CN103095412A - Precoding method for joint cell relay in broadcast channel based on relay cooperation - Google Patents

Abstract

The invention discloses a precoding method for a joint cell relay in a broadcast channel based on relay cooperation. The method comprises precoding of an information source, relay processing and receiving of information sinks. First, the information source initializes an information source precoding matrix and a relay processing matrix, and then the information source establishes a receiving matrix for each information sink based on the previous two initialized matrixes. Then, the optimized information source precoding matrix and each information sink receiving matrix are presumed to be unchanged, the relay processing matrix is optimized. In a similar way, the optimized information source and the relay processing matrix are presumed to be unchanged, and the receiving matrix of each information sink is optimized. Iteration of the above process goes on until convergence. Finally, the information source transmits the relay processing matrix to the relay and broadcasts the precoding matrix of the information source and the relay processing matrix to an information sink group. By means of the method, a direct link from the information source to the information sink group and a relay link between the information source and the relay and the information sink group are utilized innovatively so that total capacity of a system is improved.

Description

Method for precoding based on the combined signal source relaying in the broadcast channel of relay cooperative

Technical field

The present invention relates to wireless communication field, specifically a kind of based on the combined signal source in the downlink broadcast channel of relay cooperative transmission and the linear pre-coding method of via node.

Background technology

Along with the development of the communication technology, people are more and more higher to performance requirements such as transmission rate, transmission reliability, resource utilization and communication securities.The people such as van der Meulen have proposed the concept of three terminal relay collaboration communications in 1971, its core concept be allow via node to from the information of information source end through simple amplification or compressed encoding or decoding again after coding again void pass to the stay of two nights, purpose is to help the information source transmission information to the stay of two nights.But to the eighties in last century, the research of relevant cooperative relay system reduces gradually, until just again cause researcher's extensive concern the beginning of this century.Because relay cooperative communication can to anti-multipath fading, be one of important means that improves communication quality.In junction network, information source sends message to the stay of two nights, and via node also receives this message simultaneously, and it is transmitted to the stay of two nights (or through forwarding again after certain processing).All message that last stay of two nights recycling is received are decoded.Under such mode, via node has played certain synergism to the communication between information source-stay of two nights, is equal to configure under the condition of many antennas not needing, and certain space diversity is provided, and has improved communication quality.

But the collaboration communication mode of the single relaying list of the single information source stay of two nights not too conforms to topological structure in present cellular network.The situation that single many antennas information source (base station) and single many antenna relay station are served a plurality of single antenna mobile subscribers simultaneously is a kind of the most practical collaboration communication pattern.By certain precoding strategy, can suppress the decoding difficult problem that causes to different user due to interference that cochannel transmission causes.Obviously, this way can allow a multi-antenna base station and antenna relay more than serve simultaneously a plurality of single antenna mobile subscribers.The people such as Chan-Byoung Chae were published in the article " MIMO Relaying With Linear Processing for Multiuser Transmission in Fixed Relay Networks " on IEEE Trans.Sig.Proc. in 2008, considered that early a base station takes model in a plurality of users simultaneously by a relay station.Out a kind of precoding algorithm that can satisfy the QoS service given in " Joint Beamforming and Power Control for Multi-antenna Relay Broadcast Channel with QoS Constraints " literary composition that the people such as Rui Zhang delivered on IEEE Trans.Sig.Proc. in 2009." Joint Precoding Optimization for Multiuser Multi-Antenna Relaying Downlinks Using Quadratic Programming " literary composition that the people such as Wei Xu delivered on IEEETrans.Commun. in 2011 provides a kind of precoding algorithm based on double optimization.This just is that theoretical foundation has been established in the research that a plurality of mobile subscribers are served simultaneously by a relay station in a base station.But, above research all just considered the base station to relay station again to the link information between the mobile subscriber, do not consider that the base station is to the direct link information between the mobile subscriber.

