CN106788642A - A kind of mixing Precoding Design method for the extensive mimo system in actual broadband - Google Patents

Abstract

The present invention proposes a kind of mixing Precoding Design method for the extensive mimo system in actual broadband, comprises the following steps：In radio-frequency head, first assume that the simulation precoding on all carrier waves is identical, to maximize system spectral efficiency as criterion, preferably simulation pre-coding matrix is calculated using complete channel condition information；Then the phase shifter characteristic in real system, it is determined that the simulation pre-coding matrix on each carrier wave in practice, and in numeric field designed phase correction matrix, correct the phase offset that the simulation precoding on different carrier waves in practice occurs, to approach preferably simulation precoding, phase only pupil filter matrix is multiplied Ji Wei designed simulation pre-coding matrix with the simulation pre-coding matrix that phase offset occurs on each carrier wave；Finally, it is multiplied with simulation precoding is previously obtained using the channel condition information design digital precode matrix of equivalent low dimensional at base band, obtains final product mixing Precoding Design scheme.

Description

Hybrid precoding design method for actual broadband large-scale MIMO system

the technical field is as follows:

the invention belongs to the technical field of wireless communication, and particularly relates to a hybrid precoding design method for an actual broadband large-scale MIMO system.

Background art:

hybrid precoding is a research hotspot of massive MIMO, and when the number of antennas is large (hundreds to thousands), it is impossible to equip each antenna with a specific radio frequency link (RF) due to power consumption and cost, so that it becomes very meaningful to research deployment of massive MIMO with a small amount of RF. The hybrid precoding uses a low-cost phase shifter to control the phase of a signal on a transmitting antenna at a radio frequency end, so that the analog precoding is realized, the hardware cost is reduced, and the number of radio frequencies required by a system is reduced; digital precoding is achieved at baseband with equivalent low-dimensional Channel State Information (CSI) control signal amplitude and phase. Hybrid precoding can therefore enable massive MIMO with a much smaller RF number than the number of antennas. Currently, research on hybrid precoding is mainly focused on a single carrier system, research work on hybrid precoding of a wideband system is relatively small, and the current design of wideband hybrid precoding is researched under the condition that analog precoding on each subcarrier is assumed to be the same, which is only suitable for the case that the ratio of bandwidth to center carrier frequency is relatively small. However, considering future millimeter wave (30-300 GHz) applications, the above assumption is often unrealistic in practice. This is because the phase shifters in wideband beamforming networks are usually implemented using delay lines, which results in the same delay producing different phases on different carriers, i.e. although we have only set one analog precoding matrix, it produces phase shifts on different carriers. In practice the analog precoding on different carriers is therefore different, which results in a non-negligible performance loss.

In the research of the current broadband system hybrid precoding design, the non-ideal characteristic of hardware implementation in the actual system is rarely considered; in the past, where the bandwidth was narrow, the phase of the phase shifter did not change much with frequency, and its effect was negligible. The current millimeter wave technology is ultra-wideband, for example, in the 60GHz band, the bandwidth is typically 2G. On the other hand, in such a high frequency band, in order to ensure the performance of the system, the requirement on the processing precision of hardware is very fine, which results in a very high price of millimeter wave devices, and thus an intermediate frequency link is required. The common intermediate frequency is 2.75GHz, which results in a bandwidth-to-carrier ratio of approximately 0.5-1.5, and therefore the effect of phase offset on analog precoding becomes non-negligible.

In summary, it is necessary to research a hybrid precoding design method suitable for a practical wideband massive MIMO system.

The invention content is as follows:

the invention aims to provide a hybrid precoding design method for an actual broadband large-scale MIMO system, aiming at the problems, and the method can improve the hybrid precoding performance of the actual broadband system.

In order to achieve the purpose, the invention adopts the following technical scheme to realize the purpose:

a hybrid precoding design method for an actual broadband massive MIMO system comprises the following steps:

1) at a radio frequency end, firstly, assuming that analog precoding on all carriers is the same, and designing an ideal analog precoding matrix by using complete channel state information by taking the maximum system spectrum efficiency as a criterion;

2) determining a simulated precoding matrix with phase shift on each carrier according to the characteristics of a phase shifter in practice, designing a phase correction matrix in a digital domain, correcting the phase shift of the simulated precoding on different carriers in practice to approximate to the ideal simulated precoding matrix obtained in the step 1), and multiplying the phase correction matrix and the simulated precoding matrix with the phase shift on each carrier to obtain the designed simulated precoding matrix;

3) designing a digital precoding matrix by using equivalent low-dimensional channel state information at a baseband;

4) multiplying the analog precoding obtained in the step 2) and the digital precoding obtained in the step 3) to obtain a mixed precoding design scheme.

