WO2011085550A1 - 多输入多输出系统中有限码本闭环反馈的方法及装置 - Google Patents

多输入多输出系统中有限码本闭环反馈的方法及装置 Download PDF

Info

Publication number
WO2011085550A1
WO2011085550A1 PCT/CN2010/070166 CN2010070166W WO2011085550A1 WO 2011085550 A1 WO2011085550 A1 WO 2011085550A1 CN 2010070166 W CN2010070166 W CN 2010070166W WO 2011085550 A1 WO2011085550 A1 WO 2011085550A1
Authority
WO
WIPO (PCT)
Prior art keywords
feature value
data
correlation matrix
spatial correlation
information
Prior art date
Application number
PCT/CN2010/070166
Other languages
English (en)
French (fr)
Inventor
宋扬
吕荻
陈晋辉
Original Assignee
上海贝尔股份有限公司
阿尔卡特朗讯
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 上海贝尔股份有限公司, 阿尔卡特朗讯 filed Critical 上海贝尔股份有限公司
Priority to CN2010800344841A priority Critical patent/CN102474331A/zh
Priority to PCT/CN2010/070166 priority patent/WO2011085550A1/zh
Publication of WO2011085550A1 publication Critical patent/WO2011085550A1/zh

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
    • H04B7/0413MIMO systems
    • H04B7/0417Feedback systems
    • 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

