CN102868433B - Signal transmission method based on antenna selection in multiple-input multiple-output Y channel - Google Patents

Abstract

The invention discloses a signal transmission method based on antenna selection in a multiple-input multiple-output Y channel, mainly for solving the problem that the quantity of antennas is strictly limited in prior art. The signal transmission method comprises the steps that a relay node selects a receiving antenna; the relay node determines a multiple-address access time slot effective pre-encoding vector; a user node sends a signal to the relay node; the relay node decodes the received signal; the relay node selects a sending antenna; the relay node determines a broadcast time slot effect pre-encoding vector; the relay node broadcasts the signal to the user node; and the user node recovers useful data. Compared with the conventional signal transmission method, the signal transmission method provided by the invention has the advantage that the error rate property is greatly reduced.

Description

Based on the method for transmitting signals of sky line options in multiple-input and multiple-output Y-channel

Technical field

The invention belongs to wireless communication technology field, relate to interference alignment and network code, further relate to the method for transmitting signals of a kind of day line options, can be used for multiple-input and multiple-output Y-channel.

Background technology

Traditional bidirectional relay channel be a pair user or multipair user by the mutual photos and sending messages of relaying, under this model, each user sends information only to one of them user, receives the information sent from this user simultaneously.In actual applications, each user often needs information different from other user interactions.In this case, occurred multidirectional trunk channel model, multiple-input and multiple-output Y-channel is exactly wherein a kind of important communication channel model.

Multiple-input and multiple-output Y-channel model comprises K user node and a via node, wherein each user node equipment M root antenna, via node equipment N root antenna.Direct link is not had between different user nodes, and each user node respectively sends a unicast info to an other K-1 user node by via node, concrete communication process is as follows: first, K user node sends data to via node simultaneously, and this time slot is called multiple access access slot; Then, via node is broadcast to K user node after processing the information received, and this time slot is called time slot; Finally, by K user node respectively according to oneself Received signal strength and oneself translate in the information that multiple access access slot sends the information that an other K-1 user node issues oneself.

Multiple-input and multiple-output Y-channel can obtain than multi-user multiple-input and multiple-output and the higher degree of freedom of time division multiplexing, but the antenna number of existing multiple-input and multiple-output Y-channel to each node has very strict restriction.Korea S scholar Kwangwon Lee etc. analyze the condition of each node antennas equipment number of multiple-input and multiple-output Y-channel in " Feasibility Conditions of Signal Space Alignmentfor NetworkCoding on K-user MIMO Y channels; IEEE Trans.Inform.Theory " at article, adopt signal space alignment techniques at multiple access access slot and adopt the interference cancellation beam forming technique of coding Network Based at time slot, demonstrate each node antennas and need meet M>=K-1 and the condition of 2M-N>=1.Therefore, can the situation that the antenna number for each node does not meet the demands, design corresponding spatial alignment technology, to reach the maximum degree of freedom, is the problem needing at present to solve.

Summary of the invention:

The present invention is directed to the deficiency of technology, method for transmitting signals based on sky line options in a kind of multiple-input and multiple-output Y-channel is proposed, to solve when antenna arrangement number does not meet 2M-N >=1 condition, make system can obtain the maximum degree of freedom and reduce error rate of system.

Realize object technical scheme of the present invention, comprise the steps:

(1) trunk node selection reception antenna

(1.1) for each user's multiple access access slot channel matrix, all p is left out ₁..., p _{n-N '}row vector obtains respective multiple access access slot channel submatrix, and wherein, N is relay antenna number, the antenna number of N ' selected by relaying, p ₁..., p _{n-N '}∈ 1 ..., N} and p ₁≠ ... ≠ p _{n-N '}, 2M-N '=1, M>=K-1, K are user node number, and M is the antenna number of each user;

(1.2) according to each user's multiple access access slot channel matrix, adopt signal space alignment schemes, determine respective multiple access access slot precoding vector and multiple access access slot alignment vector;

(1.3) according to multiple access access slot alignment vector form multiple access access slot alignment matrix U _f,

$U_f=\left[\begin{array}{ccccc}{u}_f^1& ...& u_f^{\mathrm{\π}(i,j)}& ...& u_f^E\end{array}\right],$

Wherein, subscript π (i, j) is an index function, meets π (i, j)=π (j, i), and has f=τ (p ₁..., p _{n-N '}) be another index function, will $\left(\begin{array}{c}N\\ N-N^{\′}\end{array}\right)$ Plant combinations of values (p ₁..., p _{n-N '}) be mapped to one by one $[1,\left(\begin{array}{c}N\\ N-N^{\′}\end{array}\right)]$ Integer on, $f\∈\left(1,...,\left(\begin{array}{c}N\\ N-N^{\′}\end{array}\right)\right),$ i，j∈{1，...，K}，i≠j；

(1.4) according to above-mentioned multiple access access slot alignment matrix U _f, determine wherein, [] _{t, t}represent the element of the capable t row of matrix t, t ∈ 1 ..., E};

