CN107809275A - A kind of Limited Feedback mixing method for precoding based on millimeter wave mimo system - Google Patents
A kind of Limited Feedback mixing method for precoding based on millimeter wave mimo system Download PDFInfo
- Publication number
- CN107809275A CN107809275A CN201711233176.1A CN201711233176A CN107809275A CN 107809275 A CN107809275 A CN 107809275A CN 201711233176 A CN201711233176 A CN 201711233176A CN 107809275 A CN107809275 A CN 107809275A
- Authority
- CN
- China
- Prior art keywords
- matrix
- precoding
- codebook
- base station
- user terminal
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04B—TRANSMISSION
- H04B7/00—Radio transmission systems, i.e. using radiation field
- H04B7/02—Diversity systems; Multi-antenna system, i.e. transmission or reception using multiple antennas
- H04B7/04—Diversity systems; Multi-antenna system, i.e. transmission or reception using multiple antennas using two or more spaced independent antennas
- H04B7/0413—MIMO systems
- H04B7/0456—Selection of precoding matrices or codebooks, e.g. using matrices antenna weighting
Abstract
The invention discloses a kind of Limited Feedback based on millimeter wave mimo system to mix method for precoding, comprises the following steps：S1：Constructing analog predefines codebook set and its code word searching method；S2：The simulation of millimeter wave mimo system is predefined into codebook set and the predefined codebook set of numeral is stored in base station and user terminal simultaneously；S3：User terminal obtains simulation pre-coding matrix FRF；S4：Further according to simulation pre-coding matrix FRFAnd channel condition information obtains base band pre-coding matrix FBB；S5:Will simulation pre-coding matrix FRFWith digital pre-coding matrix FBBBeam index feeds back to base station；S6：Base station obtains pre-coding matrix F=FRFFBB, and send data to user using pre-coding matrix F.The present invention can obtain preferable performance using the building method of the predefined code book of simulation.
Description
Technical Field
The invention belongs to the technical field of mobile communication and multi-antenna, relates to a Frequency Division Duplex (FDD) system in a millimeter wave frequency band, and provides a limited feedback mixed pre-coding method of a millimeter wave MIMO (Multiple-Input Multiple-Output) system with a large-scale antenna array applied to a single-user scene.
Background
Currently, with the rapid development of wireless communication technology and mobile internet, people are continuously making higher demands on mobile communication rate. However, system resources such as available spectrum and transmit power in wireless communication systems are limited and cannot meet the increasing rate requirements. Millimeter-wave communications have a large and underdeveloped millimeter-wave bandwidth compared to microwave communications applied to most wireless applications and cellular systems. Although millimeter wave signals experience more free space path loss, due to the short wavelength of the millimeter wave bandwidth, a large number of antenna elements can be packaged in a certain size, that is, the transmitting end and the receiving end of the millimeter wave system have large-scale antenna arrays, so the millimeter wave system can utilize the beam forming gain generated by the large-scale antenna arrays to compensate the free space loss caused by the large-scale antenna arrays. The propagation characteristic of the millimeter-wave channel is its high free-space path loss, which results in limited spatial selectivity or limited scatterers. I.e., the mmwave system has the characteristic of spatially sparse scatterer, the mmwave communication has become one of the most promising candidates for the future cellular wireless system. For the millimeter wave MIMO system, it is difficult to implement in practical scenarios using the same all-digital transceiving architecture as the traditional low-frequency MIMO system. First, the all-digital transceiver architecture design requires one rf chain per antenna element, and the high cost and high power consumption of millimeter wave signal mixing components makes an all-digital precoding matrix impractical in terms of current semiconductor technology. Second, millimeter-wave transceivers are characterized by large, closely-stacked antenna arrays, which can result in a high level of antenna correlation.
In addition, in general, the limited feedback is to feed back the channel state information to the base station, and then the base station performs precoding design according to the channel state information, but in the millimeter wave MIMO system or the low frequency MIMO system, the number of dimensions of the channel matrix H becomes large due to the increase of the number of antennas, and the computational complexity of the limited feedback is very high. Meanwhile, the method cannot be realized in an actual scene.
Therefore, it is necessary to provide a solution to the above-mentioned drawbacks in the prior art.
Disclosure of Invention
In view of this, it is necessary to provide a limited feedback hybrid precoding method based on the millimeter wave MIMO system, which adopts a two-stage hybrid precoding method, combines analog/digital precoding by adopting a new construction method of an analog predetermined codebook and a structure of a dual codebook, and significantly reduces codeword search complexity, thereby effectively reducing computational complexity and obtaining better performance, and compared with other millimeter wave limited feedback methods in the prior art, the method improves 1-2dB.