Definitions

  • the present invention relates to wireless communication technologies, and in particular, to a finite codebook closed loop feedback method and apparatus in a multiple input multiple output system. Background technique
  • the data transmitting device can employ precoding processing to improve reception performance.
  • the signal received by the data receiving device can be expressed as:
  • r is the received signal
  • H is the channel transmission matrix
  • the dimension is Nr x Nt
  • W is the precoding matrix
  • the dimension is Nt x Mt
  • X is the signal carried by the transmitted data stream
  • n is the Gaussian white noise vector
  • Nt is The number of transmit antennas
  • Nr is the number of receive antennas
  • Mt is the number of data streams that need to be transmitted.
  • the optimal choice of W is the right singular matrix of the H matrix. Since the H matrix is observed by the data receiving device, the precoding matrix W also needs to be fed back to the data transmitting device by the data receiving device.
  • a limited precoding codebook closed-loop feedback scheme is widely used in the prior art, such as Long Term Evolution (LTE) or IEEE 802.16e.
  • Figure 1 shows a prior art system block diagram. For each transmit antenna scale, a set of precoding matrices or vectors is constructed, and this set of matrices or vectors (hereinafter referred to as matrices) is called a "codebook" and the data transmitting device and the data receiving device are made.
  • the data receiving device detects the channel and selects the optimal codeword (precoding matrix) for the current time, and then feeds back the q-bit index of the codeword to the data transmitting device.
  • the codebook in the 8th edition of the Long Term Evolution Project Protocol Standard (Release 8) is used in all scenarios and all user equipment.
  • the actual channel observed by a data receiving device may be very different and may change both in the time and frequency domains. Can not be well characterized by a codeword in a fixed codebook.
  • Current research shows that the transmission spatial correlation helps to form a changing codebook, which is derived from the basic codebook, can better adapt to the characteristics of the user channel, and further improve system performance.
  • the data transmitting device must know the transmission spatial correlation information that can be detected and fed back by the data receiving device.
  • the current draft IEEE 802.16m protocol uses a transform codebook-based precoding mode.
  • the Precoder Matrix Index ( ⁇ ) fed back by the data receiving device should be able to reflect one of the transformed codebooks, wherein the transformation of the basic codebook is performed by a type that can be characterized by
  • the spatial correlation matrix of the time channel characteristics is obtained.
  • the broadband transmit spatial correlation matrix is normalized and then element level quantized to reduce feedback overhead. Since the emission spatial correlation matrix has conjugate symmetry, only the upper triangular elements can be quantized and fed back.
  • the normalized diagonal elements are all positive real numbers and are quantized using 1 bit respectively.
  • the non-diagonal lines are complex numbers, which are respectively quantized by 4 bits.
  • the dimension of the spatial correlation matrix of the transmission is 2 x 2
  • the feedback amount of the spatial correlation matrix of the wideband transmission is 6 bits.
  • the spatial correlation matrix has a dimension of 4 x 4 and each feedback amount is 28 bits.
  • the feedback spatial correlation matrix has a long feedback period, such as feedback every 20 milliseconds to reduce feedback overhead.
  • the transformed codebook mode can be used for single-user single-stream multiple-input multiple-output transmission or multi-user single-stream multiple-input multiple-output transmission.
  • the transmit spatial correlation matrix not every element contributes the same. There is still a need to further reduce the feedback overhead, or to further improve the accuracy of the reconstructed transmit correlation matrix with equal feedback overhead. Summary of the invention
  • the present invention proposes a technical solution for adding information related to the eigenvalues and eigenvectors of the spatial correlation matrix based on the precoding matrix index feedback of the finite codebook.
  • the eigenvectors corresponding to different eigenvalues have different importance.
  • the main eigenvector corresponding to the main eigenvalue carries the most important information of the spatial correlation matrix.
  • Corresponding to a feature vector The larger the modulus value of the eigenvalue, the higher the importance of the eigenvector. The more important the feature vector plays a more important role in reconstructing the accuracy of the spatial correlation matrix, thus the greater the effect on improving system performance. Therefore, lower quantization precision can be adopted for the lesser eigenvectors, and even some unimportant eigenvectors are not fed back, which will not have a significant impact on the improvement of system performance, and can reduce the feedback overhead.
  • a method for feeding back channel information to a data transmitting device in a data receiving device of a multiple input multiple output system comprising the steps of: a. determining said data transmitting device to said a transmission spatial correlation matrix of the data receiving device; b. determining quantization information related to the feature value and the feature vector of the transmission spatial correlation matrix, comprising: a first number of bits for quantizing the first feature corresponding to the first feature value Information of the vector; wherein the first feature value is a main feature value; c. transmitting the quantized information determined in step b to the data transmitting device.
  • a method for transmitting data to a data receiving device in a data transmitting device of a multiple input multiple output system comprising the steps of: A. receiving and transmitting from the data receiving device Quantization information related to the spatial correlation matrix; wherein the quantization information includes: information of a first number of bits used to quantize a first eigenvector corresponding to a first eigenvalue of the transmit spatial correlation matrix; wherein the first eigenvalue is a main eigenvalue; B. reconstructing a transmit spatial correlation matrix according to the quantized information received in step A, and performing precoding processing on the data to be transmitted according to the reconstructed transmit spatial correlation matrix and the precoded codeword index .
  • a feedback apparatus for feeding back channel information to a data transmitting device in a data receiving device of a multiple input multiple output system, comprising: first determining means, configured to determine the data transmission a transmission spatial correlation matrix of the device to the data receiving device; second determining means, configured to determine quantization information related to the feature value and the feature vector of the transmission spatial correlation matrix, comprising: a first number of bits for quantization Information of a first feature vector corresponding to a feature value; wherein the first feature value is a main feature value; signaling transmitting means, configured to send the quantized information determined by the second determining device to the data transmitting device .
  • a signaling receiving apparatus configured to receive, from the data receiving device, quantization information related to a transmit spatial correlation matrix; where the quantized information includes: a first number of bits of a first eigenvalue for quantizing a transmit spatial correlation matrix Corresponding information of the first feature vector; wherein the first feature value is a main feature value; and precoding means, configured to reconstruct a transmit spatial correlation matrix according to the received quantized information, and according to the reconstructed transmit The spatial correlation matrix and the precoded codeword index precode the data to be transmitted.
  • the feedback overhead can be reduced while maintaining or improving the system performance of the MIMO system, or the system performance can be further improved with the same feedback overhead as the prior art.
  • Figure 1 shows a block diagram of a prior art system
  • FIG. 2 shows a system block diagram of a multiple input multiple output system in accordance with one embodiment of the present invention
  • FIG. 3 is a flow chart showing a method for feeding back channel information to a data transmitting device in a data receiving device of a multiple input multiple output system according to an embodiment of the present invention
  • FIG. 4 shows an embodiment in accordance with the present invention. Flowchart of a method for transmitting data to a data receiving device in a data transmitting device of a multiple input multiple output system
  • Figure 5 is a block diagram showing the structure of a feedback device for feeding back channel information to a data transmitting device in a data receiving device of a multiple input multiple output system according to an embodiment of the present invention
  • Figure 6 is a block diagram showing the structure of a data transmitting apparatus for transmitting data to a data receiving apparatus in a data transmitting apparatus of a multiple input multiple output system according to an embodiment of the present invention
  • the data transmitting device 20 transmits data to the data receiving device 10 through the MIMO interface, and the data receiving device 10 feeds back the signaling information to the data transmitting device 20.
  • the data transmitting device 20 may be a base station or an evolved Node B (eNB), and the data receiving device 10 may be a user equipment; for the uplink, the data transmitting device 20 may be a user equipment.
  • the data receiving device may be a base station or an evolved Node B.
  • FIG. 3 is a flow chart showing a method for feeding back channel information to a data transmitting device in a data receiving device of a multiple input multiple output system according to an embodiment of the present invention. As shown, the method includes three steps Sl l, S12, S13. Without loss of generality, the data receiving device 10 and the data transmitting device 20 will be described below as an example.
  • step S11 the data receiving device 10 will determine the transmission spatial correlation matrix of the data transmitting device 20 to the data receiving device 10.
  • the basic expression of the instantaneous transmission spatial correlation matrix on one subcarrier is H W * H ; where H represents the channel transmission matrix of the subcarrier observed by the data receiving device 10, and the dimension is Nr x Nt, Nt For the number of transmit antennas, Nr is the number of receive antennas; superscript / ⁇ indicates conjugate transpose.
  • the wideband transmission spatial correlation matrix used in the closed-loop feedback is determined by averaging in the frequency domain and the time domain, for example, averaging instantaneous transmission spatial correlation matrices in multiple subcarriers and multiple frames to obtain a broadband transmission space. Correlation matrix.
  • the emission spatial correlation matrix is a conjugate symmetric matrix of dimension Nt x Nt, which is the Hermitian matrix.
  • step S12 the data receiving device 10 determines the quantization information related to the feature value and the feature vector of the transmission spatial correlation matrix, and includes: a first number of bits for quantizing the first feature vector corresponding to the first feature value Information; wherein the first feature value is a primary feature value.
