CN105827293A - A linear zero-forcing precoding method for multi-user generalized spatial modulation systems - Google Patents

Abstract

The invention discloses a multi-user generalized space modulation system linearity zero-forcing precoding method. The method comprises the following steps: establishing a multi-user generalized space modulation system, and grouping antennas at a link sending end, wherein each group of sending antennas serve one receiving end by use of generalized space modulation; performing dimension reduction on a channel gain matrix of a sending signal matrix in a system so as to obtain an equivalent channel gain matrix and an equivalent sending signal matrix; and by use of a dimension reduction matrix, performing zero-forcing precoding processing on the sending signal matrix at the sending end, and transmitting signals by use of the equivalent channel gain matrix so as to eliminate interference between users and interference between channels at the same time. The method provided by the invention can effectively improve error code performance of the system and the capacity of the system.

Description

Multi-user generalized spatial modulation system linear zero-forcing precoding method

Technical Field

The invention relates to the field of wireless communication, in particular to a linear zero-forcing precoding method of a multi-user generalized spatial modulation system.

Background

The spatial modulation technology effectively solves the problem of inter-channel interference by activating an antenna at a link sending end to transmit information, mapping part of the sent information to a traditional digital modulation constellation diagram, and mapping the rest to a spatial dimension generated by antenna serial numbers. The generalized spatial modulation technology can enable a link sending end to activate two or more antennas, so that the limitation that the number of the antennas at the link sending end in spatial modulation must be power of 2 is eliminated, diversity gain is increased, the spectrum efficiency is effectively improved, and channel interference is inevitably introduced. The zero-forcing precoding technology is widely applied to a multi-user MIMO system, effectively eliminates the interference in a lower link and improves the system capacity.

In the prior art, J.Wu et al researches precoding of a multi-user MIMO system in combination with spatial modulation, the system is equivalent to a plurality of parallel single-user MIMO systems through block diagonal precoding operation, internal interference of a channel is eliminated through spatial modulation, and finally a gradient descent algorithm is adopted to update a power distribution factor of each user until the error rate of the system is reduced to the minimum, so that a closed-form solution is obtained.

Wu et al, after block diagonalization precoding processing on a multi-user MIMO system, eliminate inter-channel interference through spatial modulation, but this method is only applicable to a multi-user spatial modulation system, i.e., a special case when only one antenna at a transmitting end is activated in a single-user MIMO system, and for a multi-user generalized spatial modulation system, because there are two or more active antennas, inter-channel interference still exists. If the zero-forcing precoding technology is directly applied to the multi-user generalized spatial modulation system, the multi-user generalized spatial modulation system cannot directly use the zero-forcing precoding method to eliminate the inter-channel interference because the channel gain matrix is not full-rank and the channel gain matrix does not have generalized inverse.

Disclosure of Invention

In order to solve the problem that the conventional zero-forcing precoding method cannot be applied to a multi-user generalized spatial modulation system at present, the invention provides a linear zero-forcing precoding method applied to a downlink of the multi-user generalized spatial modulation system.

The invention provides a linear zero-forcing precoding method of a multi-user generalized spatial modulation system, which comprises the following steps:

establishing a multi-user generalized spatial modulation system, grouping the antennas at the transmitting end of a link, wherein each group of transmitting antennas adopts generalized spatial modulation to serve one receiving end;

reducing dimensions of a channel gain matrix and a sending signal matrix in the multi-user generalized spatial modulation system to obtain an equivalent channel gain matrix and an equivalent sending signal matrix;

and performing zero-forcing pre-coding processing on an equivalent sending signal matrix of the link sending end of the multi-user generalized spatial modulation system by using the dimension reduction matrix, and transmitting signals by using the equivalent channel gain matrix so as to eliminate interference between users and interference between channels at the same time.

