CN103826295A - Resource optimization algorithm applicable to two-way relay system - Google Patents

Abstract

A resource optimization algorithm suitable for two-way relay system, involving relay selection and power allocation. The two transceivers use the wireless channel to send the pilot sequence respectively; each relay node estimates its own two-hop channel coefficient, sets and starts its own timer and is in the listening mode, once it detects that there is a relay accessing the network , all other relays immediately stop the timer and exit; the one connected to the network is the optimal relay, which broadcasts its own channel state information to the two transceivers; the two transceivers calculate their respective optimal transmission powers, and simultaneously broadcast their respective information After the optimal relay receives the mixed signal of the two transceivers, it uses PNC mapping to obtain the transmitted signal, calculates the optimal transmission power, and broadcasts the mapped signal; the two transceivers receive the optimal relay to send After the data, the information sent by the other transceiver is respectively obtained through PNC demodulation processing; the present invention improves the transmission rate of the system by allocating resources.

Description

A kind of resource optimization algorithm that is applicable to bidirectional relay system

Technical field

The present invention relates to the resource optimization of bidirectional relay system, is specifically a kind of power division and relay selection algorithm that is applicable to two-way cooperative transmission system.

Background technology

Bi-directional relaying is the efficient protocal that promotes spectrum efficiency, and in bi-directional relaying mechanism, one or more via nodes are configured between two transceivers to set up the transfer of data of both direction.Transmitted in both directions has 3 kinds of implementations: 4 traditional one step process comprise two one-way junction mechanism, and it does not also rely on network code; TDBC (time division broadcast, the time-division multicast) mode of 3 steps needs 3 transmission time slots to exchange two groups of information symbols, has therefore promoted the bandwidth availability of 4 one step process; In order further to promote spectrum efficiency, there is again the MABC (multiple access broadcast, multiple access multicast) of 2 steps, it only needs to take two time slots and can complete the exchange of two groups of information symbols.According to via node different processing mode to received signal, the mode of operation of MABC can be divided into again: ANC (Analog Network Coding, analog network coding) and PNC (Physical-layer Network coding, physical-layer network coding).

In addition studies show that, for specific node pair, only select an optimum relaying for it and can make cooperating relay mechanism obtain maximum value with minimum expense.Document " Relay Selection in Dual-Hop Vehicular Networks " has been studied double bounce DF (Decode-and-Forward, decoding forwards) mobile network's relay selection." Performance Analysis of Hybrid Relay Selection in Cooperative Wireless Systems " considered that via node not only can be operated in DF but also can be operated in AF (Amplify-and-Forward, amplification forwarding) when pattern, the mixed mode relay selection that to maximize received signal to noise ratio be target.And from the analysis that has document, by tasked to source and selected relaying dividing of limited power optimized, the performance of system also can get a promotion.S.Talwar has discussed joint relay selection and the power division of bilateral relay network while using ANC in document " Joint Relay Selection and Power Allocation for Two-Way Relay Networks ".Due to relay forwarding be that in above-mentioned document, in each direction, the transmission of data can exist some power losses from the copy after the mixed signal convergent-divergent of two transceivers, even this loss still exists under optimal power allocation.By contrast, PNC can make up this defect.

As shown in Figure 1, for using the system model of two-way cooperation transmission of PNC, comprise two transceiver (S ₁and S ₂), M via node (R _m, m=1 ..., M), each node is all equipped with an antenna and is operated in semiduplex mode.Imagine direct link between two transceivers very weak to such an extent as to can ignore, they only can complete communication by the help of via node so.This sight may occur in when direct link is by barrier, while blocking as high mountain etc.Every link in hypothetical network all experiences smooth slow fading, and channel remains unchanged in multiple transmission time slots, and mark is relayed to transceiver S from m respectively ₁and S ₂between multiple symmetric channel coefficient be h _mand g _m.