  • the data receiving device 10 will calculate the feature values of the transmitted spatial correlation matrix and their corresponding feature vectors.
  • the first feature value is a main feature value, that is, a feature value having the largest modulus value.
  • the first feature vector is a feature vector corresponding to the first feature value, That is, the main feature vector.
  • the first feature vector is the most important feature vector.
  • the first feature vector is regarded as an indispensable content in the feedback, and thus the quantization bit of the first feature vector is included in the quantization information.
  • the first feature vector If only one feature vector, that is, the first feature vector, is fed back, it is not necessary to feed back the first feature value, because only the reconstructed normalized spatial correlation matrix needs to be reconstructed in the transmitting and receiving parties, and the default first feature value of the transmitting and receiving parties is 1 .
  • step S13 the data receiving device 10 transmits the quantized information determined in step S12 to the data transmitting device 20.
  • the data receiving device 10 will feed back at least two feature vectors of the spatial correlation matrix and corresponding feature values.
  • the quantization information determined by the data receiving device 10 includes: a number of bits of information for quantizing the first feature vector corresponding to the first feature value, and a second number of bits for quantizing the second feature value.
  • the first feature value is a primary feature value, and the first number is not less than the second number.
  • the first number is for example but not limited to 6, the second number is for example but not limited to 4; in other words, the first eigenvector may be quantized with 6 bits and the second eigenvector may be quantized with 4 bits.
  • the above-described quantization bits are exemplary and not limiting. In different application scenarios, the number of quantization bits (accuracy) of each feature vector can be changed according to different needs.
  • the information related to the first feature value and the second feature value can take a variety of quantization forms.
  • the first feature value and the second feature value may be quantized separately.
  • the ratio of the other feature values other than the first feature value to the first feature value may be quantized; the ratio of the first feature value to itself, that is, 1, may not be quantized and fed back, thereby reducing the feedback overhead.
  • the quantized form of the information related to the feature value is pre-defined, and the data receiving device 10 and the data transmitting device 20 have a common understanding of this.
  • the second feature value may be any feature value other than the main feature value.
  • the second eigenvalue may be the second largest eigenvalue of the modulo value, or the third largest eigenvalue of the modulo value, or 66
  • the data receiving device 10 will feed back to the data transmitting device 20 a plurality of feature vectors including the main feature vector and corresponding feature values of the spatial correlation matrix.
  • the plurality of feature vectors may be all or part of the feature vector of the transmitted spatial correlation matrix.
  • the data receiving device 10 quantizes a plurality of feature vectors that need to be fed back, respectively. All the feature vectors that need feedback are sorted in descending order of the modulus values of the corresponding feature values, and the quantization bit number of the top ranked feature vector is greater than or equal to the number of quantization bits of the ordered feature vector.
  • the ratio of the other feature values other than the main feature value to the main feature value may be quantized, or each feature value may be separately quantized.
  • the quantization of each feature vector of the transmitted spatial correlation matrix includes vector quantization or element quantization in the vector.
  • Element quantization is the quantification of each element in a feature vector. For example, element quantization is performed on a feature vector including four complex elements, and the real part and the imaginary part of each element are respectively quantized into 2 bits, and the quantization information of the feature vector includes a total of 16 bits.
  • Vector quantization includes: constructing a plurality of vector samples to find a sample closest to a feature vector, and the quantization information is an index of the sample. Usually, each vector sample is normalized. The definition of "closest" here can be the smallest spatial distance. Taking 6-bit quantization as an example, the quantized sample space of the feature vector may include 64 vector samples. In general, the quantized sample space of the feature vector should be predefined, and the data receiving device 10 and the data transmitting device 20 have a consensus on this.
  • a number of transmit antennas is 4, with 3-bit quantization, and a discrete Fourier transform vector sample space with a sample size of 8 can be expressed as:
  • the data receiving device 10 performs vector quantization on the quantization of each feature vector of the transmission spatial correlation matrix.
  • the quantization information includes: information for quantizing the first feature vector of the first number of bits, and information for quantizing the second feature vector of the second number of bits, wherein the first number is not less than the second number.
  • the number of samples of the quantized sample space of the first feature vector is greater than or equal to the number of samples of the quantized sample space of the second feature vector, and the quantized sample space of the second feature vector may be equal to one subspace of the quantized sample space of the first feature vector.
  • the data receiving device 10 determines the transmission spatial correlation matrix by using a frequency domain average, a time domain long time average, or a time domain moving average mode.
  • Frequency domain averaging includes calculating a spatial correlation matrix of transmissions of different frequency components and then averaging, which may be for one subcarrier, one resource block, multiple resource blocks, or an average of the entire frequency band.
  • the time domain long time average can average the instantaneous transmit spatial correlation matrix of multiple subframes to determine the currently used transmit spatial correlation matrix.
  • a time-domain smoothing average is defined as follows:
  • R (Bu, - 1 + c ⁇ HH, ; where R / represents the transmission spatial correlation matrix determined in the f-1th subframe, 0 is a time domain averaging factor, ⁇ , indicating that the data receiving device 10 is The channel transmission matrix observed on a certain subcarrier of the fth subframe, H*H represents the instantaneous transmission spatial correlation matrix on the subcarrier, and S represents the set of subcarriers for calculating the average transmission spatial correlation matrix.
  • the transmission spatial correlation matrix determined in the fth subframe With the time domain smoothing average, the data receiving device 10 only needs to store the previously determined transmission spatial correlation matrix without storing the instantaneous data of multiple subframes, thereby saving system resources.
  • a method for feeding back channel information to a data transmitting device in a data receiving device of a multiple input multiple output system in addition to the above step S11,
  • the data receiving device 10 reconstructs the transmission spatial correlation matrix according to the quantization information determined in step S12, and determines a precoding codeword index that needs to be fed back according to the reconstructed transmission spatial correlation matrix.
  • the data receiving device 10 transmits the precoded codeword index that needs to be fed back to the data transmitting device 20.
  • the data receiving device 10 may employ different feedback periods for the precoded codeword index and the quantized information related to the feature values and feature vectors of the transmitted spatial correlation matrix.
  • the pre-coded codeword index can be fed back more frequently, for example once every 5 milliseconds.
  • Quantization information relating to the eigenvalues and eigenvectors of the spatial correlation matrix of the transmission may be used for a longer feedback period, e.g., every 20 milliseconds.
  • the quantization information determined at step S12 as the data receiving apparatus 10 includes: a first feature vector quantization bit Ul, the second feature vector u 2 of the quantization bit ratio of the first and second feature values eigenvalue The quantization bit of d.
  • the data receiving device 10 determines the quantized sample of the first feature vector U1 and the second feature vector u 2 according to the quantization information determined in step S12, which may be obtained according to the above data.
  • An existing precoding codebook such as the codebook defined in the eighth edition of the Long Term Evolution Project Protocol Standard, is referred to as a basic codebook, denoted as B, and any one of the codewords is denoted as b.
  • the data receiving device 10 multiplies the reconstructed transmit spatial correlation matrix ⁇ with each codeword in the basic codebook B to form a transformed codeword to construct a transformed codebook S, wherein any transformed codeword can be expressed as:
  • the basic codeword corresponds to an index value for the basic codebook B.
  • the transformed codeword corresponds to an index value for transforming the codebook ⁇ .
  • the basic codeword and its transform code The index values corresponding to the words are the same.
  • the optimal selection of the encoded codeword to be fed back is the (right) primary singular vector V1 of the channel transmission matrix H.
  • the data receiving device 10 selects a pre-encoded codeword to be fed back from the transformed codebook ⁇ according to a certain rule, such as a minimum chord distance rule, and the expression is:
  • Precoding code word index is the index required to be fed to the codeword value n b.
  • Figure 4 shows the representation of a multiple input multiple output system in accordance with one embodiment of the present invention, the method comprising two steps S21, S22.
  • the data receiving device 10 and the data transmitting device 20 will be described below as an example.
  • step S21 the data transmitting device 20 receives the quantization information related to the transmission spatial correlation matrix from the data receiving device 10; wherein the quantization information includes: a first number of bits of the first eigenvalue for quantizing the transmission spatial correlation matrix Corresponding information of the first feature vector; wherein the first feature value is a main feature value.
  • step S22 the data transmitting device 20 reconstructs a transmission spatial correlation matrix according to the quantization information received in step S21, and performs data to be transmitted according to the reconstructed transmission spatial correlation matrix and the precoding codeword index. Precoding processing.
  • the data transmitting device 20 selects the precoded codeword b hug 4 from the basic codebook B according to the precoding codeword index, and then multiplies and normalizes the reconstructed transmit spatial correlation matrix ⁇ with the codeword. Forming a transformed precoded codeword whose expression is
  • the data transmitting device 20 can precode the data to be transmitted to the data receiving device 10 based on the code word.