In order to solve the technical problems in the prior art, the technical scheme of the invention is as follows:
a limited feedback mixed precoding method based on a millimeter wave MIMO system comprises the following steps:
step S1: an analog predefined codebook set and a code word searching method thereof are constructed, and a digital predefined codebook adopts a conventional random vector quantization codebook set. Because the analog precoder of the millimeter wave MIMO system is usually implemented by using an analog phase shifter, the modulus of each element of the analog precoding matrix is constant. The predefined codebook F is designed taking this constraint into account. The simulated predefined codebook design includes a design method and a design criterion of the codebook. In the design of simulating the predefined codebook, the concept of a rotation matrix is added, and based on the coding principle, good equivalent mapping can be carried out, so that a better codebook is constructed. Since the relevant hardware of the baseband precoding matrix has no constraint, the digital predefined codebook W employs a conventional Random Vector Quantization (RVQ) codebook set.
Step S2: and simultaneously storing the analog predefined codebook set and the digital predefined codebook set of the millimeter wave MIMO system in the base station and the user terminal.
And step S3: in the first stage of feedback, the user terminal is based on the estimationThe channel state information is subjected to code word searching operation, namely, an analog pre-coding matrix F is obtained firstly RF . The user terminal estimates the channel state information according to the pilot training sequence sent by the base station, processes the CSI to obtain the optimal digital pre-coding matrix F of the data stream opt And then, according to the space sparse optimization problem of the millimeter waves, a non-convex optimization problem of hybrid precoding can be obtained similarly:
s.t.F RF ∈F，and is obtained by least square methodThe optimization problem described above can be written from:
s.t.F RF ∈F，wherein, N s Is the number of data streams transmitted. The user terminal searches the precoding codebook F according to the optimization problem, thereby obtaining the simulation precoder F RF 。
And step S4: in the second phase, the user terminal then performs precoding based on the analog precoding matrix F RF And carrying out code word searching operation on the channel state information so as to obtain a baseband precoding matrix F BB . The baseband precoding codebook is a random vector quantization codebook W, and a digital precoding matrix F is obtained by vector quantization of the random vector quantization codebook W BB 。
Step S5: the user terminal will simulate a precoding matrix F RF And a digital precoding matrix F BB The beam index is fed back to the base station. The code word searching is carried out in the first stage and the second stage to obtain a pre-coding matrix F RF And baseband precoding matrix F BB And returns its index numberTo the base station.
Step S6: base band precoding matrix F designed by mixing in base station BB And an analog precoding matrix F RF Combining to obtain a precoding matrix F = F RF F BB And transmits data to the user by using the precoding matrix F.
Preferably, the method further comprises the following steps in step S1:
the construction method of the simulated predefined codebook F is as follows: let L be the size F of the codebook, i.e., L =2 b Where b is the number of feedback bits;
let L = L 1 L 2 …L s Wherein, L 1 L 2 8230L is a undetermined positive integer; the number of the antennas is M, and the codebook F adopts the following design method: f = { phi = 1 ,φ 2 ,…,φ L }; whereinθ i Is a rotation matrix of the construction and,
0≤u i1 ,u i2 ,…,u iM ≤L i -1 is M pending integers; phi is a 0 Unitary matrix of dimension M x r, i.e. satisfy(I L An identity matrix of L dimension), r is the rank of the codebook; and u i1 ,u i2 ,…,u iM The size of the code book is determined according to a method of maximizing the minimum angle of the code book; phi is a 0 Selecting front r columns from a DFT matrix with dimension of M multiplied by M;
simulating design criteria of a predefined codebook: by phi i Is generated by the column vector of<φ i >, it is an r-dimensional subspace in an M-dimensional space, thus,<φ i >, and<φ j >, r angles exist between the two, and the r angles are respectively recorded as omega ij,1 ≤ω ij,2 ≤…≤ω ij,r Let us note λ ij,1 ≤λ ij,2 ≤…≤λ ij,r Is a matrixThen their calculation method is as follows:
when the index s&At gt, M/2, there is omega s =0; thus, when r ≦ M/2, the codebook is selected such that min 1≤i<j≤L ω ij,r The larger the better; when r is&When gt, M/2, selecting codebook to make min 1≤i<j≤L ω ij,t Wherein t = M/2.
Preferably, in step S3,
a base station sends downlink pilot frequency training;
the user terminal calculates Channel State Information (CSI) according to the received pilot frequency sequence;
the user terminal utilizes the estimated channel matrix H to carry out beam searching operation and pre-coding processing;
preferably, in step S5 the data is transmitted,
in the double-codebook limited feedback, the computation complexity can be greatly reduced by utilizing the analog predefined codebook and the digital predefined codebook.
Preferably, in step S6,
the base station sends a symbol s, and the base station sends data to the user terminal by using the precoding matrix F.
Preferably, the method further comprises the following steps:
and the user terminal receives and decodes the data transmitted by the base station.
Compared with the prior art, the invention adopts the mixed analog/digital pre-coding method based on the limited feedback, thereby effectively reducing the calculation complexity and obtaining better performance, and compared with the limited feedback method of other millimeter waves in the prior artCompared with the prior art, the improved speed is improved by 1-2dB. And simulation results show that when N s =1, the performance of our proposed method is significantly better than the general limited feedback method; when N is s The performance of our proposed method will also improve when =2, but without N s Performance is evident when =1.
