WO2011038697A1 - 一种获取预编码矩阵指示的方法和装置 - Google Patents

一种获取预编码矩阵指示的方法和装置 Download PDF

Info

Publication number
WO2011038697A1
WO2011038697A1 PCT/CN2010/077545 CN2010077545W WO2011038697A1 WO 2011038697 A1 WO2011038697 A1 WO 2011038697A1 CN 2010077545 W CN2010077545 W CN 2010077545W WO 2011038697 A1 WO2011038697 A1 WO 2011038697A1
Authority
WO
WIPO (PCT)
Prior art keywords
matrix
differential
precoding matrix
codebook
indication
Prior art date
Application number
PCT/CN2010/077545
Other languages
English (en)
French (fr)
Inventor
王建国
周永行
Original Assignee
华为技术有限公司
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 华为技术有限公司 filed Critical 华为技术有限公司
Priority to EP10819926.6A priority Critical patent/EP2485408B1/en
Priority to JP2012531227A priority patent/JP5489029B2/ja
Publication of WO2011038697A1 publication Critical patent/WO2011038697A1/zh
Priority to US13/434,540 priority patent/US8553799B2/en

Links

Classifications

    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04BTRANSMISSION
    • H04B7/00Radio transmission systems, i.e. using radiation field
    • H04B7/02Diversity systems; Multi-antenna system, i.e. transmission or reception using multiple antennas
    • H04B7/04Diversity systems; Multi-antenna system, i.e. transmission or reception using multiple antennas using two or more spaced independent antennas
    • H04B7/06Diversity systems; Multi-antenna system, i.e. transmission or reception using multiple antennas using two or more spaced independent antennas at the transmitting station
    • H04B7/0613Diversity systems; Multi-antenna system, i.e. transmission or reception using multiple antennas using two or more spaced independent antennas at the transmitting station using simultaneous transmission
    • H04B7/0615Diversity systems; Multi-antenna system, i.e. transmission or reception using multiple antennas using two or more spaced independent antennas at the transmitting station using simultaneous transmission of weighted versions of same signal
    • H04B7/0619Diversity systems; Multi-antenna system, i.e. transmission or reception using multiple antennas using two or more spaced independent antennas at the transmitting station using simultaneous transmission of weighted versions of same signal using feedback from receiving side
    • H04B7/0636Feedback format
    • H04B7/0641Differential feedback
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04BTRANSMISSION
    • H04B7/00Radio transmission systems, i.e. using radiation field
    • H04B7/02Diversity systems; Multi-antenna system, i.e. transmission or reception using multiple antennas
    • H04B7/04Diversity systems; Multi-antenna system, i.e. transmission or reception using multiple antennas using two or more spaced independent antennas
    • H04B7/06Diversity systems; Multi-antenna system, i.e. transmission or reception using multiple antennas using two or more spaced independent antennas at the transmitting station
    • H04B7/0613Diversity systems; Multi-antenna system, i.e. transmission or reception using multiple antennas using two or more spaced independent antennas at the transmitting station using simultaneous transmission
    • H04B7/0615Diversity systems; Multi-antenna system, i.e. transmission or reception using multiple antennas using two or more spaced independent antennas at the transmitting station using simultaneous transmission of weighted versions of same signal
    • H04B7/0619Diversity systems; Multi-antenna system, i.e. transmission or reception using multiple antennas using two or more spaced independent antennas at the transmitting station using simultaneous transmission of weighted versions of same signal using feedback from receiving side
    • H04B7/0621Feedback content
    • H04B7/0626Channel coefficients, e.g. channel state information [CSI]
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04BTRANSMISSION
    • H04B7/00Radio transmission systems, i.e. using radiation field
    • H04B7/02Diversity systems; Multi-antenna system, i.e. transmission or reception using multiple antennas
    • H04B7/04Diversity systems; Multi-antenna system, i.e. transmission or reception using multiple antennas using two or more spaced independent antennas
    • H04B7/06Diversity systems; Multi-antenna system, i.e. transmission or reception using multiple antennas using two or more spaced independent antennas at the transmitting station
    • H04B7/0613Diversity systems; Multi-antenna system, i.e. transmission or reception using multiple antennas using two or more spaced independent antennas at the transmitting station using simultaneous transmission
    • H04B7/0615Diversity systems; Multi-antenna system, i.e. transmission or reception using multiple antennas using two or more spaced independent antennas at the transmitting station using simultaneous transmission of weighted versions of same signal
    • H04B7/0619Diversity systems; Multi-antenna system, i.e. transmission or reception using multiple antennas using two or more spaced independent antennas at the transmitting station using simultaneous transmission of weighted versions of same signal using feedback from receiving side
    • H04B7/0636Feedback format
    • H04B7/0639Using selective indices, e.g. of a codebook, e.g. pre-distortion matrix index [PMI] or for beam selection
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04BTRANSMISSION
    • H04B7/00Radio transmission systems, i.e. using radiation field
    • H04B7/02Diversity systems; Multi-antenna system, i.e. transmission or reception using multiple antennas
    • H04B7/04Diversity systems; Multi-antenna system, i.e. transmission or reception using multiple antennas using two or more spaced independent antennas
    • H04B7/06Diversity systems; Multi-antenna system, i.e. transmission or reception using multiple antennas using two or more spaced independent antennas at the transmitting station
    • H04B7/0613Diversity systems; Multi-antenna system, i.e. transmission or reception using multiple antennas using two or more spaced independent antennas at the transmitting station using simultaneous transmission
    • H04B7/0615Diversity systems; Multi-antenna system, i.e. transmission or reception using multiple antennas using two or more spaced independent antennas at the transmitting station using simultaneous transmission of weighted versions of same signal
    • H04B7/0619Diversity systems; Multi-antenna system, i.e. transmission or reception using multiple antennas using two or more spaced independent antennas at the transmitting station using simultaneous transmission of weighted versions of same signal using feedback from receiving side
    • H04B7/0636Feedback format
    • H04B7/0645Variable feedback
    • H04B7/065Variable contents, e.g. long-term or short-short
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04BTRANSMISSION
    • H04B7/00Radio transmission systems, i.e. using radiation field
    • H04B7/02Diversity systems; Multi-antenna system, i.e. transmission or reception using multiple antennas
    • H04B7/04Diversity systems; Multi-antenna system, i.e. transmission or reception using multiple antennas using two or more spaced independent antennas
    • H04B7/06Diversity systems; Multi-antenna system, i.e. transmission or reception using multiple antennas using two or more spaced independent antennas at the transmitting station
    • H04B7/0613Diversity systems; Multi-antenna system, i.e. transmission or reception using multiple antennas using two or more spaced independent antennas at the transmitting station using simultaneous transmission
    • H04B7/0615Diversity systems; Multi-antenna system, i.e. transmission or reception using multiple antennas using two or more spaced independent antennas at the transmitting station using simultaneous transmission of weighted versions of same signal
    • H04B7/0619Diversity systems; Multi-antenna system, i.e. transmission or reception using multiple antennas using two or more spaced independent antennas at the transmitting station using simultaneous transmission of weighted versions of same signal using feedback from receiving side
    • H04B7/0658Feedback reduction
    • H04B7/0663Feedback reduction using vector or matrix manipulations