  • step S21 the data transmitting device 20 will also receive the precoded codeword index fed back from the data receiving device 10.
  • step S22 the data transmitting device 20 according to the precoded codeword index received in step S21.
  • the reconstructed transmit spatial correlation matrix ⁇ obtains the transformed precoded codeword b curatft , and then precodes the data to be transmitted to the data receiving device 10 according to the codeword.
  • the quantization information related to the transmission spatial correlation matrix received from the data receiving device 10 by the data transmitting device 20 includes: a first number of bits for quantizing the transmission spatial correlation matrix Information of the first feature vector corresponding to the first feature value, information of the second number of bits for quantizing the second feature vector corresponding to the second feature value, and related to the first feature value and the second feature value
  • the first feature value is a primary feature value, and the first number is not less than the second number.
  • the information fed back by the data receiving device 10 includes: a quantization bit of the first feature vector U1, a quantization bit of the second feature vector u 2 , and a quantization bit of a ratio d of the second feature value to the first feature value.
  • step S22 the data transmission apparatus 20 determines the bit quantized feature vector from a first quantized sample space of the first feature vector quantized in accordance with the received first eigenvector of ⁇ Ul, the second feature of the received vector
  • the quantized bits of 11 2 determine the quantized second eigenvector ⁇ 2 from the quantized sample space of the second eigenvector, and determine the quantized ratio J from the quantized bits of the ratio d.
  • the reconstructed transmit spatial correlation matrix ⁇ can be obtained according to the above data, and the expression is:
  • Fig. 5 is a block diagram showing the construction of a feedback means for feeding back channel information to a data transmitting apparatus in a data receiving apparatus of a multiple input multiple output system according to an embodiment of the present invention.
  • the feedback device 100 includes: a first determining device 101, a second determining device 102, and a signaling transmitting device 103.
  • Feedback device 100 is typically disposed within data receiving device 10. Without loss of generality, the data receiving device 10 and the data transmitting device 20 will be described below as an example.
  • the first determining means 101 is operative to perform step S1 l in the aforementioned method.
  • the first determining means 101 in the data receiving device 10 will determine the transmission spatial correlation matrix of the data transmitting device 20 to the data receiving device 10.
  • the second determining means 102 is operative to perform step S12 of the aforementioned method.
  • the second determining means 102 in the data receiving device 10 will determine a matrix associated with the transmitting space
  • the eigenvalue and the eigenvector related quantization information comprising: a first number of bits of information used to quantize the first eigenvector corresponding to the first eigenvalue; wherein the first eigenvalue is a main eigenvalue.
  • the signaling transmitting means 103 is for performing step S13 in the aforementioned method.
  • the signaling transmitting means 103 in the data receiving device 10 transmits the quantized information determined by the second determining means 102 to the data transmitting device 20.
  • the data receiving device 10 will feed back at least two feature vectors of the spatial correlation matrix and corresponding feature values.
  • the quantization information determined by the second determining device 102 in the feedback device 100 in the data receiving device 10 includes: a first number of bits of information for quantizing the first feature vector corresponding to the first feature value, and a second number of bits Information for quantizing the second feature vector corresponding to the second feature value, and information related to the first feature value and the second feature value.
  • the first feature value is a primary feature value, and the first number is not less than the second number.
  • the first number is for example but not limited to 6, and the second number is for example but not limited to 4; in other words, the first eigenvector may be quantized with 6 bits and the second eigenvector may be quantized with 4 bits.
  • the above-described quantization bits are exemplary and not limiting. In different application scenarios, the number of quantization bits (accuracy) of each feature vector can be changed according to different needs.
  • the information related to the first feature value and the second feature value can take a variety of quantization forms.
  • the first feature value and the second feature value may be quantized separately.
  • the ratio of the other feature values other than the first feature value to the first feature value may be quantized; the ratio of the first feature value to itself, that is, 1, may not be quantized and fed back, thereby reducing the feedback overhead.
  • the quantized form of the information related to the feature value is pre-defined, and the data receiving device 10 and the data transmitting device 20 have a common understanding of this.
  • the second feature value may be any feature value other than the main feature value.
  • the second characteristic value may be the second largest eigenvalue of the modulus value, or the third largest eigenvalue of the modulus value, or others.
  • the feedback device 100 in addition to the first determining device 101, the second determining device 102, and the signaling transmitting device 103, the feedback device 100 further includes a third Determine the device.
  • the third determining means is configured to: reconstruct a transmit spatial correlation matrix according to the quantized information determined by the second determining means 102, and determine a precoded codeword index to be fed back according to the reconstructed transmit spatial correlation matrix.
  • the signaling device 103 can also be used to send the pre-coded codewords that need to be fed back to the data transmitting device 20.
  • the data receiving device 10 may employ different feedback periods for the precoded codeword index and the quantized information related to the feature values and feature vectors of the transmitted spatial correlation matrix.
  • the pre-coded codeword index can be fed back more frequently, for example once every 5 milliseconds.
  • Quantization information relating to the eigenvalues and eigenvectors of the spatial correlation matrix of the transmission may be used for a longer feedback period, e.g., every 20 milliseconds.
  • Fig. 6 is a block diagram showing the structure of a data transmitting apparatus for transmitting data to a data receiving apparatus in a data transmitting apparatus of a multiple input multiple output system according to an embodiment of the present invention.
  • the data transmitting apparatus 200 includes: a signaling receiving apparatus 201, and a precoding apparatus 202.
  • the data transmitting device 200 is typically disposed in the data transmitting device 20. Without loss of generality, the data receiving device 10 and the data transmitting device 20 will be described below as an example.
  • the signaling receiving apparatus 201 is for performing step S21 in the foregoing method.
  • the signaling receiving apparatus 201 in the data transmitting device 20 receives the quantization information related to the transmission spatial correlation matrix from the data receiving device 10; wherein the quantization information includes: a first number of bits for quantizing the transmission spatial correlation matrix Information of the first feature vector corresponding to the first feature value; wherein the first feature value is a primary feature value.
  • the precoding device 202 is operative to perform step S22 in the foregoing method.
  • the precoding apparatus 202 in the data transmitting device 20 reconstructs a transmission spatial correlation matrix according to the received quantization information, and performs data to be transmitted according to the reconstructed transmission spatial correlation matrix and a precoding codeword index. Precoding processing.
  • the signaling receiving apparatus 201 in the data transmitting apparatus 200 in the data transmitting apparatus 20 is further configured to: receive a precoded codeword index fed back from the data receiving apparatus 10.
  • the precoding device 202 in the data transmitting device 20 prefetches the data to be transmitted according to the precoding codeword index received by the signaling receiving device 201. Encoding processing.
  • Figure 7 shows a simulation comparison of the technical solutions of five finite codebook closed-loop feedbacks.
  • the simulation conditions are set as follows:
  • the channel model adopts the urban microcellular model;
  • the antenna is set as four vertically polarized antennas of the data transmitting device, and the adjacent antennas are separated by half wavelength, and the data receiving device has two vertically polarized antennas, between adjacent antennas.
  • the half-wavelength is separated;
  • the basic codebook uses the 4-bit single-stream codebook in the eighth edition of the Long Term Evolution Project Protocol Standard;
  • the feedback period of the transmit spatial correlation matrix is 20 milliseconds;
  • the feedback delay is 6 sec;
  • the data transmission equipment and data The data flow between the receiving devices is a single stream.
  • the spatial correlation matrix is not fed back.
  • the second scheme feeds back a non-quantized transmit spatial correlation matrix.
  • the third is the technical solution in the aforementioned IEEE 802.16m, which performs element quantization and feedback on a transmission spatial correlation matrix of dimension 4 x 4.
  • the total number of bits of the feedback spatial correlation matrix is 28 bits.
  • the first feature vector of the feedback spatial correlation matrix is quantized by 6-bit vector
  • the second eigenvector is quantized by 6-bit vector
  • the ratio is quantized by 3 bits
  • the quantized vector sample space is constructed by the discrete Fourier transform method described above.
  • the total number of feedback bits of the transmit spatial correlation matrix is 15 bits.
  • the only difference between the fifth and fourth schemes is that the second eigenvector is quantized using 4-bit vectors.
  • the total number of feedback bits of the transmitted spatial correlation matrix is 13 bits.
  • the items of simulation comparison are: The quantization error of the main singular vector V1 of the multiplexed multi-input multi-output channel transmission matrix on 60 consecutive subcarriers.
  • the abscissa is the quantization error
  • the ordinate is the cumulative distribution function
  • the dashed line is the simulation curve of the first scheme
  • the solid line is the simulation curve of the second scheme
  • the solid line with the triangle mark is the third scheme.
  • the simulation curve, the solid line with the square mark is the simulation curve of the fourth scheme
  • the solid line with the circular mark is the simulation curve of the fifth scheme.
  • the simulation results show that the fourth and fifth schemes have better system performance than the third scheme, and the feedback overhead is lower than the third scheme.
  • the device referred to in the present invention may be implemented by a software function module, may also be implemented by a hardware module, or may be implemented by a combination of hardware and software.