Drawings
Fig. 1 is a schematic diagram of an application scenario of the present invention.
Fig. 2 is a schematic diagram of an angle of arrival of a user terminal signal at a base station in the present invention.
Fig. 3 is a flow chart of a limited feedback hybrid precoding method applied to a large-scale MIMO system in a millimeter wave frequency band according to the present invention.
FIG. 4 is a system diagram of a large-scale millimeter wave system with a hybrid preprocessing architecture for the method of the present invention
Fig. 5 is a performance comparison of four precoding methods in embodiment 1 of the present invention.
Fig. 6 is a performance comparison of four precoding methods in embodiment 2 of the present invention.
Fig. 7 is a performance comparison of four precoding methods in embodiment 3 of the present invention.
Fig. 8 is a performance comparison of four precoding methods in embodiment 4 of the present invention. .
The following specific embodiments will further illustrate the invention in conjunction with the above-described figures.
Detailed Description
The technical solution provided by the present invention will be further explained with reference to the accompanying drawings.
Before describing the specific technical scheme of the invention, partial abbreviations and symbols are defined and introduced into a system model. (. Cndot.) -1 、(·) T And (·) * Respectively representing the inverse, transpose, and conjugate transpose operations (Hermite operators); | A | non-conducting phosphor F Is the Frobenius norm of a; tr (A), wherein A represents the trace and determinant of A; [ A | B ]]Indicating waterFlat linking; I.C. A N×N A unit array representing N × N;
referring to fig. 1, a schematic diagram of an application scenario of the present invention is shown. Consider a downstream single-user millimeter wave system. Base station side equipment N t Root transmitting antenna andan RF chain, a user terminal is provided with N r Root receiving antenna andan RF chain. The invention is shownIs applied to transmit information symbol vectorsIn the base band ofAfter the baseband pre-coding and RF processing steps of each RF chain, the base station side carries out the transmission signalThe analog precoding processing of (1). Given F = F RF F BB As a combination of base station precoding matrices for users. The transmit-end signal of the base station can be expressed as:
x＝Fs＝F RF F BB s
whereinAndassuming that the transmitted information symbols have unity power, i.e.Because F RF Is realized by using an analog phase shifterOf (F) RF Is to satisfyWherein (·) l,l Representing the ith diagonal element of the matrix, i.e. F RF The modular length of all elements is one. Meanwhile, in order to satisfy the burden of the total power P of the base station, the constraint of the precoding matrix F is Tr { FF * }≤P。
The downlink channel from the base station to the user terminal isThe received signal of the user terminal can be modeled as
y＝Hx+z＝HF RF F BB s+z
Wherein y is N r X 1 received signal vector, z is assumed to be independently and identically distributedA noise vector.
For convenience of explaining the technical scheme of the present invention, referring to fig. 2, a schematic diagram of a scatterer of a millimeter wave MIMO system is shown, an extended Saleh-Valenzuela channel model is adopted, and a matrix channel H of a trunking channel model is assumed to be N cl A plurality of scatterer clusters, each scatterer cluster having N ray Sum of contributions of individual propagation paths, i.e. L = N cl ×N ray . Thus, the discrete-time narrowband channel H can be written as:
where L is the number of propagation paths; alpha is alpha l Is the complex gain of the l-th path, including the path loss and alpha l ～CN(0，σ α )；Andits (azimuth, pitch) arrival angle and departure angle, respectively. Then, vectorAndare respectively shown inAndnormalized receive and transmit array response vectors over (azimuth, pitch) angles of (c),is a receiving array matrix andis a transmit array matrix; additional normalization factorIs to satisfy
Note the array response vectorAndtwo commonly used antenna array configurations are a Uniform Linear Array (ULA) and a Uniform Planar Array (UPA), depending only on the receive and transmit antenna array configurations. For the case of Uniform Planar Arrays (UPAs) having W and H elements in the yz plane in the y and z axes, respectively, the array response vector can be written as
Wherein m is 0 or more and W is 0 or less and N is 0 or more and H is the index number of the antenna elements of the y-axis and the z-axis, respectively, and the size of the antenna array is N = WH.
Referring to fig. 3 to 4, shown are a flow chart diagram of a limited feedback hybrid precoding method applied to a large-scale MIMO system in a millimeter wave frequency band and a millimeter wave system diagram, specifically including the following steps:
step S1: an analog predefined codebook set and a code word searching method thereof are constructed, and a digital predefined codebook adopts a conventional random vector quantization codebook set. Because the analog precoder of the millimeter wave MIMO system is usually implemented by using an analog phase shifter, the modulus of each element of the analog precoding matrix is constant. The predefined codebook F is designed taking this constraint into account. The simulated predefined codebook design includes a design method and a design criterion of the codebook. In the design of simulating the predefined codebook, the concept of a rotation matrix is added, and based on the coding principle, good equivalent mapping can be carried out, so that a better codebook is constructed. Since the relevant hardware of the baseband precoding matrix has no constraint, the digital predefined codebook W employs a conventional Random Vector Quantization (RVQ) codebook set.
