise Ratio to source node for the purpose node, γ_RD=k_rdP_rFor the purpose of The Signal to Interference plus Noise Ratio to via node for the node；

(5) under power-limited condition, according to the relation maximizing system channel capacity and system Signal to Interference plus Noise Ratio, set up The object function of bigization Signal to Interference plus Noise Ratio：

5a) for γ=min { γ_SR,γ_RDSystem Signal to Interference plus Noise Ratio, according to maximizing, minima is theoretical, obtain only with Via node power P_rRelated fresh target function：

S.t.0 ＜ P_s+P_r≤P_t

0 ＜ P_s≤P_max, 0 ＜ P_r≤P_max,

5b) for γ=min { γ_SR,γ_SD+γ_RDSystem Signal to Interference plus Noise Ratio, according to maximizing, minima is theoretical, obtains Only with via node power P_rRelated fresh target function：

S.t.0 ＜ P_s+P_r≤P_t

0 ＜ P_s≤P_max, 0 ＜ P_r≤P_max；

(6) use the fresh target function gamma obtaining in (5) to relaying node power P_rDerivation, obtains via node power P_r's Optimal value, further according to the source node power P maximizing the theoretical foundation of minima_sWith via node power P_rRelation, obtain source section Point power P_sOptimal value.

The present invention compared with prior art has advantages below：

First, the present invention, is carried out to it as two kinds of situations of interference signal and useful signal due to according to tie link Optimal power allocation, can be flexibly applied to network transmission apart from actual scene of different sizes compared to prior art.

Second, in the case of resource-constrained, the present invention considers under the single power of joint and general power confined condition Excellent power distribution, compared to the scene only considering limited by single power limited or general power, improves the utilization rate of limited resources.

3rd, compared to half-duplex cooperative relay network, via node receives and sending signal respectively in orthogonal timeslots, this Invention is directed to full duplex cooperative relay network, and via node receives and sending signal in same time slot, it is possible to achieve higher frequency is composed Efficiency, the overall performance of lifting cooperative system.

4th, the via node in collaborative network of the present invention, using decoding forwarding strategy, compared to amplification forwarding strategy, subtracts Sonic propagation of making an uproar in system, the reliability of lift system are lacked.

Brief description

Fig. 1 is the flowchart of the present invention；

Fig. 2 is tie link to be used as the system block diagram of interference signal in the present invention；

Fig. 3 is tie link to be used as the system block diagram of useful signal in the present invention；

Fig. 4 be to present invention tie link as interference signal system channel capacity analogous diagram；

Fig. 5 be to present invention tie link as useful signal system channel capacity analogous diagram.

Specific embodiment

Below in conjunction with accompanying drawing, embodiment of the present invention is described in further detail.

With reference to Fig. 2, in this full duplex cooperative relay network, source node and destination node are operated in semiduplex mode and all It is provided with an antenna, via node is received and sending signal using full-duplex mode simultaneously, and be provided with two antennas.Source node pair Transmission signal is modulated, then in each time slot to via node and destination node sending signal；Via node is using decoding Forward DF mode that the signal receiving is decoded, then re-encoding forwards；Destination node is by the way of maximum-ratio combing Receive tie link and the signal forwarding by repeated link, due to the long transmission distance of source node and destination node, by road Footpath loss and the impact of shadow fading, the signal of tie link transmission is as interference signal.

With reference to Fig. 3, in this full duplex cooperative relay network, source node and destination node are operated in semiduplex mode and all It is provided with an antenna, via node is received and sending signal using full-duplex mode simultaneously, and be provided with two antennas.Source node pair Transmission signal is modulated, then in each time slot to via node and destination node sending signal；Via node is using decoding Forward DF mode that the signal receiving is decoded, then re-encoding forwards；Destination node is by the way of maximum-ratio combing Receive tie link and the signal forwarding by repeated link, because the transmission range of source node and destination node is not far, signal Intensity enough destination obtain collaboration diversity, tie link transmission signal as useful signal.

With reference to Fig. 1, the network system based on Fig. 2 and 3 for the present invention, realize optimal power allocation method include as follows：

Step 1, system initialization：

Source node, via node and destination node utilize known sending signal sequence, by least mean-square error channel Method of estimation, estimates the shock response characteristic of each channel, including the channel parameter h of source node to via node_SR, relaying section Point arrives the channel parameter h of destination node_RD, the channel parameter h of source node to destination node_SD, via node loop channel parameter h_RR.