Landscapes

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

Description

多输入多输出系统中有限码本闭环反馈的方法及装置 技术领域
本发明涉及无线通信技术, 尤其涉及多输入多输出系统中有限 码本闭环反馈方法及装置。 背景技术
在多输入多输出系统中, 数据发送设备可以采用预编码处理来 提高接收性能。 数据接收设备接收到的信号可以表示为:
r = HWx + n;
其中, r表示接收信号; H表示信道传输矩阵, 维度为 Nr x Nt; W 表示预编码矩阵, 维度为 Nt x Mt; X表示发送数据流所承载的信号; n表示高斯白噪声向量; Nt为发射天线数量, Nr为接收天线数量, Mt为所需传送的数据流的数量。
如果不考虑反馈开销, W的最优选择是 H矩阵的右奇异矩阵。 因为 H矩阵是由数据接收设备观测到的,所以预编码矩阵 W也需由 数据接收设备反馈给数据发送设备。 为降低反馈开销, 现有技术, 例如长期演进项目 (Long Term Evolution, LTE ) 或 IEEE 802.16e, 中广泛采用有限预编码码本闭环反馈方案。 图 1 示出了一种现有技 术的系统框图。 对于每一发射天线尺度, 构建一组预编码矩阵或向 量, 将这一组矩阵或向量 ( 以下均称为矩阵) 称之为 "码本 ( codebook ) " , 并使数据发送设备和数据接收设备双方均知道这 一码本, 该码本表示为 P = {P PL} , 其中的每一个矩阵称之为 一个 "码字 (codeword ) " 。 如果 = 2表示码本的大小, 则 q就是 对码本进行索引所需要的比特数。 数据接收设备检测信道并选择用 于当前时刻的最优码字 (预编码矩阵) , 然后将该码字的 q 比特索 引反馈到数据发送设备。
在长期演进项目协议标准第八版( Release 8 )中的码本是在所有 场景、 所有用户设备所使用。 然而, 由一个数据接收设备所观测到 的实际信道可能非常不同, 并且在时域和频域都发生变化, 因而可 能无法由固定码本中的一个码字很好的表征。 当前研究显示, 发送 空间相关性有助于形成变化的码本, 该变化的码本源自于基本码本, 可以更好的适应用户信道的特征, 并进一步改善系统性能。 在此情 况下, 数据发送设备必须了解可以被数据接收设备检测并反馈的发 送空间相关性信息。
目前制定的 IEEE 802.16m协议草案采用了一种变换的基于码本 的预编码模式。 在变换的码本模式下, 数据接收设备所反馈的预编 码矩阵索引 ( Precoder Matrix Index , ΡΜΙ )应能体现变换的码本中的 一项, 其中基本码本的变换是由一种能够刻画长时信道特性的发射 空间相关矩阵所得到的。 对宽带发射空间相关矩阵进行归一化, 然 后进行元素级量化, 以减少反馈开销。 因为发射空间相关矩阵具有 共轭对称性, 仅量化并反馈上三角元素即可。 归一化的对角线元素 均为正实数, 分别采用 1 比特量化。 非对角线为复数, 分别采用 4 比特量化。 当发射天线为 2根时, 发射空间相关矩阵的维度为 2 x 2 , 每次宽带发射空间相关矩阵的反馈量为 6比特。 当发射天线为 4根 时, 发射空间相关矩阵的维度为 4 x 4, 每次反馈量为 28比特。 发射 空间相关矩阵的反馈周期较长, 例如每 20毫秒反馈一次, 以降低反 馈开销。 变换的码本模式既可以用于单用户单流多输入多输出传输, 也可以用于多用户单流多输入多输出传输。 然而, 在发射空间相关 矩阵中, 并非每一个元素的贡献都是相同的。 仍然存在进一步减小 反馈开销的需求, 或者在同等反馈开销的情况下进一步改善重构的 发射相关矩阵的精确性的需求。 发明内容
为了克服现有技术的上述缺陷, 本发明提出了一种技术方案, 在有限码本的预编码矩阵索引反馈的基础上, 附加反馈与发射空间 相关矩阵的特征值和特征向量有关的信息。
经过大量研究表明, 在发射空间相关矩阵中, 对应于不同特征 值的特征向量具有不同的重要性, 例如, 对应于主特征值的主特征 向量携带了发射空间相关矩阵最重要的信息。 一个特征向量所对应 的特征值的模值越大, 则该特征向量的重要性越高。 越重要的特征 向量对重构发射空间相关矩阵的准确性起到的作用越大, 从而对改 善系统性能的作用越大。 因此, 可以对于较为次要的特征向量采用 较低的量化精度, 甚至不反馈某些不重要的特征向量, 这样既不会 对系统性能的改善带来明显的影响, 又可以降低反馈开销。
根据本发明的第一方面, 提供了一种在多输入多输出系统的数 据接收设备中用于向数据发送设备反馈信道信息的方法, 包括以下 步骤: a. 确定所述数据发送设备至所述数据接收设备的发射空间相 关矩阵; b. 确定与所述发射空间相关矩阵的特征值和特征向量有关 的量化信息, 包括: 第一数目比特的用于量化第一特征值所对应的 第一特征向量的信息; 其中所述第一特征值为主特征值; c. 将步骤 b中确定的所述量化信息发送给所述数据发送设备。
根据本发明的第二方面, 提供了一种在多输入多输出系统的数 据发送设备中用于向数据接收设备发送数据的方法, 包括以下步骤: A.接收来自所述数据接收设备的与发射空间相关矩阵有关的量化信 息; 其中所述量化信息包括: 第一数目比特的用于量化发射空间相 关矩阵的第一特征值所对应的第一特征向量的信息; 其中所述第一 特征值为主特征值; B. 根据步骤 A 中接收到的所述量化信息重构 发射空间相关矩阵, 并根据所述重构的发射空间相关矩阵以及预编 码码字索引对将要发送的数据进行预编码处理。
根据本发明的第三方面, 提供了一种在多输入多输出系统的数 据接收设备中用于向数据发送设备反馈信道信息的反馈装置, 包括: 第一确定装置, 用于确定所述数据发送设备至所述数据接收设备的 发射空间相关矩阵; 第二确定装置, 用于确定与所述发射空间相关 矩阵的特征值和特征向量有关的量化信息, 包括: 第一数目比特的 用于量化第一特征值所对应的第一特征向量的信息; 其中所述第一 特征值为主特征值; 信令发送装置, 用于将第二确定装置确定的所 述量化信息发送给所述数据发送设备。
根据本发明的第四方面, 提供了一种在多输入多输出系统的数 信令接收装置, 用于接收来自所述数据接收设备的与发射空间相关 矩阵有关的量化信息; 其中所述量化信息包括: 第一数目比特的用 于量化发射空间相关矩阵的第一特征值所对应的第一特征向量的信 息; 其中所述第一特征值为主特征值; 预编码装置, 用于根据接收 到的所述量化信息重构发射空间相关矩阵, 并根据所述重构的发射 空间相关矩阵以及预编码码字索引对将要发送的数据进行预编码处 理。
通过使用本发明提供的方法及装置, 可以在保持或提高多输入 多输出系统的系统性能的基础上减小反馈开销, 或者在与现有技术 同等反馈开销的情况下进一步改善系统性能。 附图说明
参考下面的图和说明, 将更好地理解该系统。 图中的元件不一 定按比例绘制, 而是重点用于说明典型模型的原理。 在图中, 贯穿 不同的示图, 类似的参考标号表示对应的特征。
图 1示出了一种现有技术的系统框图;
图 2 示出了根据本发明的一个实施例的多输入多输出系统的系 统框图;
图 3 示出了根据本发明的一个实施例的在多输入多输出系统的 数据接收设备中用于向数据发送设备反馈信道信息的方法流程图; 图 4 示出了根据本发明的一个实施例的在多输入多输出系统的 数据发送设备中用于向数据接收设备发送数据的方法流程图;
图 5 示出了根据本发明的一个实施例的在多输入多输出系统的 数据接收设备中用于向数据发送设备反馈信道信息的反馈装置的结 构框图;
图 6 示出了根据本发明的一个实施例的在多输入多输出系统的 数据发送设备中用于向数据接收设备发送数据的数据发送装置的结 构框图;
图 7示出了几种技术方案的仿真结果对比图。 具体实施方式
图 2 示出了根据本发明的一个实施例的多输入多输出系统的系 统框图。 如图所示, 数据发送设备 20通过多输入多输出信道向数据 接收设备 10发送数据, 数据接收设备 10向数据发送设备 20反馈信 令信息。 本领域技术人员应能理解, 对于下行链路, 数据发送设备 20可以是基站或者演进节点 B ( eNB ) , 数据接收设备 10可以是用 户设备; 对于上行链路, 数据发送设备 20可以是用户设备, 数据接 收设备可以是基站或者演进节点 B。
不失一般性地, 以下实施例均以下行链路为例。
图 3 示出了根据本发明的一个实施例的在多输入多输出系统的 数据接收设备中用于向数据发送设备反馈信道信息的方法流程图。 如图所示, 该方法包括三个步骤 Sl l、 S12、 S13。 不失一般性地, 以下将以数据接收设备 10和数据发送设备 20为例进行说明。
在步骤 S11 中, 数据接收设备 10将确定数据发送设备 20至数 据接收设备 10的发射空间相关矩阵。
具体地, 在一个子载波上的瞬时发射空间相关矩阵的基本表达 式为 HW * H ; 其中 H表示数据接收设备 10所观测到的该子载波的信 道传输矩阵, 维度为 Nr x Nt, Nt为发射天线数量, Nr为接收天线数 量; 上标/ ί表示共轭转置。 通常, 在闭环反馈中采用的宽带发射空 间相关矩阵是在频域、 时域上取平均来确定的, 例如对多个子载波、 多个帧中的瞬时发射空间相关矩阵取平均以得到宽带发射空间相关 矩阵。 发射空间相关矩阵是一个维度为 Nt x Nt的共轭对称矩阵, 亦 即厄密特矩阵 ( Hermitian matrix ) 。