Step S2: and simultaneously storing the analog predefined codebook set and the digital predefined codebook set of the millimeter wave MIMO system in the base station and the user terminal.
And step S3: in the first stage of feedback, the user terminal performs codeword search operation according to the estimated channel state information, i.e. first obtains the analog precoding matrix F RF . The user terminal estimates the channel state information according to the pilot training sequence sent by the base station, and processes the CSI to obtain the optimal digital pre-coding matrix F of the data stream opt And then, according to the space sparse optimization problem of the millimeter wave, a non-convex optimization problem of hybrid precoding can be obtained similarly:
s.t.F RF ∈F，and is obtained by least square methodThe optimization problem described above can be written as:
s.t.F RF ∈F，wherein N is s Is the number of data streams transmitted. The user terminal searches the precoding codebook F according to the optimization problem, thereby obtaining the simulation precoder F RF 。
And step S4: in the second phase, the user terminal then performs precoding based on the analog precoding matrix F RF And carrying out code word searching operation on the channel state information so as to obtain a baseband precoding matrix F BB . The baseband precoding codebook is a random vector quantization codebook W, and a digital precoding matrix F is obtained by vector quantization of the random vector quantization codebook W BB 。
Step S5: the user terminal will simulate a precoding matrix F RF And a digital precoding matrix F BB The beam index is fed back to the base station. The code word searching is carried out in the first stage and the second stage to obtain a pre-coding matrix F RF And baseband precoding matrix F BB And returns its index number to the base station.
Step S6: base band precoding matrix F designed by mixing in base station BB And an analog precoding matrix F RF Combining to obtain a precoding matrix F = F RF F BB And transmits data to the user by using the precoding matrix F.
The key point of the present invention is the construction of the analog precoding codebook set, and in a preferred embodiment, the step S1 further includes the following steps:
simulation ofThe construction method of the predefined codebook F is as follows: let L be the size F of the codebook, i.e., L =2 b Wherein b is the number of feedback bits;
let L = L 1 L 2 …L s Wherein L is 1 L 2 8230and L is a undetermined positive integer; the number of the antennas is M, and the codebook F adopts the following design method: f = { phi = 1 ,φ 2 ,…,φ L }; whereinθ i Is a rotation matrix of the construction and,
0≤u i1 ,u i2 ,…,u iM ≤L i -1 is M pending integers; phi is a 0 Unitary matrix of dimension M x r, i.e. satisfy(I L An identity matrix of L dimension), r is the rank of the codebook; and u i1 ,u i2 ,…,u iM The size of the code book is determined according to a method of maximizing the minimum angle of the code book; phi is a 0 The structure of the method is that front r columns are selected from a DFT matrix with M multiplied by M dimensions;
simulating design criteria of a predefined codebook: by phi i Is generated by the column vector of<φ i >, it is an r-dimensional subspace in an M-dimensional space, thus,<φ i >, and<φ j >, r angles exist between the two, and the r angles are respectively recorded as omega ij,1 ≤ω ij,2 ≤…≤ω ij,r Let us note λ ij,1 ≤λ ij,2 ≤…≤λ ij,r Is a matrixThen their calculation method is as follows:
when the index s&At gt, M/2, there is omega s =0; thus, when r ≦ M/2, the codebook is selected such that min 1≤i<j≤L ω ij,r The larger the better; when r is&Selecting codebook for min at gt, M/2 1≤i<j≤L ω ij,t Where t = M/2.
In a preferred embodiment, in step S3,
a base station sends downlink pilot frequency training;
the user terminal calculates Channel State Information (CSI) according to the received pilot frequency sequence;
the user terminal utilizes the estimated channel matrix H to perform beam searching operation and precoding processing;
in a preferred embodiment, in step S5,
in the double-codebook limited feedback, the computation complexity can be greatly reduced by utilizing the analog predefined codebook and the digital predefined codebook.
Preferably, in step S6,
the base station sends symbols for s the base station sends data to the user terminal by using the precoding matrix F.
Preferably, the method further comprises the following steps:
and a step of receiving and decoding the data transmitted by the base station by the user terminal.
In a preferred embodiment, the method further comprises the following steps:
and the user terminal receives and decodes the data transmitted by the base station. It is the inverse process of the encoding process, and some specific decoding processes are briefly introduced here:
for the signal y received by the user terminal, the matrix W is combined by simulation RF And a digital equalization matrix W BB After the receiving process, the modulus operation is carried out, and then the signals are processedPerforming decoding processing。
The following describes in detail a finite feedback hybrid precoding method applied to a massive MIMO system in a millimeter wave band by using a specific example.