(1.5) according to above-mentioned determine obtain f ^*independent variable (p corresponding in index function τ ₁ ^*..., p _{n-N '} ^*), relaying r selects except p ₁ ^*..., p _{n-N '} ^*antenna beyond root is as reception antenna, obtain the multiple access access slot active matrix of each user, wherein, multiple access access slot active matrix represents that multiple access access slot user transmitting antenna draws the channel coefficient matrix of reception antenna to relay selection, and arg min represents variate-value when making target function get minimum value;

(2) via node is according to each user's multiple access access slot efficient channel matrix, adopts signal space alignment schemes, determines that multiple access access slot efficient precoding vector and multiple access access slot effectively align vector;

(3) each user carries out binary phase modulation to respective information to be sent, obtains modulation signal, and adopts the above-mentioned respective multiple access access slot efficient precoding vector determined to send after modulation signal weighting;

(4) signal of each user's transmission of relay reception, obtains relaying by squeeze theorem and waits to signal;

(5) trunk node selection transmitting antenna

(5.1) for each user's time slot channel matrix, all q is left out ₁... q _{n-N '}column vector obtains respective time slot channel submatrix, wherein, and q ₁..., q _{n-N '}∈ 1 ..., N} and q ₁≠ ... ≠ q _{n-N '}, $g\∈\left(1,...,\left(\begin{array}{c}N\\ N-N^{\′}\end{array}\right)\right);$

(5.2) according to each user's time slot channel matrix, adopt signal space alignment schemes, determine respective decoding vector and time slot alignment vector;

(5.3) according to above-mentioned time slot alignment vector, time slot precoding vector is determined;

(5.4) according to the decoding vector of above-mentioned user i and user j time slot alignment vector and time slot precoding vector determine ${\mathrm{\λ}}_{\min }^g=\underset{i,j}{\min }\left\{{\mathrm{\λ}}_g^{[i,j]}\right\},$ Wherein, ${\mathrm{\λ}}_g^{[i,j]}=\frac{{\left|\right|{\left(z_g^{\mathrm{\π}(i,j)}\right)}^H{v}_g^{[\mathrm{\π}(i,j),r]}\left|\right|}^2}{{\left|\right|w_g^{[i,j]}\left|\right|}^2},$ I, j ∈ 1 ..., K}, i ≠ j, () ^hrepresent complex conjugate transpose operation, || || ²represent two norms of vector;

(5.5) according to above-mentioned determine obtain g ^*independent variable (q corresponding in index function τ ₁ ^*..., q _{n-N '} ^*), via node r selects except q ₁ ^*..., q _{n-N '} ^*antenna beyond root, as transmitting antenna, obtains the time slot active matrix of each user, and wherein, the time slot active matrix of each user represents that selected by time slot relaying, transmitting antenna is to the channel coefficient matrix of all reception antennas of each user;

(6) via node is according to each user's time slot efficient channel matrix, determines each user's efficient coding vector, time slot effectively aligns vector and time slot efficient precoding vector;

(7) relaying adopts above-mentioned determining to send after time slot efficient precoding vector is by signal weighting pending for relaying at time slot;

(8) user node recovers raw information

(8.1) each user interference of utilizing respective efficient coding vector to suppress between other users in Received signal strength couple, obtains desired signal;

(8.2) self interference is suppressed according to above-mentioned each user's desired signal and respective transmission modulation signal, and adopt squeeze theorem again demodulation recover the raw information that other users send to oneself, wherein, the demodulation mode that demodulation employing is corresponding with the modulation in step (3).

The present invention compared with prior art tool has the following advantages:

1) antenna selecting method of the present invention owing to have employed at multiple access access slot and time slot, overcome the qualification of antenna number, reduce the requirement of multiple-input and multiple-output Y-channel to configuration number of antennas, thus expand multiple-input and multiple-output Y-channel range of application in practice.

2) the present invention is owing to adopting multi-antenna technology, take full advantage of the condition that number of antennas increases, have selected at multiple access access slot and can obtain one group of antenna of maximum signal to noise ratio as reception antenna at this time slot, have selected at time slot equally and can obtain another group antenna of maximum signal to noise ratio as transmitting antenna at this time slot, thus obtain potential diversity gain, reduce the error rate of system.

Accompanying drawing explanation

Fig. 1 is the multiple-input and multiple-output Y-channel model that the present invention is suitable for;

Fig. 2 is flow chart of the present invention;

Fig. 3 be the present invention multiple access access slot have noise and time slot noiseless time BER Simulation figure;

Fig. 4 be the present invention time slot have noise and multiple access access slot noiseless time BER Simulation figure;

Fig. 5 is the BER Simulation figure of the present invention when multiple access access slot and time slot have noise.

Embodiment

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

Method for transmitting signals proposed by the invention is applicable to the multiple-input and multiple-output Y-channel model described in Fig. 1, this channel model comprises a K user node and a via node, wherein each user node equipment M root antenna, via node equipment N root antenna, each user node respectively sends a unicast info to an other K-1 user node by via node; Whole communication process is divided into two time slots, first time slot is multiple access access slot, second time slot is time slot, concrete communication process is as follows: at multiple access access slot, K user node sends information to via node simultaneously, and via node carries out process to the received signal and obtains time slot and wait to signal; At time slot, relay node broadcasts information is to K user node, and each user node goes out according to the signal oneself received and the signal recuperation oneself sent the information that an other K-1 user node issues oneself respectively.