The invention provides a linear zero-forcing pre-coding method applied to a lower link of a multi-user generalized spatial modulation system, which decomposes the multi-user generalized spatial modulation system into a plurality of independent single-user generalized spatial modulation systems by grouping antennas at a link transmitting end, generates a dimension reduction matrix according to the characteristic that silent antennas exist in generalized spatial modulation under the condition of assuming that the number of receiving antennas of a user is the same as the number of activated antennas of a corresponding transmitting antenna group so as to reduce the dimensions of a channel gain matrix and a transmitting signal matrix, extracts the transmitting information and channel state information of active antennas by the obtained equivalent channel gain matrix and the equivalent transmitting signal matrix, and finally eliminates interference by using a channel inverse zero-forcing pre-coding method. The invention processes the sending signal matrix by applying the pre-coding matrix at the sending end, because the interference between users and the interference between channels are eliminated, the receiving end can more accurately detect the antenna serial number and the sending symbol by utilizing a maximum likelihood detection method or a maximum combining ratio detection method, so the error code performance and the system capacity of the system can be further improved, wherein, the maximum combining ratio detection method judges the sending antenna serial number and determines the corresponding sending bit by detecting the position of the energy maximum symbol; the maximum likelihood detection method can detect the antenna serial number and the transmitted symbol at the same time, it finds out the bit information closest to the received signal by the exhaustion method, and it can be seen from fig. 4 that the maximum combination ratio detection method can obtain better error code performance.

Drawings

Fig. 1 is a flowchart of a linear zero-forcing precoding method applied to a downlink of a multi-user generalized spatial modulation system according to the present invention.

Fig. 2 is a block diagram of a multi-user generalized spatial modulation system used in the present invention.

Fig. 3 is a graph of a simulation of rate performance using the method of the present invention.

Fig. 4 is a simulation diagram of error performance using the method of the present invention.

Detailed Description

In order to make the objects, technical solutions and advantages of the present invention more apparent, the present invention is described in further detail below with reference to specific embodiments and the accompanying drawings.

The invention provides a linear zero-forcing precoding method applied to a downlink of a multi-user generalized spatial modulation system, which is applied to the multi-user generalized spatial modulation system shown in figure 2. As shown in fig. 1, the linear zero-forcing precoding method applied to the downlink of the multi-user generalized spatial modulation system includes the following steps:

step S1: establishing a multi-user generalized spatial modulation system, grouping transmitting end antennas in a downlink of the multi-user generalized spatial modulation system, wherein each group of transmitting antennas adopts generalized spatial modulation to serve a specific user, namely a receiving end;

for a multi-user generalized spatial modulation system, it is assumed that the channel state information is known at the transmitting end, and it is considered that the base station at the transmitting end has N_tHaving a root transmit antenna to serve K users simultaneously, as shown in FIG. 2, assume that user i ∈ {1, …, K } has M_iThe total number of receiving antennas isIn a multi-user generalized spatial modulation system, N_r≤N_t. Will N_tThe transmitting antennas are divided into K groups, each group of antennas adopts generalized spatial modulation to serve one special antennaA fixed user, which is equivalent to a multi-single-user generalized spatial modulation system, assuming that each transmit antenna group has N_tAnd the ith group of transmitting antennas serves the user i, the number of receiving antennas of the user side is the same as that of the activated antennas in the corresponding transmitting end antenna group, and the symbol information transmitted by the activated antennas in one group of transmitting antennas on each time slot is the same.

For the multi-user generalized spatial modulation system, the received signal at the receiving end is represented as: y is f.HPx + n, f is a power limiting factor, H represents a Rayleigh flat fading channel gain matrix,is the precoding matrix of the transmitting end, x represents the transmitted signal, and n represents the additive white gaussian noise.

The received signal may be expanded to:

$y=\left[\begin{array}{c}{y}^{\left(1\right)}\\ y^{\left(2\right)}\\ .\\ .\\ .\\ y^{\left(K\right)}\end{array}\right]=f\·HP\left[\begin{array}{c}{x}^{\left(1\right)}\\ x^{\left(2\right)}\\ .\\ .\\ .\\ x^{\left(K\right)}\end{array}\right]+\left[\begin{array}{c}{n}^{\left(1\right)}\\ n^{\left(2\right)}\\ .\\ .\\ .\\ n^{\left(K\right)}\end{array}\right],$

the rayleigh flat fading channel H can be represented in the form of a block matrix:

sub-matrixRepresenting a channel gain matrix between the nth group of transmitting antennas and the user i;a received signal representing user i; x is the number of⁽ⁱ⁾Is the symbol vector transmitted by the ith group of transmit antennas,wherein N is_u⁽ⁱ⁾Representing the number of active antennas in the ith set of transmit antennas,representing signals from the jth active antenna in the ith group of transmit antennas, s_qRepresenting the q-th symbol in the M-QAM constellation;representing an additive white gaussian noise vector for user i. Order toRepresents a radicalA transmit symbol vector of a station, satisfiesP_TIs the total transmitted power of the base station, E [. cndot]The mean value is calculated, and the Eucliden norm is calculated by | · | |; the total power limit of the transmitting end can be expressed as: t is_r(PP^H)≤P_TAnd Tr (-) represents the trace-seeking operation.

Step S2: according to the characteristic of generalized spatial modulation, reducing the dimensions of a channel gain matrix and a transmission signal matrix in the multi-user generalized spatial modulation system to obtain an equivalent channel gain matrix and an equivalent transmission signal matrix;

in the multi-user generalized spatial modulation system, because only the active antennas transmit information and the inactive antennas remain silent, the columns corresponding to the silent antennas in the channel gain matrix H and the rows corresponding to the silent antennas in the transmitted signal x are zero vectors. Aiming at the characteristic, a dimension reduction matrix for reducing the channel gain matrix H and the x dimension of a transmission signal matrix are respectively establishedAndsuppose the number M of active antennas in the ith group of transmit antennas_iNumber of receiving antennas N with the i-th user_u⁽ⁱ⁾Same, i.e. M_i＝N_u⁽ⁱ⁾First, when the transmission signal is not precoded, the received signal at the receiving end is considered to be:

$y=Hx+n=\left[\begin{array}{ccccccccccc}0& \mathrm{...}& 0& h^{(l_1,1)}& 0& \mathrm{...}& 0& h^{(l_{{N_u}^{\left(K\right)}},K)}& 0& \mathrm{...}& 0\end{array}\right]\left[\begin{array}{c}0\\ .\\ .\\ .\\ 0\\ s_q^{(l_1,1)}\\ 0\\ .\\ .\\ .\\ 0\\ s_q^{(l_{{N_u}^{\left(K\right)}},K)}\\ 0\\ .\\ .\\ .\\ 0\end{array}\right]+\left[\begin{array}{c}{n}^{\left(1\right)}\\ n^{\left(2\right)}\\ .\\ .\\ .\\ n^{\left(k\right)}\end{array}\right]$

wherein,representing a channel state information matrix column corresponding to a jth active antenna in an ith group of transmit antennas, where a silent antenna does not perform information transmission, and when only channel information and transmit information of an active antenna are considered, a received signal may be further written as:

$\left[\begin{array}{c}{y}_1\\ .\\ .\\ .\\ y_{N_r}\end{array}\right]=\left[\begin{array}{ccc}{h}^{(l_1,1)}& \mathrm{...}& h^{(l_{{N_u}^{\left(K\right)}},K)}\end{array}\right]\left[\begin{array}{c}{s}_q^{(l_1,1)}\\ .\\ .\\ .\\ s_q^{(l_{{N_u}^{\left(K\right)}},K)}\end{array}\right]+\left[\begin{array}{c}{n}^{\left(1\right)}\\ n^{\left(2\right)}\\ .\\ .\\ .\\ n^{\left(K\right)}\end{array}\right],$

the equivalent channel gain matrix and the equivalent transmission signal matrix are respectively:

$\tilde{h}=\left[\begin{array}{ccc}{h}^{(l_1,1)}& \mathrm{...}& h^{(l_{{N_u}^{\left(K\right)}},K)}\end{array}\right]=\left[\begin{array}{ccccccccccc}0& \mathrm{...}& 0& h^{(l_1,1)}& 0& \mathrm{...}& 0& h^{(l_{{N_u}^{\left(K\right)}},K)}& 0& \mathrm{...}& 0\end{array}\right]P_1$