Example 1
In a millimeter wave MIMO system of a single user, the number of base station antennas is set to be 8 multiplied by 8 Uniform Plane Array (UPA), namely N t =64 andnumber of RF chains, and the number of user antennas set to 4 x 4 Uniform Planar Array (UPA), i.e., N r =16 anda radio frequency chain. Assume that the channel of each user contains 8 clusters, and there are 10 propagation paths in each cluster, i.e., the total number of propagation paths of the users is L =80. The number of data streams sent by the base station to the user is set to be N s =1. All channel path gains alpha l The assumption is that the variance of the independent co-distributions isGaussian distribution of (a). And assuming azimuth angles of the beam arrival angle and departure angle are [0;2 pi]Uniformly distributed in the inner part and at a pitch angle ofThe inner parts are uniformly distributed. Setting the noise variance to σ 2 =1, SNR at the time of drawing is defined as
Assuming feedback of channel state information of channel matrix H and feedback of optimal digital precoding matrix F opt The feedback bit number of the channel state information is b =10bit; in the design of the dual-codebook hybrid beam forming, the feedback bit number b1=4bit of the analog predefined codebook is set, and the feedback bit number of the digital predefined codebook is b2=6bit is. And when the total bit number of the feedback links is the same, performing simulation comparison on the feedback links.
Referring to fig. 5, which is a performance simulation diagram of embodiment 1, the spectral efficiency of the system obtained by respectively adopting 4 precoding methods is shown, and it can be seen that the spectral efficiency of the method provided by the present invention is improved by about 3dB compared with the performance of directly feeding back a channel matrix H; and direct feedback digital precoding matrix F opt Compared with the performance of the prior art, the performance is improved by about 0.4dB, and meanwhile, the computational complexity is greatly reduced.
Example 2
In the millimeter wave MIMO system of single user, the number of base station antennas is set to be 8 multiplied by 8 Uniform Plane Array (UPA), namely N t =64 andnumber of RF chains, and the number of user antennas set to 4 x 4 Uniform Planar Array (UPA), i.e., N r =16 anda radio frequency chain. Assume that the channel of each user contains 8 clusters, and there are 10 propagation paths in each cluster, i.e., the total number of propagation paths of the users is L =80. The number of data streams sent by the base station to the user is set to be N s And (2). All channel path gains alpha l The assumption is that the variance of the independent co-distributions isGaussian distribution of (a). And assuming that the azimuth angles of the beam arrival angle and departure angle are [0;2 pi]Uniformly distributed in the inner part and at a pitch angle ofThe inner parts are uniformly distributed. Setting the noise variance to σ 2 =1, SNR at the time of drawing is defined as
Assuming feedback of channel state information of channel matrix H and feedback of optimal digital precoding matrix F opt The feedback bit number of the channel state information is b =10bit; in the design of the dual-codebook hybrid beam forming, the feedback bit number b1=4 bits of the analog predefined codebook is set, and the feedback bit number of the digital predefined codebook is b2=6 bits. And when the total bit number of the feedback links is the same, performing simulation comparison on the feedback links.
Referring to fig. 6, which is a performance simulation diagram of embodiment 2, the spectral efficiency of the system obtained by respectively adopting 4 precoding methods is shown, and it can be seen that the spectral efficiency of the method provided by the present invention is improved by about 3.5dB compared with the performance of directly feeding back the channel matrix H; and direct feedback digital precoding matrix F opt Compared with the performance of the prior art, the performance is improved by about 1.6dB, and meanwhile, the computational complexity is greatly reduced.
Example 3
In the millimeter wave MIMO system of single user, the number of base station antennas is set to be 8 multiplied by 8 Uniform Plane Array (UPA), namely N t =64 andnumber of RF chains, and the number of user antennas set to 4 x 4 Uniform Planar Array (UPA), i.e., N r =16 anda radio frequency chain. Assume that the channel of each user contains 8 clusters, and there are 10 propagation paths in each cluster, i.e., the total number of propagation paths of the users is L =80. The number of data streams sent by the base station to the user is set to be N s =1. All channel path gains a l The assumption is that the variances of the independent equal distributions areA gaussian distribution of (a). And assuming that the azimuth angles of the beam arrival angle and departure angle are [0;2 pi]Uniformly distributed in the inner part and at a pitch angle ofThe inner parts are uniformly distributed. Setting the noise variance to σ 2 =1, SNR at the time of drawing is defined as
Assuming feedback of channel state information of channel matrix H and feedback of optimal digital precoding matrix F opt The feedback bit number of the channel state information is b =8bit; in the design of the dual-codebook hybrid beam forming, the feedback bit number b1=4bit of the analog predefined codebook is set, and the feedback bit number of the digital predefined codebook is b2=4bit. And when the total bit number of the feedback links is the same, performing simulation comparison on the feedback links.
