CN107171702A - Extensive mimo channel feedback method based on PCA evolution - Google Patents

Extensive mimo channel feedback method based on PCA evolution Download PDF

Info

Publication number
CN107171702A
CN107171702A CN201710333865.3A CN201710333865A CN107171702A CN 107171702 A CN107171702 A CN 107171702A CN 201710333865 A CN201710333865 A CN 201710333865A CN 107171702 A CN107171702 A CN 107171702A
Authority
CN
China
Prior art keywords
channel
matrix
vector
sparse
antenna
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.)
Pending
Application number
CN201710333865.3A
Other languages
Chinese (zh)
Inventor
廖勇
陈玲
张舒敏
沈轩帆
胡异
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Chongqing University
Original Assignee
Chongqing University
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Chongqing University filed Critical Chongqing University
Priority to CN201710333865.3A priority Critical patent/CN107171702A/en
Publication of CN107171702A publication Critical patent/CN107171702A/en
Pending legal-status Critical Current

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
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04B—TRANSMISSION
    • H04B17/00—Monitoring; Testing
    • H04B17/30—Monitoring; Testing of propagation channels
    • H04B17/391—Modelling the propagation channel
    • 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
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L25/00—Baseband systems
    • H04L25/02—Details ; arrangements for supplying electrical power along data transmission lines
    • H04L25/0202—Channel estimation
    • H04L25/024—Channel estimation channel estimation algorithms

Landscapes

  • Engineering & Computer Science (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Signal Processing (AREA)
  • Physics & Mathematics (AREA)
  • Electromagnetism (AREA)
  • Power Engineering (AREA)
  • Radio Transmission System (AREA)

Abstract

The present invention proposes a kind of extensive mimo channel feedback method based on PCA evolution.First, it is considered to the characteristic of channel in two domains in spatial domain and frequency domain, down channel relevant featuring parameters are obtained, channel model is set up；Secondly, the channel condition information after vector quantization is subjected to sub-clustering, the correlation and compressibility of channel condition information is improved, so as to which higher-dimension channel condition information is reduced into low-dimensional；Then, the covariance matrix and sparse matrix of channel condition information in the every cluster of characterisitic parameter channel model are calculated in transmitting-receiving two-end；Furthermore, sparse vector main component is extracted by selection matrix in receiving terminal, selected condition of sparse channel vector is met the maximum principle of each element sum；Finally, the codewords indexes of selected condition of sparse channel vector and selection matrix are fed back into transmitting terminal.This method, which is used, is based on PCA thoughts and sub-clustering thought, realizes and substantial amounts of feedback quantity and reduction computation complexity are reduced while degree of precision is obtained.