In Fig. 1, show the MABC of two steps of carrying out relay selection, in the first step, two transceiver S ₁and S ₂send data to via node, this can be called MAC (multiple access access) time slot simultaneously.Make x _irepresent by S _ithe information symbol sending, and meet E{|x _i| ²}=P _i, E{} represents mathematic expectaion here, P _irepresent signal x _itransmitted power, the reception signal of m relaying can be expressed as:

$y_{R_m}=h_m{x}_1+g_m{x}_2+n_{R_m}$

Wherein

relaying R _m, m=1 ..., the noise at M place.In second step, in M relaying one, supposes that it is m relaying without loss of generality, chosen it is received after signal is modulated and is transmitted to two transceivers.This is called BC (broadcast) time slot.The PNC modulation that via node uses, it does not need to carry out to received signal decoding, has shone upon (PNC mapping process can referring to document " Hot topic:Physical-layer network coding ") and only need simply to apply a kind of PNC.Therefore the signal that two transceivers receive is respectively:

${\mathrm{<figure-callout\; id="1"\; label="y"\; filenames="HDA0000460909910000021.png,HDA0000460909910000022.png"\; state="\{\{state\}\}">y</figure-callout>}}_1=h_m{x}_{R_m}+{\mathrm{<figure-callout\; id="1"\; label="n"\; filenames="HDA0000460909910000021.png,HDA0000460909910000022.png"\; state="\{\{state\}\}">n</figure-callout>}}_1$

${\mathrm{<figure-callout\; id="2"\; label="y"\; filenames="HDA0000460909910000021.png,HDA0000460909910000022.png"\; state="\{\{state\}\}">y</figure-callout>}}_2=g_m{x}_{R_m}+{\mathrm{<figure-callout\; id="2"\; label="n"\; filenames="HDA0000460909910000021.png,HDA0000460909910000022.png"\; state="\{\{state\}\}">n</figure-callout>}}_2$

Wherein relaying R _msymbol after the PNC of place modulation, meets

n _i, i=1,2 is the noises at i transceiver place.Afterwards, because each transceiver is known the own data that send before, by PNC demodulation process, they can obtain the data of another transceiver.

For the sake of simplicity, we suppose that noise sample all in network is independent identically distributed AWGN (additive white Gaussian noise), obey and distribute

and by transmission bandwidth unitization.According to document " Spectral efficient protocols for half-duplex fading relay channels " and " Hot topic:Physical-layer network coding ", from S ₁to S ₂and from S ₂to S ₁, use R _machievable rate during as selected relaying is expressed as:

$I_{12}^m=\frac{1}{2}\min (C({\mathrm{<figure-callout\; id="1"\; label="P"\; filenames="HDA0000460909910000021.png,HDA0000460909910000022.png"\; state="\{\{state\}\}">P</figure-callout>}}_1{\left|h_m\right|}^2/{\mathrm{\&sigma;}}^2),C(P_{R_m}{\left|g_m\right|}^2/{\mathrm{\&sigma;}}^2))$

$I_{21}^m=\frac{1}{2}\min (C({\mathrm{<figure-callout\; id="2"\; label="P"\; filenames="HDA0000460909910000021.png,HDA0000460909910000022.png"\; state="\{\{state\}\}">P</figure-callout>}}_2{\left|g_m\right|}^2/{\mathrm{\&sigma;}}^2),C(P_{R_m}{\left|h_m\right|}^2/{\mathrm{\&sigma;}}^2))$

Wherein C (x)=log ₂(1+x), exist the factor 1/2 to be because the transmission reality of every information symbol all takies two time slots, cause the spectrum efficiency drop by half in each direction, efficiency of transmission is low.

Summary of the invention

For the defect existing in prior art, the object of the present invention is to provide a kind of resource optimization algorithm that is applicable to bidirectional relay system, improve the transmission rate of system by Resources allocation.

For reaching above object, the technical scheme that the present invention takes is: a kind of resource optimization algorithm that is applicable to bidirectional relay system, comprise two transceiver and multiple potential via nodes that have information interaction, comprise the steps: that S1. two transceivers utilize wireless channel to send respectively pilot frequency sequence, each via node is received after pilot frequency sequence, estimates respectively the channel coefficients h of its two hop channel _mand g _m, and according to this via node R _mdouble bounce channel coefficients h _mand g _msetting a parameter is

timer, wherein a=max (α, β), b=min (α, β), α=| h _m| ²/ σ ²represent this via node R _mthe unit received signal to noise ratio of the first hop link, β=| g _m| ²/ σ ²represent this via node R _mthe unit received signal to noise ratio of the second hop link, σ ²for noise power; S2. each via node R _mwith one and parameter θ _mthe initial value that is inversely proportional to is opened own timer, and decays in zero process all in listen mode at wait timer separately; Once listen to any one relay access network network, other all via nodes stop at once the timer of oneself and exit; S3. the via node of access network is optimum via node, and it broadcasts the channel coefficients h of its two hop channel to two transceivers _mand g _m; S4. two transceivers calculate respectively optimum transmit power separately according to described channel condition information, broadcast the information of inclusion information amount separately with optimum transmit power simultaneously; S5. described optimum via node receives after the mixed signal from the constant power of two transceivers, and application PNC mapping obtains transmitted signal, calculates optimum transmit power according to local channel state information simultaneously, and the signal after mapping is broadcasted away; S6. two transceivers are all received after the data of optimum via node transmission, according to the data that send before separately, by PNC demodulation process, obtain separately the information that another transceiver sends.