$\tilde{x}=\left[\begin{array}{c}{s}_q^{(l_1,1)}\\ .\\ .\\ .\\ s_q^{(l_{{N_u}^{\left(K\right)}},K)}\end{array}\right]=P_3{\left[\begin{array}{ccccccccccc}0& \mathrm{...}& 0& s_q^{(l_1,1)}& 0& \mathrm{...}& 0& s_q^{(l_{{N_u}^{\left(K\right)}},K)}& 0& \mathrm{...}& 0\end{array}\right]}^H$

from this, P is obtained₁In the form of_jI) line, thThe elements in the column are 1, the remaining elements are 0, and are notedThe other positions are 0, wherein l (l)_jI) represents the serial number of the j active antenna in the i group of transmitting antennas, N_u^(k)Representing the number of active antennas in the kth group of transmitting antennas; simultaneously, the following can be obtained:the other positions are 0, and take precoding matrix P ═ P₁P₂P₃Whereinwhen the signal is used as a zero-forcing pre-coding matrix, the purpose of reducing the dimension of a channel gain matrix H and a sending signal matrix x can be met.

Step S3: and performing zero-forcing precoding processing on an equivalent sending signal matrix of the link sending end of the multi-user generalized spatial modulation system, and transmitting signals by using the equivalent channel gain matrix, so that the inter-user interference and the inter-channel interference can be eliminated simultaneously.

In this step, the precoding matrix is taken as P ═ P₁P₂P₃，The received signal of the user may be expressed as:wherein, it is madeAn equivalent channel gain matrix for the transmission channel is represented. The equivalent channel gain matrix can be used for eliminating interference among multiple users and among channelsAnd adopting a zero-forcing precoding method of channel inversion, wherein:therefore, when the zero forcing linear precoding method of the multi-user generalized spatial modulation system is adopted, the precoding matrix of the sending end of the base station is represented as follows:

$P=P_1{P}_2{P}_3=P_1{\tilde{H}}^H{\left(\tilde{H}{\tilde{H}}^H\right)}^{-1}{P}_3,$

the total transmission power of the link transmitting end is P_TTherefore, the total power limit at the transmitting end of the link is T_r(PP^H)＝P_TThe power limiting factor can be found to be:

then the received signal of the u-th receiving antenna is:

$y_u=f\·{\mathrm{HP}}_1{P}_2{P}_{3}x+n=f\·HPx+n={\stackrel{\‾}{x}}_u+n_u.$

the effects of the present invention are further described below in conjunction with fig. 2-4. Referring to the simulation diagrams of the rate and error performance of the linear zero-forcing precoding method using the multi-user generalized spatial modulation system of the present invention in fig. 3-4, and the structure diagram of the multi-user generalized spatial modulation system shown in fig. 2. The simulation environment of FIGS. 3-4 is: the channel in the multi-user generalized spatial modulation system is a flat Rayleigh fading channel, the noise is zero-mean additive white Gaussian noise, the number of users K is 6, each group of antennas adopts a 4-QAM generalized spatial modulation method, and N is_tWhere M is 7 and the number of active antennas is N_u⁽ⁱ⁾2. In fig. 3-4, MU-GSM and MU-GSM-ZF respectively represent a multi-user generalized spatial modulation system that does not employ the multi-user generalized spatial modulation system linear zero-forcing precoding method and a multi-user generalized spatial modulation system that employs the multi-user generalized spatial modulation system linear zero-forcing precoding method; MU-GSM-MRC and MU-GSM-ML respectively represent a multi-user generalized spatial modulation system adopting a maximum combination ratio detection method and a multi-user generalized spatial modulation system adopting a maximum likelihood detection method; the MU-GSM-ZF-MRC represents a multi-user generalized spatial modulation system which adopts a maximum combination ratio detection method and applies a multi-user generalized spatial modulation system linear zero-forcing pre-coding method; MU-GSM-ZF-ML represents a multi-user generalized spatial modulation system that employs a maximum likelihood detection method and applies a multi-user generalized spatial modulation system linear zero-forcing precoding method, and it is apparent from fig. 3-4 that the rate and error code performance of the system can be significantly improved by the multi-user generalized spatial modulation linear zero-forcing precoding proposed by the present invention in the same signal-to-noise ratio interval, and when the maximum combining ratio detection method is used, better error code performance can be obtained than when the maximum likelihood detection method is employed.