The further improvement of the invention is that the specific implementation method of the step 1) is as follows:

considering a downlink broadband large-scale MIMO system, a base station passes N_RFRoot radio frequency, N_tRoot antenna, sending N_sA data stream, user configuration N_rA root antenna, the radio frequency number of which is the same as the number of antennas, and K subcarriers are shared; f_BB[k]Representing the digital baseband precoding matrix on the k-th carrier, F_RFIs an analog precoding matrix representing all carriers, W k at the receiving end]Represents a receive combining matrix, where N_t＞＞N_RF；

101) Received signal on the kth carrier:

y[k]＝W^H[k]H[k]F_RFF_BB[k]s[k]+W^H[k]n[k](1)

wherein, H [ k]Indicating the channel on the k-th carrier,representing the noise on the kth carrier;

102) according to the received signals in the previous step, the base station end designs analog precoding and digital precoding with the goal of maximizing mutual information as follows:

wherein,is a feasible set of analog precoding, i.e. a set of N with all elements of the same amplitude_t×N_RFA matrix set;

103) solving the optimization problem to obtain an ideal simulation precoding matrix:

where R is the channel correlation matrix.

The further improvement of the invention is that the specific implementation method of the step 2) is as follows:

201) determining an analog precoding matrix with phase shift on each carrier according to the characteristics of the actual phase shifter: taking the analog precoding matrix obtained in step 1) as an analog precoding matrix on the center carrier frequency, the analog precoding matrix on the k-th carrier in practice can be represented as:

wherein is represented by F_RFIs an analog precoding matrix at the center carrier frequency,representing the phase deviation matrix on the k-th carrier,representing the Hadamard product of matrix A and matrix B;

202) the phase offset is corrected in the digital domain: the modification can be described as:

whereinIt can be found in step 1) that when the frequency of the kth carrier is given, F is calculated from equation (6)_RF[k]Therefore, solving for P2 yields a well-known least mean square solution:

whereinRepresents the pseudo-inverse of matrix a;

203) multiplying the phase correction matrix by the analog precoding matrix with phase offset on each carrier to obtain the designed analog precoding matrix, i.e. F_RF[k]F_c[k]。

The further improvement of the invention is that the specific implementation method of the step 3) is as follows:

after the analog precoding and phase correction matrices on all carriers are determined in step 2), the equivalent low-dimensional channel state information at the baseband is represented as:

at baseband, SVD decomposition of equivalent low dimensional channel state information can yield digital precodingIn which V is_eff[k]Is an equivalent channel H_eff[k]Right singular matrix of the SVD decomposition of (a).

The further improvement of the invention is that the specific implementation method of the step 4) is as follows:

multiplying the analog precoding obtained in the step 2) and the digital precoding obtained in the step 3) to obtain a mixed precoding design scheme:

F＝F_RF[k]F_c[k]F_BB[k](10)。

compared with the prior art, the invention has the following advantages:

the invention considers the performance of the phase shifter in the broadband beam forming network which is sensitive to the frequency change for the first time and designs the mixed pre-coding scheme which is more in line with the actual system. Compared with the traditional mixed precoding scheme, the invention provides an effective scheme for setting a phase correction matrix in a digital domain to correct the phase deviation aiming at the actual situation that the simulated precoding on different carriers is different due to the phase deviation generated by a phase shifter on the basis of designing the ideal simulated precoding, thereby improving the integral frequency spectrum efficiency of the system and greatly improving the frequency efficiency of the edge carrier.

Description of the drawings:

fig. 1 is a graph of spectral efficiency versus frequency spectrum efficiency on each carrier for different schemes;

fig. 2 is a graph of average spectral efficiency over all carriers for different schemes.

The specific implementation mode is as follows:

the invention is described in further detail below with reference to the accompanying drawings:

the invention provides a hybrid precoding design method for an actual broadband large-scale MIMO system, which mainly adopts the following steps: firstly, at a radio frequency end, assuming that analog precoding on all carriers is the same, and designing an ideal analog precoding matrix by using complete channel state information; secondly, determining a simulation pre-coding matrix with phase shift on each carrier according to the characteristics of the actual phase shifter, designing a phase correction matrix in a digital domain, correcting the phase shift of simulation pre-coding on different carriers in the actual process to approximate the ideal simulation pre-coding matrix obtained in the first step, and multiplying the phase correction matrix and the simulation pre-coding matrix with the phase shift on each carrier to obtain the designed simulation pre-coding matrix; thirdly, designing a digital precoding matrix at a baseband by using equivalent low-dimensional channel state information; and fourthly, multiplying the analog precoding obtained in the second step and the digital precoding obtained in the third step to obtain a mixed precoding design scheme.