On the basis of technique scheme, in described S2, each via node is according to T _m=λ/θ _mopen the timer of oneself, wherein T _mfor with parameter θ _mthe initial value being inversely proportional to, the constant value that λ is initial setting, is used for coordinating the time that between relaying, competition expends.

On the basis of technique scheme, described two transceivers are respectively S ₁and S ₂, they and via node R _moptimum transmit power be by maximize S ₁→ R _m→ S ₂and S ₂→ R _m→ S ₁on both direction, the little person of transmission rate obtains.

On the basis of technique scheme, while optimizing transmitted power, first bi-directional relaying is regarded as to two one-way junction systems, then obtain the suboptimum power of bidirectional relay system.

On the basis of technique scheme, when described suboptimum power is assigned to three node S that participate in communication ₁, S ₂and R _mtime, then promote targeted rate by adjusting suboptimum power or redistributing useless power

On the basis of technique scheme, by described power adjustment, as α>=β, to S ₁, S ₂and relaying R _mpower adjustment meet $\mathrm{\&Delta;}{\mathrm{<figure-callout\; id="1"\; label="P"\; filenames="HDA0000460909910000021.png,HDA0000460909910000022.png"\; state="\{\{state\}\}">P</figure-callout>}}_1^{*}=\frac{\mathrm{\&alpha;}(\mathrm{\&alpha;}-\mathrm{\&beta;})}{(2\mathrm{\&alpha;}+\mathrm{\&beta;})\mathrm{\&beta;}}{\stackrel{\&OverBar;}{P}}_1,\mathrm{\&Delta;}{\mathrm{<figure-callout\; id="2"\; label="P"\; filenames="HDA0000460909910000021.png,HDA0000460909910000022.png"\; state="\{\{state\}\}">P</figure-callout>}}_2^{*}=(1+\frac{\mathrm{\&alpha;}}{\mathrm{\&beta;}})\mathrm{\&Delta;}{\mathrm{<figure-callout\; id="1"\; label="P"\; filenames="HDA0000460909910000021.png,HDA0000460909910000022.png"\; state="\{\{state\}\}">P</figure-callout>}}_1^{*},$

time power optimization problem reach optimum point, wherein

for S ₁suboptimum performance number; The optimum transmit power value of two transceivers and best relay is:

wherein P _tfor gross power restriction; Targeted rate now

if α≤β, can regard two transceiver S as ₁with S ₂exchange position.

Beneficial effect of the present invention is: using in the bidirectional relay system of physical-layer network coding, improve the transmission rate of system by Resources allocation.Underway continuing while selection, each via node only needs to know the channel coefficients between own and two transceivers, does not need to know overall CSI (channel condition information).In addition, optimum via node only need be broadcast to local CSI two transceivers and can carry out power optimization.Resource of the present invention is distributed the distributed algorithm that belongs to low expense, in the two-way cooperation transmission of assurance, in two transmission direction fairness, has promoted the transmission rate of system.

Accompanying drawing explanation

Fig. 1 is the system model figure that uses the two-way cooperation transmission of PNC;

The capacity relationship of each link in network when Fig. 2 is the present invention's application suboptimum power policy;

Fig. 3 is for maximizing the link capacity bringing because of Modulating Power for target changes schematic diagram;

When Fig. 4 is M=4 in different mechanisms two directions the little person of speed with the change curve of SNR;

When Fig. 5 is M=1, SNR=15dB in different mechanisms two directions the little person of speed with the change curve of via node position d.

Embodiment

Below in conjunction with drawings and Examples, the present invention is described in further detail.