With reference to Fig. 2, it is as follows that the present invention utilizes the channel model in Fig. 1 to carry out the step of Signal transmissions:

Step 1, trunk node selection reception antenna.

According to the multiple access access slot channel matrix H of user i to relaying ^{[r, i]}, via node selects N ' root as reception antenna from the N root antenna of equipment, and wherein, user i is to the multiple access access slot channel matrix H of relaying ^{[r, i]}each element obey the separate multiple Gaussian random variable that average is zero, variance is 1, 2M-N '=1, M>=K-1, K are user node number, and M is the number of antennas of user, i ∈ 1 ..., K}, concrete system of selection is as follows:

(1.1) removed users i is to the multiple access access slot channel matrix of relaying ${\left(H^{[r,i]}\right)}^T=\left[\begin{array}{ccccc}{h}_1^{[r,i]}& ...& h_n^{[r,i]}& ...& h_N^{[r,i]}\end{array}\right]$ The individual column vector of middle N-N ' carry out transpose operation again and obtain the multiple access access slot channel submatrix of user i to relaying wherein, () ^tfor transpose of a matrix operator, n-component column vector, i ∈ 1 ..., K}, f=τ (p ₁..., p _{n-N '}), τ (p ₁..., p _{n-N '}) be index function, it is by D kind combinations of values (p ₁..., p _{n-N '}) be mapped on the integer of [1, D] one by one, p ₁..., p _{n-N '}∈ 1 ..., N} and p ₁≠ ... ≠ p _{n-N '}, f ∈ 1 ..., D}, $D=\left(\begin{array}{c}N\\ N-N^{\′}\end{array}\right).$

(1.2) according to above-mentioned user i to relaying and user j to the multiple access access slot channel submatrix of relaying with adopt signal space alignment schemes, determine the multiple access access slot precoding vector of user j and user i and multiple access access slot alignment vector wherein signal space alignment is determined by following formula:

${\tilde{H}}_f^{[r,i]}{v}_f^{[j,i]}={\tilde{H}}_f^{[r,j]}{v}_f^{[i,j]}=u_f^{\mathrm{\π}(i,j)}$

Wherein, i, j ∈ 1 ... K} and i ≠ j, subscript π (i, j) are index functions, meets π (i, j)=π (j, i), and has $\mathrm{\π}(i,j)=\underset{x=K-\min \{i,j\}+1}{\overset{K-1}{\mathrm{\Σ}}}x+|i-j|.$

(1.3) according to above-mentioned multiple access access slot alignment vector form multiple access access slot alignment matrix U _f, $U_f=\left[\begin{array}{ccccc}{u}_f^1& ...& u_f^e& ...& u_f^E\end{array}\right],$ Wherein, i, j ∈ 1 ... K}, i ≠ j, e=π (i, j), $E=\frac{K(K-1)}{2}.$

(1.4) according to above-mentioned multiple access access slot alignment matrix U _f, determine wherein, t ∈ 1 ..., E}, [] _{t, t}represent the element of the capable t row of matrix t.

(1.5) according to above-mentioned determine obtain f ^*independent variable (p corresponding in index function τ ₁ ^*..., p _{n-N '} ^*), via node r selects except p ₁ ^*..., p _{n-N '} ^*antenna beyond root is as reception antenna, and obtaining user i to the multiple access access slot active matrix of relaying is wherein, represent the channel coefficient matrix of reception antenna selected by multiple access access slot user i all transmitting antennas to relaying, arg min represents variate-value when making target function get minimum value.

Step 2, via node determination multiple access access slot efficient precoding vector.

Via node according to user i to relaying and user j to the multiple access access slot efficient channel matrix of relaying with adopt signal space alignment schemes, determine the multiple access access slot efficient precoding vector v of user j and user i ^{[i, j]}and v ^{[j, i]}and multiple access access slot effectively aligns vectorial u ^{π (i, j)}, and by v ^{[i, j]}feed back to user node j, vector v ^{[j, i]}this precoding is sent to user node i.Wherein, signal space alignment is determined by following formula:

${\stackrel{\‾}{H}}^{[r,i]}{v}^{[j,i]}={\stackrel{\‾}{H}}^{[r,j]}{v}^{[i,j]}=u^{\mathrm{\π}(i,j)}$

I, j ∈ 1 ..., K} and i ≠ j.

Step 3, user node sends signal to via node.