The specific embodiment is as follows:

considering a downlink broadband large-scale MIMO system, a base station passes N_RFRoot radio frequency, N_t(N_t＞＞N_RF) Root antenna, sending N_sA data stream, user configuration N_rThe number of radio frequencies of the antennas is the same as the number of antennas, and K subcarriers are shared by the antennas. F_BB[k]Representing the digital baseband precoding matrix on the k-th carrier, F_RFIs an analog precoding matrix representing all carriers, W k at the receiving end]Representing a receive combining matrix;

on the basis of the above, the received signal on the k-th carrier can be expressed as

y[k]＝W^H[k]H[k]F_RFF_BB[k]s[k]+W^H[k]n[k](1)

Wherein, H [ k]Indicating the channel on the k-th carrier,representing the noise on the k-th carrier.

The spectral efficiency on each carrier and the total spectral efficiency of the system are shown in equation (2) and equation (3), respectively:

the technical means of the hybrid precoding design method applicable to the actual broadband large-scale MIMO system provided by the invention is as follows:

firstly, assuming that the analog precoding is the same on all carriers, the design of the analog precoding and the digital precoding by the base station with the goal of maximizing mutual information can be described as follows:

wherein,is a feasible set of analog precoding, i.e. a set of N with all elements of the same amplitude_t×N_RFAnd (5) matrix collection.

Solving the above optimization problem can obtain ideal analog precoding:

wherein R is a channel correlation matrix;

then, the obtained ideal simulated precoding matrix is used as a simulated precoding matrix on the center carrier frequency, and according to the characteristics of the actual phase shifter, the simulated precoding matrix on the k-th carrier wave in the actual situation can be expressed as:

wherein is represented by F_RFIs an analog precoding matrix at the center carrier frequency,representing the phase deviation matrix on the k-th carrier,representing the Hadamard product of matrix a and matrix B.

In order to improve the performance of the system, the phase offset needs to be corrected in the digital domain, and the correction scheme can be described as

WhereinFor the ideal analog precoding matrix, when the frequency of the kth carrier is given, F can be calculated by equation (6)_RF[k]. Therefore, by solving for P2, we can obtain a well-known least mean square solution:

whereinExpressing the pseudo inverse of the matrix A, multiplying the phase correction matrix with the analog precoding matrix with phase shift on each carrier to obtain the designed analog precoding matrix, namely F_RF[k]F_c[k]；

On the basis, the equivalent channel state information is utilized to design a digital precoding matrix. Through the above steps, analog precoding and phase correction matrices on all carriers are determined, and equivalent low-dimensional channel state information at the baseband can be expressed as:

at baseband, SVD decomposition of equivalent low dimensional channel state information can yield digital precodingIn which V is_eff[k]Is an equivalent channel H_eff[k]Right singular matrix of the SVD decomposition of (a);

and finally, multiplying the analog precoding and the digital precoding obtained in the steps to obtain a mixed precoding design scheme:

F＝F_RF[k]F_c[k]F_BB[k](10)

the simulation effect of the invention is as follows:

number of base station antennas N_t64, radio frequency number N_RFNumber of data streams N ═ 8_sNumber of user antennas N4_rAnd 4, the subcarrier number K is 4096, and the ratio of the broadband to the central carrier frequency is 0.5-1.5. Compared with three schemes of full-digital precoding, ideal hybrid precoding and actual hybrid precoding without phase correction, the comparison results are shown in fig. 1 and fig. 2.

Fig. 1 shows the spectral efficiency on each carrier for the different schemes when the SNR is 10 dB. It can be seen that the spectral efficiency of the edge carrier of the uncorrected phase scheme is 2.2dB lower than the ideal scheme, whereas the phase correction scheme proposed in the present invention is 1.2dB higher than the uncorrected phase scheme and is somewhat close to the ideal hybrid precoding scheme.

Fig. 2 shows the variation of the average spectral efficiency over all carriers with the signal-to-noise ratio for different schemes. It can be seen from the figure that the overall performance of the system with the uncorrected phase scheme is lower than that under the ideal condition by 0.5dB, while the phase correction algorithm proposed in the present invention can eliminate the performance loss caused by the phase offset to a certain extent, and the overall performance of the system is higher than that of the system with the uncorrected phase scheme by about 0.3 dB.