Definitions

  • the present invention relates to the field of communications technologies, and in particular, to a method and apparatus for obtaining a precoding matrix indication. Background technique
  • a data transmitting end such as an eNode B (evolved Node B)
  • a data transmitting end can be based on a pre-known RI ( Rank). Indicator, rank indication), and PMI (Precoding Matrix Indicator) fed back by the data receiving end (such as UE (User Equipment)) to obtain BF (Beam Forming) matrix/precoding matrix
  • the BF matrix/precoding matrix is used to preprocess the transmitted data, and the preprocessed data is sent to the data receiving end through multiple antennas, so that the data transmission process can adapt to the change of the channel state and improve the performance of the data transmission. Therefore, how to get the performance of PMI and acquired PMI is crucial.
  • the data receiving end obtains and feeds back the PML to the data transmitting end based on the fixed codebook.
  • the existing 3GPP LTE R8 system is mainly designed for SU-MIMO (Single User MIMO), using a fixed codebook, and the PMI of the feedback is low in accuracy. Summary of the invention
  • the embodiment of the present invention provides a method and apparatus for obtaining a precoding matrix indication.
  • the technical solution is as follows:
  • an embodiment of the present invention provides a method for obtaining a precoding matrix indication, where the method includes:
  • the first difference matrix is quantized according to a preset quantization criterion according to the first differential codebook and the pre-known first rank indicator, to obtain a differential precoding matrix indication.
  • an embodiment of the present invention provides an apparatus for obtaining a precoding matrix indication, where the apparatus includes:
  • a first rotation matrix acquisition module configured to obtain a first rotation matrix according to the first channel information
  • a first difference matrix acquisition module configured to: after the first rotation matrix acquisition module obtains the first rotation matrix, according to the first a rotation matrix and a currently obtained instantaneous beamforming matrix/precoding matrix to obtain a first difference matrix
  • a differential precoding matrix indicating acquisition module configured to: after the first differential matrix obtaining module obtains the difference matrix, according to the first differential codebook and the pre-known first rank indication, according to a preset quantization criterion, A difference matrix is quantized to obtain a differential precoding matrix indication.
  • an embodiment of the present invention provides a data sending apparatus, where the apparatus includes: a second rotation matrix acquiring module, configured to obtain a second rotation matrix according to the second channel information; and a second difference matrix acquiring module, And obtaining, according to the received differential precoding matrix indication, a second difference matrix by using a second differential codebook and a pre-known second rank indicator, where the differential precoding matrix indicates that the data receiving apparatus obtains according to the second a rotation matrix and an instantaneous beamforming matrix/precoding matrix, using a first differential codebook and a pre-known first rank indication
  • the first rotation matrix is obtained according to the first channel message, and the second channel information is consistent with the first channel information; the second differential codebook and the second rank indication, Consistent with the first differential codebook and the first rank indication, respectively;
  • a precoding matrix reconstruction module configured to reconstruct the beamforming matrix/precoding according to the second rotation matrix obtained by the second rotation matrix acquisition module and the difference matrix obtained by the second difference matrix acquisition module matrix.
  • an embodiment of the present invention provides a method for beamforming matrix/precoding matrix reconstruction, where the method includes:
  • the differential precoding matrix indicates the first obtained by the data receiving apparatus according to the second differential matrix a rotation matrix and an instantaneous beamforming matrix/precoding matrix, obtained by using a first differential codebook and a pre-known first rank indication, wherein the first rotation matrix is obtained according to a first channel message, and the second The channel information is consistent with the first channel information; the second differential codebook and the second rank indication are respectively consistent with the first differential codebook and the first rank indication;
  • the differential PMI is obtained, and the differential PMI is used for feedback.
  • the existing information in the channel can be utilized, which not only reduces the overhead, but also makes full use of the correlation between the channel in the frequency domain and the time domain, thereby improving the feedback precision.
  • FIG. 1 is a flowchart of a method for obtaining a precoding matrix indication according to Embodiment 1 of the present invention
  • FIG. 2 is a flowchart of a method for obtaining a precoding matrix indication according to Embodiment 2 of the present invention
  • 3 is a flowchart of a method for obtaining a precoding matrix indication according to Embodiment 3 of the present invention
  • FIG. 4 is a schematic structural diagram of an apparatus for acquiring a precoding matrix indication according to Embodiment 7 of the present invention.
  • FIG. 5 is a schematic structural diagram of another apparatus for obtaining a precoding matrix indication according to Embodiment 7 of the present invention.
  • FIG. 6 is a schematic structural diagram of another apparatus for obtaining a precoding matrix indication according to Embodiment 7 of the present invention.
  • FIG. 7 is a schematic structural diagram of a data transmitting apparatus according to Embodiment 8 of the present invention.
  • FIG. 8 is a flowchart of a method for reconstructing a beamforming matrix/precoding matrix according to Embodiment 9 of the present invention. Detailed ways
  • An embodiment of the present invention provides a method for obtaining a precoding matrix indication. Referring to FIG. 1, the method includes:
  • the obtaining the first rotation matrix according to the first channel information may specifically include:
  • the obtaining the first rotation matrix according to the first channel information may specifically include:
  • the reference precoding matrix indicates that the precoding matrix column corresponding to the pre-known first rank indication is orthogonal and full rank, normalize each column of the precoding matrix, and use the normalized precoding matrix as the first Rotation matrix
  • the reference precoding matrix indicates that the precoding matrix column corresponding to the pre-known first rank indication is orthogonal and not full, but the reference precoding matrix indicates that the corresponding codebook satisfies the nesting characteristic, then selecting the reference precoding matrix indicates the corresponding
  • the full rank precoding matrix normalizes the columns of the full rank precoding matrix, and uses the normalized full rank precoding matrix as the first rotation matrix.
  • the obtaining the first rotation matrix according to the first channel information may specifically include:
  • a QR (orthogonal triangulation) decomposition is performed on the normalized precoding matrix to obtain a first rotation matrix.
  • the first differential codebook adopts a differential dual-polarization codebook, and the differential dual-polarization codebook is based on a single-polarization differential code including 2 codewords.
  • the differential dual-polarization codebook includes 2 L+1 codewords, wherein 2 L codewords in the differential dual-polarization codebook are the same as the differential single-polarization codebook, and there are 2 L codes.
  • the word is obtained by multiplying each codeword in the differential single-polarization codebook by a preset diagonal matrix, where N is an even number greater than or equal to 2, and L represents one of the feedback differential single-polarization codebooks. The number of information bits occupied by the code word.
  • the method further includes: A codebook is selected from the power limited differential codebook and the power unrestricted differential codebook as the first differential codebook according to the local power headroom and the preset power headroom threshold.
  • the method further includes: feeding back the differential precoding matrix indication to the data transmitting end, so that the data transmitting end uses the second rotation matrix according to the differential precoding matrix indication and the acquired a second differential codebook and a pre-known second rank indication, reconstructing a beamforming matrix/precoding matrix, wherein the second rotation matrix is obtained by the data transmitting end according to the second channel information, and the second channel information,
  • the two rotation matrix, the second differential codebook, and the second rank indication are respectively consistent with the first channel information, the first rotation matrix, the first differential codebook, and the first rank indication.
  • the data receiving end may feed back the first channel information, the first rotation matrix, the first differential codebook or the first rank indication to the data sending end, and the data sending end respectively uses the received feedback information as the second channel.
  • Information, a second rotation matrix, a second differential codebook, or a second rank indication such that the second channel information, the second rotation matrix, the second differential codebook, and the second rank indication are respectively associated with the first channel information, the first rotation
  • the matrix, the first differential codebook and the first rank indication are consistent; and/or the second channel information, the second rotation matrix, the second differential codebook or the second rank indication may be fed back to the data receiving end by the data transmitting end, the data Receiving, by the receiving end, the received feedback information as the second channel information, the second rotation matrix, the second differential codebook, or the second rank indication, respectively, so that the second channel information, the second rotation matrix, the second differential codebook, and The second rank indication is consistent with the first channel information, the first rotation matrix, the first differential codebook, and the first rank indication, respectively; and
  • the data transmitting end may be configured to: according to the differential precoding matrix indication and the second rotation matrix that is obtained by using the second differential codebook and the second known rank indicator, the beamforming matrix/precoding matrix may include: Obtaining a second rotation matrix according to the second channel information;
  • a beamforming matrix/precoding matrix is reconstructed according to the second difference matrix and the second rotation matrix.
  • the obtaining the second rotation matrix according to the second channel information may specifically include:
  • the plurality of channel matrices obtained on the average are obtained; and the feature matrix of the second long-term channel covariance matrix is used as the second rotation matrix.
  • obtaining the second rotation matrix may specifically include:
  • the reference precoding matrix indication and the pre-known second rank indication corresponding to the precoding matrix column are orthogonal and full rank, normalize each column of the precoding matrix, and use the normalized precoding matrix as the second Rotation matrix
  • the reference precoding matrix indicates that the precoding matrix column corresponding to the pre-known second rank indication is orthogonal and not full, but the reference precoding matrix indicates that the corresponding codebook satisfies the nesting characteristic, then selecting the reference precoding matrix indicates the corresponding
  • the full rank precoding matrix normalizes the columns of the full rank precoding matrix, and uses the normalized full rank precoding matrix as the second rotation matrix.
  • obtaining the second rotation matrix may specifically include:
  • the method further includes: acquiring, by using a differential precoding matrix indication or a non-differential precoding matrix indication, performing feedback cancellation Interest rate
  • the first non-differential codebook adopts a dual-polarization non-differential codebook, and the dual-polarization non-differential codebook is based on a monopole containing 2 codewords.
  • the non-differential codebook is obtained, and the non-differential dual-polarization codebook includes 2 L+1 codewords, wherein 2 L codewords in the non-differential dual-polarization codebook are the same as the non-differential single-polarization codebook.
  • 2 L are obtained, where N is an even number greater than or equal to 2, and L represents the number of information bits occupied when feeding back one codeword in the non-differential single-polarization codebook.
  • the method further includes:
  • a codebook is selected from the power limited non-differential codebook and the power unrestricted non-differential codebook as the first non-differential codebook according to the local power headroom and the preset power headroom threshold.
  • the method for obtaining the indication of the precoding matrix obtains the differential PMI according to the channel information, and uses the differential PMI for feedback, which can utilize the existing information in the channel, which can not only reduce the overhead, but also fully utilize the channel in the frequency.
  • the correlation between the domain and the time domain improves the feedback accuracy and can meet the requirements of technologies such as MU-MIMO (Multiple User MIMO) or CoMP (Coordinated Multiple Point Transmission).
  • MU-MIMO Multiple User MIMO
  • CoMP Coordinatd Multiple Point Transmission
  • An embodiment of the present invention provides a method for obtaining a precoding matrix indication, which is based on a long term.
  • the difference PMI of the channel covariance matrix is fed back.
  • the following line ie, the eNode B sends data to the UE, and the UE feeds back the precoding matrix indication to the eNode B
  • the method includes:
  • the eNode B instructs the UE to use the non-differential PMI or the feedback based on the differential PMI of the first long-term channel covariance matrix R UE . If the eNode B instructs the UE to use the non-differential PMI for feedback, perform 208; otherwise, perform 202.
  • the eNode B can indicate to the UE by using high layer signaling or a downlink physical control channel. Moreover, it can be distinguished by different identifiers whether to indicate whether to use a non-differential PMI or a differential PMI based on a first long-term channel covariance matrix, for example: 0 can be used to indicate the use of a non-differential PMI, and 1 is used to indicate the use of the first long-term channel. Difference of covariance matrix
  • PML is not limited to this method, and can be flexibly set according to actual application conditions. Moreover, it can be selected according to the actual data to be transmitted, the channel state, etc., whether feedback is performed using a non-differential PMI, or a differential PMI based on a first long-term channel covariance matrix, or a combination of the two.
  • the first long-term channel covariance matrix R UE is obtained by averaging a plurality of channel matrices acquired by the UE on a single or multiple sub-carriers in a preset time period, and may be preset according to actual application conditions, such as Set to 1 second, 2 seconds, etc.
  • the average (1) is used for averaging
  • the first long-term channel covariance matrix R UE is obtained as follows:
  • 11 represents a channel matrix on different time slots or subcarriers i obtained in a preset time period, and represents a weighting coefficient of ⁇ ; used for moving average and normalization.
  • the first long-term channel covariance matrix R ra is not limited to being obtained by the method of the formula (1).
  • the UE performs eigenvalue decomposition on the first long-term channel covariance matrix R ra to obtain a feature matrix of the first long-term channel covariance matrix R ra , and uses the feature matrix as the first rotation matrix.
  • ⁇ ra represents wherein R ra value to diagonal elements of the diagonal matrix composed, expressed as a first long-term channel covariance matrix of eigenvectors for all columns in a matrix configuration, for ease of description will be referred to as ⁇ ⁇ A feature matrix of a long-term channel covariance matrix R UE .
  • the UE obtains the first difference matrix D according to the first rotation matrix and the currently obtained instantaneous BF matrix/precoding matrix V.
  • the first difference matrix D is obtained according to formula (3):
  • V represents the currently obtained instantaneous BF matrix/precoding matrix.
  • V is an ideal BF matrix/precoding matrix that can reflect the channel state of the UE, which can be obtained according to the specific situation of the UE channel state.
  • V can be obtained by decomposing the singular value of the instantaneous channel matrix of the UE, and can also be obtained by other various methods in the prior art, and details are not described herein again.
  • the UE quantizes the first difference matrix D according to the preset quantization rule according to the first differential codebook and the pre-known first RI, to obtain a differential precoding matrix 0.
  • the first RI may be known in advance according to various manners in the prior art, and details are not described herein.
  • the first difference matrix D is quantized by using the quantization criterion shown in the formula (4) to obtain:
  • the first differential codebook herein may select to use any one of the prior art differential codebooks according to actual application conditions.
  • the quantization criterion of the first difference matrix D is not limited to the criterion shown in the formula (4), and may be based on actual application conditions. Choose other criteria, such as maximum capacity criteria and various distance-based minimization guidelines.
  • the UE feeds back the differential PMI dlf to the eNode B.
  • the eNode B performs eigenvalue decomposition on the second long-term channel covariance matrix to obtain the feature matrix 11 of the second long-term channel covariance matrix R, and uses the feature matrix Ua3 ⁇ 4 as the second rotation matrix; the eNode B receives the differential PMI dlf , and according to The differential PMI dlf uses the second differential codebook and the second known RI to obtain the second difference matrix ⁇ .
  • the second differential codebook and the second RI used in the step are consistent with the first differential codebook and the first RI used in step 204.
  • the second long-term channel covariance matrix R is obtained by averaging a plurality of channel matrices acquired by the eNode B on a single or multiple sub-carriers in a preset time period.
  • the same method as the first long-term channel covariance matrix R UE is used to ensure that the R acquired by the UE and the eNode B can be substantially consistent, for example:
  • the UE obtains by using the formula (1).
  • the eNode B obtains the R by using the formula (1) to make the two are consistent through other signaling interactions.
  • denotes a diagonal matrix composed of diagonal elements of the eigenvalue of R, and a matrix consisting of all the eigenvectors of the second long-term channel covariance matrix R as a column, and the U-book is referred to for convenience of description. It is the feature matrix of the second long-term channel covariance matrix R book.
  • the eNode B reconstructs the step according to the second rotation matrix and the second difference matrix ⁇
  • the BF matrix/precoding matrix V is reconstructed by the equation (6):
  • the UE obtains the non-differential PMI by using the first non-differential codebook and the pre-known third RI according to the currently obtained instantaneous BF matrix/precoding matrix, and feeds the non-differential PMI to the eNode B.
  • the UE uses the first non-differential codebook and the pre-known third RI to quantize according to the currently obtained instantaneous BF matrix/precoding matrix by using a maximum capacity criterion and various distance-based minimization criteria.
  • Differential PMI Differential PMI
  • the non-differential PMI is obtained by using the method in the 3GPP LTE R8 system.
  • a non-differential codebook used is as shown in Table 1, and a non-differential can be selected from Table 1 according to actual application conditions. PMI for feedback.
  • I is a unit matrix of 4 x 4, given by Table 1 above .
  • the eNode B receives and according to the non-differential PMI, obtains the BF matrix/precoding matrix V obtained in step 208 by using the second non-differential codebook and the pre-known fourth RI.
  • the second non-differential codebook and the fourth RI are consistent with the first non-differential codebook and the third RI, respectively.
  • the method corresponding to step 208 is used to obtain the BF matrix/precoding matrix V, which is not described here.
  • the eNode B uses the obtained BF matrix/precoding matrix to preprocess the data to be sent, and sends the preprocessed data to be sent to the UE through the transmitting antenna.
  • the UE receives the received signal and performs data detection on the received signal y.
  • the received signal y received by the UE is as shown in the formula (7):
  • y represents the received signal received by the UE
  • H represents the channel matrix
  • s represents the data to be transmitted
  • n represents additive white Gaussian noise.
  • the PMI when the PMI is fed back, whether the feedback is a differential PMI or a non-differential PMI, the PMI may be fed back to the entire system bandwidth, or the system may be divided into multiple BPs (Bandwidth Part). Each BP contains multiple sub-bands, and one PMI is fed back for each sub-band.
  • the method in this embodiment when the method in this embodiment is applied to the uplink (that is, the UE sends data to the eNode B and the eNode B feeds back the precoding matrix to the UE), only the eNode B needs to be in step 201. It is determined that the non-differential PMI or the differential PMI based on the long-term channel covariance matrix is used for feedback, and the UE is notified through the high-layer signaling or the downlink physical control channel, and the UE can learn according to the notification, and other processes are similar to the above steps. No longer - repeat.
  • the method for obtaining the indication of the precoding matrix obtains the differential PMI according to the long-term channel covariance matrix, and uses the differential PMI for feedback, which can utilize the existing information in the channel, which can not only reduce the overhead but also fully utilize the channel.
  • the correlation in the frequency domain and the time domain, thereby improving the feedback accuracy can meet the requirements of technologies such as MU-MIMO or CoMP; And you can choose to use non-differential feedback or differential feedback to improve feedback flexibility.
  • An embodiment of the present invention provides a method for obtaining a precoding matrix indication, which uses a differential PMI based on a reference PMI (represented by PMI ref ) to perform feedback.