As shown in Figure 1, first need to draw the optimum transmit power that maximizes the little person of transmission rate in two directions.Maximize two transmission direction S ₁→ S ₂and S ₂→ S ₁with R _mthe little person of transmission rate during as via node

can be regarded as speed equalization problem, can be described as

$\begin{array}{c}\underset{P_{}}{\mathrm{<figure-callout\; id="1"\; label="max\; P"\; filenames="HDA0000460909910000021.png,HDA0000460909910000022.png"\; state="\{\{state\}\}">max</figure-callout>}}{I}_{\min }^m\\ s.t.{\mathrm{<figure-callout\; id="1"\; label="P"\; filenames="HDA0000460909910000021.png,HDA0000460909910000022.png"\; state="\{\{state\}\}">P</figure-callout>}}_1+{\mathrm{<figure-callout\; id="2"\; label="P"\; filenames="HDA0000460909910000021.png,HDA0000460909910000022.png"\; state="\{\{state\}\}">P</figure-callout>}}_2+P_{R_m}\&le;P_T\end{array}$

Wherein P ₁, P ₂represent respectively S ₁and S ₂transmitted power, P _tbe gross power restriction, because the target function in above formula is not continuously differentiable, therefore find optimum transmit power value according to above formula just so not directly perceived.Q.Zhang has introduced the power division of unidirectional DF relay system in document " Power allocation for regenerative relay channel with Rayleigh fading ", and points out distribute more power to the node with poor channel.

As a simple expansion, first the present invention regards bilateral relay network as two one-way junction systems, and obtains suboptimum power by solving following two optimization subproblems:

$\begin{array}{c}\underset{P_{}}{\mathrm{<figure-callout\; id="1"\; label="max\; P"\; filenames="HDA0000460909910000021.png,HDA0000460909910000022.png"\; state="\{\{state\}\}">max</figure-callout>}}{I}_{12}^m=\frac{1}{2}\min (C({\mathrm{<figure-callout\; id="1"\; label="P"\; filenames="HDA0000460909910000021.png,HDA0000460909910000022.png"\; state="\{\{state\}\}">P</figure-callout>}}_1{\left|h_m\right|}^2/{\mathrm{\&sigma;}}^2),C(P_{R_m}{\left|g_m\right|}^2/{\mathrm{\&sigma;}}^2))\\ s.t.{\mathrm{<figure-callout\; id="1"\; label="P"\; filenames="HDA0000460909910000021.png,HDA0000460909910000022.png"\; state="\{\{state\}\}">P</figure-callout>}}_1+P_{R_m}\&le;P_T-{\mathrm{<figure-callout\; id="2"\; label="P"\; filenames="HDA0000460909910000021.png,HDA0000460909910000022.png"\; state="\{\{state\}\}">P</figure-callout>}}_2\end{array}$

With

$\begin{array}{c}\underset{P_{}}{\mathrm{<figure-callout\; id="2"\; label="max\; P"\; filenames="HDA0000460909910000021.png,HDA0000460909910000022.png"\; state="\{\{state\}\}">max</figure-callout>}}{I}_{21}^m=\frac{1}{2}\min (C({\mathrm{<figure-callout\; id="2"\; label="P"\; filenames="HDA0000460909910000021.png,HDA0000460909910000022.png"\; state="\{\{state\}\}">P</figure-callout>}}_2{\left|g_m\right|}^2/{\mathrm{\&sigma;}}^2),C(P_{R_m}{\left|h_m\right|}^2/{\mathrm{\&sigma;}}^2))\\ s.t.{\mathrm{<figure-callout\; id="2"\; label="P"\; filenames="HDA0000460909910000021.png,HDA0000460909910000022.png"\; state="\{\{state\}\}">P</figure-callout>}}_2+P_{R_m}\&le;P_T-{\mathrm{<figure-callout\; id="1"\; label="P"\; filenames="HDA0000460909910000021.png,HDA0000460909910000022.png"\; state="\{\{state\}\}">P</figure-callout>}}_1\end{array}$

The solution of two formulas is respectively above

With

Wherein α=| h _m| ²/ σ ², β=| g _m| ²/ σ ².The suboptimum power of bidirectional relay system can obtain by merging two formulas above

${\stackrel{\&OverBar;}{P}}_1=\frac{{\mathrm{\&beta;}}^2}{{\mathrm{\&alpha;}}^2+{\mathrm{\&beta;}}^2+\mathrm{\&alpha;\&beta;}}{P}_T$