The above-mentioned embodiments are intended to illustrate the objects, technical solutions and advantages of the present invention in further detail, and it should be understood that the above-mentioned embodiments are only exemplary embodiments of the present invention, and are not intended to limit the present invention, and any modifications, equivalents, improvements and the like made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims (10)

1. A multi-user generalized spatial modulation system linear zero-forcing precoding method is characterized by comprising the following steps:

establishing a multi-user generalized spatial modulation system, grouping the antennas at the transmitting end of a link, wherein each group of transmitting antennas adopts generalized spatial modulation to serve one receiving end;

reducing dimensions of a channel gain matrix and a sending signal matrix in the multi-user generalized spatial modulation system to obtain an equivalent channel gain matrix, an equivalent sending signal matrix and a corresponding dimension reduction matrix;

and performing zero-forcing pre-coding processing on a signal matrix sent by the link sending end of the multi-user generalized spatial modulation system by using the dimensionality reduction matrix, and transmitting signals by using the equivalent channel gain matrix so as to eliminate interference between users and interference between channels at the same time.

2. The method of claim 1, wherein in the multi-user generalized spatial modulation system, the number of receiving antennas at a receiving end is the same as the number of active antennas in the corresponding transmitting-end antenna group.

3. The method of claim 1, wherein the symbol information transmitted by the active antennas of each group of transmit antennas in each time slot is the same in the multi-user generalized spatial modulation system.

4. The method according to claim 1, wherein in the step of reducing the dimensions of the channel gain matrix and the transmit signal matrix in the multi-user generalized spatial modulation system:

according to the multi-user generalized spatial modulation system, the column of the channel gain matrix corresponding to the silent antenna and the row of the transmission signal corresponding to the silent antenna are zero vectors, and a dimension reduction matrix P for reducing the dimension of the channel gain matrix and the transmission signal matrix is constructed₁And P₃。

5. The method of claim 4, wherein the dimension reduction matrix P₁In the first (l)_jI) line I)The elements in the column are 1 and the other elements are 0, wherein l (l)_jI) represents the serial number of the j active antenna in the i group of transmitting antennas, N_u^(k)Representing the number of active antennas in the kth group of transmitting antennas, i is 1, …, K is the number of receiving ends, j is 1, …,N_u⁽ⁱ⁾。

6. The method of claim 4, wherein the dimension reduction matrix P₃In the first placeLine l (l)_jI) the elements in column 1 and the remaining elements 0, where l (l)_jI) represents the serial number of the j active antenna in the i group of transmitting antennas, N_u⁽ⁱ⁾Representing the number of active antennas in the ith group of transmitting antennas, i is 1, …, K is the number of receiving ends, j is 1, …, N_u⁽ⁱ⁾。

7. The method according to claim 4, wherein in the step of performing zero-forcing precoding processing on the transmission signal matrix at the transmitting end of the multi-user generalized spatial modulation system link, zero-forcing precoding processing is performed on an equivalent transmission signal matrix by using a precoding matrix P, wherein the precoding matrix P is used for performing zero-forcing precoding processing on the equivalent transmission signal matrixRepresents the equivalent channel gain matrix and H represents the rayleigh flat fading channel gain matrix.

8. The method according to claim 7, wherein after performing zero-forcing precoding processing on an equivalent transmit signal matrix at a transmit end of the multi-user generalized spatial modulation system link and performing transmission of signals by using the equivalent channel gain matrix, a received signal of a u-th receive antenna is represented as:

y_u＝f·HPx+n，

where f is a power limiting factor, x represents a transmit signal matrix, and n represents additive white gaussian noise.

9. The method of claim 8, wherein the power limiting factor f is expressed as:wherein, P_TTr (-) represents the tracking operation for the total transmit power of the transmitting end.

10. The method of claim 1, wherein the receiving end performs signal detection by using a maximum combining ratio detection method or a maximum likelihood detection method.