Referring to fig. 7, a performance simulation diagram of embodiment 3 is shown, and the spectral efficiency of the system obtained by respectively adopting 4 precoding methods, it can be seen that the spectral efficiency of the method provided by the present invention is improved by about 2.5dB compared with the performance of directly feeding back the channel matrix H; and direct feedback digital precoding matrix F opt Compared with the performance of the prior art, the performance is improved by about 0.4dB, and meanwhile, the computational complexity is greatly reduced.
Example 4
In the millimeter wave MIMO system of single user, the number of base station antennas is set to be 8 multiplied by 8 Uniform Plane Array (UPA), namely N t =64 andnumber of RF chains, and the number of user antennas set to 4 x 4 Uniform Planar Array (UPA), i.e., N r =16 anda radio frequency chain. Assume that the channel of each user contains 8 clusters, and there are 10 propagation paths in each cluster, i.e., the total number of propagation paths of the users is L =80. The number of data streams sent by the base station to the user is set to be N s And (5) =2. All channel path gains alpha l Suppose to be independent ofVariance of the distribution isA gaussian distribution of (a). And assuming that the azimuth angles of the beam arrival angle and departure angle are [0;2 pi]Uniformly distributed in the inner part and at a pitch angle ofThe inner parts are uniformly distributed. Setting the noise variance to σ 2 =1, SNR at the time of drawing is defined as
Assuming feedback of channel state information of channel matrix H and feedback of optimal digital precoding matrix F opt The feedback bit number of the channel state information is b =8bit; in the design of the dual-codebook hybrid beam forming, the feedback bit number b1=4bit of the analog predefined codebook is set, and the feedback bit number of the digital predefined codebook is b2=4bit. And when the total bit number of the feedback links is the same, performing simulation comparison on the feedback links.
Referring to fig. 8, a performance simulation diagram of embodiment 4 is shown, and the spectral efficiency of the system obtained by respectively using 4 precoding methods shows that the spectral efficiency of the method provided by the present invention is improved by about 3dB compared with the performance of directly feeding back the channel matrix H; and direct feedback digital precoding matrix F opt Compared with the performance of the prior art, the performance is improved by about 2.5dB, and meanwhile, the computational complexity is greatly reduced.
The above description of the embodiments is only intended to facilitate the understanding of the method of the invention and its core idea. It should be noted that, for those skilled in the art, without departing from the principle of the present invention, it is possible to make various improvements and modifications to the present invention, and those improvements and modifications also fall within the scope of the claims of the present invention.
The previous description of the disclosed embodiments is provided to enable any person skilled in the art to make or use the present invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other embodiments without departing from the spirit or scope of the invention. Thus, the present invention is not intended to be limited to the embodiments shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.
Claims (6)
1. A limited feedback mixed precoding method based on a millimeter wave MIMO system is characterized by comprising the following steps:
step S1: constructing a simulation predefined codebook set and a code word searching method thereof;
s2, simultaneously storing an analog predefined codebook set and a digital predefined codebook set of the millimeter wave MIMO system in a base station and a user terminal;
and step S3: in the first stage of feedback, the user terminal performs codeword search operation according to the estimated channel state information, and first obtains an analog precoding matrix F RF (ii) a The specific process is as follows:
the user terminal estimates the channel state information according to the pilot training sequence sent by the base station, and processes the CSI to obtain the optimal digital pre-coding matrix F of the data stream opt ；
Obtaining a non-convex optimization problem of hybrid precoding from a space sparse optimization problem of millimeter waves:
s.t.F RF ∈F，
then obtained by Least Squares (LS)The above optimization problem is converted into:
s.t.F RF ∈F，wherein N is s Is the number of data streams transmitted;
user terminal precoding codebook according to optimization problemSearching is carried out to obtain an analog precoder F RF ；
And step S4: in the second stage feedback, the user terminal is based on the simulation pre-coding matrix F RF And carrying out code word searching operation on the channel state information so as to obtain a baseband precoding matrix F BB ；
Step S5: the user terminal will simulate the precoding matrix F RF And a digital precoding matrix F BB The beam index is fed back to the base station;
step S6: base band precoding matrix F designed by mixing in base station BB And an analog precoding matrix F RF Combining to obtain a precoding matrix F = F RF F BB And transmits data to the user by using the precoding matrix F.