Each user modulates respective transmission information, obtains the multiple access access slot modulation signal s that user i sends to user j ^{[j, i]}, modulator approach adopts binary phase modulation or quaternary phase-modulation or quadrature amplitude modulation, adopts binary phase modulation in the embodiment of the present invention.And adopt multiple access access slot efficient precoding vector v ^{[j, i]}by s ^{[j, i]}weighting, the transmission signal obtaining user i is:

$x_i=\underset{j=1,j\≠i}{\overset{K}{\mathrm{\Σ}}}{v}^{[j,i]}{s}^{[j,i]}$

At multiple access access slot, respective transmission signal is sent to via node by each user.Wherein, s ^{[j, i]}represent the modulation signal that user i sends to user j, v ^{[j, i]}represent the multiple access access slot efficient precoding vector of user i, i, j ∈ 1 ..., K} and i ≠ j.

Step 4, via node carries out decoding to received signal.

(4.1) reception antenna that via node r step (1) is selected carrys out Received signal strength, obtains Received signal strength and is:

$\begin{array}{c}{y}_r=\underset{i=1}{\overset{K}{\mathrm{\Σ}}}{\stackrel{\‾}{H}}^{[r,i]}{x}_i+n_r\\ =\underset{i=1}{\overset{K}{\mathrm{\Σ}}}{\stackrel{\‾}{H}}^{[r,i]}\underset{j=1,j\≠i}{\overset{K}{\mathrm{\Σ}}}{v}^{[j,i]}{s}^{[j,i]}+n_r\\ =\underset{\mathrm{\π}(i,j)=1}{\overset{T}{\mathrm{\Σ}}}{u}^{\mathrm{\π}(i,j)}{s}^{\left[\mathrm{\π}(j,i)\right]}+n_r\\ ={\mathrm{Us}}_{\mathrm{tra}}+n_r\end{array}$

Wherein, y _rfor received signal vector, n _rfor the noise vector that relay reception arrives, each element of this vector is average is zero, variance is separate multiple Gaussian random variable, s ^{[π (i, j)]}=s ^{[i, j]}+ s ^{[j, i]}represent user i modulation signal s ^{[j, i]}with user j modulation signal s ^{[i, j]}superposed signal, u ^{π (i, j)}for multiple access access slot effectively aligns vector, U=[u ¹u ^{π (i, j)}u ^e] be the effective alignment matrix of multiple access access slot, s _tra=[s ¹s ^{π (i, j)}s ^e] ^tfor aligned signal column vector,

(4.2) via node r y to received signal _rcarry out squeeze theorem, be separated aligned signal column vector s _tra, obtaining via node signal vector to be sent is:

s _rec＝U ⁺y _r＝s _tra+U ⁺n _r

In formula, $s_{\mathrm{rec}}={\left[\begin{array}{ccccc}{s}_{\mathrm{rec}}^1& ...& s_{\mathrm{rec}}^{\mathrm{\π}(i,j)}& ...& s_{\mathrm{rec}}^E\end{array}\right]}^T$ For relaying signal vector to be sent, for signal to be sent, () ⁺for group inverse matrices operator, i, j ∈ 1 ..., K} and i ≠ j,

Step 5, trunk node selection transmitting antenna.

According to the time slot channel matrix H being relayed to user i ^{[i, r]}, via node selects N ' root as transmitting antenna from the N root antenna of equipment, wherein, is relayed to the time slot channel matrix H of user i ^{[i, r]}each element obey the separate multiple Gaussian random variable that average is zero, variance is 1, i ∈ 1 ..., K}, concrete system of selection is as follows:

(5.1) the time slot channel matrix being relayed to user i is left out $H^{[i,r]}=\left[\begin{array}{ccccc}{h}_1^{[i,r]}& ...& h_n^{[i,r]}& ...& h_N^{[i,r]}\end{array}\right]$ The individual column vector of middle N-N ' obtain the time slot channel submatrix being relayed to user i wherein, for H ^{[i, r]}the n-th column vector, g=τ (q ₁... q _{n-N '}), q ₁... q _{n-N '}∈ 1 ..., N} and q ₁≠ ... ≠ q _{n-N "}, $g\∈\left(1,...,\left(\begin{array}{c}N\\ N-N^{\′}\end{array}\right)\right).$

(5.2) according to the above-mentioned time slot channel submatrix being relayed to user i and user j with adopt signal space alignment schemes, determine the decoding vector of user j and user i and time slot alignment vector wherein signal space alignment is determined by following formula:

${\left(w_g^{[j,i]}\right)}^H{\tilde{H}}_g^{[i,r]}={\left(w_g^{[i,j]}\right)}^H{\tilde{H}}_g^{[j,r]}=z_g^{\mathrm{\π}(i,j)}$

Wherein i, j ∈ 1 ..., K} and i ≠ j.

(5.3) according to above-mentioned time slot alignment vector composition time slot alignment matrix $Z_g^{\left[r\right]}=\left[\begin{array}{ccccc}{z}_g^{\mathrm{\π}\left(\mathrm{1,1}\right)}& ...& z_g^{\mathrm{\π}(i,j)}& ...& z_g^{\mathrm{\π}(K-1,K)}\end{array}\right],$ Corresponding time slot precoding vector is determined by following formula $v_g^{[\mathrm{\π}(i,j),r]}:$

$v_g^{[\mathrm{\π}(i,j),r]}=\mathrm{null}\left({\left({\tilde{Z}}_g^{\mathrm{\π}(i,j)}\right)}^H\right),$

Wherein, for matrix remove column vector the submatrix obtained, the kernel operator that null () is matrix, g ∈ (1 ..., D), i, j ∈ 1 ..., K} and i ≠ j.