Description

Extensive mimo channel feedback method based on PCA evolution
Technical field
The application is related to wireless communication technology field channel state information feedback method, and more particularly to one kind is drilled based on PCA The extensive mimo channel status information compressed feedback method entered.
Background technology
Multiple-input and multiple-output (Multiple-Input Multiple-Output, MIMO) technology refers in transmitting terminal and connect Receiving end uses multiple transmitting antennas and reception antenna respectively, signal is transmitted and is connect by multiple antennas of transmitting terminal and receiving terminal Receive, so as to improve communication quality.It can make full use of space resources, and MIMO is realized by multiple antennas, not increase frequency In the case of spectrum resource and antenna transmission power, system channel capacity can be increased exponentially, is considered as next generation mobile communication Core technology.OFDM (Orthogonal Frequency Division Multiplexing, OFDM) is one Effective modulation scheme is planted, it breaks a channel into some orthogonal sub-channels, high-speed data signal is converted into parallel low speed Data flow, is modulated to and is transmitted on each of the sub-channels.Receiving terminal is separated orthogonal signalling using correlation technique, is reduced Interfering between subchannel.Signal bandwidth on per sub-channels is less than the correlation bandwidth of channel, therefore per sub-channels On can regard as flatness decline, so as to eliminate intersymbol interference.
In the extensive MIMO-OFDM systems of mobile communication, to realize that energy is focused on into target using precoding technique moves In dynamic terminal, the channel information estimated is typically fed back into transmitting terminal by feedback channel by receiving terminal, makes transmitting terminal real When the accurate down channel of acquisition channel information.But as extensive MIMO technology is promoted, antenna amount can be significantly increased, If downlink channel information is directly fed back into transmitting terminal completely, capacitance loss can be significantly greatly increased, so to channel information Compressed feedback mode turn into research focus.
Principal component analysis (Principal Component Analysis, PCA), also referred to as principal component analysis or matrix data Analysis, a kind of technology of data analysis, main thought is that high dimensional data is projected to the master that polynary things is extracted compared with lower dimensional space Factor is wanted, its substantive characteristics is disclosed.It can efficiently find out the major part in data, by original complex data dimensionality reduction, The noise and redundancy in whole data are removed, related variable will be changed into some incoherent overall target variables, be a kind of Optimal transformation in lowest mean square meaning, it is therefore an objective to remove in the correlation between input random vector, prominent initial data Implied feature.The advantage of principal component analytical method is data compression and carries out dimensionality reduction to multidimensional data, simple to operate, and does not have There is parameter limitation, in that context it may be convenient to applied to each occasion.Therefore, the PCA channel status for being used for extensive mimo system is believed In breath compression, it is possible to reduce feedback overhead.But, it is proposed that use PCA the method for channel information compression is only considered mostly Correlation in spatial domain or frequency domain one domain, and need dynamic to update for the selection matrix that main component is extracted at present, This undoubtedly adds feedback quantity and improves computation complexity from another point of view.
To sum up, channel state information feedback method in the extensive mimo system occurred now, is calculated for them The problems such as method complexity is high, feedback quantity is big, feedback accuracy is relatively low, it is proposed that based on the extensive of principal component analysis evolution MIMO-OFDM channel information feedback methods, it is considered to which channel information is thought in the correlation of two dimensions of spatial domain frequency domain with reference to sub-clustering Think, by large volumes of channels status information compressed feedback to transmitting terminal and operation is reconstructed.
The content of the invention
It is contemplated that at least solving technical problem present in prior art, especially innovatively propose one kind and be based on The extensive MIMO-OFDM system channels feedback method of PCA evolution.
Goal of the invention：It is anti-after being compressed to channel condition information in order to realize in extensive MIMO-OFDM systems Feedback, the present invention proposes a kind of channel information compression method based on PCA evolution come the related channel condition information of feed back height. This method considers the characteristic of channel in two domains in spatial domain and frequency domain first, and letter is set up by obtaining down channel relevant featuring parameters Road model；Then it is that channel vector carries out sub-clustering by the channel condition information after vector quantization, improves the correlation of channel condition information Property and compressibility, to be reduced to low-dimensional data to the higher-dimension channel condition information in every cluster；Secondly, calculated in transmitting-receiving two-end The covariance matrix and sparse matrix of channel condition information, in receiving terminal, pass through choosing in the every cluster of characterisitic parameter channel model Select matrix and extract sparse vector main component, selected condition of sparse channel vector is met the maximum principle of each element sum；Finally will Selected channel information and the codewords indexes of selection matrix feed back to the reconstruct that transmitting terminal carries out channel condition information.
In order to realize the above-mentioned purpose of the present invention, the invention provides a kind of extensive MIMO-OFDM based on PCA evolution System channel feedback method, its feature includes：
S1, obtains down channel relevant featuring parameters, sets up channel model.