${\stackrel{\&OverBar;}{P}}_2=\frac{{\mathrm{\&alpha;}}^2}{{\mathrm{\&alpha;}}^2+{\mathrm{\&beta;}}^2+\mathrm{\&alpha;\&beta;}}{P}_T$

${\stackrel{\&OverBar;}{P}}_{R_m}=\frac{\mathrm{\&alpha;\&beta;}}{{\mathrm{\&alpha;}}^2+{\mathrm{\&beta;}}^2+\mathrm{\&alpha;\&beta;}}{P}_T$

In order to understand more intuitively the gap between initial optimization problem and suboptimum power, we utilize the method for drawing to be described.Meanwhile, because bidirectional relay system has symmetry, we only consider here | h _m|>=| g _m| situation, under symmetric case, | h _m|≤| g _m|, can be counted as two transceivers and exchanged their position.

As shown in Figure 2, shown in the time that suboptimum power is assigned to 3 nodes that participate in communication the relation of 4 link capacities in system.In the drawings, we are I by the capacity marking of link u → v _uv, wherein u, v ∈ { S ₁, S ₂, R _m.In addition, figure link capacity divides into groups according to the data flow in each direction, and the height of each grid has been indicated the size of link capacity.Can find out, suboptimum power policy has caused useless power consumption, and they are not to targeted rate

increase play positive acting.Therefore, can promote by adjusting suboptimum power or redistributing those useless power

Mark

for the suboptimum power according to above-mentioned divides timing link S ₁→ S ₂achievable rate,

for the suboptimum power according to above-mentioned divides timing link S ₂→ S ₁achievable rate, △ P _i>=0, i=1,2 with

represent respectively terminal S _iwith R _mthe performance number of adjusting.Due to

as seen from Figure 2, target capacity now be limited to first link pair.Therefore should meet P to the adjustment of power ₂decline, P ₁rise,

rise, increase here

in order to promote moreover, transceiver S ₂the power of saving should all be assigned to S ₁and R _m, meet

like this, former optimization problem can be equivalent to

$\begin{array}{c}\underset{\mathrm{\&Delta;}{\mathrm{<figure-callout\; id="1"\; label="P"\; filenames="HDA0000460909910000021.png,HDA0000460909910000022.png"\; state="\{\{state\}\}">P</figure-callout>}}_1,\mathrm{\&Delta;}{\mathrm{<figure-callout\; id="2"\; label="P"\; filenames="HDA0000460909910000021.png,HDA0000460909910000022.png"\; state="\{\{state\}\}">P</figure-callout>}}_2,\mathrm{\&Delta;}{P}_{R_m}}{\max }{I}_{\min }^m=\min (I_{12}^m,I_{21}^m)\\ s.t.\mathrm{\&Delta;}{\mathrm{<figure-callout\; id="2"\; label="P"\; filenames="HDA0000460909910000021.png,HDA0000460909910000022.png"\; state="\{\{state\}\}">P</figure-callout>}}_2=\mathrm{\&Delta;}{\mathrm{<figure-callout\; id="1"\; label="P"\; filenames="HDA0000460909910000021.png,HDA0000460909910000022.png"\; state="\{\{state\}\}">P</figure-callout>}}_1+\mathrm{\&Delta;}{P}_{R_m}\end{array}$

Here meet ${\mathrm{<figure-callout\; id="1"\; label="P"\; filenames="HDA0000460909910000021.png,HDA0000460909910000022.png"\; state="\{\{state\}\}">P</figure-callout>}}_1={\stackrel{\&OverBar;}{P}}_1+\mathrm{\&Delta;}{\mathrm{<figure-callout\; id="1"\; label="P"\; filenames="HDA0000460909910000021.png,HDA0000460909910000022.png"\; state="\{\{state\}\}">P</figure-callout>}}_1,{\mathrm{<figure-callout\; id="2"\; label="P"\; filenames="HDA0000460909910000021.png,HDA0000460909910000022.png"\; state="\{\{state\}\}">P</figure-callout>}}_2={\stackrel{\&OverBar;}{P}}_2-\mathrm{\&Delta;}{\mathrm{<figure-callout\; id="2"\; label="P"\; filenames="HDA0000460909910000021.png,HDA0000460909910000022.png"\; state="\{\{state\}\}">P</figure-callout>}}_2$ And $P_{R_m}={\stackrel{\&OverBar;}{P}}_{R_m}+\mathrm{\&Delta;}{P}_{R_m}.$ Obviously, the optimal solution of above formula will be