Step 2, source node is to via node and destination node sending signal.

Source node is modulated to transmission signal, simultaneously to via node and destination node transmission signal x (i), modulation methods Formula can adopt QPSK mode or BPSK mode or QAM mode, and the signal that via node receives source node transmission in i time slot is：g (i)=h_SRX (i), destination node i time slot receive source node send signal be：D (i)=h_SDX (i+ τ), source node send out Penetrating power is P_r=E [| t (i) |²]；

Wherein, i is integer, i.e. i ＞ 0, τ are the relative time delay between tie link and repeated link, and E is mathematic expectaion.

Step 3, the receipt signal of via node and forward signal.

3a) via node receives, in i time slot, signal g (i) that source node sends, and receives what via node itself forwarded simultaneously Signal f (i), obtaining receipt signal r (i) is：R (i)=g (i)+f (i)+n_R(i), wherein n_RWhen () is via node receiving terminal i i The Gaussian noise of gap, obeying average is 0, and variance is σ_R ²Multiple Gauss distribution；

3b) via node, using decoding forwarding strategy, is decoded to receipt signal r (i), by the signal of decoding more again Coding, obtaining encoded signal is：T (i)=r (i-1), and forward it to destination node, the transmission power of via node is P_r =E [| t (i) |²].

Step 4, destination node receipt signal.

Destination node receives signal d (i) and via node transmission from source node by the way of maximum-ratio combing Signal h_RDT (i), it obtains final receipt signal expression formula and is：

Y (i)=d (i)+h_RDt(i)+n_D(i)；

Wherein, n_DI () is the Gaussian noise of purpose node receiving terminal i time slot, obeying average is 0, and variance is σ_D ²Multiple height This distribution；The source node signal that d (i) receives for purpose node, according to transmission range network of different sizes, it is divided into following two The situation of kind：

The first situation is in the larger network of source node and destination node transmission range, due to path loss and shade The impact of decline, so that signal intensity attenuation, interferes to destination node demodulation, so tie link is considered interference letter Number, the interference signal that is, d (i) receives for purpose node；

Second situation is in the little network of source node and destination node transmission range, the signal of tie link transmission Intensity demodulates helpful to destination node, it is possible to achieve collaboration diversity, so tie link is considered useful signal, that is, d (i) is The useful signal that destination node receives.

Step 5, the network according to being used determines Signal to Interference plus Noise Ratio.

5a) in tie link as in the network of interference signal, via node to the Signal to Interference plus Noise Ratio of source node is Destination node to the Signal to Interference plus Noise Ratio of via node isSystem Signal to Interference plus Noise Ratio is γ=min { γ_SR,γ_RD, its In, k_srFor the channel gain of source node to via node, k_rdFor the channel gain of via node to destination node, k_sdFor source section Point arrives channel gain and the k of destination node_rrFor via node loop channel gain；

5b) in tie link as in the network of useful signal, via node to the Signal to Interference plus Noise Ratio of source node isDestination node is γ to the Signal to Interference plus Noise Ratio of source node_SD=k_sdP_s, the Signal to Interference plus Noise Ratio to via node for the destination node For γ_RD=k_rdP_r, system Signal to Interference plus Noise Ratio is γ=min { γ_SR,γ_SD+γ_RD}.

Step 6, according to system Signal to Interference plus Noise Ratio γ, sets up system channel capacity as follows with the relational expression of system Signal to Interference plus Noise Ratio：

S.t.0 ＜ P_s+P_r≤P_t

0 ＜ P_s≤P_max, 0 ＜ P_r≤P_max,

Wherein, C is system channel capacity, and B is system bandwidth, P_tFor the general power higher limit of system, P_max=wP_tFor single work( Rate higher limit, w is to control the maximum coefficient limiting scope of single power, 0 ＜ w ＜ 1,0 ＜ P_s+P_r≤P_tLimited by system total power Constraints, 0 ＜ P_s≤P_maxFor the constraints of source node power limited, 0 ＜ P_r≤P_maxFor via node power limited Constraints.

Step 7, according to the relational expression of system channel capacity and Signal to Interference plus Noise Ratio, sets up based on the target maximizing Signal to Interference plus Noise Ratio Function.