S2, by channel vector hkSub-clustering is carried out, the association of every cluster channel condition information is calculated according to characteristic of channel parameter Variance matrix, and sparse matrix is obtained by the Eigenvalues Decomposition of covariance matrix.
S3, predefines a transmitting-receiving two-end known binary system selection matrix code book, according to selected condition of sparse channel to Measure the maximum principle of each element sum and choose optimal selection matrix.
The described extensive MIMO-OFDM system channels feedback method based on PCA evolution, it is characterised in that the S1 Including：
The present invention considers an extensive MIMO-OFDM system, and transmitting terminal is configuration NtThe linear array of root uniform antenna, Receiving terminal is single-antenna subscriber, it is considered to which frequency domain has NcIndividual subcarrier；In extensive MIMO-OFDM systems, transmitting terminal and k-th In spatial domain and the channel vector h of frequency domain between userkIt is expressed as：
Wherein N=Nt×Nc, vec (A) is represented a × b matrix A vector quantization, becomes the column vector of ab × 1, hk (n) it is the spatial domain channel vector of k-th of user, n-th of subcarrier, setting up model is：
In extensive mimo system, due to transmitting terminal configuration antenna number it is more when, antenna spacing is smaller, between antenna present Stronger spatial coherence, therefore, the channel between transmitting terminal and k-th of single-antenna subscriber can be modeled as：
Wherein,Represent to send correlation matrix, n represents n-th of subcarrier, n ∈ { 1 ..., Nc}；It is 0 to represent average, and variance is independently distributed multiple Gauss random vector, H for 1iid,k(n) (u, v) is individual Element represents the channel gain being made up of between transmitting antenna u and reception antenna v path loss and multipath fading.
Because transmitting terminal configures NtRoot uniform linear array antenna, spatial correlation matrix can be obtained by Jakes models, Coefficient correlation between pth root antenna and q root antennas is expressed as：
Wherein J0() represents first kind zero Bessel function, dpqRepresent between pth root antenna and q root antennas away from From λ represents carrier wavelength.
The described extensive MIMO-OFDM system channels feedback method based on PCA evolution, it is characterised in that the S2 Including：
S2-1, by channel vector hkCarry out sub-clustering.
Continuous channel condition information is divided into G cluster, i.e. h by receiving terminalk T=[hk,1 T…hk,G T], wherein G≤N；hk,gTable Show the channel condition information of g-th of cluster, instead of N number of channel condition information element, then included with G cluster per clusterIndividual channel State information elements.
S2-2, calculates covariance matrix of the channel per cluster channel condition information, and pass through association side according to characterisitic parameter The Eigenvalues Decomposition of poor matrix obtains sparse matrix；The channel covariance matrices of k-th of user's g cluster are expressed as：
Eigenvalues Decomposition is carried out to the covariance matrix of g cluster channel condition informations, sparse matrix is obtained：
Wherein,Represent hk,gCovariance matrix,Represent to covariance matrixCarry out characteristic value point Solution, i.e. Ck,g=Ψk,gΛΨk,g H；Λ is diagonal matrix, and its diagonal element is covariance matrixCharacteristic value, Ψk,g's Column vector is covariance matrixCharacteristic vector；Converted by Carlow south, Ψk,gSparse matrix is represented simultaneously.
The described extensive MIMO-OFDM system channels feedback method based on PCA evolution, it is characterised in that the S3 Including：
S3-1, predefines the known binary system selection matrix code book C of a transmitting-receiving two-endP：
CP={ P1…PL}
Wherein PiI-th of code word of selection matrix code book, i.e. selection matrix are represented, L represents code book CPInclude L selection square Battle array.
S3-2, chooses optimal selection matrix Pi。
Pass through selection matrix PiFrom sparse vector sk,gMiddle selection M most important feedback informations, wherein
s′k,g=Pisk,g
Global search selection matrix code book, as selected condition of sparse channel vector s 'k,gWhen each element sum is maximum, receiving terminal It is determined that the selection matrix P usedi。
I=argimax|sk,gPi|
Determine selection matrix PiAfterwards, by s 'k,g=Pisk,gFrom sparse vector sk,gMiddle selection M most important feedback letters Breath.
S3-3, code book is quantified using random vector, quantifies selected condition of sparse channel vector s 'k,g, s 'k,gCodewords indexes pass through Below equation is obtained：
J=argjmax|s′k,gwj|
Wherein wjIt is j-th of code word that random vector quantifies code book；Then by s 'k,gCodewords indexes and selection matrix Pi's Codewords indexes feed back to transmitting terminal together.
In summary, by adopting the above-described technical solution, the beneficial effects of the invention are as follows：
It is extensive MIMO uniform linear arrays according to transmitting terminal, for single-antenna subscriber, this scene establishes phase to receiving terminal The channel model of pass, sub-clustering is carried out by spatial domain and the continuous N number of channel condition information of frequency domain.Calculated first according to the characteristic of channel Per the covariance matrix and sparse matrix of cluster channel condition information；Then, an identical two is predefined in transmitting-receiving two-end to enter Selection matrix code book C processedP, selection matrix is determined according to the maximum principle of the vectorial each element sum of selected condition of sparse channel, selected Fixed condition of sparse channel vector；The vectorial random vector of selected condition of sparse channel and selection matrix index are finally fed back into transmitting Hold and operation is reconstructed.Continuous N number of channel condition information is carried out sub-clustering by the present invention, improves channel condition information related Property and the compression performance based on PCA；Channel covariance matrices and sparse matrix are obtained and in transmitting-receiving two-end by the characteristic of channel Predefined binary system selection matrix code book, reduces substantial amounts of feedback quantity and reduces complexity.