(5.4) according to the decoding vector of above-mentioned user j and user i time slot alignment vector and time slot precoding vector determine wherein, i, j ∈ 1 ..., K}, i ≠ j, () ^hrepresent complex conjugate transpose operation, || || ²represent two norms of vector.

(5.5) according to above-mentioned determine obtain g ^*independent variable (q corresponding in index function τ ₁ ^*..., q _{n-N '} ^*), via node r selects except q ₁ ^*..., q _{n-N '} ^*antenna beyond root is as transmitting antenna, and the time slot active matrix obtaining being relayed to user i is wherein, represent that selected by time slot relaying, transmitting antenna is to the channel coefficient matrix of all reception antennas of user i.

Step 6, via node determination time slot efficient precoding vector.

Via node is according to the time slot efficient channel matrix being relayed to user i and user j with to be alignd formula by signal space:

${\left(w^{[j,i]}\right)}^H{\stackrel{\‾}{H}}^{[i,r]}={\left(w^{[i,j]}\right)}^H{\stackrel{\‾}{H}}^{[j,r]}=z^{\mathrm{\π}(i,j)}$

Determine the efficient coding vector w of user j and user i ^{[i, j]}and w ^{[j, i]}, time slot effectively aligns vectorial z ^{π (i, j)}, and by following formula:

$v^{[\mathrm{\π}(i,j),r]}=\mathrm{null}\left({\left({\tilde{Z}}^{\mathrm{\π}(i,j)}\right)}^H\right)$

Determine time slot efficient precoding vector v ^{[π (i, j), r]}, wherein, for the effective alignment matrix Z of time slot ^[r]remove column vector z ^{π (i, j)}the submatrix obtained, the effective alignment matrix of time slot is Z ^[r]=[z ^{π (1,1)}z ^{π (i, j)}z ^{π (K-1, K)}], i, j ∈ 1 ..., K} and i ≠ j.

Step 7, via node is to user node broadcast singal.

Relaying adopts v at time slot ^{[π (i, j), r]}relaying is waited to signal send after weighting, obtain repeat broadcast signal:

$x_r=\underset{e=1}{\overset{E}{\mathrm{\Σ}}}{v}^{[e,r]}{s}_{\mathrm{rec}}^e$

Wherein, e=π (i, j), i, j ∈ 1 ..., K} and i ≠ j,

Step 8, user node recovers useful data.

(8.1) Received signal strength of user node i is:

$y_i={\stackrel{\‾}{H}}^{[i,r]}{x}_r+n_i,$

Wherein, i ∈ 1 ..., K}, n _ibe the noise vector of via node to user node i, each element of this vector is average is zero, variance is separate multiple Gaussian random variable.

(8.2) according to above-mentioned user i Received signal strength y _i, utilize (w ^{[j, i]}) ^hsuppress the interference signal between other users couple, isolate the desired signal of user i:

$\begin{array}{c}{\hat{y}}^{[i,j]}={\left(w^{[j,i]}\right)}^H{y}_i\\ ={\left(w^{[j,i]}\right)}^H{\stackrel{\‾}{H}}^{[i,r]}\underset{t=1}{\overset{T}{\mathrm{\Σ}}}{v}^{[t,r]}{s}_r^{\left[t\right]}+{\left(w^{[j,i]}\right)}^H{n}_i\\ ={\left(z^{\mathrm{\π}(i,j)}\right)}^H{v}^{[\mathrm{\π}(i,j),r]}{s}_r^{\left[\mathrm{\π}(i,j)\right]}+{\left(w^{[j,i]}\right)}^H{n}_i\end{array}$

Wherein, t=π (i, j), j ∈ 1 ..., K} and j ≠ i.

(8.3) according to the desired signal of above-mentioned user i and the transmission modulation signal s of oneself ^{[j, i]}suppress the interference of self, specifically determine by following formula:

${\hat{x}}^{[i,j]}={\hat{y}}^{[i,j]}-{\left(z^{\mathrm{\π}(i,j)}\right)}^H{v}^{[\mathrm{\π}(i,j),r]}{s}^{[j,i]}$

Wherein, for self interference suppression signal of user i, z ^{π (i, j)}for time slot effectively aligns vector, v ^{[π (i, j), r]}for time slot efficient precoding vector, j ∈ 1 ..., K}, j ≠ i.

(8.4) according to above-mentioned self interference suppression signal of user i carry out squeeze theorem and demodulation, wherein, squeeze theorem, specifically determine by following formula:

${\hat{s}}^{[i,j]}={\left({\left(z^{\mathrm{\π}(i,j)}\right)}^H{v}^{[\mathrm{\π}(i,j),r]}\right)}^{-1}{\hat{x}}^{[i,j]}$

for the modulation signal of user i user j to be restored, z ^{π (i, j)}for time slot effectively aligns vector, v ^{[π (i, j), r]}for time slot efficient precoding vector, j ∈ 1 ..., K}.The demodulation mode that demodulation employing is corresponding with the modulation in step (3).