in time, reaches.Due to P ₂↓ and

s ₂→ S ₁the transmission rate of direction can be defined as

and because P ₁with

all increase,

can not be determined intuitively.But we notice a fact, unless reached exactly

otherwise useless power can exist always.Therefore above formula will be

time,

$C\left(\mathrm{\&alpha;}({\stackrel{\&OverBar;}{P}}_1+\mathrm{\&Delta;}{P}_1)\right)=C\left(\mathrm{\&beta;}({\stackrel{\&OverBar;}{P}}_{R_m}+\mathrm{\&Delta;}{P}_{R_m})\right)=C\left(\mathrm{\&beta;}({\stackrel{\&OverBar;}{P}}_2-\mathrm{\&Delta;}{P}_2)\right)$

Arrive optimum point.Meet constraint

under condition, the optimal solution that we finally obtain power adjustment is:

$\mathrm{\&Delta;}{\mathrm{<figure-callout\; id="1"\; label="P"\; filenames="HDA0000460909910000021.png,HDA0000460909910000022.png"\; state="\{\{state\}\}">P</figure-callout>}}_1^{*}=\frac{\mathrm{\&alpha;}(\mathrm{\&alpha;}-\mathrm{\&beta;})}{(2\mathrm{\&alpha;}+\mathrm{\&beta;})\mathrm{\&beta;}}{\stackrel{\&OverBar;}{P}}_1$

$\mathrm{\&Delta;}{\mathrm{<figure-callout\; id="2"\; label="P"\; filenames="HDA0000460909910000021.png,HDA0000460909910000022.png"\; state="\{\{state\}\}">P</figure-callout>}}_2^{*}=(1+\frac{\mathrm{\&alpha;}}{\mathrm{\&beta;}})\mathrm{\&Delta;}{\mathrm{<figure-callout\; id="1"\; label="P"\; filenames="HDA0000460909910000021.png,HDA0000460909910000022.png"\; state="\{\{state\}\}">P</figure-callout>}}_1^{*}$

$\mathrm{\&Delta;}{P}_{R_m}^{*}=\frac{\mathrm{\&alpha;}}{\mathrm{\&beta;}}\mathrm{\&Delta;}{\mathrm{<figure-callout\; id="1"\; label="P"\; filenames="HDA0000460909910000021.png,HDA0000460909910000022.png"\; state="\{\{state\}\}">P</figure-callout>}}_1^{*}$

Therefore the optimum transmit power value of former problem is

${\mathrm{<figure-callout\; id="1"\; label="P"\; filenames="HDA0000460909910000021.png,HDA0000460909910000022.png"\; state="\{\{state\}\}">P</figure-callout>}}_1^{*}=\frac{\mathrm{\&beta;}}{(2\mathrm{\&alpha;}+\mathrm{\&beta;})}{P}_T$

${\mathrm{<figure-callout\; id="2"\; label="P"\; filenames="HDA0000460909910000021.png,HDA0000460909910000022.png"\; state="\{\{state\}\}">P</figure-callout>}}_2^{*}=P_{R_m}^{*}=\frac{\mathrm{\&alpha;}}{(2\mathrm{\&alpha;}+\mathrm{\&beta;})}{P}_T$

Now relaying R _mreceive from S at MAC time slot ₁with S ₂signal-to-Noise (SNR) meet

the result of visible above formula meets PNC and modulates required power control requirement.At this moment targeted rate

$I_{\min }^{m*}=C\left(\frac{\mathrm{\&alpha;\&beta;}{\mathrm{<figure-callout\; id="2"\; label="P\; T"\; filenames="HDA0000460909910000021.png,HDA0000460909910000022.png"\; state="\{\{state\}\}">P</figure-callout>}}_T}{2\mathrm{\&alpha;}+\mathrm{\&beta;}}\right)$

As shown in Figure 3, the adjustment of link capacity while having described application optimum transmit power value.Wherein

i, j=1,2 marks power adjust to the variation bringing.Although R _mplace still has some excess power to cause

but these power can not be transferred to other node, otherwise can cause

decline, then reduced

As shown in Figure 1, in the resource optimization algorithm the present invention relates to, comprise that two have the mutual transceiver of informational needs and multiple potential via node, distributed resource optimization algorithm implementation step is as follows:

S1. two transceiver S ₁and S ₂utilize wireless channel to send respectively pilot frequency sequence, each via node R _m, m=1 ..., M, is receiving after pilot frequency sequence, estimates respectively the channel coefficients h of its two hop channel _mand g _m, and according to this via node R _mdouble bounce channel coefficients h _mand g _msetting a parameter is

timer, wherein a=max (α, β), b=min (α, β), from the expression formula of optimal objective speed, that can make function # _mmaximum via node just can form optimal path between two transceivers, and its timer can be designed to reduce at first zero.