The additional aspect and advantage of the present invention will be set forth in part in the description, and will partly become from the following description Obtain substantially, or recognized by the practice of the present invention.
Brief description of the drawings
The above-mentioned and/or additional aspect and advantage of the present invention will become from description of the accompanying drawings below to embodiment is combined Substantially and be readily appreciated that, wherein：
Fig. 1 is present system structure chart；
Fig. 2 is overview flow chart of the present invention.
Embodiment
Embodiments of the invention are described below in detail, the example of the embodiment is shown in the drawings, wherein from beginning to end Same or similar label represents same or similar element or the element with same or like function.Below with reference to attached The embodiment of figure description is exemplary, is only used for explaining the present invention, and is not considered as limiting the invention.
In the description of the invention, it is to be understood that term " longitudinal direction ", " transverse direction ", " on ", " under ", "front", "rear", The orientation or position relationship of the instruction such as "left", "right", " vertical ", " level ", " top ", " bottom " " interior ", " outer " is based on accompanying drawing institutes The orientation or position relationship shown, is for only for ease of the description present invention and simplifies description, rather than indicate or imply signified dress Put or element there must be specific orientation, with specific azimuth configuration and operation, therefore it is not intended that to the limit of the present invention System.
In the description of the invention, unless otherwise prescribed with limit, it is necessary to explanation, term " installation ", " connected ", " connection " should be interpreted broadly, for example, it may be mechanically connect or electrical connection or the connection of two element internals, can To be to be joined directly together, it can also be indirectly connected to by intermediary, for the ordinary skill in the art, can basis Concrete condition understands the concrete meaning of above-mentioned term.
The present invention proposes a kind of extensive MIMO-OFDM system channels status information compressed feedback side by PCA evolution Method, the channel model of correlation is established according to consideration scene, continuous channel condition information is carried out into sub-clustering, channel shape is improved The correlation of state information and the compression performance based on PCA, by channel statistic and in transmitting-receiving two-end predefined two System selection matrix code book, reduces substantial amounts of feedback quantity and reduces complexity.
With reference to accompanying drawing 1 and accompanying drawing 2, the present invention is described in detail, mainly includes the following steps that：
Step 1：Start.
Step 2：Down channel relevant featuring parameters are obtained, correlated channels model is set up.
The present invention considers extensive MIMO-OFDM systems, and transmitting terminal is configuration NtThe antenna of root uniform linear array, is received Hold as single-antenna subscriber, there is NcIndividual subcarrier.In extensive mimo system, due to transmitting terminal configuration antenna number it is more when, day Line spacing is smaller, and stronger spatial coherence is presented between antenna, therefore, and the channel between transmitting terminal and kth single-antenna subscriber can be built Mould is：
Wherein,Represent to send correlation matrix, n represents n-th of subcarrier, n ∈ { 1 ..., Nc}；It is 0 to represent average, and variance is independently distributed multiple Gauss random vector, H for 1iid,k(n) (u, v) is individual Element represents the channel gain being made up of between transmitting antenna u and reception antenna v path loss and multipath fading.
Because transmitting terminal configures NtRoot uniform linear array antenna, spatial correlation matrix can be obtained by Jakes models, Coefficient correlation between pth root antenna and q root antennas is expressed as：
Wherein J0() represents first kind zero Bessel function, dpqRepresent between pth root antenna and q root antennas away from From λ represents carrier wavelength.
In extensive MIMO-OFDM systems, in spatial domain and the channel vector h of frequency domain between transmitting terminal and k-th of userkRepresent For：
Wherein N=Nt×Nc, vec (A) is represented a × b matrix A vector quantization, becomes the column vector of ab × 1, hk (n) it is the spatial domain channel vector of k-th of user, n-th of subcarrier, setting up model is：
Step 3：To channel vector hkCarry out sub-clustering.
Continuous N number of channel condition information is divided into G cluster by receiving terminal, i.e.,Wherein G ≤N；hk,gThe channel condition information of g-th of cluster is represented, N number of channel condition information element is replaced with G cluster, then per cluster bag ContainIndividual channel condition information element.
Step 4：According to characterisitic parameter and step 3, characterisitic parameter channel is calculated first per cluster channel condition information Covariance matrixThen the covariance matrix of every cluster channel condition information is subjected to Eigenvalues Decomposition, i.e. Ck,g=Ψk,g ΛΨk,g H, obtain sparse matrix Ψk,g, and by Ψk,gIt is stored in transmitting terminal and receiving terminal.Specific implementation process is as follows：
(1) covariance matrix of the channel per cluster channel condition information, k-th of user's g cluster are calculated according to characterisitic parameter The covariance matrix of channel condition information is expressed as：
(2) Eigenvalues Decomposition is carried out to the covariance matrix of g cluster channel condition informations, obtains sparse matrix：
Wherein,Represent to covariance matrixCarry out Eigenvalues Decomposition, i.e. Ck,g=Ψk,gΛΨk,g H。 Λ is diagonal matrix, and its diagonal element is covariance matrixCharacteristic value, Ψk,gColumn vector be covariance matrix's Characteristic vector, is converted, Ψ by Carlow southk,gSparse matrix is represented simultaneously.