The Toronto meeting of 3GPP in 2004 has proposed famous LTE(Long Term Evolution) project.This is a transition between 3G and 4G technology, is the global standards of 3.9G.It improves and has strengthened the aerial access technology of 3G, adopt OFDM and MIMO as the sole criterion of its wireless network evolution, can provide the peak rate of descending 326Mbit/s and up 86Mbit/s under the 20MHz spectral bandwidth, improved the performance of Cell Edge User, improved cell capacity and reduce system delay.The Type-II relaying be in the LTE project about a kind of important motion of relaying, such relaying does not have residential quarter ID, its application purpose is to provide the diversity to main eNB signal, perhaps with the main eNB transmission that cooperates together, to increase cell capacity.The maximum difference in theory research of Type-II relaying and Type-I relaying is, in the Type-II relay scene, exists direct communication link between information source (eNB) and the stay of two nights (UE).Therefore, according to such network topology structure, the application of scheme in multiple source Type-II relay scene that utilizes direct link information to carry out precoding just follows a well mapped-out plan.And all being based on the Type-I model for the research of information source (eNB) precoding at present, the channel information of direct link is than not being effectively utilized.The Precoding Design that whether can utilize the information of direct link to be correlated with promotes the performance of this system, not relevant research in this respect.

Summary of the invention

The object of the invention is to overcome prior art and ignore information source to the deficiency of stay of two nights group's direct link, information source is controlled and weighting factor to the channel information between stay of two nights group and power separately to the channel information between via node and via node to the direct channels information between stay of two nights group, information source according to information source, then the encoder matrix of initialization information source and via node generates stay of two nights group's receiving matrix.On this basis, suppose relay node processing matrix and stay of two nights group receiving matrix are constant, remove to optimize the information source matrix according to optimal method; Then suppose that information source pre-coding matrix and stay of two nights group receiving matrix are constant, optimize the via node processing array, suppose again the constant renewal stay of two nights of the processing array group's of information source encoder matrix and via node receiving matrix, the iteration said process is until till convergence, thereby obtains optimum information source pre-coding matrix and the transmitting-receiving processing array of via node.This method innovation ground has comprehensively been weighed information source and had been advanced via node to the repeated link information between stay of two nights group to the direct link information between stay of two nights group and information source, maximize in the hope of power system capacity, a kind of linear pre-coding method based on combined signal source and via node in the downlink broadcast channel of relay cooperative transmission is provided.

According to an aspect of the present invention, provide a kind of based on the combined signal source in the downlink broadcast channel of relay cooperative transmission and the linear pre-coding method of via node, comprise the following steps:

Step 1: information source BS arrives stay of two nights group MU according to acquired information source self ₁, MU ₂..., MU _KBetween channel information, information source and via node between channel information H _rb, the channel information H between via node and stay of two nights group _1r, H _2r..., H _Kr, information source transmitting power P _s, via node transmitting power P _r, information source antenna number M _s, via node antenna number M _r, stay of two nights number K, each stay of two nights antenna number N _kThe estimate variance 1 of the noise after number and normalization, the initial linear pre-coding matrix of initialization information source end

Via node initial treatment matrix F=ρ I, and each stay of two nights receiving matrix A that extrapolates according to initial linear pre-coding matrix P and via node initial treatment matrix F ₁, A ₂..., A _K, wherein:

I is unit matrix; P _iIt is i user's pre-coding matrix; ρ is power control factor, satisfies the power control equations of via node

Wherein, the mark of matrix A is asked in Tr (A) expression; $A_k=P_k^H{H}_k^H{(H_k{\mathrm{PP}}^H{H}_k^H+G_k{G}_k^H)}^{-1},$ Wherein, $H_k=\left[\begin{array}{c}{H}_{\mathrm{kb}}\\ H_{\mathrm{kr}}{\mathrm{FH}}_{\mathrm{rb}}\end{array}\right],$ For formula $G_k{G}_k^H=\mathrm{diag}(I,I+H_{\mathrm{kr}}{\mathrm{FF}}^H{H}_{\mathrm{kr}}^H),$ Diag (a, b) expression diagonal element is a, the diagonal matrix of b; Information source self is to stay of two nights group MU ₁, MU ₂..., MU _KBetween channel information comprise direct channels information H between information source and stay of two nights group _1b, H _2b..., H _Kb

Step 2: suppose fixed relay station initial treatment matrix F and each stay of two nights receiving matrix A ₁, A ₂..., A _K, the initial linear pre-coding matrix P of optimization information source BS;