在步骤 S12中, 数据接收设备 10将确定与所述发射空间相关矩 阵的特征值和特征向量有关的量化信息, 包括: 第一数目比特的用 于量化第一特征值所对应的第一特征向量的信息; 其中所述第一特 征值为主特征值。
具体地, 数据接收设备 10将计算出发射空间相关矩阵的各特征 值及其对应的特征向量。 其中, 第一特征值为主特征值, 亦即模值 最大的特征值。 第一特征向量是第一特征值所对应的一个特征向量, 亦即主特征向量。 如前所述, 越重要的特征向量对重构发射空间相 关矩阵的准确性起到的作用越大。 第一特征向量即为最重要的特征 向量, 在本发明的技术方案中将第一特征向量视作反馈中必不可少 的内容, 因此在量化信息中包括第一特征向量的量化比特。 如果只 反馈一个特征向量, 即第一特征向量, 则可以不必反馈笫一特征值, 因为在收发双方只需要重构归一化的发射空间相关矩阵即可, 收发 双方默认第一特征值为 1。
在步骤 S 13中, 数据接收设备 10将步骤 S 12中确定的所述量化 信息发送给数据发送设备 20。
根据本发明的一个实施例, 数据接收设备 10将反馈发射空间相 关矩阵的至少两个特征向量及相应的特征值。 在步骤 S12 中, 数据 接收设备 10确定的量化信息包括: 笫一数目比特的用于量化第一特 征值所对应的第一特征向量的信息、 第二数目比特的用于量化第二 特征值所对应的第二特征向量的信息、 以及与所述第一特征值和第 二特征值有关的信息。 其中第一特征值为主特征值, 所述第一数目 不小于所述第二数目。
第一数目例如但不限于 6, 第二数目例如但不限于 4; 换言之, 第一特征向量可以采用 6比特量化, 第二特征向量可以采用 4比特 量化。 本领域技术人员应能理解, 上述量化比特均为示例性而非限 制性的。 在不同的应用场景下, 各特征向量的量化比特数 (精度) 可以根据不同的需要而改变。
与第一特征值和第二特征值有关的信息可以采用多种量化形 式。 例如, 可以分别对第一特征值和第二特征值进行量化。 又例如, 可以对第一特征值之外的其他特征值与第一特征值的比值进行量 化; 第一特征值与自身的比值, 也就是 1, 可以不必量化和反馈, 从 而可以降低反馈开销。 通常, 与特征值有关的信息的量化形式是预 先定义好的, 并且数据接收设备 10和数据发送设备 20对此具有共 识。
第二特征值可以是主特征值之外的任一特征值。 例如, 第二特 征值可以是模值第二大的特征值、 或者模值第三大的特征值、 或者 66
其他。
根据本发明的一个实施例, 数据接收设备 10将向数据发送设备 20反馈发射空间相关矩阵的包括主特征向量在内的多个特征向量及 相应特征值。 这多个特征向量可以是发射空间相关矩阵的全部或者 部分特征向量。 在步骤 S12中, 数据接收设备 10分别对需要反馈的 多个特征向量进行量化。 对所有需要反馈的特征向量按照其对应的 特征值的模值从大到小的顺序排序, 排序靠前的特征向量的量化比 特数大于或等于排序靠后的特征向量的量化比特数。 关于需要反馈 的多个特征向量分别对应的特征值, 可以对主特征值之外的其他特 征值与主特征值的比值进行量化, 也可以对各特征值分别进行量化。
根据本发明的一个实施例, 在步骤 S12 中, 发射空间相关矩阵 的各特征向量的量化包括向量量化或者向量中的元素量化。
元素量化是对一个特征向量中的各元素分别进行量化。 例如, 对一个包括四个复数元素的特征向量进行元素量化, 每元素的实部、 虚部分别量化为 2比特,则该特征向量的量化信息总共包括 16比特。
向量量化包括: 构造多个向量样本, 寻找与一个特征向量最接 近的样本, 量化信息即为该样本的索引。 通常, 各向量样本均是归 一化的。 这里的 "最接近" 的定义可以是具有最小的某种空间距离。 以 6比特量化为例, 特征向量的量化样本空间可以包括 64个向量样 本。 通常, 特征向量的量化样本空间应是预先定义好的, 并且数据 接收设备 10和数据发送设备 20对此具有共识。
一种可行的向量样本空间采用离散傅里叶变换来构建。 对于一 个发射天线数量为 M, 采用 k比特量化, 样本数量为 N=2k的向量量 化样本空间, 其公式定义如下:
CDFT(M, N) = [Co, ··· , cN-i ]
其中, c。至 cN— !分别表示一个向量样本。 每一个向量样本包括 M个 元素, 第
Figure imgf000009_0001
例如, 一个发射天线数量为 4, 采用 3比特量化, 样本数量为 8 的离散傅里叶变换向量样本空间可以表示为:
Figure imgf000010_0001
根据本发明的一个实施例, 在步骤 S12中, 数据接收设备 10对 发射空间相关矩阵的各特征向量的量化采用向量量化。 量化信息包 括: 第一数目比特的用于量化第一特征向量的信息、 第二数目比特 的用于量化第二特征向量的信息, 其中第一数目不小于第二数目。 第一特征向量的量化样本空间的样本数量大于或者等于第二特征向 量的量化样本空间的样本数量, 则第二特征向量的量化样本空间可 以等于第一特征向量的量化样本空间的一个子空间。
根据本发明的一个实施例, 在步骤 S1 1 中, 数据接收设备 10采 用频域平均、 时域长时平均或者时域滑动平均方式来确定所述发射 空间相关矩阵。
频域平均包括计算不同频率成分的发射空间相关矩阵然后取平 均值, 其可以是对一个子载波、 一个资源块、 多个资源块、 或者对 整个频带进行平均。
时域长时平均可以对多个子帧的瞬时发射空间相关矩阵进行平 均, 以确定当前使用的发射空间相关矩阵。
时域平滑平均一种公式定义如下:
R = (卜 ,— 1 + c ∑H H, ; 其中, R/ 表示第 f-1个子帧中所确定的发射空间相关矩阵, 0是一 个时域平均因子, Η,.表示数据接收设备 10在第 f个子帧在某一个子 载波上观测到的信道传输矩阵, H * H表示该子载波上的瞬时发射空 间相关矩阵, S表示计算平均的发射空间相关矩阵的子载波的集合。
即为第 f 个子帧中所确定的发射空间相关矩阵。 采用时域平滑平 均, 数据接收设备 10只需存储之前确定的发射空间相关矩阵, 而无 需存储多个子帧的瞬时数据, 从而节省了系统资源。
根据本发明的一个实施例, 在多输入多输出系统的数据接收设 备中用于向数据发送设备反馈信道信息的方法除了上述步骤 S 11、 S 12、 S 13之外, 还包括两个步骤: 一个确定步骤和一个发送步骤。 在所述确定步骤中, 数据接收设备 10将根据步骤 S 12中确定的 量化信息来重构发射空间相关矩阵, 并根据重构的发射空间相关矩 阵来确定需要反馈的预编码码字索引。
在所述发送步骤中, 数据接收设备 10将所述需要反馈的预编码 码字索引发送给数据发送设备 20。
数据接收设备 10可以对预编码码字索引和与发射空间相关矩阵 的特征值和特征向量有关的量化信息采用不同的反馈周期。 对于预 编码码字索引可以反馈得更加频繁, 例如每 5 毫秒反馈一次。 对于 与发射空间相关矩阵的特征值和特征向量有关的量化信息可以采用 较长的反馈周期, 例如每 20毫秒反馈一次。
上述确定步骤中, 一种确定需要反馈的预编码码字索引的具体 方式如下:
具体地, 例如, 数据接收设备 10在步骤 S12中所确定的量化信 息包括: 第一特征向量 Ul的量化比特、 第二特征向量 u2的量化比特、 第二特征值与第一特征值的比值 d的量化比特。
则在上述确定步骤中, 数据接收设备 10将根据步骤 S 12中确定 的量化信息来确定第一特征向量 Ul的量化样本 、第二特征向量 u2的 , 即可根据上述数据得
Figure imgf000011_0001
现有的预编码码本, 例如在长期演进项目协议标准第八版中定 义的码本, 被称为基本码本, 表示为 B, 其中的任一码字表示为 b,。 数据接收设备 10将重构的发射空间相关矩阵 έ与基本码本 B中每一 码字相乘形成变换的码字, 以构建变换的码本 S , 其中任一变换的码 字可以表示为:
Figure imgf000011_0002
基本码字 对应于一个用于基本码本 B的索引值。 变换的码字 对应于一个用于变换码本 δ的索引值。 通常, 基本码字 与其变换码 字 对应的索引值是相同的。
在数据发送设备 20和数据接收设备 10之间传输一个数据流的 情况下, 需要反馈的与编码码字的最优选择是信道传输矩阵 H 的 (右) 主奇异向量 Vl。 数据接收设备 10根据一定的规则, 例如最小 弦距离规则,从变换的码本 δ中选择需要反馈的预编码码字 ,其表 达式为:
Rb
b„6 = argmaxv,b , 且 =
b,eB Rb
所需要反馈的预编码码字索引即为码字 的索引值 nb
图 4 示出了根据本发明的一个实施例的在多输入多输出系统的 所示, 该方法包括两个步骤 S21、 S22。 不失一般性地, 以下将以数 据接收设备 10和数据发送设备 20为例进行说明。
在步骤 S21 中, 数据发送设备 20接收来自数据接收设备 10的 与发射空间相关矩阵有关的量化信息; 其中所述量化信息包括: 第 一数目比特的用于量化发射空间相关矩阵的第一特征值所对应的第 一特征向量的信息; 其中所述第一特征值为主特征值。
在步骤 S22中, 数据发送设备 20根据步骤 S21 中接收到的所述 量化信息重构发射空间相关矩阵, 并根据所述重构的发射空间相关 矩阵以及预编码码字索引对将要发送的数据进行预编码处理。
具体地,数据发送设备 20根据预编码码字索引 从基本码本 B中 选择出预编码码字 b„4, 然后, 将重构的发射空间相关矩阵 έ与码字 相乘并进行归一化形成变换的预编码码字 , 其表达式为