Implementing of this step is maximum capacity problem to be transformed into the problem maximizing Signal to Interference plus Noise Ratio, for difference Transmission network, set up based on maximize Signal to Interference plus Noise Ratio object function：

7a) γ=min { γ in tie link is as the network of interference signal_SR,γ_RD, manage according to maximizing minima By working as γ_SR=γ_RDWhen, system Signal to Interference plus Noise Ratio is realized maximizing, i.e. γ=γ_SR；

7b) γ=min { γ in tie link is as the network of useful signal_SR,γ_SD+γ_RD, minimum according to maximizing Value is theoretical, works as γ_SR=γ_SD+γ_RDWhen, system Signal to Interference plus Noise Ratio is realized maximizing, i.e. γ=γ_SR；

7c) finally give object function as follows：

S.t.0 ＜ P_s+P_r≤P_t

0 ＜ P_s≤P_max, 0 ＜ P_r≤P_max.

Step 8, foundation and via node power P_rThe related fresh target function based on maximization Signal to Interference plus Noise Ratio.

8a) in tie link as in the network of interference signal, set up only and via node power P_rRelated based on The fresh target function of bigization Signal to Interference plus Noise Ratio：

8a1) according to γ_SR=γ_RD, obtain the relational expression of source node power and via node power：k_srk_sdP_s ²+k_srP_s= k_rrk_rdP_r ²+k_rdP_r；

8a2) the relational expression according to source node power and via node power, obtains being saved with the source that via node power represents Point power form

8a3) by 8a2) P that obtains_sRelational expression substitutes in the object function of step 7, obtain only with via node work( Rate P_rThe related fresh target function based on maximization Signal to Interference plus Noise Ratio：

S.t.0 ＜ P_s+P_r≤P_t

0 ＜ P_s≤P_max, 0 ＜ P_r≤P_max,

8b) in tie link as in the network of useful signal, set up only and via node power P_rRelated based on The fresh target function of bigization Signal to Interference plus Noise Ratio：

8b1) according to γ_SR=γ_SD+γ_RD, obtain the relation of source node power and via node power：k_srP_s-k_sdP_s= k_rrk_rdP_r ²+k_rdP_r+k_sdk_rrP_sP_r；

8b2) the relational expression according to source node power and via node power, obtains being saved with the source that via node power represents Point power form

8b3) by 8b2) P that obtains_sRelational expression substitutes in the object function of step 7, and construction is only and via node power P_rPhase The fresh target function based on maximization Signal to Interference plus Noise Ratio closing：

S.t.0 ＜ P_s+P_r≤P_t

0 ＜ P_s≤P_max, 0 ＜ P_r≤P_max.

Step 9, obtains the optimal power of via node and source node.

9a) in the full duplex relaying network that tie link is considered as interference signal or useful signal, by fresh target function pair P_rIn (0, ∞) derivation, obtain transmission power P with via node for the system Signal to Interference plus Noise Ratio_rIncrease and monotonically increasing characteristic；

9b) according to 9a) in system Signal to Interference plus Noise Ratio characteristic, the transmission power of via node chooses and meets power constraints Maximum, i.e. via node power P_rOptimal value.

9c) by P_rOptimal value substitute into step 8a2) and source node power 8b2) and via node power relational expression In, obtain source node power P_sOptimal value.

The effect of the present invention can be further illustrated by following emulation：

(1) simulated conditions

Modulation system all adopts QPSK, if the channel between each node all obey average be 0, variance be 1 quasistatic auspicious Profit distribution.System bandwidth is B=4kHz, and noise power is N₀=0dBw, each channel gain is respectively k_sr=8dB, k_rd=12dB, k_sd=0dB, k_rr=4dB, general power higher limit P_t=0:1:10dBw, controls the maximum coefficient w=limiting scope of single power 0.55, wherein k_srFor the channel gain of source node to via node, k_rdFor the channel gain of via node to destination node, k_sd Channel gain for source node to destination node and k_rrFor via node loop channel gain.

(2) content emulating and result

Emulation 1：With the inventive method and existing average power allocation method, exhaustive search method respectively in tie link Full duplex cooperative relay network as interference signal is emulated, result such as Fig. 4.

From fig. 4, it can be seen that the channel capacity curve of the inventive method and exhaustive search method are completely superposed, demonstrate this The accuracy of invention optimum allocation method and effectiveness；Under conditions of the general power upper limit is consistent, the letter that the inventive method obtains The channel capacity that road capacity obtains more than average power allocation method, illustrates that the inventive method is substantially better than average power allocation side Method.