  • the following lines ie, the UE sends data to the eNode B, and the eNode B feeds back the PMI to the UE.
  • the method includes:
  • the eNode B determines that the non-differential PMI or the reference PMI ref based differential PMI will be used for feedback and notifies the UE that if the eNode B informs the UE that the non-differential PMI will be used, then 308 is performed; otherwise, 302 is performed.
  • the reference may be: the eNode B recently indicates the non-differential wideband PMI or the non-differential neighboring subband PMI used by the UE as the reference PMI ref ; or in step 208 of Embodiment 2, based on the long-term channel covariance matrix
  • the feature matrix uses the non-differential PMI quantized by the non-differential codebook as the reference PMI ref .
  • the method of obtaining the long-term channel covariance matrix is the same in the eNodeB and the UE end, and the equation (1) in the second embodiment can be used. .
  • the eNode B obtains the first rotation matrix according to the reference PMI ref .
  • obtaining the first rotation matrix according to the reference PMI ref may be:
  • normalizing the precoding matrix means multiplying each column of the precoding matrix by a constant to ensure that the modulus of each column vector in the precoding matrix is 1.
  • the codebook used is as shown in Table 1.
  • the pre-known first rank indicates that the corresponding precoding matrix column is orthogonal and less than the rank, but the codebook corresponding to the reference PMI ref satisfies the nesting characteristic, and then selects the full rank precoding matrix corresponding to the PMI ref , and the full rank pre
  • the columns of the coding matrix are normalized, and the normalized full rank precoding matrix is used as the first rotation matrix.
  • each column in the low rank codeword (normalization of each column) in the codebook is a subset of each column in the high rank codeword (normalization of each column).
  • obtaining the first rotation matrix according to the reference PMI ref may also be:
  • the reference PMI ref is normalized to each column of the precoding matrix corresponding to the first known first rank indication, and the normalized precoding matrix is QR decomposed to obtain a first rotation matrix ( ⁇ .
  • the first rotation matrix Q NB can be obtained as:
  • P is a common matrix defined by the eNode B and the UE.
  • the QR decomposition in equation (10) can be obtained based on the Householdr transformation, or the Givens rotation, or the Gram-Schimdt positive exchange method, or the above method. Deformation or combination, but eNodeB and UE must It is necessary to agree to use the same QR decomposition method.
  • obtaining the first rotation matrix according to the reference PMI ref may be:
  • a plurality of different rotation matrices obtained by multiplying the reference PMI ref at different times are obtained to obtain a differential PMI.
  • the method shown in the formula (11) can be used to obtain the first rotation matrix Q NB as:
  • Q intestine and QQ can be obtained by the same method as obtaining the first rotation matrix, and Q is the current moment (or subband) rotation matrix obtained according to the reference PMI ref of the current time (or subband), Q 2 .
  • the Q amount represents a rotation matrix obtained by the QR decomposition or the nesting property of the codebook based on the differential PMI of each time (or subband) before the current time (or subband).
  • the eNode B obtains the first difference matrix D according to the first rotation matrix Q and the currently obtained instantaneous BF matrix/precoding matrix V. Similar to step 203, it will not be described here.
  • the eNode B quantizes the first difference matrix D according to a preset quantization criterion according to the second differential codebook and the second known RI, to obtain a difference PMI 0. Similar to step 204, it will not be described here.
  • the eNode B notifies the UE of the differential PMI 0.
  • UE obtained according to the reference PMI ref second rotation matrix Q UE; UE receives the differential PMI D, and the differential PMI D, using the second differential codebook and the second RI known in advance to obtain a second difference matrix ⁇ .
  • the eNode B and the UE are agreed to use the same reference PMI ref in advance, and the second rotation matrix ( ⁇ £ is obtained according to the reference PMI ref in the same manner as the first rotation matrix obtained in step 302.
  • the UE reconstructs the BF matrix/precoding matrix according to the second difference matrix ⁇ and the second rotation matrix Q RA , and then performs 310. Similar to step 207, it will not be described here.
  • step 208 Similar to step 208, it will not be described here.
  • the UE receives and according to the second non-differential PMI, obtains a BF matrix/precoding matrix by using the second non-differential codebook and the pre-obtained fourth RI. Similar to step 209, it will not be described here.
  • the UE uses the obtained BF matrix/precoding matrix to preprocess the data to be sent s, and sends the preprocessed data to be sent s to the eNode B through the transmitting antenna. Similar to step 210, it will not be described here.
  • the eNode B receives the received signal y and uses y for data detection. Similar to step 211, it will not be described here.
  • the feedback PMI whether the feedback is a differential PMI or a non-differential PMI, it may be a PMI feedback to the entire system bandwidth, or may be divided into multiple BPs, each BP contains more Sub-bands, feeding back a PMI for each sub-band.
  • the eNode B when the method in the embodiment is applied to the downlink (that is, the eNode B sends data to the UE, and the UE feeds back the precoding matrix indication to the eNode B), it only needs to be in step 301.
  • the eNode B instructs the UE to use the non-differential PMI or the reference PMI-based differential PMI for feedback.
  • the other processes are similar to the above steps, and are not repeated here.
  • the method for obtaining the indication of the precoding matrix obtains the differential PMI according to the reference PMI, and uses the differential PMI for feedback, which can utilize the existing information in the channel, and can fully utilize the correlation between the channel in the frequency domain and the time domain.
  • Sexuality thus improving feedback accuracy, can meet the requirements of MU-MIMO or CoMP technology; and can choose to use non-differential feedback or differential feedback, which can improve the flexibility of feedback.
  • Embodiments of the present invention provide a method for obtaining a precoding matrix indication, and Embodiment 2
  • the method of the embodiment 3 is different in that the embodiment of the present invention can obtain the current differential PMI based on the long-term channel covariance matrix and the previous differential PMI, specifically:
  • Equation (12) For the UE, the method shown in Equation (12) may be adopted, based on the long-term channel covariance matrix and the reference PMI ref , to obtain the rotation matrix Q UEn as:
  • t ⁇ can be obtained the same manner as in Example 2 to obtain a first rotation matrix £
  • reference may be referred to as a first rotation matrix
  • Q ⁇ .Q ⁇ Q 3 can be obtained in the first embodiment employs rotation
  • the matrix Q book is obtained by the same method, and Q ⁇ represents the current time (or sub-band) rotation matrix obtained from the reference PMI ref of the current time (or sub-band), Q UE1 ⁇ ⁇ ⁇ Q UEschreib indicates the differential PMI after Q is obtained Rotation matrix.
  • the eNode B instructs the UE to use a non-differential PMI or a long-term channel covariance matrix and a reference PMI ref .
  • the differential PMI can be used, and other processes are similar to those in Embodiment 2, and will not be further described herein.
  • the eNode B determines whether the eNode B will use a non-differential PMI or based on a long-term channel.
  • the covariance matrix and the differential PMI of the reference PMI ref are fed back, and the UE is notified, and the other processes are similar to the embodiment 3, and are not described here.
  • the method for obtaining the indication of the precoding matrix obtains the differential PMI based on the long-term channel covariance matrix and the reference PMI, and uses the differential PMI for feedback, which can utilize the existing information in the channel, thereby not only reducing the overhead, but also
  • the channel can be fully correlated in the frequency domain and the time domain to improve the feedback accuracy, which can meet the requirements of technologies such as MU-MIMO or CoMP; and can choose non-differential feedback or differential feedback to improve feedback flexibility. .
  • the embodiment of the present invention provides a method for obtaining a precoding matrix indication, and the embodiments 2, 3 and The difference between the method described in FIG. 4 is that the codebook (differential or non-differential codebook) in the embodiment of the present invention does not use the codebook in the prior art, but is configured according to the prior art single-polarized transmit antenna.
  • the corresponding codebook (which may be referred to as a single-polarization codebook) acquires a codebook corresponding to the configuration of the dual-polarized transmit antenna (which may be referred to as a dual-polarization codebook).
  • the first N/2 (1, 2, 3...N/2) antennas of the N dual-polarized transmit antenna configuration are a set of co-polarized antennas, followed by N/2 (N/2+l, N/2) +2...
  • diag ..., 1,-1,...,-1 ⁇ means that the front ⁇ /2 diagonal elements are 1, and the N/2 diagonal elements are diagonal matrices of -1.
  • the corresponding row exchange in (15) is sufficient.
  • the N-1th antenna is a set of co-polarized antennas
  • the N antennas are another set of co-polarized antennas. In this case, only the diagonal elements of the diagonal matrix are 1, -1, 1, -1, 1, -1...1, -1.
  • the method for obtaining the indication of the precoding matrix obtains the differential PMI according to the channel information, and uses the differential PMI for feedback, which can utilize the existing information in the channel, which can not only reduce the overhead, but also fully utilize the channel in the frequency. Correlation between domain and time domain, thus improving feedback accuracy, can meet the requirements of technologies such as MU-MIMO or CoMP; and can choose to use non-differential PMI feedback or differential PMI feedback to improve feedback flexibility;
  • the dual-polarized transmit antenna configuration features improved codebook performance in a dual-polarized transmit antenna configuration.
  • an embodiment of the present invention provides a precoding matrix.
  • the method of the indication differs from the method described in the embodiments 2, 3, 4 and 5 in that, in the embodiment of the present invention, the codebook may be selected according to the PH (Power Headroom) of the UE, specifically including the following Two ways:
  • the first type The UE according to its own PH and preset power headroom threshold 3 ⁇ 4 ⁇ . ⁇ , judging whether its own chirp satisfies the formula (16), and if the UE's own chirp satisfies the formula (16), the UE uses the CMF codebook; otherwise, the UE uses the CMP codebook.
  • the eNode B determines whether the PH of the UE satisfies the formula (16) according to the PH of the UE and the preset power headroom threshold h hreshold . If the formula (16) is satisfied, the CMF codebook is selected; otherwise, Use a CMP codebook. It should be noted that, in this manner, the default UE and eNode B also know the preset power headroom threshold in advance. 1 (1 information. Also, in the prior art, the UE and the eNode B know the PH information of the UE.
  • the second type the eNode B determines whether the PH of the UE satisfies the formula (16) according to the PH of the UE and the preset power headroom threshold h hreshold . If the formula (16) is satisfied, the CMF codebook is selected. The UE is notified to use the CMF codebook; otherwise, the CMP codebook is used, and the UE is notified to use the CMF codebook.
  • the method for obtaining the indication of the precoding matrix obtains the differential PMI according to the channel information, and uses the differential PMI for feedback, which can utilize the existing information in the channel, which can not only reduce the overhead, but also fully utilize the channel in the frequency. Correlation between domain and time domain, thus improving feedback accuracy, can meet the requirements of technologies such as MU-MIMO or CoMP; and can choose to use non-differential PMI feedback or differential PMI feedback to improve feedback flexibility; According to the power headroom, the CMP codebook and the CMF codebook can be adaptively switched, and the respective advantages of the CMP codebook and the CMF codebook can be exerted according to the current power usage of the UE or the eNodeB.
  • Example 7 An embodiment of the present invention provides an apparatus for obtaining a precoding matrix indication. Referring to FIG. 4, the apparatus includes:
  • a first rotation matrix acquiring module 401 configured to obtain a first rotation matrix according to the first channel information
  • the first difference matrix obtaining module 402 is configured to obtain a first difference matrix according to the first rotation matrix and the currently obtained instantaneous beamforming matrix/precoding matrix after the first rotation matrix obtaining module 401 obtains the first rotation matrix;
  • the differential precoding matrix indication obtaining module 403 is configured to: after the first difference matrix obtaining module 402 obtains the difference matrix, according to the first differential codebook and the pre-known first rank indication, according to a preset quantization criterion, the first difference The matrix is quantized to obtain a differential precoding matrix indication.
  • the first rotation matrix acquiring module 401 may specifically include:
  • a feature matrix acquiring unit configured to perform eigenvalue decomposition on the first long-term channel covariance matrix, to obtain a feature matrix of the first long-term channel covariance matrix, wherein the first long-term channel covariance matrix is locally localized at a preset time
  • a plurality of channel matrices acquired on a single or multiple subcarriers in the segment are averaged;
  • the first rotation matrix acquiring unit is configured to use, after the feature matrix acquiring unit obtains the feature matrix, the feature matrix of the first long-term channel covariance matrix as the first rotation matrix.
  • the first rotation matrix obtaining module 401 may specifically include: when the first channel information is the reference precoding matrix indication,
  • a second rotation matrix acquiring unit configured to normalize each column of the precoding matrix when the reference precoding matrix indication and the pre-known first rank indication corresponding to the precoding matrix column are orthogonal and full rank a precoding matrix as a first rotation matrix;
  • a third rotation matrix acquiring unit configured to: when the reference precoding matrix indication and the pre-known first rank indication corresponding to the precoding matrix column are orthogonal and not full, but the reference precoding matrix indicates that the corresponding codebook satisfies the nesting characteristic , selecting a reference precoding matrix to indicate a corresponding full rank precoding moment Array, normalize each column of the full rank precoding matrix, and use the normalized full rank precoding matrix as the first rotation matrix.
  • the first rotation matrix obtaining module 401 may specifically include: when the first channel information is the reference precoding matrix indication,
  • a fourth rotation matrix acquiring unit configured to normalize each column of the precoding matrix corresponding to the reference precoding matrix indication and the pre-known first rank indication, and perform QR decomposition on the normalized precoding matrix to obtain the first A rotation matrix.
  • the apparatus may further include:
  • the selecting module 404 is configured to: when the power limited differential codebook and the power unrestricted differential codebook exist locally, according to the local power headroom and the preset power headroom threshold, the power limited differential codebook and A codebook is selected in the power unrestricted differential codebook as the first differential codebook.
  • the device may further include:
  • the feedback module 405 is configured to: after the differential precoding matrix indication obtaining module 403 obtains the differential precoding matrix indication, feed the differential precoding matrix indication to the data sending end, so that the data sending end according to the differential precoding matrix indication and the obtained a second rotation matrix, which is obtained by reconstructing a beamforming matrix/precoding matrix by using a second differential codebook and a second known rank indicator, wherein the second rotation matrix is obtained by the data transmitting end according to the second channel information, and The second channel information, the second rotation matrix, the second differential code, and the second rank indication are respectively consistent with the first channel information, the first rotation matrix, the first differential code, and the first rank indication.
  • the device for obtaining the indication of the precoding matrix obtains the differential PMI according to the channel information, and uses the differential PMI for feedback, which can utilize the existing information in the channel, can not only reduce the overhead, but also fully utilize the channel in the frequency. Correlation between domain and time domain, thus improving feedback accuracy, can meet the requirements of technologies such as MU-MIMO or CoMP.
  • An embodiment of the present invention provides a data sending apparatus.
  • the apparatus includes: a second rotation matrix acquiring module 501, configured to obtain a second rotation matrix according to the second channel information;
  • a second difference matrix obtaining module 502 configured to obtain, according to the received differential precoding matrix indication, a second difference matrix by using a second differential codebook and a pre-known second rank indicator, where the differential precoding matrix indicates the data
  • the receiving device obtains, according to the first rotation matrix and the instantaneous beamforming matrix/precoding matrix, the first differential codebook and the first known first rank indication, wherein the first rotation matrix is obtained according to the first channel message, And the second channel information is consistent with the first channel information; the second differential codebook and the second rank indication are respectively consistent with the first differential codebook and the first rank indication;
  • the precoding matrix reconstruction module 503 is configured to reconstruct a beamforming matrix/precoding matrix according to the difference matrix obtained by the second rotation matrix obtained by the second rotation matrix acquisition module 501 and the second difference matrix acquisition module 502.
  • the obtaining the second rotation matrix according to the second channel information may specifically include:
  • the feature matrix of the second long-term channel covariance matrix is taken as the second rotation matrix.
  • obtaining the second rotation matrix may specifically include:
  • the precoding matrix indicates that the precoding matrix column corresponding to the pre-known second rank indication is orthogonal and full rank, the precoding matrix is normalized, and the normalized precoding matrix is used as the second rotation matrix;
  • the reference precoding matrix indicates that the precoding matrix column corresponding to the pre-known second rank indication is orthogonal and not full, but the reference precoding matrix indicates that the corresponding codebook satisfies the nesting characteristic, then selecting the reference precoding matrix indicates the corresponding
  • the full rank precoding matrix normalizes the columns of the full rank precoding matrix, and uses the normalized full rank precoding matrix as the second rotation matrix.
  • obtaining the second rotation matrix may specifically include:
  • PMI reconstructing the beamforming matrix/precoding matrix
  • PMI can utilize the existing information in the channel, not only can reduce the overhead, but also make full use of the correlation of the channel in the frequency domain and the time domain, thereby improving the reconstructed
  • the accuracy of the beamforming matrix/precoding matrix can meet the requirements of technologies such as MU-MIMO or CoMP.
  • An embodiment of the present invention provides a beamforming matrix/precoding matrix reconstruction method. Referring to FIG. 8, the method includes:
  • the obtaining the second rotation matrix according to the second channel information may specifically include:
  • the feature matrix of the second long-term channel covariance matrix is taken as the second rotation matrix.
  • obtaining the second rotation matrix may specifically include:
  • the reference precoding matrix indication and the pre-known second rank indication corresponding to the precoding matrix column are orthogonal and full rank, normalize each column of the precoding matrix, and use the normalized precoding matrix as the second Rotation matrix If the reference precoding matrix indicates that the precoding matrix column corresponding to the pre-known second rank indication is orthogonal and not full, but the reference precoding matrix indicates that the corresponding codebook satisfies the nesting characteristic, then selecting the reference precoding matrix indicates the corresponding The full rank precoding matrix normalizes the columns of the full rank precoding matrix, and uses the normalized full rank precoding matrix as the second rotation matrix.
  • obtaining the second rotation matrix may specifically include:
  • the differential precoding matrix indicates a first rotation matrix and an instantaneous beamforming matrix/precoding matrix acquired by the data receiving device according to the first differential codebook and the first known first rank indication, wherein the first The rotation matrix is obtained according to the first channel message, and the second channel information is consistent with the first channel information; the second differential codebook and the second rank indication are respectively consistent with the first differential codebook and the first rank indication;
  • the beamforming matrix/precoding matrix reconstruction method reconstructs a beamforming matrix/precoding matrix by using a differential PMI obtained from channel information, and can utilize existing information in the channel, not only
  • the overhead can be reduced, and the correlation between the channel in the frequency domain and the time domain can be fully utilized, thereby improving the accuracy of the reconstructed beamforming matrix/precoding matrix, which can meet the requirements of technologies such as MU-MIMO or CoMP.
  • All or part of the technical solutions provided by the above embodiments may be implemented by software programming, and the software program is stored in a readable storage medium, such as: Hard disk, CD or floppy disk.
  • a readable storage medium such as: Hard disk, CD or floppy disk.