〜 Rb
Rb
然后, 数据发送设备 20 即可根据码字 对将要发送给数据接收设 备 10的数据进行预编码处理。
根据本发明的一个实施例, 在步骤 S21 中, 数据发送设备 20还 将接收到来自数据接收设备 10所反馈的预编码码字索引 。 在步骤 S22中, 数据发送设备 20根据步骤 S21中接收到的预编码码字索引 和重构的发射空间相关矩阵 ά得到变换的预编码码字 b„ft , 然后根 据码字 对将要发送给数据接收设备 10的数据进行预编码处理。
根据本发明的一个实施例, 在步驟 S21 中, 数据发送设备 20接 收到的来自数据接收设备 10的与发射空间相关矩阵有关的量化信息 包括: 第一数目比特的用于量化发射空间相关矩阵的第一特征值所 对应的第一特征向量的信息、 第二数目比特的用于量化第二特征值 所对应的第二特征向量的信息、 以及与所述第一特征值和第二特征 值有关的信息; 其中所述第一特征值为主特征值, 所述第一数目不 小于所述第二数目。
具体地, 例如, 数据接收设备 10所反馈的信息包括: 第一特征 向量 Ul的量化比特、 第二特征向量 u2的量化比特、 第二特征值与第一 特征值的比值 d的量化比特。
则在步骤 S22中, 数据发送设备 20将根据接收到的第一特征向 量 Ul的量化比特从第一特征向量的量化样本空间中确定量化的第一 特征向量^,根据接收到的第二特征向量112的量化比特从第二特征向 量的量化样本空间中确定量化的第二特征向量 ΰ2 , 并根据比值 d 的 量化比特确定量化的比值 J。 然后, 即可根据上述数据得到重构的发 射空间相关矩阵 έ, 其表达式为:
Figure imgf000013_0001
图 5 示出了根据本发明的一个实施例的在多输入多输出系统的 数据接收设备中用于向数据发送设备反馈信道信息的反馈装置的结 构框图。 如图所示, 反馈装置 100包括: 第一确定装置 101、 第二确 定装置 102、 信令发送装置 103。 反馈装置 100典型地设置于数据接 收设备 10之中。 不失一般性地, 以下将以数据接收设备 10和数据 发送设备 20为例进行说明。
第一确定装置 101用于执行前述方法中的步骤 S l l。 例如, 数据 接收设备 10中的第一确定装置 101将确定数据发送设备 20至数据 接收设备 10的发射空间相关矩阵。
第二确定装置 102用于执行前述方法中的步骤 S 12。 例如, 数据 接收设备 10中的第二确定装置 102将确定与所述发射空间相关矩阵 的特征值和特征向量有关的量化信息, 包括: 笫一数目比特的用于 量化第一特征值所对应的第一特征向量的信息; 其中所述第一特征 值为主特征值。
信令发送装置 103用于执行前述方法中的步骤 S 13。 例如, 数据 接收设备 10中的信令发送装置 103将第二确定装置 102确定的所述 量化信息发送给数据发送设备 20。
根据本发明的一个实施例, 数据接收设备 10将反馈发射空间相 关矩阵的至少两个特征向量及相应的特征值。 其中, 数据接收设备 10中的反馈装置 100中的第二确定装置 102确定的量化信息包括: 第一数目比特的用于量化第一特征值所对应的第一特征向量的信 息、 第二数目比特的用于量化第二特征值所对应的第二特征向量的 信息、 以及与所述第一特征值和第二特征值有关的信息。 其中第一 特征值为主特征值, 所述第一数目不小于所述第二数目。
第一数目例如但不限于 6 , 第二数目例如但不限于 4; 换言之, 第一特征向量可以采用 6比特量化, 第二特征向量可以采用 4比特 量化。 本领域技术人员应能理解, 上述量化比特均为示例性而非限 制性的。 在不同的应用场景下, 各特征向量的量化比特数(精度) 可以根据不同的需要而改变。
与第一特征值和第二特征值有关的信息可以采用多种量化形 式。 例如, 可以分别对第一特征值和第二特征值进行量化。 又例如, 可以对第一特征值之外的其他特征值与第一特征值的比值进行量 化; 第一特征值与自身的比值, 也就是 1, 可以不必量化和反馈, 从 而可以降低反馈开销。 通常, 与特征值有关的信息的量化形式是预 先定义好的, 并且数据接收设备 10和数据发送设备 20对此具有共 识。
第二特征值可以是主特征值之外的任一特征值。 例如, 第二特 征值可以是模值第二大的特征值、 或者模值第三大的特征值、 或者 其他。
根据本发明的一个实施例, 反馈装置 100 中除了第一确定装置 101、 第二确定装置 102、 信令发送装置 103之外, 还包括一个第三 确定装置。
第三确定装置用于: 根据第二确定装置 102 确定的量化信息来 重构发射空间相关矩阵, 并根据重构的发射空间相关矩阵来确定需 要反馈的预编码码字索引。
信令发送装置 103 还可以用于将所述需要反馈的预编码码字索 引发送给数据发送设备 20。
数据接收设备 10可以对预编码码字索引和与发射空间相关矩阵 的特征值和特征向量有关的量化信息采用不同的反馈周期。 对于预 编码码字索引可以反馈得更加频繁, 例如每 5 毫秒反馈一次。 对于 与发射空间相关矩阵的特征值和特征向量有关的量化信息可以采用 较长的反馈周期, 例如每 20毫秒反馈一次。
图 6 示出了根据本发明的一个实施例的在多输入多输出系统的 数据发送设备中用于向数据接收设备发送数据的数据发送装置的结 构框图。 如图所示, 数据发送装置 200包括: 信令接收装置 201、 预 编码装置 202。 数据发送装置 200典型地设置于数据发送设备 20之 中。 不失一般性地, 以下将以数据接收设备 10 和数据发送设备 20 为例进于说明。
信令接收装置 201用于执行前述方法中的步骤 S21。 例如, 数据 发送设备 20中的信令接收装置 201接收来自数据接收设备 10的与 发射空间相关矩阵有关的量化信息; 其中所述量化信息包括: 第一 数目比特的用于量化发射空间相关矩阵的第一特征值所对应的第一 特征向量的信息; 其中所述第一特征值为主特征值。
预编码装置 202用于执行前述方法中的步骤 S22。 例如, 数据发 送设备 20中的预编码装置 202根据接收到的所述量化信息重构发射 空间相关矩阵, 并根据所述重构的发射空间相关矩阵以及预编码码 字索引对将要发送的数据进行预编码处理。
根据本发明的一个实施例, 数据发送设备 20中的数据发送装置 200中的信令接收装置 201还用于: 接收来自数据接收设备 10所反 馈的预编码码字索引。 数据发送设备 20中的预编码装置 202根据信 令接收装置 201 接收到的预编码码字索引对将要发送的数据进行预 编码处理。
图 7 示出了五种有限码本闭环反馈的技术方案的仿真结杲对比 图。 仿真条件设置如下: 信道模型采用城市微蜂窝模型; 天线设置 为数据发送设备 4 个垂直极化天线, 相邻天线之间间隔半波长, 数 据接收设备 2 个垂直极化天线, 相邻天线之间间隔半波长; 基本码 本采用长期演进项目协议标准第八版中的 4 比特的单流码本; 发射 空间相关矩阵的反馈周期为 20毫秒; 反馈时延为 6亳秒; 数据发送 设备与数据接收设备之间的数据流为单流。
五种技术方案分别为:
第一种方案, 不反馈发射空间相关矩阵。
第二种方案, 反馈非量化的发射空间相关矩阵。
第三种即为前述 IEEE 802.16m 中的技术方案, 对维度为 4 x 4 的发射空间相关矩阵进行元素量化及反馈。 发射空间相关矩阵的反 馈总比特数为 28比特。
第四种方案, 反馈发射空间相关矩阵的第一特征向量、 第二特 征向量以及第二特征值与第一特征值的比值; 其中第一特征值与第 二特征值分別为主特征值和模值第二大的特征值; 第一特征向量采 用 6比特向量量化, 第二特征向量采用 6比特向量量化, 比值采用 3 比特均勾量化; 量化向量样本空间采用前述离散傅里叶变换方式构 建。 发射空间相关矩阵的反馈总比特数为 15比特。
第五种方案与第四种方案的唯一区别在于: 第二特征向量采用 4 比特向量量化。 发射空间相关矩阵的反馈总比特数为 13比特。
仿真比较的项目为: 在 60个连续的子载波上取平均的多输入多 输出信道传输矩阵的主奇异向量 Vl的量化误差。
在图 7 中, 横坐标为量化误差, 纵坐标为累积分布函数, 虚线 是第一种方案的仿真曲线, 实线是第二种方案的仿真曲线, 带三角 标记的实线是第三种方案的仿真曲线, 带方块标记的实线是第四种 方案的仿真曲线, 带圓形标记的实线是第五种方案的仿真曲线。
仿真结果显示, 第四种方案和第五种方案具有比第三种方案更 好的系统性能, 而反馈开销还要低于第三种方案。 本发明所称的装置, 可以由软件功能模块实现, 也可以由硬件 模块实现, 还可以由软硬件的结合来实现。
本领域技术人员应能理解, 上述实施例均是示例性而非限制性 的。 在不同实施例中出现的不同技术特征可以进行组合, 以取得有 益效果。 本领域技术人员在研究附图、 说明书及权利要求书的基础 上, 应能理解并实现所揭示的实施例的其他变化的实施例。 在权利 要求书中, 术语 "包括" 并不排除其他装置或步驟; 不定冠词 "一 个" 不排除多个; 术语 "第一" 、 "第二" 用于标示名称而非用于 表示任何特定的顺序。 权利要求中的任何附图标记均不应被理解为 对保护范围的限制。 权利要求中出现的多个部分的功能可以由一个 单独的硬件或软件模块来实现。 某些技术特征出现在不同的从属权 利要求中并不意味着不能将这些技术特征进行组合以取得有益效 果。