Step 3: suppose fixedly initial linear pre-coding matrix P and each stay of two nights receiving matrix A of information source BS ₁, A ₂..., A _K, optimize via node initial treatment matrix F;

Step 4: suppose fixedly initial linear pre-coding matrix P and the via node initial treatment matrix F of information source BS, optimize each stay of two nights receiving matrix A ₁, A ₂..., A _K

Step 5: the described step 2 of iteration is to step 4 until till convergence, obtain the information source initial linear pre-coding matrix P of optimum via node initial treatment matrix F, optimum, then, information source BS is transferred to via node to the via node initial treatment matrix F of the optimum that obtains, and the information source initial linear pre-coding matrix P of the optimum that obtains and optimum via node initial treatment matrix F are broadcast to each stay of two nights together, so far, cataloged procedure finishes, and begins to carry out data information transfer.

Preferably, described step 2 is specially: suppose fixed relay station initial treatment matrix F and each stay of two nights receiving matrix A ₁, A ₂..., A _K, the initial linear pre-coding matrix P=[P of optimization information source BS _l..., P _k, P _K], k P wherein _kBe the function about power control factor λ:

$P_k\left(\mathrm{\λ}\right)=w_k{(\underset{k=1}{\overset{K}{\mathrm{\Σ}}}{w}_k{H}_k^H{A}_k^H{W}_k{A}_k{H}_k+\mathrm{\λI})}^{-1}{H}_k^H{A}_k^H{W}_k,$

Wherein, A ^HRepresent the conjugate transpose of A, the mark of matrix A, w are asked in Tr (A) expression _kIf the weight factor that represents k the stay of two nights is the same w of the weight of each stay of two nights ₁=w ₂=...=w _K=1; I represents unit matrix; λ is that power control factor makes Tr (P (λ) P ^H(λ))≤P _s

And

$E_k=(I-A_k{H}_k{P}_k){(I-A_k{H}_k{P}_k)}^H+\underset{i\≠k}{\mathrm{\Σ}}{A}_k{H}_k{P}_i{P_i}^H{H}_k^H{A}_k^H+A_k{G}_k{G}_k^H{A}_k^H.$

Preferably, described step 3 is specially: suppose fixedly initial linear pre-coding matrix P and each stay of two nights receiving matrix A of information source BS _l, A ₂..., A _K, optimize via node initial treatment matrix F, wherein F is the function about power control factor μ:

$F\left(\mathrm{\μ}\right)={(\underset{k=1}{\overset{K}{\mathrm{\Σ}}}{w}_k{H}_{\mathrm{kr}}^H{A}_{2k}^H{W}_k{A}_{2k}{H}_{\mathrm{kr}}+\mathrm{\μI})}^{-1}(\underset{k=1}{\overset{K}{\mathrm{\Σ}}}-w_k{\mathrm{\Δ}}_k){(H_{\mathrm{rb}}{\mathrm{PP}}^H{H}_{\mathrm{rb}}^H+I)}^{-1}$ , wherein, ${\mathrm{\Δ}}_k=H_{\mathrm{kr}}^H{A}_{2k}^H{W}_k{A}_{1k}{H}_{\mathrm{kb}}{\mathrm{PP}}^H{H}_{\mathrm{rb}}^H-H_{\mathrm{kr}}^H{A}_{2k}^H{W}_k{P}_k^H{H}_{\mathrm{rb}}^H,$ A _1kBe the receiving matrix of user k at the first time slot, A _2kBe the receiving matrix of user k at the second time slot, power control factor μ satisfies $\mathrm{Tr}\left(F\left(\mathrm{\μ}\right)(H_{\mathrm{rb}}{\mathrm{PP}}^H{H}_{\mathrm{rb}}^H+I)F{\left(\mathrm{\μ}\right)}^H\right)\≤P_r.$

Operation principle of the present invention is as follows:

The precoding mechanism that comprises direct link: precoding mechanism in the past be only all the channel information of considering between information source-relaying-stay of two nights group, and the direct link information between information source-stay of two nights group is left in the basket, thereby causes the loss of direct link capacity.The present invention proposes a kind of Iterative Design method, then information source sets up initial stay of two nights group receiving matrix according to initialized information source pre-coding matrix and via node transmitting-receiving processing array at first according to the channel information initialization information source pre-coding matrix between the direct link information between information source-stay of two nights group and information source-relaying-stay of two nights group and the processing receiving matrix of via node; Then set up iterative process: suppose that at first the processing array of via node and stay of two nights group receiving matrix are constant, optimize the pre-coding matrix of information source, then the receiving matrix of supposing the pre-coding matrix of information source and stay of two nights group is constant, optimize the transmitting-receiving processing array of via node, final updating stay of two nights group's receiving matrix, the iteration said process is until convergence.