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

1. simulated conditions:

The user node number K=3 of setting multiple-input and multiple-output Y-channel, the antenna number M=2 of each user node configuration, the antenna number N of via node configuration is 3,4,5,6 these four kinds of situations, if the noise variance at each user node and via node place is σ ², namely i ∈ 1 ..., K}.

2. emulate content and simulation result

A () hypothesis multiple access access slot has noise and under time slot noise free conditions, to the inventive method at N=4,5, when 6 and existing need meet number of antennas require the error performance of method for transmitting signals when N=3 compare, simulation result is as Fig. 3, obtain abscissa in Fig. 3 and represent transmission signal to noise ratio, ordinate represents the error rate of multiple-input and multiple-output Y-channel.As seen from Figure 3, when the number of antennas of relaying node equipment increases thus does not meet number of antennas requirement, sky line options proposed by the invention still can apply the method for transmitting signals of signal space alignment at multiple access access slot, and be significantly improved than the error performance of existing method for transmitting signals, and increasing along with via node number of antennas, improve more obvious.

(b) hypothesis multiple access access slot noiseless and under time slot has noise conditions, to the inventive method at N=4,5, when 6 and existing need meet number of antennas require the error performance of method for transmitting signals when N=3 compare, simulation result is as Fig. 4, obtain abscissa in Fig. 4 and represent transmission signal to noise ratio, ordinate represents the error rate of multiple-input and multiple-output Y-channel.As seen from Figure 4, when the number of antennas of relaying node equipment increases thus does not meet number of antennas requirement, sky line options proposed by the invention still can apply the method for transmitting signals of signal space alignment at time slot, and be significantly improved than the error performance of existing method for transmitting signals, and increasing along with via node number of antennas, improve more obvious.

C () hypothesis multiple access access slot and time slot have noise conditions under, to the inventive method at N=4,5, when 6 and existing need meet number of antennas require the error performance of method for transmitting signals when N=3 compare, simulation result is as Fig. 5, obtain abscissa in Fig. 5 and represent received signal to noise ratio, ordinate represents the error rate of multiple-input and multiple-output Y-channel.As seen from Figure 5, when the number of antennas of relaying node equipment increases thus does not meet number of antennas requirement, sky line options proposed by the invention can apply the method for transmitting signals of signal space alignment, and be significantly improved than the error performance of existing method for transmitting signals, and increasing along with via node number of antennas, improve more obvious.

Claims (10)

1. in multiple-input and multiple-output Y-channel based on a method for transmitting signals for sky line options, comprise the steps:

(1) trunk node selection reception antenna

(1.1) for each user's multiple access access slot channel matrix, all p is left out ₁..., p _{n-N '}row vector obtains respective multiple access access slot channel submatrix, and wherein, N is relay antenna number, the antenna number of N ' selected by relaying, p ₁..., p _{n-N '}∈ 1 ..., N} and p ₁≠ ... ≠ p _{n-N '}, 2M-N '=1, M>=K-1, K are user node number, and M is the antenna number of each user;

(1.2) according to each user's multiple access access slot channel matrix, adopt signal space alignment schemes, determine respective multiple access access slot precoding vector and multiple access access slot alignment vector;

(1.3) according to multiple access access slot alignment vector form multiple access access slot alignment matrix U _f,

$U_f=\left[\begin{array}{ccccc}{u}_f^1& ...& u_f^{\mathrm{\π}(i,j)}& ...& u_f^E\end{array}\right],$

Wherein, subscript π (i, j) is an index function, meets π (i, j)=π (j, i), and has f=τ (p ₁..., p _{n-N '}) be another index function, will $\left(\begin{array}{c}N\\ N-N^{\′}\end{array}\right)$ Plant combinations of values (p ₁..., p _{n-N '}) be mapped to one by one $[1,\left(\begin{array}{c}N\\ N-N^{\′}\end{array}\right)]$ Integer on, $f\∈(1,...,\left(\begin{array}{c}N\\ N-N^{\′}\end{array}\right)),$ i,j∈{1,...,K}，i≠j；

(1.4) according to above-mentioned multiple access access slot alignment matrix U _f, determine wherein, [] _t,trepresent the element of the capable t row of matrix t, t ∈ 1 ..., E};

(1.5) according to above-mentioned determine obtain f ^*independent variable (p corresponding in index function τ ₁ ^*..., p _{n-N '} ^*), relaying r selects except p ₁ ^*..., p _{n-N '} ^*antenna beyond root is as reception antenna, obtain the multiple access access slot active matrix of each user, wherein, multiple access access slot active matrix represents that multiple access access slot user transmitting antenna draws the channel coefficient matrix of reception antenna to relay selection, and arg min represents variate-value when making target function get minimum value;

(2) via node is according to each user's multiple access access slot efficient channel matrix, adopts signal space alignment schemes, determines that multiple access access slot efficient precoding vector and multiple access access slot effectively align vector;