Claims

权 利 要 求 书
1. 一种在多输入多输出系统的数据接收设备中用于向数据发送 设备反馈信道信息的方法, 包括以下步骤:
a. 确定所述数据发送设备至所述数据接收设备的发射空间相关 矩阵;
b. 确定与所述发射空间相关矩阵的特征值和特征向量有关的量 化信息, 包括: 第一数目比特的用于量化第一特征值所对应的第一 特征向量的信息; 其中所述第一特征值为主特征值;
c 将步骤 b中确定的所述量化信息发送给所述数据发送设备。
2. 根据权利要求 1所述的方法, 其特征在于, 所述量化信息还 包括: 第二数目比特的用于量化第二特征值所对应的第二特征向量 的信息、 以及与所述第一特征值和第二特征值有关的信息; 其中所 述第一数目不小于所述第二数目。
3. 根据权利要求 1或 2所述的方法, 其特征在于, 还包括步骤:
- 根据步骤 b中确定的量化信息重构发射空间相关矩阵,并根据 重构的发射空间相关矩阵确定需要反馈的预编码码字索引;
- 将所述需要反馈的预编码码字索引发送给所述数据发送设备。
4. 根据权利要求 2所述的方法, 其特征在于, 所述第二特征值 为第二大特征值。
5. 根据权利要求 1或 2所述的方法, 其特征在于, 所述步骤 b 中特征向量的量化包括向量量化或向量中的元素量化。
6. 根据权利要求 5所述的方法, 其特征在于, 所述第二特征向 量的量化样本空间包括所述第一特征向量的量化样本空间的一个子 空间。
7. 根据权利要求 1或 2所述的方法, 其特征在于, 所述步骤 a 中采用频域平均、 时域长时平均或者时域滑动平均方式来确定所述 发射空间相关矩阵。
8. 一种在多输入多输出系统的数据发送设备中用于向数据接收 设备发送数据的方法, 包括以下步骤: A. 接收来自所述数据接收设备的与发射空间相关矩阵有关的量 化信息; 其中所述量化信息包括: 第一数目比特的用于量化发射空 间相关矩阵的第一特征值所对应的第一特征向量的信息; 其中所述 第一特征值为主特征值;
B. 根据步骤 A 中接收到的所述量化信息重构发射空间相关矩 阵, 并根据所述重构的发射空间相关矩阵以及预编码码字索引对将 要发送的数据进行预编码处理。
9. 根据权利要求 8所述的方法, 其特征在于,
所述步骤 A还包括: 接收来自所述数据接收设备的预编码码字 索引;
所述步骤 B中根据所述步骤 A中接收到的所述预编码码字索引 对将要发送的数据进行预编码处理。
10. 根据权利要求 8或 9所述的方法, 其特征在于, 所述量化 信息还包括: 第二数目比特的用于量化第二特征值所对应的第二特 征向量的信息、 以及与所述第一特征值和第二特征值有关的信息; 其中所述第一数目不小于所述第二数目。
1 1. 一种在多输入多输出系统的数据接收设备中用于向数据发 送设备反馈信道信息的反馈装置, 包括:
第一确定装置, 用于确定所述数据发送设备至所述数据接收设 备的发射空间相关矩阵;
第二确定装置, 用于确定与所述发射空间相关矩阵的特征值和 特征向量有关的量化信息, 包括: 第一数目比特的用于量化第一特 征值所对应的第一特征向量的信息; 其中所述第一特征值为主特征 值;
信令发送装置, 用于将第二确定装置确定的所述量化信息发送 给所述数据发送设备。
12. 根据权利要求 1 1所述的反馈装置, 其特征在于, 所述第二 确定装置所确定的所述量化信息还包括: 第二数目比特的用于量化 第二特征值所对应的第二特征向量的信息、 以及与所述第一特征值 和第二特征值有关的信息; 其中所述第一数目不小于所述第二数目。
13. 一种在多输入多输出系统的数据发送设备中用于向数据接 收设备发送数据的数据发送装置, 包括:
信令接收装置, 用于接收来自所述数据接收设备的与发射空间 相关矩阵有关的量化信息; 其中所述量化信息包括: 第一数目比特 的用于量化发射空间相关矩阵的第一特征值所对应的第一特征向量 的信息; 其中所述第一特征值为主特征值;
预编码装置, 用于根据接收到的所述量化信息重构发射空间相 关矩阵, 并根据所述重构的发射空间相关矩阵以及预编码码字索引 对将要发送的数据进行预编码处理。
14. 根据权利要求 13所述的数据发送装置, 其特征在于, 所述信令接收装置还用于: 接收来自所述数据接收设备的预编 码码字索引;
所述预编码装置根据所述信令接收装置接收到的所述预编码码 字索引对将要发送的数据进行预编码处理。
PCT/CN2010/070166 2010-01-13 2010-01-13 多输入多输出系统中有限码本闭环反馈的方法及装置 WO2011085550A1 (zh)

Priority Applications (2)

Application Number Priority Date Filing Date Title
CN2010800344841A CN102474331A (zh) 2010-01-13 2010-01-13 多输入多输出系统中有限码本闭环反馈的方法及装置
PCT/CN2010/070166 WO2011085550A1 (zh) 2010-01-13 2010-01-13 多输入多输出系统中有限码本闭环反馈的方法及装置

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
PCT/CN2010/070166 WO2011085550A1 (zh) 2010-01-13 2010-01-13 多输入多输出系统中有限码本闭环反馈的方法及装置

Publications (1)

Publication Number Publication Date
WO2011085550A1 true WO2011085550A1 (zh) 2011-07-21

Family

ID=44303806

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2010/070166 WO2011085550A1 (zh) 2010-01-13 2010-01-13 多输入多输出系统中有限码本闭环反馈的方法及装置

Country Status (2)

Country Link
CN (1) CN102474331A (zh)
WO (1) WO2011085550A1 (zh)

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20040203473A1 (en) * 2002-03-29 2004-10-14 Jung-Tao Liu Method for closed-loop subspace transmission and reception in a two transmit n-receive antenna system
CN101594323A (zh) * 2008-05-27 2009-12-02 鼎桥通信技术有限公司 一种多输入多输出系统的发送预编码方法
CN101674115A (zh) * 2009-10-19 2010-03-17 中兴通讯股份有限公司 协方差相关矩阵的反馈传输方法及用户终端

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20040203473A1 (en) * 2002-03-29 2004-10-14 Jung-Tao Liu Method for closed-loop subspace transmission and reception in a two transmit n-receive antenna system
CN101594323A (zh) * 2008-05-27 2009-12-02 鼎桥通信技术有限公司 一种多输入多输出系统的发送预编码方法
CN101674115A (zh) * 2009-10-19 2010-03-17 中兴通讯股份有限公司 协方差相关矩阵的反馈传输方法及用户终端

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
HUAWEI: "R1-091820, Adaptive Codebook Designs for DL MIMO", 3GPP TSG RAN WG1 MEETING #57, 8 May 2009 (2009-05-08) *

Also Published As

Publication number Publication date
CN102474331A (zh) 2012-05-23

Similar Documents

Publication Publication Date Title
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
JP5698256B2 (ja) ファクタ化プリコーディングを使用するための方法と装置
JP5236753B2 (ja) 開ループ空間多重化モードでの信号送受信方法
US9083411B2 (en) Device and method of channel information feedback for multi-cell MIMO
US20130201912A1 (en) Method and apparatus for compressing channel state information based on path location information
KR101363969B1 (ko) 다중 안테나 시스템, 및 다중 안테나 시스템에서 정보를 송수신하기 위한 방법 및 장치
WO2012041107A1 (zh) 信道状态信息反馈方法及终端
WO2011098015A1 (zh) 基于码书的信道信息反馈方法、设备和系统
WO2017076347A1 (zh) 一种信道状态信息量化反馈方法及终端
JP2016040946A (ja) 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端末装置及びその方法
WO2012068880A1 (zh) 一种信道状态信息反馈方法及装置
WO2011137726A1 (zh) 一种信道状态信息上报方法及其装置
TW201001950A (en) Method and apparatus for performing multiple-input multiple-output wireless communications
WO2011160451A1 (zh) 信道状态信息的反馈方法及终端
CN111865377B (zh) 指示和确定预编码矩阵的方法以及通信装置
WO2020221117A1 (zh) 一种用于构建预编码矩阵的系数指示方法和通信装置
WO2010009640A1 (zh) 多输入多输出系统中选取预编码矩阵的方法、设备和系统
WO2012041103A1 (zh) 信道信息反馈方法及终端
CN111757382B (zh) 指示信道状态信息的方法以及通信装置
WO2020211681A1 (zh) 一种信道测量方法和通信装置
WO2011157217A1 (zh) 一种基站,以及一种信道信息获取方法及系统
WO2011160453A1 (zh) 信道状态信息的反馈方法
CN111756422B (zh) 指示信道状态信息的方法以及通信装置
WO2011124024A1 (zh) 用于在多输入多输出系统中反馈和构建相关矩阵的方法和设备

Legal Events

Date Code Title Description
WWE Wipo information: entry into national phase

Ref document number: 201080034484.1

Country of ref document: CN

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

Ref document number: 10842834

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 10842834

Country of ref document: EP

Kind code of ref document: A1