On strategy, direct link information and the channel information between information source-relaying-stay of two nights group comprehensively weighed between information source-stay of two nights group design information source pre-coding matrix and via node reception ﹠ disposal matrix, make power system capacity maximize.

Description of drawings

By reading the detailed description of non-limiting example being done with reference to the following drawings, it is more obvious that other features, objects and advantages of the present invention will become:

Fig. 1 is 1-1-K repeat broadcast network abstract graph.

Fig. 2 is the concrete enforcement example of 1-1-K repeat broadcast network.

Embodiment

The present invention is described in detail below in conjunction with specific embodiment.Following examples will help those skilled in the art further to understand the present invention, but not limit in any form the present invention.Should be pointed out that to those skilled in the art, without departing from the inventive concept of the premise, can also make some distortion and improvement.These all belong to protection scope of the present invention.

As shown in Figure 1, the application scenarios of the present embodiment is " 1-1-K " repeat broadcast network, namely includes the BS of antenna information source node more than 1, antenna relay node R S more than 1 and K many antennas stay of two nights MU in network ₁, MU ₂..., MU _K

As shown in Figure 2, the present embodiment adopts channel model shown in this figure.1 information source (BS) use Base Station(base station) expression, via node RS Relay Station(relay station) expression, stay of two nights group uses MU ₁, MU ₂..., MU _K(user) expression.Wherein there is M the base station _sIndividual antenna, relay station has M _rIndividual antenna, K user, k user only has N _kIndividual antenna.We further suppose

This complete signals transmission divides two orthogonal channels, can be the time-division orthogonal channel, or the frequency division orthogonal channel.Suppose that two of short duration orthogonal channels remain unchanged in same once transmission.Carried out Precoding Design before the signal transmission.Carry out according to the following steps at every turn:

Step 1: information source BS arrives stay of two nights group MU according to acquired information source self ₁, MU ₂..., MU _KBetween direct channels information matrix H _1b, H _2b..., H _Kb, channel information H between information source and via node _rb, the channel information H between via node and stay of two nights group _1r, H _2r..., H _Kr, initialization information source initial linear pre-coding matrix

Via node initial treatment matrix F=ρ I and each stay of two nights receiving matrix A ₁, A ₂..., A _K

Wherein: I is unit matrix; P _iIt is i user's pre-coding matrix; ρ is the power control equations that power control factor satisfies via node

Wherein, the mark of matrix A is asked in Tr (A) expression.

Step 2: suppose fixed relay station initial treatment matrix F and each stay of two nights receiving matrix A ₁, A ₂..., A _K, the initial linear pre-coding matrix P of optimization information source BS;

Step 3: suppose fixedly initial linear pre-coding matrix P and each stay of two nights receiving matrix A of information source BS ₁, A ₂..., A _K, optimize via node initial treatment matrix F;

Step 4: suppose fixedly initial linear pre-coding matrix P and the via node initial treatment matrix F of information source BS, optimize each stay of two nights receiving matrix A ₁, A ₂..., A _K

Step 5: the described step 2 of iteration is to step 4 until till convergence, obtain the information source initial linear pre-coding matrix P of optimum via node initial treatment matrix F, optimum.Then, information source BS is transferred to via node to the via node initial treatment matrix F of the optimum that obtains, and the information source initial linear pre-coding matrix P of the optimum that obtains and optimum via node initial treatment matrix F are broadcast to each stay of two nights together, so far, cataloged procedure finishes, and begins to carry out data information transfer.