Landscapes

  • Engineering & Computer Science (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Signal Processing (AREA)
  • Physics & Mathematics (AREA)
  • Mathematical Physics (AREA)
  • Radio Transmission System (AREA)
  • Compression, Expansion, Code Conversion, And Decoders (AREA)

Description

一种获取预编码矩阵指示的方法和装置
本申请要求于 2009 年 9 月 30 日提交中国专利局、 申请号为 200910235286.0、发明名称为"一种获取预编码矩阵指示的方法和装置"的中 国专利申请的优先权, 其全部内容通过引用结合在本申请中。 技术领域
本发明涉及通信技术领域, 特别涉及一种获取预编码矩阵指示的方法 和装置。 背景技术
随着通信技术的不断发展, 在 MIMO ( Multiple Input Multiple Output, 多输入多输出)无线系统中, 数据发送端 (如 eNode B ( evolved Node B, 演进型基站))可以根据预先获知的 RI ( Rank Indicator , 秩指示), 及数据 接收端(如 UE( User Equipment, 用户设备 ) )反馈的 PMI( Precoding Matrix Indicator, 预编码矩阵指示), 得到 BF ( Beam Forming , 波束赋形)矩阵 / 预编码矩阵, 然后利用 BF矩阵 /预编码矩阵对待发送数据进行预处理, 通 过多天线将预处理后的数据发送到数据接收端, 这样可以使数据发送过程 自适应信道状态的变化, 改善数据传输的性能。 因此, 如何获取 PMI和获 取的 PMI的性能至关重要。
目前, 3GPP LTE R8 ( 3rd Generation Partnership Project Long Term Evolution Release 8 , 第三代合作伙伴计划长期演进发布版本 8 ) 系统中, 数据接收端基于固定的码本, 得到并向数据发送端反馈 PML
然而, 在实现本发明的过程中, 发明人发现现有技术至少存在以下问 题:
现有 3GPP LTE R8系统, 主要是为 SU-MIMO ( Single User MIMO , 单用户 MIMO )设计, 采用固定码本, 反馈的 PMI的精度低。 发明内容
为了在反馈 PMI时, 提高反馈的精度, 本发明实施例提供了一种获取 预编码矩阵指示的方法和装置。 所述技术方案如下:
一方面, 本发明实施例提供了一种获取预编码矩阵指示的方法, 所述 方法包括:
根据第一信道信息, 得到第一旋转矩阵;
根据所述第一旋转矩阵和当前得到的瞬时波束赋形矩阵 /预编码矩阵, 得到第一差分矩阵;
根据第一差分码本和预先获知的第一秩指示, 按照预设的量化准则, 对所述第一差分矩阵进行量化, 得到差分预编码矩阵指示。
另一方面, 本发明实施例提供了一种获取预编码矩阵指示的装置, 所 述装置包括:
第一旋转矩阵获取模块, 用于根据第一信道信息, 得到第一旋转矩阵; 第一差分矩阵获取模块, 用于在所述第一旋转矩阵获取模块得到第一 旋转矩阵后, 根据所述第一旋转矩阵和当前得到的瞬时波束赋形矩阵 /预编 码矩阵, 得到第一差分矩阵;
差分预编码矩阵指示获取模块, 用于在所述第一差分矩阵获取模块得 到差分矩阵后, 根据第一差分码本和预先获知的第一秩指示, 按照预设的 量化准则, 对所述第一差分矩阵进行量化, 得到差分预编码矩阵指示。
又一方面, 本发明实施例提供了一种数据发送装置, 所述装置包括: 第二旋转矩阵获取模块, 用于根据第二信道信息, 得到第二旋转矩阵; 第二差分矩阵获取模块, 用于根据接收到的差分预编码矩阵指示, 利 用第二差分码本和预先获知的第二秩指示, 得到第二差分矩阵, 其中, 所 述差分预编码矩阵指示由数据接收装置根据其获取的第一旋转矩阵和瞬时 波束赋形矩阵 /预编码矩阵, 利用第一差分码本和预先获知的第一秩指示得 到, 其中, 所述第一旋转矩阵根据第一信道消息得到, 并且, 所述第二信 道信息与所述第一信道信息相一致; 所述第二差分码本和所述第二秩指示, 分别与所述第一差分码本和第一秩指示相一致;
预编码矩阵重构模块, 用于根据所述第二旋转矩阵获取模块得到的第 二旋转矩阵和所述第二差分矩阵获取模块得到的差分矩阵, 重构出所述波 束赋形矩阵 /预编码矩阵。
又一方面, 本发明实施例提供了一种波束赋形矩阵 /预编码矩阵重构的 方法, 所述方法包括:
根据第二信道信息, 得到第二旋转矩阵;
根据接收到的差分预编码矩阵指示, 利用第二差分码本和预先获知的 第二秩指示, 得到第二差分矩阵, 其中, 所述差分预编码矩阵指示由数据 接收装置根据其获取的第一旋转矩阵和瞬时波束赋形矩阵 /预编码矩阵, 利 用第一差分码本和预先获知的第一秩指示得到, 其中, 所述第一旋转矩阵 根据第一信道消息得到, 并且, 所述第二信道信息与所述第一信道信息相 一致; 所述第二差分码本和所述第二秩指示, 分别与所述第一差分码本和 第一秩指示相一致;
根据所述第二旋转矩阵和所述第二差分矩阵, 重构出所述波束赋形矩 阵 /预编码矩阵。
本发明实施例提供的技术方案的有益效果是:
根据信道信息, 得到差分 PMI, 使用差分 PMI进行反馈, 可以利用信 道中已存在的信息, 不仅可以减少开销, 而且可以充分利用信道在频域和 时间域上的相关性, 从而提高反馈精度。 附图说明
图 1是本发明实施例 1提供的一种获取预编码矩阵指示的方法流程图; 图 2是本发明实施例 2提供的一种获取预编码矩阵指示的方法流程图; 图 3是本发明实施例 3提供的一种获取预编码矩阵指示的方法流程图; 图 4是本发明实施例 7提供的一种获取预编码矩阵指示的装置结构示 意图;
图 5是本发明实施例 7提供的另一种获取预编码矩阵指示的装置结构 示意图;
图 6是本发明实施例 7提供的又一种获取预编码矩阵指示的装置结构 示意图;
图 7是本发明实施例 8提供的一种数据发送装置的结构示意图。
图 8是本发明实施例 9提供的一种波束赋形矩阵 /预编码矩阵重构的方 法流程图。 具体实施方式
为使本发明的目的、 技术方案和优点更加清楚, 下面将结合附图对本 发明实施方式作进一步地详细描述。
实施例 1
本发明实施例提供了一种获取预编码矩阵指 示的方法, 参见图 1, 该 方法包括:
101 : 根据第一信道信息, 得到第一旋转矩阵。
102: 根据第一旋转矩阵和当前得到的瞬时波束赋形矩阵 /预编码矩阵, 得到第一差分矩阵。
103: 根据第一差分码本和预先获知的第一秩指示, 按照预设的量化准 则, 对第一差分矩阵进行量化, 得到差分预编码矩阵指示。
其中, 当第一信道信息为第一长期信道协方差矩阵时, 根据第一信道 信息, 得到第一旋转矩阵具体可以包括:
对第一长期信道协方差矩阵进行特征值分解, 得到第一长期信道协方 差矩阵的特征矩阵, 其中, 第一长期信道协方差矩阵由本地对其在预设的 时间段内在单个或多个子载波上获取的多个信道矩阵进行平均得到; 将第一长期信道协方差矩阵的特征矩阵作为第一旋转矩阵。
其中, 当第一信道信息为参考预编码矩阵指示时, 根据第一信道信息, 得到第一旋转矩阵具体可以包括:
如果参考预编码矩阵指示和预先获知的第一秩指示对应的预编码矩阵 列正交且满秩, 则对预编码矩阵的各列进行归一化, 将归一化的预编码矩 阵作为第一旋转矩阵;
如果参考预编码矩阵指示和预先获知的第一秩指示对应的预编码矩阵 列正交且不满秩, 但是参考预编码矩阵指示对应的码本满足嵌套特性, 则 选择参考预编码矩阵指示对应的满秩预编码矩阵, 对满秩预编码矩阵的各 列进行归一化, 将归一化的满秩预编码矩阵作为第一旋转矩阵。
其中, 当第一信道信息为参考预编码矩阵指示时, 根据第一信道信息, 得到第一旋转矩阵具体可以包括:
将参考预编码矩阵指示和预先获知的第一秩指示对应的预编码矩阵的 各列进行归一化;
对归一化的预编码矩阵进行 QR (正交三角化)分解, 得到第一旋转矩 阵。
其中, 当本地所在的系统存在 N个双极化发射天线配置时, 第一差分 码本采用差分双极化码本, 差分双极化码本根据包含有 2 个码字的单极化 差分码本得到, 差分双极化码本中包含有 2L+1个码字, 其中, 差分双极化 码本中有 2L个码字与差分单极化码本相同, 另外有 2L个码字为分别将差 分单极化码本中的每个码字左乘以一个预设的对角矩阵得到, 其中, N 为 大于等于 2的偶数, L表示反馈差分单极化码本中的一个码字时占用的信息 比特数。
进一步地, 当本地存在功率受限差分码本和功率不受限差分码本时, 该方法还包括: 根据本地的功率余量和预设的功率余量门限值, 从功率受限差分码本 和功率不受限差分码本中选择一个码本, 作为第一差分码本。
进一步地, 在得到差分预编码矩阵指示之后, 该方法还包括: 将差分预编码矩阵指示反馈给数据发送端, 使数据发送端根据差分预 编码矩阵指示和其获取的第二旋转矩阵, 利用第二差分码本和预先获知的 第二秩指示, 重构出波束赋形矩阵 /预编码矩阵, 其中, 第二旋转矩阵由数 据发送端根据第二信道信息得到, 并且, 第二信道信息、 第二旋转矩阵、 第二差分码本和第二秩指示, 分别与第一信道信息、 第一旋转矩阵、 第一 差分码本和第一秩指示相一致。 具体的, 可以通过数据接收端向数据发送 端反馈第一信道信息、 第一旋转矩阵、 第一差分码本或第一秩指示, 数据 发送端将接收到的上述反馈的信息分别作为第二信道信息、 第二旋转矩阵、 第二差分码本或第二秩指示, 以使得第二信道信息、 第二旋转矩阵、 第二 差分码本和第二秩指示分别与第一信道信息、 第一旋转矩阵、 第一差分码 本和第一秩指示相一致; 和 /或可以通过数据发送端向数据接收端反馈第二 信道信息、 第二旋转矩阵、 第二差分码本或第二秩指示, 数据接收端将接 收到的上述反馈的信息分别作为第二信道信息、 第二旋转矩阵、 第二差分 码本或第二秩指示, 使得第二信道信息、 第二旋转矩阵、 第二差分码本和 第二秩指示分别与第一信道信息、 第一旋转矩阵、 第一差分码本和第一秩 指示相一致; 和 /或可以通过预先在数据接收端和数据发送端设置第二信道 信息、 第二旋转矩阵、 第二差分码本或第二秩指示, 以及第一信道信息、 第一旋转矩阵、 第一差分码本或第一秩指示, 使得第二信道信息、 第二旋 转矩阵、 第二差分码本和第二秩指示, 分别与第一信道信息、 第一旋转矩 阵、 第一差分码本和第一秩指示相一致。
其中, 数据发送端根据差分预编码矩阵指示和其获取的第二旋转矩阵, 利用第二差分码本和预先获知的第二秩指示, 重构出波束赋形矩阵 /预编码 矩阵具体可以包括: 根据第二信道信息, 得到第二旋转矩阵;
接收差分预编码矩阵指示, 利用第二差分码本和预先获知的第二秩指 示得到第二差分矩阵;
根据第二差分矩阵和第二旋转矩阵,重构出波束赋形矩阵 /预编码矩阵。 其中, 当第二信道信息为第二长期信道协方差矩阵时, 根据第二信道 信息, 得到第二旋转矩阵具体可以包括:
对第二长期信道协方差矩阵进行特征值分解, 得到第二长期信道协方 差矩阵的特征矩阵, 其中, 第二长期信道协方差矩阵由本地对其在预设的 时间段内在单个或者多个子载波上获取的多个信道矩阵进行平均得到; 将第二长期信道协方差矩阵的特征矩阵作为第二旋转矩阵。
其中, 当第二信道信息为参考预编码矩阵指示时, 根据第二信道信息, 得到第二旋转矩阵具体可以包括:
如果参考预编码矩阵指示和预先获知的第二秩指示对应的预编码矩阵 列正交且满秩, 则对预编码矩阵的各列进行归一化, 将归一化的预编码矩 阵作为第二旋转矩阵;
如果参考预编码矩阵指示和预先获知的第二秩指示对应的预编码矩阵 列正交且不满秩, 但是参考预编码矩阵指示对应的码本满足嵌套特性, 则 选择参考预编码矩阵指示对应的满秩预编码矩阵, 对满秩预编码矩阵的各 列进行归一化, 将归一化的满秩预编码矩阵作为第二旋转矩阵。
其中, 当第二信道信息为参考预编码矩阵指示时, 根据第二信道信息, 得到第二旋转矩阵具体可以包括:
将参考预编码矩阵指示和预先获知的第二秩指示对应的预编码矩阵的 各列进行归一化; 对归一化的预编码矩阵进行 QR (正交三角化)分解, 得 到第二旋转矩阵。
进一步地, 根据第一信道信息, 得到第一旋转矩阵之前还包括: 获取使用差分预编码矩阵指示或非差分预编码矩阵指示进行反馈的消 息;
当获取到使用差分预编码矩阵指示进行反馈的消息时, 执行根据第一 信道信息, 得到第一旋转矩阵的步骤;
当获取到使用非差分预编码矩阵指示进行反馈的消息时, 根据当前得 到的瞬时波束赋形矩阵 /预编码矩阵, 利用第一非差分码本和预先获知的第 三秩指示, 得到非差分预编码矩阵指示。
其中, 当本地所在的系统存在 N个双极化发射天线配置时, 第一非差 分码本采用双极化非差分码本, 双极化非差分码本根据包含有 2 个码字的 单极化非差分码本得到, 非差分双极化码本中包含有 2L+1个码字, 其中, 非差分双极化码本中有 2L个码字与非差分单极化码本相同, 另外有 2L个 得到, 其中, N为大于等于 2的偶数, L表示反馈非差分单极化码本中的一 个码字时占用的信息比特数。
进一步地, 当本地存在功率受限非差分码本和功率不受限非差分码本 时, 所述方法还包括:
根据本地的功率余量和预设的功率余量门限值, 从功率受限非差分码 本和功率不受限非差分码本中选择一个码本, 作为第一非差分码本。
本发明实施例所述的获取预编码矩阵指示的方法, 根据信道信息, 得 到差分 PMI, 使用差分 PMI进行反馈, 可以利用信道中已存在的信息, 不 仅可以减少开销, 而且可以充分利用信道在频域和时间域上的相关性, 从 而提高反馈精度, 可以满足 MU-MIMO ( Multiple User MIMO , 多用户 MIMO )或 CoMP ( Coordinated Multiple Point transmission ,协作多点传输 ) 等技术的要求。 实施例 2
本发明实施例提供了一种获取预编码矩阵指示的方法, 使用基于长期 信道协方差矩阵的差分 PMI进行反馈, 下面以下行(即 eNode B向 UE发 送数据, UE向 eNode B反馈预编码矩阵指示 ) 为例进行说明, 参见图 2, 该方法包括:
201 : eNode B指示 UE使用非差分 PMI或是基于第一长期信道协方差 矩阵 RUE的差分 PMI进行反馈, 如果 eNode B指示 UE使用非差分 PMI进 行反馈, 则执行 208; 否则, 执行 202。
具体地, eNode B可以通过高层信令或者下行物理控制信道对 UE进行 指示。 并且, 可以通过不同的标识区分是指示使用非差分 PMI还是使用基 于第一长期信道协方差矩阵 的差分 PMI, 例如: 可以用 0表示指示使用 非差分 PMI, 用 1 表示指示使用基于第一长期信道协方差矩阵 的差分
PML 并不限于此种方式, 可以根据实际应用状况进行灵活设置。 并且, 可 以根据实际要发送的数据、 信道状态等情况来选择, 是使用非差分 PMI进 行反馈, 还是使用基于第一长期信道协方差矩阵 的差分 PMI进行反馈, 或者将二者结合起来使用。
其中, 第一长期信道协方差矩阵 RUE是 UE对其在预设的时间段内在单 个或者多个子载波上获取的多个信道矩阵进行平均得到, 可以根据实际应 用状况设置预设的时间,如设置为 1秒、 2秒等。本发明实施例中采用式( 1 ) 进行平均, 得到第一长期信道协方差矩阵 RUE如下所示:
Figure imgf000011_0001
其中, 11;表示预设的时间段内获取的不同时隙或者子带内的子载波 i 上的信道矩阵, 表示 Η;的加权系数, 用于滑动平均和归一化。 并且需要 说明的是, 并不限于通过式(1 )的方式获取第一长期信道协方差矩阵 Rra
202: UE对第一长期信道协方差矩阵 Rra进行特征值分解, 得到第一长 期信道协方差矩阵 Rra的特征矩阵 , 将特征矩阵 作为第一旋转矩阵。
具体地, 对第一长期信道协方差矩阵 进行特征值分解如式(2 )所
Figure imgf000012_0001
其中, ∑ra表示以 Rra的特征值为对角线元素构成的对角矩阵, 表 示以第一长期信道协方差矩阵 所有的特征向量为列构成的矩阵, 为了便 于描述将 υυΕ称为第一长期信道协方差矩阵 RUE的特征矩阵。
203: UE根据第一旋转矩阵 和当前得到的瞬时 BF矩阵 /预编码矩阵 V , 得到第一差分矩阵 D。
具体地, 根据式(3 )得到第一差分矩阵 D为:
Figure imgf000012_0002
其中, V表示当前得到的瞬时 BF矩阵 /预编码矩阵。 V是可以反映 UE 的信道状态的一个理想的 BF矩阵 /预编码矩阵,可才艮据 UE信道状态的具体 情况得到。例如,可以通过对 UE的瞬时信道矩阵的奇异值进行分解得到 V, 还可通过现有技术的其他各种方法得到, 此处不再赘述。
204: UE根据第一差分码本和预先获知的第一 RI, 按照预设的量化准 则, 对第一差分矩阵 D进行量化, 得到差分预编码矩阵 0。
其中, 第一 RI可以根据现有技术中的各种方式预先获知, 此处不再赘 述。
具体地, 根据第一差分码本和预先获知的第一 RI, 采用式(4 )所示的 量化准则, 对第一差分矩阵 D进行量化得到 £>为:
D = argmax||DJiD!.|| (4) 其中, 0表示差分预编码矩阵,与之对应的差分 ΡΜΙ表示为 PMIdlf; || ||F 表示 Frobenius范数; CD表示 eNodeB和 UE共知的第一差分码本; D;表示 第一差分码本 Crf中的码字。 需要说明的是, 此处的第一差分码本可以根据 实际应用状况, 选择使用现有技术中的任一种差分码本。 此外, 第一差分 矩阵 D的量化准则并不限于式(4 )所示的准则, 还可以根据实际应用状况 选择其他准则, 如最大容量准则以及各种基于距离最小化的准则等。
205: UE将差分 PMIdlf反馈给 eNode B。
206: eNode B对第二长期信道协方差矩阵 进行特征值分解, 得到 第二长期信道协方差矩阵 R 的特征矩阵 11 , 将特征矩阵 Ua¾作为第二旋 转矩阵; eNode B接收差分 PMIdlf, 并根据差分 PMIdlf, 利用第二差分码本 和预先获知的第二 RI, 得到第二差分矩阵 δ。
其中, 该步骤中采用的第二差分码本和第二 RI, 与步骤 204中采用的 第一差分码本和第一 RI相一致。
其中, 第二长期信道协方差矩阵 R 是 eNode B对其在预设的时间段 内在单个或者多个子载波上获取的多个信道矩阵进行平均得到。 在获取第 二长期信道协方差矩阵 RNB时, 采用与获取第一长期信道协方差矩阵 RUE相 同的方法, 以保证 UE获取的 和 eNode B获取的 R 可以保持基本一致, 例如: 如果在步骤 201中, UE采用式( 1 )获取 那么该步骤中, eNode B采用式(1 )获取 R 者通过其它信令交互使得二者相一致。
具体地, 对第二长期信道协方差矩阵 进行特征值分解如式(5 )所 示:
RJVB - Ujyg U (5)
其中, ∑^表示以 R 的特征值为对角线元素构成的对角矩阵, υ溯表示 以第二长期信道协方差矩阵 R 所有的特征向量为列构成的矩阵,为了便于 描述将 U冊称为第二长期信道协方差矩阵 R冊的特征矩阵。
207: eNode B 根据第二旋转矩阵 和第二差分矩阵 δ, 重构出步骤
203得到的 BF矩阵 /预编码矩阵 , 然后执行 210。
具体地, 根据第二旋转矩阵 和第二差分矩阵5, 通过式(6 ) 重构 出 BF矩阵 /预编码矩阵 V为:
= UOTD (6) 208: UE根据当前得到的瞬时 BF矩阵 /预编码矩阵, 利用第一非差分 码本和预先获知的第三 RI,得到非差分 PMI,并将非差分 PMI反馈给 eNode B。
具体地, UE利用第一非差分码本和预先获知的第三 RI,采用最大容量 准则以及各种基于距离最小化的准则,对根据当前得到的瞬时 BF矩阵 /预编 码矩阵进行量化, 得到非差分 PMI,
或者, 采用 3GPP LTE R8系统中的方法得到非差分 PMI,在 3GPP LTE R8系统中, 所使用的一种非差分码本如表 1所示, 可以根据实际应用状况 从表 1中选择一个非差分 PMI来进行反馈。
Figure imgf000014_0001
其中, ff„W表示由 PMI为 n对应的矩阵 „ = / - 2unun Hlun Hun的列集合 w构成 的矩阵, I为 4x 4的单位阵, 由上表 1给出。
或者采用现有技术中的其他方法得到非差分 PMI。 209: eNode B接收并根据非差分 PMI, 利用第二非差分码本和预先获 知的第四 RI, 得到步骤 208得到的 BF矩阵 /预编码矩阵 V。
具体地,第二非差分码本和第四 RI,分别与第一非差分码本和第三 RI 相一致。并且采用与步骤 208中相对应的方法得到 BF矩阵 /预编码矩阵 V即 可, 此处不再赘述。
210: eNode B利用得到的 BF矩阵 /预编码矩阵 ,对待发送数据 s进行 预处理, 并将预处理后的待发送数据 s通过发射天线发送给 UE。
211 : UE接收到接收信号 , 并对接收信号 y进行数据检测。
具体地, UE接收到的接收信号 y如式 (7)所示:
y - HVs +n (7)
其中, y表示 UE接收到的接收信号; H表示信道矩阵; 表示 BF矩 阵 /预编码矩阵; s表示待发送数据; n表示加性高斯白噪声。
需要说明的是, 上述在反馈 PMI时, 无论反馈的是差分 PMI还是非差 分 PMI, 均可以是对整个系统带宽反馈一个 PMI, 也可以是将系统分为多 个 BP(Bandwidth Part, 带宽部分), 每个 BP中含有多个子带 (sub-band), 为 每个子带反馈一个 PMI。
并且需要说明的是, 当将本方面实施例所述的方法, 应用于上行(即 UE向 eNode B发送数据, eNode B向 UE反馈预编码矩阵 ) 时, 只需在步 骤 201中由 eNode B 自行确定将使用非差分 PMI或是基于长期信道协方差 矩阵的差分 PMI 进行反馈, 并通过高层信令或者下行物理控制信道通知 UE, UE根据该通知得知即可, 其他过程与上述步骤类似, 此处不再—— 赘述。
本发明实施例所述的获取预编码矩阵指示的方法, 根据长期信道协方 差矩阵得到差分 PMI, 使用差分 PMI进行反馈, 可以利用信道中已存在的 信息, 不仅可以减少开销, 而且可以充分利用信道在频域和时间域上的相 关性, 从而提高反馈精度, 可以满足 MU-MIMO或 CoMP等技术的要求; 并且可以选择使用非差分反馈或是差分反馈, 可以提高反馈的灵活性。 实施例 3
本发明实施例提供了一种获取预编码矩阵指示的方法, 使用基于参考 PMI (用 PMIref表示)的差分 PMI进行反馈, 下面以上行(即 UE向 eNode B发送数据, eNode B向 UE反馈 PMI )为例进行说明, 参见图 3, 该方法 包括:
301 : eNode B 确定将使用非差分 PMI或是基于参考 PMIref的差分 PMI 进行反馈, 并通知 UE, 如果 eNode B通知 UE将使用非差分 PMI, 则执行 308; 否则, 执行 302。
其中参考 ^11^可以是: 将 eNode B最近指示 UE使用的非差分宽带 PMI或非差分相邻子带 PMI作为参考 PMIref; 或是将实施例 2步骤 208中, 基于长期信道协方差矩阵的特征矩阵, 利用非差分码本量化得到的非差分 PMI作为参考 PMIref。 当使用基于长期信道协方差矩阵得到的非差分 PMI, 作为参考 PMIref时, 在 eNodeB和 UE端各自得到长期信道协方差矩阵的方 法相同, 如均可以采用实施例 2中的式(1 )等。
302: eNode B根据参考 PMIref得到第一旋转矩阵 。
具体地, 根据参考 PMIref得到第一旋转矩阵 可以是:
1 )如果参考 PMIre^。预先获知的第一秩指示对应的预编码矩阵列正交 且满秩, 则可以对该预编码矩阵的各列进行归一化, 将归一化的预编码矩 阵作为第一旋转矩阵 Q冊 。
其中, 对预编码矩阵归一化是指将预编码矩阵的各列分别乘以某个常 数以保证预编码矩阵中的每个列矢量的模为 1。
例如: 参考 ^11^为 3, 预先获知的第一秩指示为 4, 则采用的码本如 表 1所示, 表 1中与 PMI=3, RI=1对应的满秩预编码矩阵为^ 214}/2, 则对
W3 {m4) /2进行归一化得到第一旋转矩阵 QNB如式 (8)所示: Q 3,/2]*2 = 3, (8)
2 )如果参考 PMIre^。预先获知的第一秩指示对应的预编码矩阵列正交 且不满秩, 但是参考 PMIref对应的码本满足嵌套特性, 则选择参考 PMIref 对应的满秩预编码矩阵, 对该满秩预编码矩阵的各列进行归一化, 将归一 化的满秩预编码矩阵作为第一旋转矩阵 。
其中, 码本满足嵌套特性, 是该码本中低秩码字 (各列归一化) 中的 各列是高秩码字 (各列归一化) 中各列的子集。
例如: 参考 ^11^为 6, 预先获知的第一秩指示为 2, 则采用的码本如 表 1所示, 表 1中与 PMI=6, RI=2对应预编码矩阵为
Figure imgf000017_0001
6 il3i /V2 不满秩,则选择 PMI=6, RI=4对应的满秩预编码矩阵 ^{1324ί/2,对^ /2进 行归一化得到第一旋转矩阵 Q 如式 (9)所示:
Figure imgf000017_0002
具体地, 根据参考 PMIref得到第一旋转矩阵 还可以是:
将参考 PMIref和预先获知的第一秩指示对应的预编码矩阵的各列进行 归一化, 对归一化后的预编码矩阵进行 QR分解, 得到第一旋转矩阵 (^。
具体地, 假定与参考 PMIref对应的各列归一化后的预编码矩阵为 Ϋκί, 则对归一化后的预编码矩阵 ref进行 QR分解如式 (10)所示:
Figure imgf000017_0003
从而可以得到第一旋转矩阵 QNB为:
Figure imgf000017_0004
其中, P为 eNode B和 UE之间共同约定的酉矩阵。
需要说明的是, 式 (10)中的 QR分解可以基于 Householder (豪斯霍尔 德) 变换, 或者 Givens (吉文斯)旋转, 或者 Gram-Schimdt (施密特)正 交换方法得到, 或者上述方法的变形或者组合得到, 但是 eNodeB和 UE必 须约定使用相同的 QR分解方法。
具体地, 根据参考 PMIref得到第一旋转矩阵 可以是:
将根据不同时刻的参考 PMIref, 得到的多个不同的旋转矩阵进行相乘, 得到差分 PMI,具体地可以采用如式 (11)所示的方法,得到第一旋转矩阵 QNB 为:
Q _ Q Q VBl · QN · · · Q ( 11 )
其中, Q腸 、 Q Q 可以采用与得到第一旋转矩阵 相同的方法 得到, Q 表示根据当前时刻(或子带)的参考 PMIref,得到的当前时刻(或 子带)旋转矩阵, Q 2 . . . Q量 表示根据当前时刻 (或子带)之前的每一时 刻 (或子带) 的差分 PMI利用 QR分解或者码本的嵌套特性得到的旋转矩 阵。
303: eNode B根据第一旋转矩阵 Q 和当前得到的瞬时 BF矩阵 /预编 码矩阵 V, 得到第一差分矩阵 D。 与步骤 203类似, 此处不再赘述。
304: eNode B根据第二差分码本和预先获知的第二 RI, 按照预设的量 化准则, 对第一差分矩阵 D进行量化, 得到差分 PMI 0。 与步骤 204类似, 此处不再赘述。
305: eNode B将差分 PMI 0通知给 UE。
306: UE根据参考 PMIref得到第二旋转矩阵 QUE; UE接收差分 PMI D, 并根据差分 PMI D , 利用第二差分码本和预先获知的第二 RI, 得到第二差 分矩阵 δ。
其中, 事先约定 eNode B和 UE使用相同的参考 PMIref ,并且, 采用与 步骤 302中得到第一旋转矩阵 相同的方法, 根据参考 PMIref得到第二旋 转矩阵 (^£
307: UE根据第二差分矩阵 δ和第二旋转矩阵 QRA, 重构出 BF矩阵 /预 编码矩阵 , 然后执行 310。 与步骤 207类似, 此处不再赘述。 308: eNode B才艮据 BF矩阵 /预编码矩阵, 利用第一非差分码本和预先 获知的第三 RI, 得到非差分 PMI, 并将非差分 PMI通知给 UE。
与步骤 208类似, 此处不再赘述。
309: UE接收并根据第二非差分 PMI, 利用第二非差分码本和预先获 知的第四 RI, 得到 BF矩阵 /预编码矩阵 。 与步骤 209类似, 此处不再赘 述。
310: UE利用得到的 BF矩阵 /预编码矩阵 , 对待发送数据 s进行预处 理,并将预处理后的待发送数据 s通过发射天线发送给 eNode B。与步骤 210 类似, 此处不再赘述。
311 : eNode B接收接收信号 y, 并利用 y进行数据检测。 与步骤 211类 似, 此处不再赘述。
需要说明的是, 上述在反馈 PMI时, 无论反馈的是差分 PMI还是非差 分 PMI, 均可以是对整个系统带宽反馈一个 PMI, 也可以是将系统分为多 个 BP, 每个 BP中含有多个子带, 为每个子带反馈一个 PMI。
并且需要说明的是, 当将本方面实施例所述的方法, 应用于下行(即 eNode B向 UE发送数据, UE向 eNode B反馈预编码矩阵指示)的情况时, 只需在步骤 301中由 eNode B指示 UE使用非差分 PMI或是基于参考 PMI 的差分 PMI进行反馈即可, 其他过程与上述步骤类似, 此处不再——赘述。
本发明实施例所述的获取预编码矩阵指示的方法, 根据参考 PMI得到 差分 PMI, 使用差分 PMI进行反馈, 可以利用信道中已存在的信息, 可以 充分利用信道在频域和时间域上的相关性, 从而提高反馈精度, 可以满足 MU-MIMO或 CoMP等技术的要求; 并且可以选择使用非差分反馈或是差 分反馈, 可以提高反馈的灵活性。 实施例 4
本发明实施例提供了一种获取预编码矩阵指示的方法, 与实施例 2和 实施例 3 所述方法的区别在于, 本发明实施例可以基于长期信道协方差矩 阵和此前的差分 PMI, 得到当前的差分 PMI, 具体地:
对于 UE而言, 可以采用如式 (12)所示的方法, 基于长期信道协方差矩 阵和参考 PMIref, 得到旋转矩阵 QUEn为:
QlJEn _ UUE · QUEi · QuE2 · · · QuEn ( 12)
其中, t ^可以采用与实施例 2 中得到第一旋转矩阵 £相同的方法得 到, 可将 称为第一参考旋转矩阵; Q^.Q^^Q 可以采用与实施例 3 中得到第一旋转矩阵 Q冊相同的方法得到, Q ^表示根据当前时刻(或子带 ) 的参考 PMIref得到的当前时刻 (或子带)旋转矩阵, QUE1 · · · QUE„表示 Q 之 后的差分 PMI得到的旋转矩阵. 表示根据当前时刻(或子带)之前的每 一时刻 (或子带)的参考 PMIref, 得到的当前时刻 (或子带)之前的每一时 刻(或子带)的参考旋转矩阵, (3[/£2...(3 —1示(3[/£1之后的差分?^11利用 QR 分解或者码本的嵌套特性得到的旋转矩阵。
相应地, 对于 eNodeB而言, 可以采用如式 (13)所示的方法, 基于长期 信道协方差矩阵和参考 PMIref, 得到旋转矩阵 QOT„为:
QJ H _ U NB · QJ · QN · · · QJ H ( 13) 其中, 可以采用与实施例 2 中得到第二旋转矩阵 相同的方法得 到, Q腦、 Q 2...Q量可以采用与实施例 3 中得到第一旋转矩阵 相同的 方法得到。 具体含义与式( 12)类似, 此处不再赘述。
当在实施例 2 中使用本实施例的方法, 获取旋转矩阵时, 只需在实施 例 2的步骤 201中, eNode B指示 UE使用非差分 PMI或是基于长期信道协 方差矩阵和参考 PMIref的差分 PMI即可, 其他过程与实施例 2相似, 此处 不再一一赘述。
当在实施例 3使用本实施例的方法, 获取旋转矩阵时, 只需在实施例 3 的步骤 301中 eNode B 确定 eNode B将使用非差分 PMI或是基于长期信道 协方差矩阵和参考 PMIref的差分 PMI进行反馈, 并通知 UE即可, 其他过 程与实施例 3相似, 此处不再——赘述。
本发明实施例所述的获取预编码矩阵指示的方法, 基于长期信道协方 差矩阵和参考 PMI, 得到差分 PMI, 使用差分 PMI进行反馈, 可以利用信 道中已存在的信息, 不仅可以减少开销, 而且可以充分利用信道在频域和 时间域上的相关性, 从而提高反馈精度, 可以满足 MU-MIMO或 CoMP等 技术的要求; 并且可以选择使用非差分反馈或是差分反馈, 可以提高反馈 的灵活性。 实施例 5
随着发射天线的增加, 双极化发射天线将成为一种普遍的配置。 现有 LTE R8系统的码本并没有专门针对双极化天线配置进行优化。 目前 3GPP LTE-A ( LTE-Advance, LTE高级演进)下行系统中, 发射天线数将进一步 增加, 为此本发明实施例提供了一种获取预编码矩阵指示的方法, 与实施 例 2、 3和 4中所述的方法的区别在于, 本发明实施例中的码本(差分或非 差分码本)并不是采用现有技术中的码本, 而是根据现有技术中单极化发 射天线配置对应的的码本(可称为单极化码本), 获取双极化发射天线配置 对应的码本(可称为双极化码本)。
具体地, 假定 N ( N为偶数)单极化发射天线配置对应的码本中包含 2L个码字(预编码矩阵), 其中 L表示反馈其中的一个码字时占用的信息比 特数(反馈单极化码本中单个码字需要 L比特信息, 则反馈双极化码本中 单个码字需要 L+1比特信息)。且其中第 i个码字用 = 0, 1...2L-1表示。 假定 N双极化发射天线配置的前 N/2 ( 1、 2、 3...N/2 )个天线为一组同极 化天线, 后 N/2 ( N/2+l、 N/2+2... N )个天线为另外一组同极化天线, 则 N 双极化发射天线配置对应的码本(可称为双极化码本) 中第 i个码字 W/可 以表示为如式( 14 )所示: = 0,1,···,2£ - 1 ( 14 ) 或可以表示为如式(15 )所示:
Figure imgf000022_0001
其中, diag …, 1,-1,…,- 1}表示前 Ν/2个对角线元素为 1, 后 N/2个对角 线元素为 -1的对角矩阵。
需要说明的是, 如果天线编号顺序发生变化, 则对(15 ) 中进行相应 的行交换即可。 如 N个双极化发射天线的中的第 1个、 第 3个、 第 5个… 第 N-1个天线为一组同极化天线, 第 2个、 第 4个、 第 6个…第 N个天线 为另外一组同极化天线, 则此时只需对角矩阵的对角元素为 1, -1, 1, -1, 1, -1...1 , -1即可。
本发明实施例中其他地方与实施例 2、 3和 4中相同, 此处不再——赘 述。
本发明实施例所述的获取预编码矩阵指示的方法, 根据信道信息, 得 到差分 PMI, 使用差分 PMI进行反馈, 可以利用信道中已存在的信息, 不 仅可以减少开销, 而且可以充分利用信道在频域和时间域上的相关性, 从 而提高反馈精度, 可以满足 MU-MIMO或 CoMP等技术的要求; 并且可以 选择使用非差分 PMI反馈或是差分 PMI反馈, 可以提高反馈的灵活性; 另 充分利用双极化发射天线配置的特点, 提高双极化发射天线配置下码本的 性能。 实施例 6
目前 3GPP LTE-A上行系统中,存在两种类型的码本, 即 CMF ( Cubic Metric Friendly , 立方度量友好)码本和 CMP ( Cubic Metric Preserving, 立 方度量保持)码本。 其中, CMF码本适用于功率不受限的场景, 而 CMP码 本适用于功率受限的场景。 为此本发明实施例提供了一种获取预编码矩阵 指示的方法, 与实施例 2、 3、 4和 5中所述的方法的区别在于, 本发明实施 例中, 可以根据 UE的 PH ( Power Headroom, 功率余量) , 选择码本, 具体 包括下面两种方式:
第一种: UE根据自身的 PH和预设的功率余量门限值 ¾^。ω, 判断自身 的 ΡΗ是否满足式(16) , 如果 UE自身的 ΡΗ满足式(16), 则 UE使用 CMF码本; 否则 UE使用 CMP码本。
ΡΗ > ¾reshold (16)
同时, eNode B也根据预先得知的 UE的 PH和预设的功率余量门限 threshold , 判断 UE的 PH是否满足式 (16), 如果满足式 (16), 则选择使用 CMF 码本; 否则,使用 CMP码本。需要说明的是,该种方式默认 UE和 eNode B 也预先获知预设的功率余量门限 ¾^。1(1信息。 并且, 在现有技术中 UE和 eNode B共知 UE的 PH信息。
第二种: eNode B根据预先得知的 UE的 PH和预设的功率余量门限 threshold , 判断 UE的 PH是否满足式 (16), 如果满足式 (16), 则选择使用 CMF 码本, 并通知 UE使用 CMF码本; 否则, 使用 CMP码本, 并通知 UE使用 CMF码本。
本发明实施例所述的获取预编码矩阵指示的方法, 根据信道信息, 得 到差分 PMI, 使用差分 PMI进行反馈, 可以利用信道中已存在的信息, 不 仅可以减少开销, 而且可以充分利用信道在频域和时间域上的相关性, 从 而提高反馈精度, 可以满足 MU-MIMO或 CoMP等技术的要求; 并且可以 选择使用非差分 PMI反馈或是差分 PMI反馈, 可以提高反馈的灵活性; 另 外, 可以根据功率余量在 CMP码本和 CMF码本之间自适应切换, 可以根 据 UE或 eNode B当前的功率使用情况, 发挥 CMP码本和 CMF码本各自 的优势。 实施例 7 本发明实施例提供了一种获取预编码矩阵指示的装置, 参见图 4, 该装 置包括:
第一旋转矩阵获取模块 401, 用于根据第一信道信息,得到第一旋转矩 阵;
第一差分矩阵获取模块 402,用于在第一旋转矩阵获取模块 401得到第 一旋转矩阵后, 根据第一旋转矩阵和当前得到的瞬时波束赋形矩阵 /预编码 矩阵, 得到第一差分矩阵;
差分预编码矩阵指示获取模块 403, 用于在第一差分矩阵获取模块 402 得到差分矩阵后, 根据第一差分码本和预先获知的第一秩指示, 按照预设 的量化准则, 对第一差分矩阵进行量化, 得到差分预编码矩阵指示。
其中, 当第一信道信息为第一长期信道协方差矩阵时, 第一旋转矩阵 获取模块 401, 具体可以包括:
特征矩阵获取单元, 用于对第一长期信道协方差矩阵进行特征值分解, 得到第一长期信道协方差矩阵的特征矩阵, 其中, 第一长期信道协方差矩 阵由本地对其在预设的时间段内在单个或者多个子载波上获取的多个信道 矩阵进行平均得到;
第一旋转矩阵获取单元, 用于在特征矩阵获取单元得到特征矩阵后, 将第一长期信道协方差矩阵的特征矩阵作为第一旋转矩阵。
其中, 当第一信道信息为参考预编码矩阵指示时, 第一旋转矩阵获取 模块 401, 具体可以包括:
第二旋转矩阵获取单元, 用于当参考预编码矩阵指示和预先获知的第 一秩指示对应的预编码矩阵列正交且满秩时, 对预编码矩阵的各列进行归 一化, 将归一化的预编码矩阵作为第一旋转矩阵;
第三旋转矩阵获取单元, 用于当参考预编码矩阵指示和预先获知的第 一秩指示对应的预编码矩阵列正交且不满秩, 但是参考预编码矩阵指示对 应的码本满足嵌套特性时, 选择参考预编码矩阵指示对应的满秩预编码矩 阵, 对满秩预编码矩阵的各列进行归一化, 将归一化的满秩预编码矩阵作 为第一旋转矩阵。
其中, 当第一信道信息为参考预编码矩阵指示时, 第一旋转矩阵获取 模块 401, 具体可以包括:
第四旋转矩阵获取单元, 用于对参考预编码矩阵指示和预先获知的第 一秩指示对应的预编码矩阵的各列进行归一化, 对归一化的预编码矩阵进 行 QR分解, 得到第一旋转矩阵。
进一步地, 参见图 5, 该装置还可以包括:
选择模块 404,用于当本地存在功率受限差分码本和功率不受限差分码 本时, 根据本地的功率余量和预设的功率余量门限值, 从功率受限差分码 本和功率不受限差分码本中选择一个码本, 作为第一差分码本。
进一步地, 参见图 6, 该装置还可以包括:
反馈模块 405,用于在差分预编码矩阵指示获取模块 403得到差分预编 码矩阵指示后, 将差分预编码矩阵指示反馈给数据发送端, 使数据发送端 根据差分预编码矩阵指示和其获取的第二旋转矩阵, 利用第二差分码本和 预先获知的第二秩指示, 重构出波束赋形矩阵 /预编码矩阵, 其中, 第二旋 转矩阵由数据发送端根据第二信道信息得到, 并且, 第二信道信息、 第二 旋转矩阵、 第二差分码和第二秩指示, 分别与第一信道信息、 第一旋转矩 阵、 第一差分码和第一秩指示相一致。
本发明实施例所述的获取预编码矩阵指示的装置, 根据信道信息, 得 到差分 PMI, 使用差分 PMI进行反馈, 可以利用信道中已存在的信息, 不 仅可以减少开销, 而且可以充分利用信道在频域和时间域上的相关性, 从 而提高反馈精度, 可以满足 MU-MIMO或 CoMP等技术的要求。 实施例 8
本发明实施例提供了一种数据发送装置, 参见图 7, 该装置包括: 第二旋转矩阵获取模块 501, 用于根据第二信道信息,得到第二旋转矩 阵;
第二差分矩阵获取模块 502, 用于根据接收到的差分预编码矩阵指示, 利用第二差分码本和预先获知的第二秩指示, 得到第二差分矩阵, 其中, 差分预编码矩阵指示由数据接收装置根据其获取的第一旋转矩阵和瞬时波 束赋形矩阵 /预编码矩阵,利用第一差分码本和预先获知的第一秩指示得到, 其中, 第一旋转矩阵根据第一信道消息得到, 并且, 第二信道信息与第一 信道信息相一致; 第二差分码本和第二秩指示, 分别与第一差分码本和第 一秩指示相一致;
预编码矩阵重构模块 503,用于根据第二旋转矩阵获取模块 501得到的 第二旋转矩阵和第二差分矩阵获取模块 502得到的差分矩阵, 重构出波束 赋形矩阵 /预编码矩阵。 其中, 当第二信道信息为第二长期信道协方差矩阵 时, 根据第二信道信息, 得到第二旋转矩阵具体可以包括:
对第二长期信道协方差矩阵进行特征值分解, 得到第二长期信道协方 差矩阵的特征矩阵, 其中, 第二长期信道协方差矩阵由本地对其在预设的 时间段内在多个子载波上获取的多个信道矩阵进行平均得到;
将第二长期信道协方差矩阵的特征矩阵作为第二旋转矩阵。
其中, 当第二信道信息为参考预编码矩阵指示时, 根据第二信道信息, 得到第二旋转矩阵具体可以包括:
如果参考预编码矩阵指示和预先获知的第二秩指示对应的预编码矩阵 列正交且满秩, 则对预编码矩阵进行归一化, 将归一化的预编码矩阵作为 第二旋转矩阵;
如果参考预编码矩阵指示和预先获知的第二秩指示对应的预编码矩阵 列正交且不满秩, 但是参考预编码矩阵指示对应的码本满足嵌套特性, 则 选择参考预编码矩阵指示对应的满秩预编码矩阵, 对满秩预编码矩阵的各 列进行归一化, 将归一化的满秩预编码矩阵作为第二旋转矩阵。 其中, 当第二信道信息为参考预编码矩阵指示时, 根据第二信道信息, 得到第二旋转矩阵具体可以包括:
对参考预编码矩阵指示和预先获知的第二秩指示对应的预编码矩阵的 各列进行归一化, 对归一化后的预编码矩阵进行 QR (正交三角化)分解, 得到第二旋转矩阵。
本发明实施例所述的数据发送装置, 利用根据信道信息得到的差分
PMI, 重构出波束赋形矩阵 /预编码矩阵, 可以利用信道中已存在的信息, 不仅可以减少开销, 而且可以充分利用信道在频域和时间域上的相关性, 从而提高重构出的波束赋形矩阵 /预编码矩阵的精度, 可以满足 MU-MIMO 或 CoMP等技术的要求。 实施例 9
本发明实施例提供了一种波束赋形矩阵 /预编码矩阵重构方法, 参见图 8, 该方法包括:
601 : 根据第二信道信息, 得到第二旋转矩阵。
其中, 当第二信道信息为第二长期信道协方差矩阵时, 根据第二信道 信息, 得到第二旋转矩阵具体可以包括:
对第二长期信道协方差矩阵进行特征值分解, 得到第二长期信道协方 差矩阵的特征矩阵, 其中, 第二长期信道协方差矩阵由本地对其在预设的 时间段内在单个或多个子载波上获取的多个信道矩阵进行平均得到;
将第二长期信道协方差矩阵的特征矩阵作为第二旋转矩阵。
其中, 当第二信道信息为参考预编码矩阵指示时, 根据第二信道信息, 得到第二旋转矩阵具体可以包括:
如果参考预编码矩阵指示和预先获知的第二秩指示对应的预编码矩阵 列正交且满秩, 则对预编码矩阵的各列进行归一化, 将归一化的预编码矩 阵作为第二旋转矩阵; 如果参考预编码矩阵指示和预先获知的第二秩指示对应的预编码矩阵 列正交且不满秩, 但是参考预编码矩阵指示对应的码本满足嵌套特性, 则 选择参考预编码矩阵指示对应的满秩预编码矩阵, 对满秩预编码矩阵的各 列进行归一化, 将归一化的满秩预编码矩阵作为第二旋转矩阵。
其中, 当第二信道信息为参考预编码矩阵指示时, 根据第二信道信息, 得到第二旋转矩阵具体可以包括:
对参考预编码矩阵指示和预先获知的第二秩指示对应的预编码矩阵的 各列进行归一化, 对归一化的预编码矩阵进行 QR (正交三角化)分解, 得 到第二旋转矩阵。
602: 根据接收到的差分预编码矩阵指示, 利用第二差分码本和预先获 知的第二秩指示, 得到第二差分矩阵。
其中, 差分预编码矩阵指示由数据接收装置根据其获取的第一旋转矩 阵和瞬时波束赋形矩阵 /预编码矩阵, 利用第一差分码本和预先获知的第一 秩指示得到, 其中, 第一旋转矩阵根据第一信道消息得到, 并且, 第二信 道信息与第一信道信息相一致; 第二差分码本和第二秩指示, 分别与第一 差分码本和第一秩指示相一致;
603: 根据第二旋转矩阵和第二差分矩阵, 重构出波束赋形矩阵 /预编码 矩阵。
本发明实施例所述的波束赋形矩阵 /预编码矩阵重构方法, 利用根据信 道信息得到的差分 PMI, 重构出波束赋形矩阵 /预编码矩阵, 可以利用信道 中已存在的信息, 不仅可以减少开销, 而且可以充分利用信道在频域和时 间域上的相关性, 从而提高重构出的波束赋形矩阵 /预编码矩阵的精度, 可 以满足 MU-MIMO或 CoMP等技术的要求。 以上实施例提供的技术方案中的全部或部分内容可以通过软件编程实 现, 其软件程序存储在可读取的存储介质中, 存储介质例如: 计算机中的 硬盘、 光盘或软盘。 以上所述仅为本发明的较佳实施例, 并不用以限制本发明, 凡在本发 明的精神和原则之内, 所作的任何修改、 等同替换、 改进等, 均应包含在 本发明的保护范围之内。

Claims

权利要求
1、 一种获取预编码矩阵指示的方法, 其特征在于, 所述方法包括: 根据第一信道信息, 得到第一旋转矩阵;
根据所述第一旋转矩阵和当前得到的瞬时波束赋形矩阵 /预编码矩阵, 得到第一差分矩阵;
根据第一差分码本和预先获知的第一秩指示, 按照预设的量化准则, 对所述第一差分矩阵进行量化, 得到差分预编码矩阵指示。
2、 根据权利要求 1所述的方法, 其特征在于, 所述第一信道信息为第 一长期信道协方差矩阵, 所述根据第一信道信息, 得到第一旋转矩阵具体 包括:
对所述第一长期信道协方差矩阵进行特征值分解, 得到所述第一长期 信道协方差矩阵的特征矩阵, 其中, 所述第一长期信道协方差矩阵由本地 对其在预设的时间段内在单个或多个子载波上获取的多个信道矩阵进行平 均得到;
将所述第一长期信道协方差矩阵的特征矩阵作为第一旋转矩阵。
3、 根据权利要求 1所述的方法, 其特征在于, 所述第一信道信息为参 考预编码矩阵指示, 所述根据第一信道信息, 得到第一旋转矩阵具体包括: 如果所述参考预编码矩阵指示和所述预先获知的第一秩指示对应的预 编码矩阵列正交且满秩, 则对所述预编码矩阵的各列进行归一化, 将归一 化的预编码矩阵作为第一旋转矩阵;
如果所述参考预编码矩阵指示和所述预先获知的第一秩指示对应的预 编码矩阵列正交且不满秩, 但是所述参考预编码矩阵指示对应的码本满足 嵌套特性, 则选择所述参考预编码矩阵指示对应的满秩预编码矩阵, 对所 述满秩预编码矩阵的各列进行归一化, 将归一化的满秩预编码矩阵作为第 一旋转矩阵;
或, 所述第一信道信息为参考预编码矩阵指示, 所述根据第一信道信 息, 得到第一旋转矩阵具体包括:
将所述参考预编码矩阵指示和所述预先获知的第一秩指示对应的预编 码矩阵的各列进行归一化;
对归一化的预编码矩阵进行 QR分解, 得到第一旋转矩阵。
4、 根据权利要求 1-3中任一权利要求所述的方法, 其特征在于, 所述得到差分预编码矩阵指示之后, 所述方法还包括:
将所述差分预编码矩阵指示反馈给数据发送端, 使所述数据发送端根 据所述差分预编码矩阵指示和其获取的第二旋转矩阵, 利用第二差分码本 和预先获知的第二秩指示, 重构出所述波束赋形矩阵 /预编码矩阵, 其中, 所述第二旋转矩阵由所述数据发送端根据第二信道信息得到, 并且, 所述 第二信道信息、 所述第二旋转矩阵、 所述第二差分码本和所述第二秩指示, 分别与所述第一信道信息、 所述第一旋转矩阵、 所述第一差分码本和所述 第一秩指示相一致。
5、根据权利要求 1所述的方法, 其特征在于, 所述根据第一信道信息, 得到第一旋转矩阵之前还包括:
获取使用差分预编码矩阵指示或非差分预编码矩阵指示进行反馈的消 息;
当获取到使用差分预编码矩阵指示进行反馈的消息时, 执行所述根据 第一信道信息, 得到第一旋转矩阵的步骤;
当获取到使用非差分预编码矩阵指示进行反馈的消息时, 根据当前得 到的瞬时波束赋形矩阵 /预编码矩阵, 利用第一非差分码本和预先获知的第 三秩指示, 得到非差分预编码矩阵指示。
6、 根据权利要 5所述的方法, 其特征在于:
所述非差分预编码矩阵指示为对整个系统带宽反馈的或对每个子带反 馈的;
所述差分预编码矩阵指示为对整个系统带宽反馈的或对每个子带反馈 的。
7、 根据权利要 5所述的方法, 其特征在于, 当本地所在的系统存在 N 个双极化发射天线配置时, 所述第一非差分码本采用双极化非差分码本, 所述双极化非差分码本根据包含有 2 个码字的单极化非差分码本得到, 所 述非差分双极化码本中包含有 2L+1个码字, 其中, 所述非差分双极化码本 中有 2L个码字与所述非差分单极化码本相同, 另外有 2L个码字为分别将 所述非差分单极化码本中的每个码字左乘以一个预设的对角矩阵得到, 其 中, N为大于等于 2的偶数, L表示反馈所述非差分单极化码本中的一个码 字时占用的信息比特数。
8、 根据权利要求 5所述的方法, 其特征在于, 当本地存在功率受限非 差分码本和功率不受限非差分码本时, 所述方法还包括:
根据本地的功率余量和预设的功率余量门限值, 从所述功率受限非差 分码本和所述功率不受限非差分码本中选择一个码本, 作为第一非差分码 本
9、 一种获取预编码矩阵指示的装置, 其特征在于, 所述装置包括: 第一旋转矩阵获取模块, 用于根据第一信道信息, 得到第一旋转矩阵; 第一差分矩阵获取模块, 用于在所述第一旋转矩阵获取模块得到第一 旋转矩阵后, 根据所述第一旋转矩阵和当前得到的瞬时波束赋形矩阵 /预编 码矩阵, 得到第一差分矩阵; 差分预编码矩阵指示获取模块, 用于在所述第一差分矩阵获取模块得 到差分矩阵后, 根据第一差分码本和预先获知的第一秩指示, 按照预设的 量化准则, 对所述第一差分矩阵进行量化, 得到差分预编码矩阵指示。
10、 根据权利要求 9所述的装置, 其特征在于, 当所述第一信道信息 为第一长期信道协方差矩阵时, 所述第一旋转矩阵获取模块, 具体包括: 特征矩阵获取单元, 用于对所述第一长期信道协方差矩阵进行特征值 分解, 得到所述第一长期信道协方差矩阵的特征矩阵, 其中, 所述第一长 取的多个信道矩阵进行平均得到;
第一旋转矩阵获取单元, 用于在所述特征矩阵获取单元得到特征矩阵 后, 将所述第一长期信道协方差矩阵的特征矩阵作为第一旋转矩阵。
11、 根据权利要求 9所述的装置, 其特征在于, 当所述第一信道信息 为参考预编码矩阵指示时, 所述第一旋转矩阵获取模块, 具体包括:
第二旋转矩阵获取单元, 用于当所述参考预编码矩阵指示和所述预先 获知的第一秩指示对应的预编码矩阵列正交且满秩时, 对所述预编码矩阵 的各列进行归一化, 将归一化的预编码矩阵作为第一旋转矩阵;
第三旋转矩阵获取单元, 用于当所述参考预编码矩阵指示和所述预先 获知的第一秩指示对应的预编码矩阵列正交且不满秩, 但是所述参考预编 码矩阵指示对应的码本满足嵌套特性时, 选择所述参考预编码矩阵指示对 应的满秩预编码矩阵, 对所述满秩预编码矩阵的各列进行归一化, 将归一 化的满秩预编码矩阵作为第一旋转矩阵;
或, 当所述信道信息为参考预编码矩阵指示时, 所述第一旋转矩阵获 取模块, 具体包括:
第四旋转矩阵获取单元, 用于对所述参考预编码矩阵指示和所述预先 获知的第一秩指示对应的预编码矩阵的各列进行归一化, 对归一化的预编 码矩阵进行 QR分解, 得到第一旋转矩阵。
12、 根据权利要求 9所述的装置, 其特征在于, 所述装置还包括: 选择模块, 用于当本地存在功率受限差分码本和功率不受限差分码本 时, 根据本地的功率余量和预设的功率余量门限值, 从所述功率受限差分 码本和所述功率不受限差分码本中选择一个码本, 作为第一差分码本。
13、 根据权利要求 9-12中任一权利要求所述的装置, 其特征在于, 所 述装置还包括:
反馈模块, 用于在所述差分预编码矩阵指示获取模块得到差分预编码 矩阵指示后, 将所述差分预编码矩阵指示反馈给数据发送端, 使所述数据 发送端根据所述差分预编码矩阵指示和其获取的第二旋转矩阵, 利用第二 差分码本和预先获知的第二秩指示,重构出所述波束赋形矩阵 /预编码矩阵, 其中, 所述第二旋转矩阵由所述数据发送端根据第二信道信息得到, 并且, 所述第二信道信息、 所述第二旋转矩阵、 所述第二差分码和所述第二秩指 示, 分别与所述第一信道信息、 所述第一旋转矩阵、 所述第一差分码和所 述第一秩指示相一致。
14、 一种数据发送装置, 其特征在于, 所述装置包括:
第二旋转矩阵获取模块, 用于根据第二信道信息, 得到第二旋转矩阵; 第二差分矩阵获取模块, 用于根据接收到的差分预编码矩阵指示, 利 用第二差分码本和预先获知的第二秩指示, 得到第二差分矩阵, 其中, 所 述差分预编码矩阵指示由数据接收装置根据其获取的第一旋转矩阵和瞬时 波束赋形矩阵 /预编码矩阵, 利用第一差分码本和预先获知的第一秩指示得 到, 其中, 所述第一旋转矩阵根据第一信道消息得到, 并且, 所述第二信 道信息与所述第一信道信息相一致; 所述第二差分码本和所述第二秩指示, 分别与所述第一差分码本和第一秩指示相一致;
预编码矩阵重构模块, 用于根据所述第二旋转矩阵获取模块得到的第 二旋转矩阵和所述第二差分矩阵获取模块得到的差分矩阵, 重构出所述波 束赋形矩阵 /预编码矩阵。
15、 一种波束赋形矩阵 /预编码矩阵重构的方法, 其特征在于, 所述方法 包括: 根据第二信道信息, 得到第二旋转矩阵;
根据接收到的差分预编码矩阵指示, 利用第二差分码本和预先获知的 第二秩指示, 得到第二差分矩阵, 其中, 所述差分预编码矩阵指示由数据 接收装置根据其获取的第一旋转矩阵和瞬时波束赋形矩阵 /预编码矩阵, 利 用第一差分码本和预先获知的第一秩指示得到, 其中, 所述第一旋转矩阵 根据第一信道消息得到, 并且, 所述第二信道信息与所述第一信道信息相 一致; 所述第二差分码本和所述第二秩指示, 分别与所述第一差分码本和 第一秩指示相一致;
根据所述第二旋转矩阵和所述第二差分矩阵, 重构出所述波束赋形矩阵 /预编码矩阵。
16、 根据权利要求 15所述的方法, 其特征在于:
当所述第二信道信息为接收自所述数据接收装置时, 所述第二信道信 息为对整个系统带宽反馈的或对每个子带反馈的;
所述差分预编码矩阵指示为对整个系统带宽反馈的或对每个子带反馈 的。
PCT/CN2010/077545 2009-09-30 2010-09-30 一种获取预编码矩阵指示的方法和装置 WO2011038697A1 (zh)

Priority Applications (3)

Application Number Priority Date Filing Date Title
EP10819926.6A EP2485408B1 (en) 2009-09-30 2010-09-30 Method and device for acquiring precoding matrix indicator
JP2012531227A JP5489029B2 (ja) 2009-09-30 2010-09-30 プレコーディング行列インジケータを取得する方法および装置
US13/434,540 US8553799B2 (en) 2009-09-30 2012-03-29 Method and apparatus for obtaining precoding matrix indicator

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN2009102352860A CN102035626B (zh) 2009-09-30 2009-09-30 一种获取预编码矩阵指示的方法和装置
CN200910235286.0 2009-09-30

Related Child Applications (1)

Application Number Title Priority Date Filing Date
US13/434,540 Continuation US8553799B2 (en) 2009-09-30 2012-03-29 Method and apparatus for obtaining precoding matrix indicator

Publications (1)

Publication Number Publication Date
WO2011038697A1 true WO2011038697A1 (zh) 2011-04-07

Family

ID=43825592

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2010/077545 WO2011038697A1 (zh) 2009-09-30 2010-09-30 一种获取预编码矩阵指示的方法和装置

Country Status (5)

Country Link
US (1) US8553799B2 (zh)
EP (1) EP2485408B1 (zh)
JP (1) JP5489029B2 (zh)
CN (1) CN102035626B (zh)
WO (1) WO2011038697A1 (zh)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2013541259A (ja) * 2010-08-23 2013-11-07 インテル コーポレイション チャネルフィードバック方法、通信局及び差分量子化方法

Families Citing this family (26)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
KR101092686B1 (ko) * 2010-01-08 2011-12-09 서울대학교산학협력단 직교 주파수 분할 다중 접속 기반 다중안테나 무선 시스템에서 송신 빔포밍 방법 및 장치
CN102158971B (zh) 2010-02-11 2014-11-05 华为技术有限公司 实现半持续调度业务或类似半持续调度业务的方法及设备
CN102957496A (zh) * 2011-08-19 2013-03-06 中兴通讯股份有限公司 一种实现码本重排的方法和系统
CN102412942B (zh) * 2011-10-18 2015-07-15 电信科学技术研究院 Ri确定方法和设备
US8737513B2 (en) * 2011-11-11 2014-05-27 Qualcomm Incorporated Apparatus and method for jointly selecting precoding matrices in a MIMO multi-flow system
CN103326768B (zh) * 2012-03-23 2016-06-22 华为技术有限公司 信道信息反馈的方法、装置及系统
CN102801455B (zh) * 2012-07-31 2015-12-16 华为技术有限公司 波束码本生成方法、波束搜索方法及相关装置
CN103780366A (zh) * 2012-10-17 2014-05-07 上海贝尔股份有限公司 协同多点传输的方法及装置
JP6006428B2 (ja) * 2012-10-23 2016-10-12 エルジー エレクトロニクス インコーポレイティド 無線通信システムにおいてチャネル状態情報をフィードバックする方法およびそのための装置
WO2014134876A1 (zh) * 2013-03-08 2014-09-12 华为技术有限公司 预编码矩阵指示的反馈方法、接收端和发射端
ES2781106T3 (es) * 2013-06-29 2020-08-28 Huawei Tech Co Ltd Procedimiento y aparato para determinar un indicador de matriz de precodificación, equipo de usuario, y estación base
JP6165645B2 (ja) * 2014-01-29 2017-07-19 日本電信電話株式会社 無線通信方法、無線通信システムおよび制御局装置
CN105940615B (zh) * 2014-03-05 2019-07-09 Lg电子株式会社 用于在无线通信系统中测量基于大规模天线阵列的波束成形的信道变化的方法和设备
US10169816B2 (en) * 2015-01-06 2019-01-01 Morgan Stanley Services Group Inc. Trading anomaly kill switch
CN105871441B (zh) * 2015-01-21 2018-11-30 华邦电子股份有限公司 多输入多输出无线通信系统及其通道分解方法
US20180026694A1 (en) * 2015-02-13 2018-01-25 Lg Electronics Inc. Method and apparatus for communication based on common feedback information in multiple antenna system
US10581462B2 (en) 2015-12-01 2020-03-03 Huawei Technologies Co., Ltd. Signature-enabled polar encoder and decoder
WO2017164623A2 (en) 2016-03-22 2017-09-28 Samsung Electronics Co., Ltd. Method and apparatus for transmitting uplink control information in wireless communication system
KR102458077B1 (ko) * 2016-03-22 2022-10-25 삼성전자 주식회사 이동 통신 시스템에서의 상향링크 제어 신호 전송 방법 및 장치
US10772049B2 (en) 2016-08-03 2020-09-08 Lg Electronics Inc. Method for terminal for reporting power headroom in wireless communication system, and terminal utilizing method
CN107612599A (zh) * 2017-08-04 2018-01-19 深圳市金立通信设备有限公司 一种指示频率选择性预编码信息的方法、基站及用户设备
CN110768701B (zh) * 2018-07-27 2022-10-28 中兴通讯股份有限公司 信道状态处理方法及装置、系统、终端、基站、存储介质
US10404339B1 (en) 2018-08-09 2019-09-03 At&T Intellectual Property I, L.P. Precoding matrix indicator determination in wireless communication systems
CN111464260B (zh) * 2019-01-21 2023-07-11 华为技术有限公司 一种信号发送、接收方法及设备
CN115441911A (zh) * 2021-06-01 2022-12-06 华为技术有限公司 通信处理方法和通信处理装置
WO2022267972A1 (en) * 2021-06-21 2022-12-29 Mediatek Singapore Pte. Ltd. High spatial resolution mimo precoding for uplink communication

Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101036332A (zh) * 2004-08-17 2007-09-12 德州仪器公司 用于提供闭环传输预编码的方法及设备
CN101262456A (zh) * 2007-02-12 2008-09-10 美国博通公司 用于处理通信信号的方法和系统
US20080232501A1 (en) * 2007-03-20 2008-09-25 Nec Laboratories America, Inc. Static and differential precoding codebook for mimo systems
WO2008138165A1 (en) * 2007-05-10 2008-11-20 Alcatel Shanghai Bell Co., Ltd. Method and device for pre-processing data to be transmitted in multi input communication system
CN101340219A (zh) * 2007-07-04 2009-01-07 华为技术有限公司 信道状态信息反馈方法及无线收发装置
CN101471907A (zh) * 2007-12-28 2009-07-01 三星电子株式会社 多入多出系统中的预编码方法及使用该方法的装置
CN101512929A (zh) * 2006-08-17 2009-08-19 交互数字技术公司 用于在mimo无线通信系统中提供有效预编码反馈的方法和设备

Family Cites Families (11)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20060039489A1 (en) * 2004-08-17 2006-02-23 Texas Instruments Incorporated Method and apparatus for providing closed-loop transmit precoding
US7885348B2 (en) * 2006-02-09 2011-02-08 Intel Corporation MIMO communication system and method for beamforming using polar-cap codebooks
TWI470957B (zh) * 2006-10-30 2015-01-21 Interdigital Tech Corp 無線通信系統中處理反饋方法及裝置
JP4709820B2 (ja) * 2006-10-30 2011-06-29 株式会社エヌ・ティ・ティ・ドコモ 受信装置、送信装置及びプリコーディング情報を提供する方法
JP4840088B2 (ja) * 2006-11-08 2011-12-21 ソニー株式会社 無線通信システム、並びに無線通信装置及び無線通信方法
CN101146078A (zh) * 2006-12-27 2008-03-19 中兴通讯股份有限公司 一种多输入多输出空间复用预编码矩阵的选择方法
US8687715B2 (en) 2007-02-12 2014-04-01 Broadcom Corporation Method and system for rate reduction pre-coding matrices
EP1995979B1 (en) * 2007-05-23 2009-06-03 NTT DoCoMo, Inc. Subchannel allocation apparatus and corresponding method
US9225397B2 (en) * 2007-07-10 2015-12-29 Texas Instruments Incorporated Efficient rank and precoding matrix feedback for MIMO systems
CN101400074A (zh) * 2007-09-28 2009-04-01 夏普株式会社 减少预编码矩阵索引号反馈量的方法、用户设备及系统
CN101931440B (zh) 2009-06-19 2014-04-02 华为技术有限公司 信道状态信息获取方法及装置

Patent Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101036332A (zh) * 2004-08-17 2007-09-12 德州仪器公司 用于提供闭环传输预编码的方法及设备
CN101512929A (zh) * 2006-08-17 2009-08-19 交互数字技术公司 用于在mimo无线通信系统中提供有效预编码反馈的方法和设备
CN101262456A (zh) * 2007-02-12 2008-09-10 美国博通公司 用于处理通信信号的方法和系统
US20080232501A1 (en) * 2007-03-20 2008-09-25 Nec Laboratories America, Inc. Static and differential precoding codebook for mimo systems
WO2008138165A1 (en) * 2007-05-10 2008-11-20 Alcatel Shanghai Bell Co., Ltd. Method and device for pre-processing data to be transmitted in multi input communication system
CN101340219A (zh) * 2007-07-04 2009-01-07 华为技术有限公司 信道状态信息反馈方法及无线收发装置
CN101471907A (zh) * 2007-12-28 2009-07-01 三星电子株式会社 多入多出系统中的预编码方法及使用该方法的装置

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2013541259A (ja) * 2010-08-23 2013-11-07 インテル コーポレイション チャネルフィードバック方法、通信局及び差分量子化方法

Also Published As

Publication number Publication date
EP2485408A1 (en) 2012-08-08
EP2485408B1 (en) 2017-06-14
CN102035626B (zh) 2013-06-12
JP2013506360A (ja) 2013-02-21
US20120189075A1 (en) 2012-07-26
US8553799B2 (en) 2013-10-08
EP2485408A4 (en) 2012-08-08
JP5489029B2 (ja) 2014-05-14
CN102035626A (zh) 2011-04-27

Similar Documents

Publication Publication Date Title
WO2011038697A1 (zh) 一种获取预编码矩阵指示的方法和装置
CN102725967B (zh) 用于信息反馈以及预编码的方法和装置
US9124328B2 (en) System and method for channel information feedback in a wireless communications system
EP2985924B1 (en) Method and apparatus for acquiring a precoding matrix indicator and a precoding matrix
EP2622758B1 (en) Method and apparatus for feedback in multi-user multiple-input multiple-output (mu-mimo) communication system
JP5745017B2 (ja) ワイヤレス通信ネットワーク中の複数ユーザのmimoのためのスケジューリング
WO2020221582A1 (en) Methods and apparatuses for csi reporting in a wireless communication system
US8630311B2 (en) System and method for reporting quantized feedback information for adaptive codebooks
US9800312B2 (en) Method and terminal for feeding back channel state information
EP2442509A1 (en) System and method for channel status information feedback in a wireless communications system that utilizes multiple-input multiple-output (MIMO) transmission
WO2012068880A1 (zh) 一种信道状态信息反馈方法及装置
WO2012037772A1 (zh) 一种信道状态信息反馈方法和终端
WO2012041102A1 (zh) 反馈信道信息的方法及终端
US20110243207A1 (en) System and Method for Adapting Codebooks
WO2012041103A1 (zh) 信道信息反馈方法及终端
WO2011123972A1 (zh) 执行空间信息辅助的分级反馈的方法和设备
US8995551B2 (en) Apparatus and method for generating codebook in a wireless communication system
US8576776B2 (en) Apparatus and method for overhead reduction of feedback in closed loop MIMO system

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 10819926

Country of ref document: EP

Kind code of ref document: A1

WWE Wipo information: entry into national phase

Ref document number: 2012531227

Country of ref document: JP

NENP Non-entry into the national phase

Ref country code: DE

REEP Request for entry into the european phase

Ref document number: 2010819926

Country of ref document: EP

WWE Wipo information: entry into national phase

Ref document number: 2010819926

Country of ref document: EP