S2. each via node R _mwith a channel quality θ with them _mthe initial value T being inversely proportional to _m=λ/θ _mopen the timer of oneself, the constant value that λ is initial setting here, is used for coordinating the time that between relaying, competition expends, and its unit depends on θ _munit.And each relaying wait for timer separately decay to zero process in listen mode, once listen to via node access network, other all via nodes stop at once the timer of oneself and exit.

S3. the via node of access network is optimum via node, and it broadcasts its double bounce channel condition information h to two transceivers _mand g _m.

S4. useful signal transmission starts, and two transceivers calculate respectively optimum transmit power separately according to described channel condition information, broadcasts the information that needs transmission with optimum transmit power simultaneously, and this is called multiple access access slot.

S5. described optimum via node receives after the mixed signal from the constant power of two transceivers, and application PNC mapping obtains transmitted signal, calculates optimum transmit power according to local channel state information simultaneously, and the signal after mapping is broadcasted away.

S6. two transceivers are received after the data that optimum via node sends, and according to the data that send before separately, by PNC demodulation process, obtain the information that another transceiver sends respectively.

We assess the performance of described resource optimization algorithm by emulation.A linear one dimension geometric network that comprises M via node for example, and by S ₁to S ₂parasang.Mark S ₁to R _mdistance be d _m, channel is modeled as

and

wherein

be generated as independent identically distributed multiple Gaussian random variable, κ=3rd, path loss index.By noise power unitization and define SNR=P _t/ σ ².Mechanism relatively comprises random selection via node (RRS, random relay selection), mean allocation transmitted power (EPA, equal power allocation), the joint relay selection that is applicable to two-way ANC junction network and power distribution algorithm (JRSPA) that the document " Joint Relay Selection and Power Allocation for Two-Way Relay Networks " of mentioning in suboptimum power division (SOPA, suboptimal power allocation) and background technology proposes.And we also will show that the embodiment of the present invention is as the performance of the described one-way junction system of document " Power allocation for regenerative relay channel with Rayleigh fading " of spread foundation as benchmark.

As shown in Figure 4, while comparing M=4, each mechanism can reach the little person of transmission rate with the situation of change of SNR on both direction.Apart from d _m, m=1 ..., M is produced as and independently between 0 to 1, obeys equally distributed stochastic variable." OPA+RS " is power division and the relay selection algorithm that the present invention proposes, and the optimal solution that its power obtains according to the present invention is estimated.First we observe, and OPA strategy is compared with SOPA mechanism with EPA within the scope of the whole SNR of emulation, can obtain the rate gain of 1-2dB and 3-4dB always, and select can obtain compared with RRS the gain of 4-5dB based on the distributed relay of OPA.And, in Fig. 4, show, be applicable to the JRSPA algorithm performance of two-way ANC relay system far away not as good as the OPA+RS of the present invention's proposition, description in background technology conforms to us for this.In addition,, compared with one-way junction transmission, the performance advantage of the bi-directional relaying of use PNC is along with the increase of SNR is more and more obvious.

As shown in Figure 5, describe each mechanism in the time of M=1, SNR=15dB and on both direction, can reach the little person of transmission rate with the variation of relaying position d.In Fig. 5, clearly show, no matter where relaying is positioned at, the OPA strategy that the present invention proposes can present best performance.By contrast, SOPA poor-performing, the performance of succeeding to the throne in the middle of especially while putting near two transceivers is poorer.We further observe, and because targeted rate has been considered the fairness of two transmission directions, the mid point of two transceivers is optimum positions of relaying.

The present invention is not limited to above-mentioned execution mode, for those skilled in the art, under the premise without departing from the principles of the invention, can also make some improvements and modifications, within these improvements and modifications are also considered as protection scope of the present invention.The content not being described in detail in this specification belongs to the known prior art of professional and technical personnel in the field.

Claims (6)

1. be applicable to a resource optimization algorithm for bidirectional relay system, comprise two transceiver and multiple potential via nodes that have information interaction, it is characterized in that, comprise the steps:

S1. two transceivers utilize wireless channel to send respectively pilot frequency sequence, and each via node is received after pilot frequency sequence, estimates respectively the channel coefficients h of its two hop channel _mand g _m, and according to this via node R _mdouble bounce channel coefficients h _mand g _msetting a parameter is

timer, wherein a=max (α, β), b=min (α, β), α=| h _m| ²/ σ ²represent this via node R _mthe unit received signal to noise ratio of the first hop link, β=| g _m| ²/ σ ²represent this via node R _mthe unit received signal to noise ratio of the second hop link, σ ²for noise power;

S2. each via node R _mwith one and parameter θ _mthe initial value that is inversely proportional to is opened own timer, and decays in zero process all in listen mode at wait timer separately; Once listen to any one relay access network network, other all via nodes stop at once the timer of oneself and exit;

S3. the via node of access network is optimum via node, and it broadcasts the channel coefficients h of its two hop channel to two transceivers _mand g _m;

S4. two transceivers calculate respectively optimum transmit power separately according to described channel condition information, broadcast the information of inclusion information amount separately with optimum transmit power simultaneously;

S5. described optimum via node receives after the mixed signal from the constant power of two transceivers, and application PNC mapping obtains transmitted signal, calculates optimum transmit power according to local channel state information simultaneously, and the signal after mapping is broadcasted away;

S6. two transceivers are all received after the data of optimum via node transmission, according to the data that send before separately, by PNC demodulation process, obtain separately the information that another transceiver sends.

2. a kind of resource optimization algorithm that is applicable to bidirectional relay system as claimed in claim 1, is characterized in that: in described S2, each via node is according to T _m=λ/θ _mopen the timer of oneself, wherein T _mfor with parameter θ _mthe initial value being inversely proportional to, the constant value that λ is initial setting, is used for coordinating the time that between relaying, competition expends.

3. a kind of resource optimization algorithm that is applicable to bidirectional relay system as claimed in claim 1, is characterized in that: described two transceivers are respectively S ₁and S ₂, they and via node R _moptimum transmit power be by maximize S ₁→ R _m→ S ₂and S ₂→ R _m→ S ₁on both direction, the little person of transmission rate obtains.

4. a kind of resource optimization algorithm that is applicable to bidirectional relay system as claimed in claim 3, is characterized in that: while optimizing transmitted power, first bi-directional relaying is regarded as to two one-way junction systems, then obtained the suboptimum power of bidirectional relay system.

5. a kind of resource optimization algorithm that is applicable to bidirectional relay system as claimed in claim 4, is characterized in that: when described suboptimum power is assigned to three node S that participate in communication ₁, S ₂and R _mtime, then promote targeted rate by adjusting suboptimum power or redistributing useless power

6. a kind of resource optimization algorithm that is applicable to bidirectional relay system as claimed in claim 5, is characterized in that: by described power adjustment, as α>=β, to S ₁, S ₂and relaying R _mpower adjustment meet $\mathrm{\&Delta;}{P}_1^{*}=\frac{\mathrm{\&alpha;}(\mathrm{\&alpha;}-\mathrm{\&beta;})}{(2\mathrm{\&alpha;}+\mathrm{\&beta;})\mathrm{\&beta;}}{\stackrel{\&OverBar;}{P}}_1,\mathrm{\&Delta;}{P}_2^{*}=(1+\frac{\mathrm{\&alpha;}}{\mathrm{\&beta;}})\mathrm{\&Delta;}{P}_1^{*},\mathrm{\&Delta;}{P}_{R_m}^{*}=\frac{\mathrm{\&alpha;}}{\mathrm{\&beta;}}\mathrm{\&Delta;}{P}_1^{*}$ Time power optimization problem reach optimum point, wherein

for S ₁suboptimum performance number; The optimum transmit power value of two transceivers and best relay is: $P_1^{*}=\frac{\mathrm{\&beta;}}{(2\mathrm{\&alpha;}+\mathrm{\&beta;})}{P}_T,P_2^{*}=P_{R_m}^{*}=\frac{\mathrm{\&alpha;}}{(2\mathrm{\&alpha;}+\mathrm{\&beta;})}{P}_T,$ Wherein P _tfor gross power restriction; Targeted rate now

If α≤β, can regard two transceiver S as ₁with S ₂exchange position.

Images