More specifically:

Step 1: channel estimating.The base station estimates that the base station to the channel information between customer group, obtains the channel information matrix H of direct link _SB=H _1b, H _2b..., H _KbRelay station is estimated the base station to the channel information matrix between relay station and relay station to the channel information matrix between customer group, thereby obtains H _rbAnd H _SR=H _1r, H _2r..., H _Kr

The base station is according to the direct channels information H between information source and stay of two nights group _1b, H _2b..., H _Kb, channel information H between information source and via node _rb, the channel information H between via node and stay of two nights group _1r, H _2r..., H _Kr, the initial linear pre-coding matrix P of initialization information source end, via node initial treatment matrix F and each stay of two nights receiving matrix A _l, A ₂..., A _K, namely:

Via node initial treatment matrix F=ρ I (ρ satisfies the power control equations of via node, namely: I is unit matrix), and the receiving matrix A of each stay of two nights of extrapolating according to initial p and F ₁, A ₂..., A _K, wherein $A_k=P_k^H{H}_k^H{(H_k{\mathrm{PP}}^H{H}_k^H+G_k{G}_k^H)}^{-1},$ Wherein $H_k=\left[\begin{array}{c}{H}_{\mathrm{kb}}\\ H_{\mathrm{kr}}{\mathrm{FH}}_{\mathrm{rb}}\end{array}\right],$ For formula

Diag (a, b) expression diagonal element is a, the diagonal matrix of b;

Step 2: suppose fixed relay station initial treatment matrix F and each stay of two nights receiving matrix A ₁, A ₂..., A _K, the initial linear pre-coding matrix P=[P of optimization information source BS _l..., P _k, P _K], k P wherein _kBe the function about power control factor λ:

$P_k\left(\mathrm{\λ}\right)=w_k{(\underset{k=1}{\overset{K}{\mathrm{\Σ}}}{w}_k{H}_k^H{A}_k^H{W}_k{A}_k{H}_k+\mathrm{\λI})}^{-1}{H}_k^H{A}_k^H{W}_k,$

Wherein, A ^HRepresent the conjugate transpose of A, the mark of matrix A, w are asked in Tr (A) expression _kWeight factor (if the same w of the weight of each stay of two nights that represents k the stay of two nights ₁=w ₂=...=w _K=1), I represents unit matrix, and λ is that power control factor makes Tr (P (λ) P ^H(λ))≤P _s,

And

$E_k=(I-A_k{H}_k{P}_k){(I-A_k{H}_k{P}_k)}^H+\underset{i\≠k}{\mathrm{\Σ}}{A}_k{H}_k{P}_i{P_i}^H{H}_k^H{A}_k^H+A_k{G}_k{G}_k^H{A}_k^H.$

Step 3: suppose fixedly initial linear pre-coding matrix P and each stay of two nights receiving matrix A of information source BS ₁, A ₂..., A _K, optimize via node initial treatment matrix F, wherein F is the function about power control factor μ:

$F\left(\mathrm{\μ}\right)={(\underset{k=1}{\overset{K}{\mathrm{\Σ}}}{w}_k{H}_{\mathrm{kr}}^H{A}_{2k}^H{W}_k{A}_{2k}{H}_{\mathrm{kr}}+\mathrm{\μI})}^{-1}(\underset{k=1}{\overset{K}{\mathrm{\Σ}}}-w_k{\mathrm{\Δ}}_k){(H_{\mathrm{rb}}{\mathrm{PP}}^H{H}_{\mathrm{rb}}^H+I)}^{-1}$ , wherein ${\mathrm{\Δ}}_k=H_{\mathrm{kr}}^H{A}_{2k}^H{W}_k{A}_{1k}{H}_{\mathrm{kb}}{\mathrm{PP}}^H{H}_{\mathrm{rb}}^H-H_{\mathrm{kr}}^H{A}_{2k}^H{W}_k{P}_k^H{H}_{\mathrm{rb}}^H,$ A _1kBe the receiving matrix of user k at the first time slot, A _2kBe the receiving matrix of user k at the second time slot, μ is power control factor, satisfies $\mathrm{Tr}\left(F\left(\mathrm{\μ}\right)(H_{\mathrm{rb}}{\mathrm{PP}}^H{H}_{\mathrm{rb}}^H+I)F{\left(\mathrm{\μ}\right)}^H\right)\≤P_r.$