Due to the channel vector h in step 4kIt is that by characteristic of channel gain of parameter, then need to only determine that channel status is believed Cease sub-clustering number G, then the covariance matrix of transmitting-receiving two-end is exactly known, and then the sparse matrix Ψ of transmitting-receiving two-endk,gIt is also Know.By sparse matrix Ψk,gTransmitting terminal and receiving terminal are stored in, then receiving terminal is not required to again to base station feedback Ψk,g, therefore reduce Feedback quantity.
Step 5：Channel estimation.
The spatial domain correlated channels of all subcarriers of k-th of user are estimated first, that is, obtain 1 × NtHk(n), n ∈ {1,…,Nc}；Then according to formula (4) by Hk(n) vector quantization is changed into Nt× 1 spatial domain channel vector hk(n)；Further according to public affairs Formula (3) construction channel vector hk.Current invention assumes that using preferable channel estimation, not considering channel estimation errors.
Step 6：Channel it is sparse.
The channel vector h for first being constructed channel estimation according to step 3kCarry out sub-clustering；Then receiving terminal g clusters channel shape State information hk,gPass through sparse matrix Ψk,gObtain a sparse vector sk,g, sk,gBy extracting hk,gMain component represent hk,g, it is expressed as
Step 7：Predefined selection matrix code book and selection optimal selection matrix.
The all known binary system selection matrix code book C of a transmitting-receiving two-end is predefined firstP, it is then sparse by what is selected Channel vector s 'k,gInterior each element sum maximal criterion chooses optimal selection matrix Pi.Specific implementation process is as follows：
(1) all known binary system selection matrix code book C of a transmitting-receiving two-end is predefinedP：
CP={ P1…PL} (8)
Wherein PiRepresent i-th of code word of selection matrix code book, i.e. selection matrix；L represents code book CPInclude L selection square Battle array.
(2) selection matrix P is passed throughiFrom sparse vector sk,gMiddle selection M most important feedback informations, wherein
s′k,g=Pisk,g (9)
Global search selection matrix code book, as selected condition of sparse channel vector s 'k,gWhen interior each element sum is maximum, receive End determines the selection matrix P usedi。
I=argimax|sk,gPi| (10)
Wherein PiRepresent i-th of code word of selection matrix code book.Traditional principal component analytical method needs dynamic to update per cluster Not selection matrix P in the same timeiAnd transmitting terminal is fed back to, add computation complexity.It is of the invention to receive and dispatch two in advance One binary system selection matrix code book C of end settingP, need to only feed back selection matrix indexes base station, reduces computation complexity And feedback overhead.
Step 8：Determine selection matrix PiAfterwards, by formula (9) from sparse vector sk,gM main component of middle extraction, that is, lead Want feedback information.Step 8 becomes the channel condition information of higher-dimension the channel condition information of low-dimensional.
Step 9：Code book is quantified using random vector, quantifies the selected condition of sparse channel vector s ' of low-dimensionalk,g, then will s′k,gCodewords indexes and step 7 selection matrix PiCodewords indexes feed back to transmitting terminal together.s′k,gCodewords indexes by with Lower formula is obtained：
J=argjmax|s′k,gwj| (11)
Wherein wjIt is j-th of code word that random vector quantifies code book.Compared with directly quantifying the channel condition information of higher-dimension, Random vector quantization code originally can be designed to smaller, reduce feedback overhead.
Step 10：Transmitting terminal reconstructs channel condition information.
Transmitting terminal is by feeding back the s ' obtainedk,gCodewords indexes j and selection matrix PiCodewords indexes i carries out channel status Signal reconstruct, specific implementation process is as follows：
(1) transmitting terminal finds correspondence g clusters s ' from random vector quantization code bookk,gCode word, from selection matrix code book CP ={ P1…PLIn find correspondence g cluster selection matrixs PiCode word, according to below equation：
Recover the higher-dimension channel condition information vector of g clusters
(2) transmitting terminal passes through selection matrix PiWith sparse matrix Ψk,gMake inverse transformation, according to below equation：
Recover the channel condition information of g clusters
(3) transmitting terminal is right successively according to sub-clustering sequence number gRearranged, reconstruct downlink channel status letter Cease hk。
Step 11：Terminate.
It is of the invention with traditional PCA be used for channel information compression feedback method compared with, its innovation is to set up channel mould The characteristic of channel in two domains in spatial domain and frequency domain is considered during type, and obtained channel vector is subjected to sub-clustering, then every cluster is believed Channel state information is based on the conversion of Carlow south and obtains sparse vector, and the main component of sparse vector is extracted simultaneously finally by selection matrix The selection matrix of static state is predefined in feedback, the present invention and is fed back in code book form, reduces feedback quantity.
In the description of this specification, reference term " one embodiment ", " some embodiments ", " example ", " specifically show The description of example " or " some examples " etc. means to combine specific features, structure, material or the spy that the embodiment or example are described Point is contained at least one embodiment of the present invention or example.In this manual, to the schematic representation of above-mentioned term not Necessarily refer to identical embodiment or example.Moreover, specific features, structure, material or the feature of description can be any One or more embodiments or example in combine in an appropriate manner.
Although an embodiment of the present invention has been shown and described, it will be understood by those skilled in the art that：Not In the case of departing from the principle and objective of the present invention a variety of change, modification, replacement and modification can be carried out to these embodiments, this The scope of invention is limited by claim and its equivalent.

Claims (4)

1. a kind of extensive MIMO-OFDM system channels feedback method based on PCA evolution, it is characterised in that including following step Suddenly：
S1, obtains down channel relevant featuring parameters, sets up channel model；
S2, by channel vector hkSub-clustering is carried out, the covariance square of every cluster channel condition information is calculated according to characteristic of channel parameter Battle array, and sparse matrix is obtained by the Eigenvalues Decomposition of covariance matrix；
S3, predefines the known binary system selection matrix code book of a transmitting-receiving two-end, according to selected condition of sparse channel vector respectively Element sum maximum principle chooses optimal selection matrix.
2. the extensive MIMO-OFDM system channels feedback method according to claim 1 based on PCA evolution, its feature It is, the S1 includes：
Consider an extensive MIMO-OFDM system, transmitting terminal is configuration NtThe linear array of root uniform antenna, receiving terminal is single Antenna user, it is considered to which frequency domain has NcIndividual subcarrier；In extensive MIMO-OFDM systems, in sky between transmitting terminal and k-th of user Domain and the channel vector h of frequency domainkIt is expressed as：
Wherein N=Nt×Nc, vec (A) is represented a × b matrix A vector quantization, becomes the column vector of ab × 1, hk(n) it is The spatial domain channel vector of k-th of user, n-th of subcarrier, setting up model is：
In extensive mimo system, due to transmitting terminal configuration antenna number it is more when, antenna spacing is smaller, is presented stronger between antenna Spatial coherence, therefore, the channel between transmitting terminal and k-th of single-antenna subscriber can be modeled as：
Wherein,Represent to send correlation matrix, n represents n-th of subcarrier, n ∈ { 1 ..., Nc}； It is 0 to represent average, and variance is independently distributed multiple Gauss random vector, H for 1iid,k(n) (u, v) individual element represents transmitting day The channel gain being made up of between line u and reception antenna v path loss and multipath fading；
Because transmitting terminal configures NtRoot uniform linear array antenna, spatial correlation matrix can be obtained by Jakes models, pth root Coefficient correlation between antenna and q root antennas is expressed as：
Wherein J0() represents first kind zero Bessel function, dpqRepresent the distance between pth root antenna and q root antennas, λ Represent carrier wavelength.
3. the extensive MIMO-OFDM system channels feedback method according to claim 1 based on PCA evolution, its feature It is, the S2 includes：
S2-1, by channel vector hkCarry out sub-clustering；
Continuous channel condition information is divided into G cluster, i.e. h by receiving terminalk T=[hk,1 T…hk,G T], wherein G≤N；hk,gRepresent g The channel condition information of individual cluster, instead of N number of channel condition information element, is then included with G cluster per clusterIndividual channel status Information element；
S2-2, calculates covariance matrix of the channel per cluster channel condition information, and pass through covariance square according to characterisitic parameter The Eigenvalues Decomposition of battle array obtains sparse matrix；The channel covariance matrices of k-th of user's g cluster are expressed as：
Eigenvalues Decomposition is carried out to the covariance matrix of g cluster channel condition informations, sparse matrix is obtained：
Wherein,Represent hk,gCovariance matrix,Represent to covariance matrixCarry out Eigenvalues Decomposition, That is Ck,g=Ψk,gΛΨk,g H；Λ is diagonal matrix, and its diagonal element is covariance matrixCharacteristic value, Ψk,gRow to Amount is covariance matrixCharacteristic vector；Converted by Carlow south, Ψk,gSparse matrix is represented simultaneously.
4. the extensive MIMO-OFDM system channels feedback method according to claim 1 based on PCA evolution, its feature It is, the S3 includes：
S3-1, predefines the known binary system selection matrix code book C of a transmitting-receiving two-endP：
CP={ P1…PL}
Wherein PiI-th of code word of selection matrix code book, i.e. selection matrix are represented, L represents code book CPInclude L selection matrix；
S3-2, chooses optimal selection matrix Pi；
Pass through selection matrix PiFrom sparse vector sk,gMiddle selection M most important feedback informations, wherein
s'k,g=Pisk,g
Global search selection matrix code book, as selected condition of sparse channel vector s'k,gWhen each element sum is maximum, receiving terminal is determined The selection matrix P usedi；
I=argi max|sk,gPi|
Determine selection matrix PiAfterwards, s' is passed throughk,g=Pisk,gFrom sparse vector sk,gMiddle selection M most important feedback informations；
S3-3, code book is quantified using random vector, quantifies selected condition of sparse channel vector s'k,g, s'k,gCodewords indexes pass through following Formula is obtained：
J=argj max|s'k,g wj|
Wherein wjIt is j-th of code word that random vector quantifies code book；Then by s'k,gCodewords indexes and selection matrix PiCode word Index feeds back to transmitting terminal together.
CN201710333865.3A 2017-05-12 2017-05-12 Extensive mimo channel feedback method based on PCA evolution Pending CN107171702A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201710333865.3A CN107171702A (en) 2017-05-12 2017-05-12 Extensive mimo channel feedback method based on PCA evolution

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201710333865.3A CN107171702A (en) 2017-05-12 2017-05-12 Extensive mimo channel feedback method based on PCA evolution

Publications (1)

Publication Number Publication Date
CN107171702A true CN107171702A (en) 2017-09-15

Family

ID=59815538

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201710333865.3A Pending CN107171702A (en) 2017-05-12 2017-05-12 Extensive mimo channel feedback method based on PCA evolution

Country Status (1)

Country Link
CN (1) CN107171702A (en)

Cited By (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2020020017A1 (en) * 2018-07-26 2020-01-30 华为技术有限公司 Channel estimation method and apparatus
US10911168B2 (en) 2018-02-02 2021-02-02 Cornell University Channel charting in wireless systems
CN112333120A (en) * 2020-11-09 2021-02-05 电子科技大学 PCA-based channel gain matrix feature extraction method
CN112514334A (en) * 2018-06-01 2021-03-16 弗劳恩霍夫应用研究促进协会 Explicit channel information feedback based on high-order PCA decomposition or PCA synthesis
CN113242063A (en) * 2021-04-29 2021-08-10 江南大学 Large-scale MIMO channel model modeling method based on random coupling
WO2023126007A1 (en) * 2021-12-31 2023-07-06 华为技术有限公司 Channel information transmission method and apparatus

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20070211836A1 (en) * 2006-03-09 2007-09-13 Interdigital Technology Corporation Wireless communication method and apparatus for performing knowledge-based and blind interference cancellation
CN102164107A (en) * 2004-01-12 2011-08-24 英特尔公司 Method and device for signaling information by modifying modulation constellations
CN103532671A (en) * 2013-10-16 2014-01-22 南通大学 MIMO-OFDM (Multiple Input Multiple Output-Orthogonal Frequency Division Multiplexing) system bit distribution method based on delay channel state information
CN103840870A (en) * 2013-12-31 2014-06-04 重庆邮电大学 Method for lowering limiting feedback expenditure of 3D MIMO channel
CN104812061A (en) * 2015-03-24 2015-07-29 成都希盟泰克科技发展有限公司 Indoor range finding and positioning method based on MIMO-OFDM channel state information

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102164107A (en) * 2004-01-12 2011-08-24 英特尔公司 Method and device for signaling information by modifying modulation constellations
US20070211836A1 (en) * 2006-03-09 2007-09-13 Interdigital Technology Corporation Wireless communication method and apparatus for performing knowledge-based and blind interference cancellation
CN103532671A (en) * 2013-10-16 2014-01-22 南通大学 MIMO-OFDM (Multiple Input Multiple Output-Orthogonal Frequency Division Multiplexing) system bit distribution method based on delay channel state information
CN103840870A (en) * 2013-12-31 2014-06-04 重庆邮电大学 Method for lowering limiting feedback expenditure of 3D MIMO channel
CN104812061A (en) * 2015-03-24 2015-07-29 成都希盟泰克科技发展有限公司 Indoor range finding and positioning method based on MIMO-OFDM channel state information

Cited By (11)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US10911168B2 (en) 2018-02-02 2021-02-02 Cornell University Channel charting in wireless systems
CN112514334A (en) * 2018-06-01 2021-03-16 弗劳恩霍夫应用研究促进协会 Explicit channel information feedback based on high-order PCA decomposition or PCA synthesis
US11876589B2 (en) 2018-06-01 2024-01-16 Fraunhofer-Gesellschaft zur Förderung der angewandten Forschung e.V. Explicit channel information feedback based on high-order PCA decomposition or PCA composition
CN112514334B (en) * 2018-06-01 2024-05-03 弗劳恩霍夫应用研究促进协会 Explicit channel information feedback based on higher order PCA decomposition or PCA synthesis
WO2020020017A1 (en) * 2018-07-26 2020-01-30 华为技术有限公司 Channel estimation method and apparatus
US11431384B2 (en) 2018-07-26 2022-08-30 Huawei Technologies Co., Ltd. Channel estimation method and apparatus
US11736157B2 (en) 2018-07-26 2023-08-22 Huawei Technologies Co., Ltd. Channel estimation method and apparatus
CN112333120A (en) * 2020-11-09 2021-02-05 电子科技大学 PCA-based channel gain matrix feature extraction method
CN112333120B (en) * 2020-11-09 2021-08-24 电子科技大学 PCA-based channel gain matrix feature extraction method
CN113242063A (en) * 2021-04-29 2021-08-10 江南大学 Large-scale MIMO channel model modeling method based on random coupling
WO2023126007A1 (en) * 2021-12-31 2023-07-06 华为技术有限公司 Channel information transmission method and apparatus

Similar Documents

Publication Publication Date Title
CN107171702A (en) Extensive mimo channel feedback method based on PCA evolution
KR101317136B1 (en) Method and system for precoding and method for constructing precoding codebook
CN102725967B (en) For the method and apparatus of information feed back and precoding
US8537914B2 (en) Multi-resolution precoding codebook
US8971434B2 (en) Precoding codebook and feedback representation
CN111147112B (en) Energy maximization collection method based on MIMO-NOMA system
US7957701B2 (en) Closed-loop multiple-input-multiple-output scheme for wireless communication based on hierarchical feedback
CN107483091B (en) Channel information feedback algorithm under FDD large-scale MIMO-OFDM system
KR101481391B1 (en) Channel state information feedback method and system thereof
CN107113040A (en) Method and apparatus for precoding channel state information reference signals
CN101136718A (en) Multi-input multi-output space multiplexing precoding method of wireless communication system
KR102273118B1 (en) Apparatus and method for feeding back channel information in wireless communication system
CN106793108B (en) Federated user selection and power distribution optimization method in mimo system downlink
KR102586755B1 (en) Method and apparatus for determining a codebook in non-orthogonal multiple access system
CN101340218A (en) Communication method and apparatus in MIMO system
CN106301496B (en) Spatial modulation system based on day line options and precoding
CN111988073A (en) Design method for semi-dynamic subarray mixed structure of broadband millimeter wave communication system
CN105933042A (en) Novel adaptive finite feedback method based on clustering in LTE system
CN102684837B (en) Dynamic updating method of covariance matrix in cooperative multipoint joint transmission
CN101394256B (en) Pre-coding method and codebook constructing method based on codebook mode
CN106899338B (en) User grouping method based on density in downlink of large-scale MIMO system
CN103368628B (en) Double-current beam forming method based on code books in TD-LTE system
CN102291201A (en) Low-complexity codebook searching method of dual-codebook-oriented structure
CN109302217B (en) Efficient MIMO system transmitting antenna selection method
CN108259072B (en) Method for reducing training sequence overhead for FDD large-scale MIMO downlink system

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
WD01 Invention patent application deemed withdrawn after publication
WD01 Invention patent application deemed withdrawn after publication

Application publication date: 20170915