Emulation 2：With the inventive method and existing average power allocation method, exhaustive search method respectively in tie link Full duplex cooperative relay network as useful signal is emulated, result such as Fig. 5.

From fig. 5, it can be seen that the channel capacity curve of the inventive method and exhaustive search method are completely superposed, demonstrate this The accuracy of invention optimum allocation method and effectiveness；Under conditions of the general power upper limit is consistent, the letter that the inventive method obtains The channel capacity that road capacity obtains more than average power allocation method, illustrates that the inventive method is substantially better than average power allocation side Method.

Contrast Figure 4 and 5 can be seen that the inventive method and apply acquisition in tie link is as the network of useful signal Performance is substantially better than to be applied in tie link as in the network of interference signal.

Claims (7)

1. a kind of optimal power allocation method of full duplex cooperative relay network, including:

(1) adopt least mean-square error channel estimation methods, respectively estimate source node to via node channel parameter h_SR, relaying Node is to the channel parameter h of destination node_RD, the channel parameter h of source node to destination node_SDWith via node loop channel ginseng Number h_RR；

(2) source node is modulated to sending signal, and signal x (i) after modulation is sent to via node and destination node, in The node that continues obtains receipt signal g (i), and destination node obtains receipt signal d (i)；

(3) via node recompiles after total receipt signal r (i) is demodulated again, and signal t (i) after coding is sent out Give destination node, destination node obtains final receipt signal：

Y (i)=d (i)+h_RDt(i)+n_D(i),

Wherein, n_DI () is the Gaussian noise of purpose node, obeying average is 0, and variance is σ_D ²Multiple Gauss distribution；

(4) set up the relational expression maximizing system channel capacity and system Signal to Interference plus Noise Ratio：

$\begin{array}{cc}\underset{(P_s,P_r)}{max}& C={\mathrm{Blog}}_2(1+\mathrm{\γ})\end{array}$

S.t.0 ＜ P_s+P_r≤P_t

0 ＜ P_s≤P_max, 0 ＜ P_r≤P_max,

Wherein, C is system channel capacity, and B is system bandwidth, P_sFor the transmission power of source node, P_rTransmitting work(for via node Rate, P_tFor the general power higher limit of system, P_max=wP_tFor single power upper limit value, w be control single power maximum limit scope be Number, 0 ＜ w ＜ 1,0 ＜ P_s+P_r≤P_tConstraints limited by system total power, 0 ＜ P_s≤P_maxFor source node power limited Constraints, 0 ＜ P_r≤P_maxConstraints for via node power limited；γ is system Signal to Interference plus Noise Ratio, and its value is according to institute The network using determines：

γ=min { γ in tie link is as the network of interference signal_SR,γ_RD,For via node, source is saved The Signal to Interference plus Noise Ratio of point,For the Signal to Interference plus Noise Ratio to via node for the purpose node, wherein, k_srFor source node to relaying section The channel gain of point, k_rdFor the channel gain of via node to destination node, k_sdChannel gain for source node to destination node And k_rrFor via node loop channel gain；

γ=min { γ in tie link is as the network of useful signal_SR,γ_SD+γ_RD,For via node pair The Signal to Interference plus Noise Ratio of source node, γ_SD=k_sdP_sFor the Signal to Interference plus Noise Ratio to source node for the purpose node, γ_RD=k_rdP_rFor purpose node pair The Signal to Interference plus Noise Ratio of via node；

(5) under power-limited condition, according to the relation maximizing system channel capacity and system Signal to Interference plus Noise Ratio, set up and maximize The object function of Signal to Interference plus Noise Ratio：

5a) for γ=min { γ_SR,γ_RDSystem Signal to Interference plus Noise Ratio, according to maximizing, minima is theoretical, obtain only with relaying section Point power P_rRelated fresh target function：

$\begin{array}{cc}\underset{(P_s,P_r)}{max}& \mathrm{\γ}=\frac{-k_{sr}+\sqrt{{k_{sr}}^2+4k_{sr}{k}_{sd}(k_{rr}{k}_{rd}{P_r}^2+k_{rd}{P}_r)}}{2k_{sd}(k_{rr}{P}_r+1)}\end{array}$