Step 4: suppose fixedly initial linear pre-coding matrix P and the via node initial treatment matrix F of information source BS, optimize each stay of two nights receiving matrix A ₁, A ₂..., A _K,

Wherein $H_k=\left[\begin{array}{c}{H}_{\mathrm{kb}}\\ H_{\mathrm{kr}}{\mathrm{FH}}_{\mathrm{rb}}\end{array}\right],$ For formula $G_k{G}_k^H=\mathrm{diag}(I,I+H_{\mathrm{kr}}{\mathrm{FF}}^H{H}_{\mathrm{kr}}^H),$ Diag (a, b) expression diagonal element is a, the diagonal matrix of b.Step 5: the described step 2 of iteration is to step 4 until till convergence, obtain the information source initial linear pre-coding matrix P of optimum via node initial treatment matrix F, optimum.Then, information source BS is transferred to via node to the via node initial treatment matrix F of the optimum that obtains, and information source initial linear pre-coding matrix P and the via node initial treatment matrix F of the optimum that obtains are broadcast to each stay of two nights together, so far, cataloged procedure finishes, and begins to carry out data information transfer.

Relay station the data-signal that receives in the first orthogonal channel and noise take advantage of with a via node initial treatment matrix F after, be broadcast to the user in second orthogonal channel.

(k receiving matrix corresponding to user is A to each user with receiving matrix _k) receive in the first orthogonal channel and after the data in the second orthogonal channel, be decrypted operation.One time the transport communication process finishes.

In the present embodiment, set two orthogonal channels, in the first orthogonal channel, the base station the data broadcast of precoding to via node and stay of two nights group.In second orthogonal channel, via node is transmitted to the user after the data that receive are processed through F.The combined signal source that the present invention proposes and the pre-coding scheme of via node, fully utilize the base station and passed through via node to the repeated link information between the user to user's direct link information and base station, original scheme of comparing, fully balance the base station contribute through the repeated link that is relayed between the user to the direct link between the user and base station, so capacity is greatly improved.

Above specific embodiments of the invention are described.It will be appreciated that, the present invention is not limited to above-mentioned specific implementations, and those skilled in the art can make various distortion or modification within the scope of the claims, and this does not affect flesh and blood of the present invention.

Claims (3)

1. the method for precoding based on the combined signal source relaying in the broadcast channel of relay cooperative, is characterized in that, comprises the following steps:

Step 1: information source BS arrives stay of two nights group MU according to acquired information source self ₁, MU ₂..., MU _KBetween channel information, information source and via node between channel information H _rb, the channel information H between via node and stay of two nights group _1r, H _2r..., H _Kr, information source transmitting power P _s, via node transmitting power P _r, information source antenna number M _s, via node antenna number M _r, stay of two nights number K, each stay of two nights antenna number N _kThe estimate variance 1 of the noise after number and normalization, the initial linear pre-coding matrix of initialization information source end

Via node initial treatment matrix F=ρ I, and each stay of two nights receiving matrix A that extrapolates according to initial linear pre-coding matrix P and via node initial treatment matrix F ₁, A ₂..., A _K, wherein:

I is unit matrix; P _iIt is i user's pre-coding matrix; ρ is power control factor, satisfies the power control equations of via node

Wherein, the mark of matrix A is asked in Tr (A) expression; $A_k=P_k^H{H}_k^H{(H_k{\mathrm{PP}}^H{H}_k^H+G_k{G}_k^H)}^{-1},$ Wherein, $H_k=\left[\begin{array}{c}{H}_{\mathrm{kb}}\\ H_{\mathrm{kr}}{\mathrm{FH}}_{\mathrm{rb}}\end{array}\right],$ For formula $G_k{G}_k^H=\mathrm{diag}(I,I+H_{\mathrm{kr}}{\mathrm{FF}}^H{H}_{\mathrm{kr}}^H),$ Diag (a, b) expression diagonal element is a, the diagonal matrix of b; Information source self is to stay of two nights group MU ₁, MU ₂..., MU _KBetween channel information comprise direct channels information H between information source and stay of two nights group _1b, H _2b..., H _Kb