Claims (5)

1. A hybrid precoding design method for an actual wideband massive MIMO system is characterized by comprising the following steps:

1) at a radio frequency end, firstly, assuming that analog precoding on all carriers is the same, and designing an ideal analog precoding matrix by using complete channel state information by taking the maximum system spectrum efficiency as a criterion;

2) determining a simulated precoding matrix with phase shift on each carrier according to the characteristics of a phase shifter in practice, designing a phase correction matrix in a digital domain, correcting the phase shift of the simulated precoding on different carriers in practice to approximate to the ideal simulated precoding matrix obtained in the step 1), and multiplying the phase correction matrix and the simulated precoding matrix with the phase shift on each carrier to obtain the designed simulated precoding matrix;

3) designing a digital precoding matrix by using equivalent low-dimensional channel state information at a baseband;

4) multiplying the analog precoding obtained in the step 2) and the digital precoding obtained in the step 3) to obtain a mixed precoding design scheme.

2. The method for designing hybrid precoding for a large-scale MIMO system with practical wideband as claimed in claim 1, wherein the specific implementation method of step 1) is as follows:

considering a downlink broadband large-scale MIMO system, a base station passes N_RFRoot radio frequency, N_tRoot antenna, sending N_sA data stream, user configuration N_rA root antenna, the radio frequency number of which is the same as the number of antennas, and K subcarriers are shared; f_BB[k]Representing the digital baseband precoding matrix on the k-th carrier, F_RFIs an analog precoding matrix representing all carriers, W k at the receiving end]Indicating the reception of the combining matrix, wherein,

101) received signal on the kth carrier:

y[k]＝W^H[k]H[k]F_RFF_BB[k]s[k]+W^H[k]n[k](1)

wherein, H [ k]Indicating the channel on the k-th carrier,representing the noise on the kth carrier;

102) according to the received signals in the previous step, the base station end designs analog precoding and digital precoding with the goal of maximizing mutual information as follows:

$\begin{array}{cc}P1:& \underset{F_{RF},F_{BB}\[k\]}{max}\underset{k=1}{\overset{K}{\Σ}}{\log }_2\det (I+\frac{1}{{\mathrm{\σ}}^2}H\[k\]F_{RF}{F}_{BB}\[k\]F_{BB}{\[k\]}^H{F}_{RF}^{H}H{\[k\]}^H)\end{array}$

$\begin{array}{cc}s.t.& \underset{k=1}{\overset{K}{\Σ}}\left|\right|F_{RF}{F}_{BB}\[k\]|{|}_F^2\≤P_{tot}\end{array}---\left(4\right)$

wherein,is a feasible set of analog precoding, i.e. a set of N with all elements of the same amplitude_t×N_RFA matrix set;

103) solving the optimization problem to obtain an ideal simulation precoding matrix:

$F_{RF}^{opt}={\[V_R\]}_{1:N_{RF}}---\left(5\right)$

where R is the channel correlation matrix.

3. The method of claim 2, wherein the specific implementation method of step 2) is as follows:

201) determining an analog precoding matrix with phase shift on each carrier according to the characteristics of the actual phase shifter: taking the analog precoding matrix obtained in step 1) as an analog precoding matrix on the center carrier frequency, the analog precoding matrix on the k-th carrier in practice can be represented as:

wherein is represented by F_RFIs an analog precoding matrix at the center carrier frequency,representing the phase deviation matrix on the k-th carrier,representing the Hadamard product of matrix A and matrix B;

202) the phase offset is corrected in the digital domain: the modification can be described as:

$\begin{array}{cc}P2:& \underset{F_c\[k\]}{min}\left|\right|F_{RF}^{opt}-F_{RF}\[k\]F_c\[k\]|{|}_F^2\end{array}---\left(7\right)$

whereinIt can be found in step 1) that when the frequency of the kth carrier is given, F is calculated from equation (6)_RF[k]Therefore, solving for P2 yields a well-known least mean square solution:

whereinRepresents the pseudo-inverse of matrix a;

203) multiplying the phase correction matrix by the analog precoding matrix with phase offset on each carrier to obtain the designed analog precoding matrix, i.e. F_RF[k]F_c[k]。

4. The method of claim 3, wherein the specific implementation method of step 3) is as follows:

after the analog precoding and phase correction matrices on all carriers are determined in step 2), the equivalent low-dimensional channel state information at the baseband is represented as:

$H_{eff}\[k\]=H\[k\]F_{RF}\[k\]F_c\[k\]{({\left(F_{RF}\[k\]F_c\[k\]\right)}^H{F}_{RF}\[k\]F_c\[k\])}^{-\frac{1}{2}}---\left(9\right)$

at baseband, SVD decomposition of equivalent low dimensional channel state information can yield digital precodingIn which V is_eff[k]Is an equivalent channel H_eff[k]Right singular matrix of the SVD decomposition of (a).

5. The method of claim 4, wherein the specific implementation method of step 4) is as follows:

multiplying the analog precoding obtained in the step 2) and the digital precoding obtained in the step 3) to obtain a mixed precoding design scheme:

F＝F_RF[k]F_c[k]F_BB[k](10)。