S.t.0 ＜ P_s+P_r≤P_t

0 ＜ P_s≤P_max, 0 ＜ P_r≤P_max,

5b) for γ=min { γ_SR,γ_SD+γ_RDSystem Signal to Interference plus Noise Ratio, according to maximizing, minima is theoretical, obtain only with Via node power P_rRelated fresh target function：

$\begin{array}{cc}\underset{(P_s,P_r)}{max}& \mathrm{\γ}=\frac{k_{sr}{k}_{rd}{P}_r}{k_{sr}-k_{sd}-k_{sd}{k}_{rr}{P}_r}\end{array}$

S.t.0 ＜ P_s+P_r≤P_t

0 ＜ P_s≤P_max, 0 ＜ P_r≤P_max；

(6) use the fresh target function gamma obtaining in (5) to relaying node power P_rDerivation, obtains via node power P_rOptimum Value, further according to the source node power P maximizing the theoretical foundation of minima_sWith via node power P_rRelation, obtain source node work( Rate P_sOptimal value.

2. receipt signal g (i) that in method according to claim 1, wherein step (2), via node obtains, is by i Signal x (i) of time slot source node transmission and the channel parameter h of source node to via node_SRBe multiplied, i.e. g (i)=h_SRx(i).

3. receipt signal d (i) that in method according to claim 1, wherein step (2), destination node obtains, is by i Source node sending signal x (i+ τ) of+τ time slot and source node are to the channel parameter h of destination node_SDBe multiplied, i.e. d (i)=h_SDx(i + τ), wherein τ is the relative time delay between tie link and repeated link.

4. in method according to claim 1, wherein step (3) via node obtain total receipt signal r (i), be by Source node sending signal g (i) receiving, self-interference signal f (i) receiving and Gaussian noise n_RI () three is added, i.e. r (i) =g (i)+f (i)+n_RI (), wherein, f (i) is self-interference signal, i.e. f (i)=h_RRT (i), t (i) are the forwarding letter of via node Number, n_RI () is the Gaussian noise of via node receiving terminal, obeying average is 0, and variance is σ_R ²Multiple Gauss distribution.

5. forward signal t (i) in method according to claim 1, wherein step (3), it relays section with i-1 time slot Total receipt signal of point is equal, i.e. t (i)=r (i-1).

6. in method according to claim 1, wherein step (5a) according to maximize minima theory obtain only with relaying Node power P_rRelated fresh target functionCarry out as follows：

(5a1) according to as γ=min { γ_SR,γ_SD+γ_RDWhen, system Signal to Interference plus Noise Ratio is maximum, i.e. γ=γ_SRCharacteristic, set up Source node and the relational expression of via node power：k_srk_sdP_s ²+k_srP_s=k_rrk_rdP_r ²+k_rdP_r；

(5a2) by all of source node power P_sAll use via node power P_rReplace and substitute into γ=γ_SR, obtain only saving with relaying Point power P_rRelated fresh target function：

$\begin{array}{cc}\underset{(P_s,P_r)}{max}& \mathrm{\γ}=\frac{-k_{sr}+\sqrt{{k_{sr}}^2+4k_{sr}{k}_{sd}(k_{rr}{k}_{rd}{P_r}^2+k_{rd}{P}_r)}}{2k_{sd}(k_{rr}{P}_r+1)}\end{array}$

S.t.0 ＜ P_s+P_r≤P_t

0 ＜ P_s≤P_max, 0 ＜ P_r≤P_max.

7. in method according to claim 1, wherein step (5b) according to maximize minima theory obtain only with relaying Node power P_rRelated fresh target functionCarry out as follows：

(5b1) basis works as γ_SR=γ_SD+γ_RDWhen, system Signal to Interference plus Noise Ratio is maximum, i.e. γ=γ_SRCharacteristic, set up source node work( Rate and the relational expression of via node power：k_srP_s-k_sdP_s=k_rrk_rdP_r ²+k_rdP_r+k_sdk_rrP_sP_r；

(5b2) by all of source node power P_sAll use via node power P_rReplace and substitute into γ=γ_SR, obtain only saving with relaying Point power P_rRelated fresh target function：

$\begin{array}{cc}\underset{(P_s,P_r)}{max}& \mathrm{\γ}=\frac{k_{sr}{k}_{rd}{P}_r}{k_{sr}-k_{sd}-k_{sd}{k}_{rr}{P}_r}\end{array}$

S.t.0 ＜ P_s+P_r≤P_t

0 ＜ P_s≤P_max, 0 ＜ P_r≤P_max.