Step 2: suppose fixed relay station initial treatment matrix F and each stay of two nights receiving matrix A ₁, A ₂..., A _K, the initial linear pre-coding matrix P of optimization information source BS;

Step 3: suppose fixedly initial linear pre-coding matrix P and each stay of two nights receiving matrix A of information source BS ₁, A ₂..., A _K, optimize via node initial treatment matrix F;

Step 4: suppose fixedly initial linear pre-coding matrix P and the via node initial treatment matrix F of information source BS, optimize each stay of two nights receiving matrix A ₁, A ₂..., A _K

Step 5: the described step 2 of iteration is to step 4 until till convergence, obtain the information source initial linear pre-coding matrix P of optimum via node initial treatment matrix F, optimum, then, information source BS is transferred to via node to the via node initial treatment matrix F of the optimum that obtains, and the information source initial linear pre-coding matrix P of the optimum that obtains and optimum via node initial treatment matrix F are broadcast to each stay of two nights together, so far, cataloged procedure finishes, and begins to carry out data information transfer.

2. the method for precoding based on the combined signal source relaying in the broadcast channel of relay cooperative according to claim 1, is characterized in that, described step 2 is specially: suppose fixed relay station initial treatment matrix F and each stay of two nights receiving matrix A ₁, A ₂..., A _K, the initial linear pre-coding matrix P=[P of optimization information source BS _l..., P _k, P _K], k P wherein _kBe the function about power control factor λ:

$P_k\left(\mathrm{\λ}\right)=w_k{(\underset{k=1}{\overset{K}{\mathrm{\Σ}}}{w}_k{H}_k^H{A}_k^H{W}_k{A}_k{H}_k+\mathrm{\λI})}^{-1}{H}_k^H{A}_k^H{W}_k,$

Wherein, A ^HRepresent the conjugate transpose of A, the mark of matrix A, w are asked in Tr (A) expression _kIf the weight factor that represents k the stay of two nights is the same w of the weight of each stay of two nights ₁=w ₂=...=w _K=1; I represents unit matrix; λ is that power control factor makes Tr (P (λ) P ^H(λ))≤P _s

And $E_k=(I-A_k{H}_k{P}_k){(I-A_k{H}_k{P}_k)}^H+\underset{i\≠k}{\mathrm{\Σ}}{A}_k{H}_k{P}_i{P_i}^H{H}_k^H{A}_k^H+A_k{G}_k{G}_k^H{A}_k^H.$

3. the method for precoding based on the combined signal source relaying in the broadcast channel of relay cooperative according to claim 2, is characterized in that, described step 3 is specially: suppose fixedly initial linear pre-coding matrix P and each stay of two nights receiving matrix A of information source BS ₁, A ₂..., A _K, optimize via node initial treatment matrix F, wherein F is the function about power control factor μ:

$F\left(\mathrm{\μ}\right)={(\underset{k=1}{\overset{K}{\mathrm{\Σ}}}{w}_k{H}_{\mathrm{kr}}^H{A}_{2k}^H{W}_k{A}_{2k}{H}_{\mathrm{kr}}+\mathrm{\μI})}^{-1}(\underset{k=1}{\overset{K}{\mathrm{\Σ}}}-w_k{\mathrm{\Δ}}_k){(H_{\mathrm{rb}}{\mathrm{PP}}^H{H}_{\mathrm{rb}}^H+I)}^{-1}$ , wherein, ${\mathrm{\Δ}}_k=H_{\mathrm{kr}}^H{A}_{2k}^H{W}_k{A}_{1k}{H}_{\mathrm{kb}}{\mathrm{PP}}^H{H}_{\mathrm{rb}}^H-H_{\mathrm{kr}}^H{A}_{2k}^H{W}_k{P}_k^H{H}_{\mathrm{rb}}^H,$ A _1kBe the receiving matrix of user k at the first time slot, A _2kBe the receiving matrix of user k at the second time slot, power control factor μ satisfies $\mathrm{Tr}\left(F\left(\mathrm{\μ}\right)(H_{\mathrm{rb}}{\mathrm{PP}}^H{H}_{\mathrm{rb}}^H+I)F{\left(\mathrm{\μ}\right)}^H\right)\≤P_r.$

Images