(3) each user carries out binary phase modulation to respective information to be sent, obtains modulation signal, and adopts the above-mentioned respective multiple access access slot efficient precoding vector determined to send after modulation signal weighting;

(4) signal of each user's transmission of relay reception, obtains relaying by squeeze theorem and waits to signal;

(5) trunk node selection transmitting antenna

(5.1) for each user's time slot channel matrix, all q is left out ₁... q _{n-N '}column vector obtains respective time slot channel submatrix, wherein, and q ₁..., q _{n-N '}∈ 1 ..., N} and q ₁≠ ... ≠ q _{n-N '} $g\∈(1,...,\left(\begin{array}{c}N\\ N-N^{\′}\end{array}\right));$

(5.2) according to each user's time slot channel matrix, adopt signal space alignment schemes, determine respective decoding vector and time slot alignment vector;

(5.3) according to above-mentioned time slot alignment vector, time slot precoding vector is determined;

(5.4) according to the decoding vector of above-mentioned user i and user j time slot alignment vector and time slot precoding vector determine ${\mathrm{\λ}}_{\min }^g=\underset{i,j}{\min }\left\{{\mathrm{\λ}}_g^{[i,j]}\right\},$ Wherein, ${\mathrm{\λ}}_g^{[i,j]}=\frac{{\left|\right|{\left(z_g^{\mathrm{\π}(i,j)}\right)}^H{v}_g^{[\mathrm{\π}(i,j),r]}\left|\right|}^2}{{\left|\right|w_g^{[i,j]}\left|\right|}^2},$ I, j ∈ 1 ..., K}, i ≠ j, () ^hrepresent complex conjugate transpose operation, ‖ ‖ ²represent two norms of vector;

(5.5) according to above-mentioned determine obtain g ^*independent variable (q corresponding in index function τ ₁ ^*..., q _{n-N '} ^*), via node r selects except q ₁ ^*..., q _{n-N '} ^*antenna beyond root, as transmitting antenna, obtains the time slot active matrix of each user, and wherein, the time slot active matrix of each user represents that selected by time slot relaying, transmitting antenna is to the channel coefficient matrix of all reception antennas of each user;

(6) via node is according to each user's time slot efficient channel matrix, determines each user's efficient coding vector, time slot effectively aligns vector and time slot efficient precoding vector;

(7) relaying adopts above-mentioned determining to send after time slot efficient precoding vector is by signal weighting pending for relaying at time slot;

(8) user node recovers raw information

(8.1) each user interference of utilizing respective efficient coding vector to suppress between other users in Received signal strength couple, obtains desired signal;

(8.2) self interference is suppressed according to above-mentioned each user's desired signal and respective transmission modulation signal, and adopt squeeze theorem again demodulation recover the raw information that other users send to oneself, wherein, the demodulation mode that demodulation employing is corresponding with the modulation in step (3).

2. in multiple-input and multiple-output Y-channel according to claim 1 based on the method for transmitting signals of sky line options, employing signal space alignment schemes wherein described in step (1.2), determine respective multiple access access slot precoding vector and multiple access access slot alignment vector, determine by following signal space alignment formula:

${\tilde{H}}_f^{[r,i]}{v}_f^{[j,i]}={\tilde{H}}_f^{[r,j]}{v}_f^{[i,j]}=u_f^{\mathrm{\π}(i,j)},$

Wherein, for user i is to the multiple access access slot channel submatrix of relaying, for the multiple access access slot precoding vector of user j and user i, for multiple access access slot alignment vector, i, j ∈ 1 ..., K} and i ≠ j.

3. in multiple-input and multiple-output Y-channel according to claim 1 based on the method for transmitting signals of sky line options, employing signal space alignment schemes wherein described in step (2), determine that multiple access access slot efficient precoding vector and multiple access access slot effectively align vector, determine by following signal space alignment formula:

${\stackrel{\‾}{H}}^{[r,i]}{v}^{[j,i]}={\stackrel{\‾}{H}}^{[r,j]}{v}^{[i,j]}=u^{\mathrm{\π}(i,j)}$

Wherein, for the multiple access access slot efficient channel matrix of user i and user j, v ^{[i, j]}for the multiple access access slot efficient precoding vector of user j, v ^{[j, i]}for the multiple access access slot efficient precoding vector of user i, i, j ∈ 1 ..., K} and i ≠ j.