2. The limited feedback hybrid precoding method based on the mmwave MIMO system of claim 1, further comprising the following steps in step S1:
the construction method of the simulated predefined codebook F is as follows:
let L be the size F of the codebook, i.e., L =2 b Wherein b is the number of feedback bits;
let L = L 1 L 2 …L s Wherein, L 1 L 2 8230and L is a undetermined positive integer; the number of the antennas is M, and the codebook F adopts the following design method: f = { phi = 1 ,φ 2 ,…,φ L }; whereinθ i Is a rotation matrix of the construction and,
0≤u i1 ,u i2 ,…,u iM ≤L i -1 is M pending integers; phi is a 0 Unitary matrix of dimension M x r, i.e. satisfy(I L An identity matrix of L dimension), r is the rank of the codebook; and u i1 ,u i2 ,…,u iM The size of the code book is determined according to a method of maximizing the minimum angle of the code book; phi is a 0 Selecting front r columns from a DFT matrix with dimension of M multiplied by M;
simulating design criteria of a predefined codebook: by phi i Is generated by the column vector of<φ i >, it is an r-dimensional subspace in an M-dimensional space, thus,<φ i >, and<φ j &r angles exist between gt and g, and the r angles are respectively recorded as omega ij,1 ≤ω ij,2 ≤…≤ω ij,r Let us denote λ ij,1 ≤λ ij,2 ≤…≤λ ij,r Is a matrixThen their calculation method is as follows:
when index s&At gt, M/2, there is omega s =0; thus, when r ≦ M/2, the codebook is selected such that min 1≤i<j≤L ω ij,r The larger the better; when r is&Selecting codebook for min at gt, M/2 1≤i<j≤L ω ij,t Where t = M/2.
3. The limited feedback hybrid precoding method based on the millimeter wave MIMO system of claim 1, wherein in step S3,
a base station sends downlink pilot frequency training;
the user terminal calculates Channel State Information (CSI) according to the received pilot frequency sequence;
and the user terminal performs beam searching operation and precoding processing by using the estimated channel matrix H.
4. The MMIME MIMO system-based limited feedback hybrid precoding method of claim 1, wherein in step S5, a dual codebook structure is adopted, wherein the first stage adopts an analog precoding codebook F, and the second stage adopts a baseband precoding codebook F
5. The limited feedback hybrid precoding method based on the millimeter wave MIMO system of claim 1, wherein in step S6,
transmitting data to a userProcessing by a preprocessing matrix F of a base station end to enable a transmitted signal to be x = Fs; the transmission signal of the base station is x and the data stream is transmitted to the user terminal by using the precoding matrix F.
6. The limited feedback hybrid precoding method based on the mmwave MIMO system of claim 1, further comprising the steps of:
and the user terminal receives and decodes the data transmitted by the base station.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711233176.1A CN107809275B (en) | 2017-11-30 | 2017-11-30 | Finite feedback hybrid precoding method based on millimeter wave MIMO system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711233176.1A CN107809275B (en) | 2017-11-30 | 2017-11-30 | Finite feedback hybrid precoding method based on millimeter wave MIMO system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN107809275A true CN107809275A (en) | 2018-03-16 |
| CN107809275B CN107809275B (en) | 2020-06-23 |
Family
ID=61581868
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201711233176.1A Active CN107809275B (en) | 2017-11-30 | 2017-11-30 | Finite feedback hybrid precoding method based on millimeter wave MIMO system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107809275B (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109245803A (en) * | 2018-11-16 | 2019-01-18 | 西安科技大学 | A kind of estimation of Fast Channel and mix method for precoding and system |
| CN111447620A (en) * | 2020-03-19 | 2020-07-24 | 重庆邮电大学 | Millimeter wave heterogeneous network resource allocation joint optimization method |
| CN111654456A (en) * | 2020-06-09 | 2020-09-11 | 江南大学 | Millimeter wave large-scale MIMO angular domain channel estimation method and device based on dimension reduction decomposition |
| CN113287265A (en) * | 2019-01-17 | 2021-08-20 | 三菱电机株式会社 | Method for enabling analog precoding and analog combining |
| CN113728558A (en) * | 2019-04-26 | 2021-11-30 | 华为技术有限公司 | Method and system for hybrid beamforming for MIMO communication |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20150188682A1 (en) * | 2013-12-26 | 2015-07-02 | Samsung Electronics Co., Ltd. | Hybrid zero-forcing beamforming method and apparatus |
| CN106130610A (en) * | 2016-06-20 | 2016-11-16 | 电子科技大学 | A kind of iteratively faster beam-forming method in millimeter wave pre-coding system |
| CN107171709A (en) * | 2017-06-23 | 2017-09-15 | 杭州电子科技大学 | A kind of extensive mimo system method for precoding being applied under aggregate users scene |
| CN107181511A (en) * | 2017-06-07 | 2017-09-19 | 深圳先进技术研究院 | The mixing method for precoding and system of a kind of millimeter wave mimo system |
| CN107294590A (en) * | 2017-06-20 | 2017-10-24 | 电子科技大学 | A kind of numerical model analysis beam form-endowing method based on up training |