4. in multiple-input and multiple-output Y-channel according to claim 1 based on the method for transmitting signals of sky line options, the squeeze theorem in wherein said step (4) determines by following formula:

s _rec＝U ⁺y _r

Wherein, U is the effective alignment matrix of multiple access access slot, y _rfor via node received signal vector, s _recfor relaying signal vector pending, $s_{\mathrm{rec}}={\left[\begin{array}{ccccc}{s}_{\mathrm{rec}}^1& ...& s_{\mathrm{rec}}^{\mathrm{\π}(i,j)}& ...& s_{\mathrm{rec}}^E\end{array}\right]}^T,$ for relaying is waited to signal, () ⁺for group inverse matrices operator, i, j ∈ 1 ..., K}, i ≠ j,

5. in multiple-input and multiple-output Y-channel according to claim 1 based on the method for transmitting signals of sky line options, employing signal space alignment schemes in wherein said step (5.2), determining respective decoding vector and time slot alignment vector, is determine by following signal space alignment formula:

${\left(w_g^{[j,i}\right)}^H{\tilde{H}}_g^{[i,r]}={\left(w_g^{[i,j]}\right)}^H{\tilde{H}}_g^{[j,r]}=z_g^{\mathrm{\π}(i,j)},$

Wherein, for being relayed to the time slot channel submatrix of user i, for the decoding vector of user j and user i, for time slot alignment vector, i, j ∈ 1 ..., K} and i ≠ j.

6. in multiple-input and multiple-output Y-channel according to claim 1 based on the method for transmitting signals of sky line options, aliging according to time slot wherein described in step (5.3) is vectorial, determining each time slot precoding vector, is determine by following formula:

$v_g^{[\mathrm{\π}(i,j),r]}=\mathrm{null}\left({\left({\tilde{Z}}_g^{\mathrm{\π}(i,j)}\right)}^H\right),$

Wherein, for waiting to signal time slot precoding vector, for time slot alignment matrix remove column vector after the submatrix that obtains,

Wherein, $Z_g^{\left[r\right]}=\left[\begin{array}{ccccc}{z}_g^{\mathrm{\π}\left(\mathrm{1,1}\right)}& ...& z_g^{\mathrm{\π}(i,j)}& ...& z_g^{\mathrm{\π}(K-1,K)}\end{array}\right],$ for time slot alignment vector, the kernel operator that null () is matrix, i, j ∈ 1 ..., K} and i ≠ j.

7. in multiple-input and multiple-output Y-channel according to claim 1 based on the method for transmitting signals of sky line options, wherein described in step (6) according to each user's time slot efficient channel matrix, determine each user's efficient coding vector, time slot effectively align vector and time slot efficient precoding vector, is determine by following formula:

$w^{[j,i]}{\stackrel{\‾}{H}}^{[r,i]}=w^{[i,j]}{\stackrel{\‾}{H}}^{[r,j]}=z^{\mathrm{\π}(i,j)},$

$v^{[\mathrm{\π}(i,j),r]}=\mathrm{null}\left({\left({\tilde{Z}}^{\mathrm{\π}(i,j)}\right)}^H\right),$

Wherein, for the time slot efficient channel matrix of user i and user j, w ^{[i, j]}for efficient coding vector during user j recovery user i transmission information, w ^{[j, i]}for efficient coding vector during user i recovery user j transmission information, z ^{π (i, j)}for time slot effectively aligns vector, for the effective alignment matrix Z of time slot ^[r]=[z ^{π (1,1)}z ^{π (i, j)}z ^{π (K-1, K)}] remove z ^{π (i, j)}after the submatrix that obtains, v ^{[π (i, j), r]}for relaying is waited to signal time slot efficient precoding vector, i, j ∈ 1 ..., K} and i ≠ j.

8. in multiple-input and multiple-output Y-channel according to claim 1 based on the method for transmitting signals of sky line options, the interference that each user wherein described in step (8.1) utilizes respective efficient coding vector to suppress between other users in Received signal strength couple, obtaining desired signal, is determine by following formula:

${\hat{y}}^{[i,j]}={\left(w^{[j,i]}\right)}^H{y}_i,$

Wherein, i, j ∈ 1 ..., K}, j ≠ i, w ^{[j, i]}for the efficient coding vector of user i, y _ifor the signal that user i receives, for the desired signal of user i.

9. in multiple-input and multiple-output Y-channel according to claim 1 based on the method for transmitting signals of sky line options, suppressing self to disturb according to each user's desired signal and respective transmission modulation signal wherein described in step (8.2), is determine by following formula:

${\hat{x}}^{[i,j]}={\hat{y}}^{[i,j]}-{\left(z^{\mathrm{\π}(i,j)}\right)}^H{v}^{[\mathrm{\π}(i,j),r]}{s}^{[j,i]},$

Wherein, for self interference suppression signal of user i, z ^{π (i, j)}for time slot effectively aligns vector, v ^{[π (i, j), r]}for time slot efficient precoding vector, s ^{[j, i]}for the modulation signal that oneself sends, i, j ∈ 1 ..., K}, j ≠ i.

10. in multiple-input and multiple-output Y-channel according to claim 1 based on the method for transmitting signals of sky line options, the squeeze theorem in wherein said step (8.2) determines by following formula:

${\hat{s}}^{[i,j]}={\left({\left(z^{\mathrm{\π}(i,j)}\right)}^H{v}^{[\mathrm{\π}(i,j),r]}\right)}^{-1}{\hat{x}}^{[i,j]},$

Wherein, for the modulation signal of user i user j to be restored, z ^{π (i, j)}for time slot effectively aligns vector, v ^{[π (i, j), r]}for time slot efficient precoding vector, i, j ∈ 1 ..., K}, j ≠ i.