| CN107332596A (en) * | 2017-05-26 | 2017-11-07 | 南京邮电大学 | A kind of millimeter-wave communication system mixing method for precoding based on ZF |
-
2017
- 2017-11-30 CN CN201711233176.1A patent/CN107809275B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20150188682A1 (en) * | 2013-12-26 | 2015-07-02 | Samsung Electronics Co., Ltd. | Hybrid zero-forcing beamforming method and apparatus |
| CN106130610A (en) * | 2016-06-20 | 2016-11-16 | 电子科技大学 | A kind of iteratively faster beam-forming method in millimeter wave pre-coding system |
| CN107332596A (en) * | 2017-05-26 | 2017-11-07 | 南京邮电大学 | A kind of millimeter-wave communication system mixing method for precoding based on ZF |
| CN107181511A (en) * | 2017-06-07 | 2017-09-19 | 深圳先进技术研究院 | The mixing method for precoding and system of a kind of millimeter wave mimo system |
| CN107294590A (en) * | 2017-06-20 | 2017-10-24 | 电子科技大学 | A kind of numerical model analysis beam form-endowing method based on up training |
| CN107171709A (en) * | 2017-06-23 | 2017-09-15 | 杭州电子科技大学 | A kind of extensive mimo system method for precoding being applied under aggregate users scene |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109245803A (en) * | 2018-11-16 | 2019-01-18 | 西安科技大学 | A kind of estimation of Fast Channel and mix method for precoding and system |
| CN113287265A (en) * | 2019-01-17 | 2021-08-20 | 三菱电机株式会社 | Method for enabling analog precoding and analog combining |
| CN113287265B (en) * | 2019-01-17 | 2024-01-12 | 三菱电机株式会社 | Method for enabling analog precoding and analog combining |
| CN113728558A (en) * | 2019-04-26 | 2021-11-30 | 华为技术有限公司 | Method and system for hybrid beamforming for MIMO communication |
| CN113728558B (en) * | 2019-04-26 | 2023-04-04 | 华为技术有限公司 | Method and system for hybrid beamforming for MIMO communication |
| CN111447620A (en) * | 2020-03-19 | 2020-07-24 | 重庆邮电大学 | Millimeter wave heterogeneous network resource allocation joint optimization method |
| CN111447620B (en) * | 2020-03-19 | 2022-05-17 | 重庆邮电大学 | Millimeter wave heterogeneous network resource allocation joint optimization method |
| CN111654456A (en) * | 2020-06-09 | 2020-09-11 | 江南大学 | Millimeter wave large-scale MIMO angular domain channel estimation method and device based on dimension reduction decomposition |
Also Published As
| Publication number | Publication date |
|---|---|
| CN107809275B (en) | 2020-06-23 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107809275B (en) | Finite feedback hybrid precoding method based on millimeter wave MIMO system | |
| Mendez-Rial et al. | Dictionary-free hybrid precoders and combiners for mmWave MIMO systems | |
| Seleem et al. | Hybrid precoding-beamforming design with Hadamard RF codebook for mmWave large-scale MIMO systems | |
| US8594161B2 (en) | Method and system for beamforming in a multiple user multiple input multiple output (MIMO) communication system using a codebook | |
| Song et al. | Codebook design for hybrid beamforming in millimeter wave systems | |
| CN107453795B (en) | Beam allocation method of multi-user millimeter wave communication system, device and system thereof | |
| CN111294095A (en) | IRS (inter-range instrumentation Standard) assisted large-scale MIMO (multiple input multiple output) wireless transmission method based on statistical CSI (channel State information) | |
| CN107332596B (en) | Zero forcing-based millimeter wave communication system hybrid precoding method | |
| CN109861731B (en) | Hybrid precoder and design method thereof | |
| CN106571858B (en) | Hybrid beam forming transmission system | |
| CN108736943B (en) | Hybrid precoding method suitable for large-scale MIMO system | |
| CN108023620A (en) | Extensive mimo system mixing method for precoding applied to millimeter wave frequency band | |
| EP1568155A2 (en) | Low complexity beamformers for multiple transmit and receive antennas | |
| Zhang et al. | Joint precoding and combining design for hybrid beamforming systems with subconnected structure | |
| Song et al. | Coordinated hybrid beamforming for millimeter wave multi-user massive MIMO systems | |
| Salh et al. | Low computational complexity for optimizing energy efficiency in mm-wave hybrid precoding system for 5G | |
| CN111953393A (en) | Large-scale MIMO hybrid precoder and matching relationship | |
| Han et al. | Efficient hybrid beamforming design in mmWave massive MU-MIMO DF relay systems with the mixed-structure | |
| CN108566236B (en) | User terminal, base station, and hybrid beamforming transmission method and system | |
| CN109067446B (en) | Mixed precoding method for multi-antenna multi-user large-scale antenna | |
| CN114900398A (en) | IRS (intelligent resilient framework) assisted cloud access network downlink beam forming method for non-ideal CSI (channel state information) | |
| CN107104719B (en) | Millimeter wave digital-analog hybrid precoding design method based on geometric construction | |
| WO2017059448A1 (en) | System and method for resolving channel sparsity in multiple-input multiple-output systems using (re)configurable antennas | |
| CN108123741B (en) | Overlapping sub-array (OSA) based beamforming method and apparatus | |
| Jiang et al. | Downlink multiuser hybrid beamforming for MmWave massive MIMO-NOMA system with imperfect CSI |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |