[go: up one dir, main page]

WO2011041965A1 - Method and device for constructing channel information quantization codebook of multi-antenna system - Google Patents

Method and device for constructing channel information quantization codebook of multi-antenna system Download PDF

Info

Publication number
WO2011041965A1
WO2011041965A1 PCT/CN2010/075994 CN2010075994W WO2011041965A1 WO 2011041965 A1 WO2011041965 A1 WO 2011041965A1 CN 2010075994 W CN2010075994 W CN 2010075994W WO 2011041965 A1 WO2011041965 A1 WO 2011041965A1
Authority
WO
WIPO (PCT)
Prior art keywords
channel
codeword
vector
rank
antenna
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Ceased
Application number
PCT/CN2010/075994
Other languages
French (fr)
Chinese (zh)
Inventor
陈艺戬
郁光辉
戴博
朱嵘彦
杨勋
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
ZTE Corp
Original Assignee
ZTE Corp
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 ZTE Corp filed Critical ZTE Corp
Publication of WO2011041965A1 publication Critical patent/WO2011041965A1/en
Anticipated expiration legal-status Critical
Ceased legal-status Critical Current

Links

Classifications

    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04LTRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L1/00Arrangements for detecting or preventing errors in the information received
    • H04L1/02Arrangements for detecting or preventing errors in the information received by diversity reception
    • H04L1/06Arrangements for detecting or preventing errors in the information received by diversity reception using space diversity
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04LTRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L1/00Arrangements for detecting or preventing errors in the information received
    • H04L1/0001Systems modifying transmission characteristics according to link quality, e.g. power backoff
    • H04L1/0023Systems modifying transmission characteristics according to link quality, e.g. power backoff characterised by the signalling
    • H04L1/0026Transmission of channel quality indication
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04LTRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L1/00Arrangements for detecting or preventing errors in the information received
    • H04L1/0001Systems modifying transmission characteristics according to link quality, e.g. power backoff
    • H04L1/0023Systems modifying transmission characteristics according to link quality, e.g. power backoff characterised by the signalling
    • H04L1/0028Formatting

Definitions

  • the present invention relates to the field of wireless communication technologies, and in particular, to a method and an apparatus for constructing a channel information quantization codebook in a multi-antenna system.
  • the receiving end feeds back the channel information of the transmitting end, and the transmitting end uses some transmitting precoding techniques according to the channel information obtained from the receiving end, thereby greatly improving the transmission performance.
  • the simple method of using the channel feature vector information for precoding is mainly used in Single-Input Multiple-Out-put (MIMO), and there are other better but more complicated methods. Used in multi-user MIMO.
  • each layer can transmit different data symbols, and the number of layers is equal to the rank of the channel matrix (Rank). If the complete channel information can be accurately known at the transmitting end, the obtained channel matrix information (CSI) information is used to perform linear or non-linear precoding on the layer data, so that the user receives the data when receiving the receiving end.
  • the noise ratio is the largest, and the inter-layer interference and inter-user interference are minimal.
  • the channel information can only be obtained directly and accurately at the receiving end, and the CSI needs to obtain the CSI information from the receiving end through the receiving end.
  • An important question is how to effectively quantify the feedback CSI information.
  • the feedback capacity provided by the system to CSI information is relatively limited, because the feedback amount of feedback on the entire channel information is very large. Therefore, the mainstream feedback methods are based on the quantification of codebooks.
  • 91 may be further divided into codebooks corresponding to multiple ranks, and each rank corresponding to multiple codewords to quantize the precoding matrix formed by the channel feature vectors under the rank. Since the number of ranks and non-zero feature vectors of the channel are equal, in general, the codeword with the rank N will have N columns. Therefore, we can divide the codebook 91 into multiple subcodebooks according to the Rank.
  • the performance of precoding according to CSI is the best. Due to the limitation of feedback overhead, we often use codebook based channel information to quantify feedback. Therefore, the most important part of the feedback method is the design of a codebook. In the actual communication system, the majority of the user's channel matrix ranks in Rank 1 and 2, and the Rank1 codebook and Rank2 codebook performance for channel information quantization feedback is transmitted to the entire system. It is very important. As shown in Table 1, the codebook is shown as a schematic structure divided by Rank.
  • the existing codebook for channel information quantization feedback mainly considers the channel applied by the single-polarized antenna at the transmitting end.
  • the channel is distinguished by correlation, and the strong correlation channel is used and independent.
  • the code word of the channel feature is used and independent.
  • One direction of the existing codebook design is that some of the codewords consider completely independent and uncorrelated channels, ie each element in the channel (the channel between each pair of transceiver antennas is represented as a channel element in a channel matrix) ) are independent and irrelevant (iid), at this time Rank is "the codebook is not in the same
  • the method of designing the channel codeword is as follows: Find multiple "columns of columns ("degenerate to vector when 1", so that the multiple matrices (ie subspaces) are evenly distributed in the whole space, the main means is to pass the Gera Grassmannian line compression or subspace compression method to find multiple codeword matrices in a uniform space.
  • Another direction of the existing codebook design is a single-polarized array antenna scenario with smaller antenna spacing.
  • the channel has strong correlation. At this time, the channel is not independent and irrelevant, and its feature vector shows a kind.
  • the eigenvector of the strongly correlated channel of the single-polarized antenna is [1 e e ... , which represents different phase values.
  • This model has the same model as the column vector of the Discrete Fourier Transform (DFT) matrix. Therefore, the column of the DFT matrix is selected as the codeword of the relevant channel.
  • DFT Discrete Fourier Transform
  • the existing codebook construction techniques are considered for the application scenario of a single-polarized antenna.
  • the codebook design method for the correlated channel and the uncorrelated channel for the single-polarized antenna cannot be dual-polarized.
  • the channel characteristics of the antenna are well matched, and the performance exhibited in the dual-polarized channel is poor. Therefore, there is a need for a good technique and codebook construction scheme for channel information feedback for dual-polarized channels.
  • the technical problem to be solved by the present invention is to provide a method and a device for constructing a channel information quantization codebook in a multi-antenna system, which are used for constructing a quantized codebook that satisfies the channel information feedback requirement of a dual-polarized channel, in a multi-antenna system. Good matching channel characteristics are obtained for the channels of the dual-polarized antenna.
  • the present invention provides a method for constructing a channel information quantization codebook in a multi-antenna system, including:
  • a channel codeword of rank 1 and a channel codeword of rank 2 of a channel matrix matched to the dual polarization for the correlation channel configuration are included.
  • the correlation characteristic of the channel is an associated channel
  • the step of constructing a channel codeword of rank 2 of the associated channel includes:
  • the codeword CV [U nl U n2 ] of rank 2 includes:
  • the relevant characteristics of the channel are related channels, and the polarization characteristics of the antenna at the transmitting end are extreme
  • w n is the correlation coefficient of the channel of any one of the M-dimensional vectors M ⁇ ⁇ is the relevant channel, and the antenna polarization characteristic of the transmitting end is - ⁇ /+ ⁇
  • the step of constructing the channel codeword of the rank 1 of the relevant channel includes: selecting a 2M-dimensional vector Where ⁇ " is the number of associated codewords to be generated in the codebook, ⁇ is half of the number of transmit antennas at the transmitting end; and the codeword of rank 1 is constructed using ⁇ 2 ⁇ -dimensional vectors M w. ff
  • wrada is any vector of the 2M-dimensional vector ⁇ ⁇ ⁇ , ⁇ ⁇ , is a real number, a vector consisting of two elements preceding the 2 ⁇ code word and a vector consisting of the following elements, one difference
  • any one of the vectors meets the following characteristics:
  • the present invention also provides a method for constructing a channel information quantization codebook in another multi-antenna system, comprising: constructing a channel codeword with a rank of 1 of a channel matrix according to a correlation characteristic of a channel, an antenna polarization characteristic of a transmitting end and a receiving end; a channel codeword of rank 2, forming a channel information quantization codebook, wherein the channel information quantization codebook includes a rank of a channel matrix matched to the dual polarization for the uncorrelated channel structure a channel codeword of 1 and a channel codeword of rank 2,
  • the correlation characteristic of the channel is an uncorrelated channel, and the antenna polarization characteristic of the transmitting end is - / + pole
  • the steps of constructing the channel codeword of the rank 2 of the uncorrelated channel include:
  • CWCH , where: is any vector of the M-dimensional vectors ⁇ ⁇ w w w w w w w w w w w w w w w w w w w w w w w w w w w w w w w w w w w w w w w w w w w w w w w w w w w w w w w w w w w w w w w w w w w w w w w w w w w w w w w w w w w w w w w w w w w w w w w w w w w w w w w w w w w w w w w w w w w w
  • the steps of the channel codeword with rank 2 of the track include: selecting 2 M M-dimensional vectors ⁇ ...! ⁇ And! ⁇ ... 1 ⁇ 4 ⁇ , where AT is the number of associated codewords to be generated in the codebook, and M is half the number of transmit antennas at the transmitting end, ⁇ ...! ⁇ uniformly distributed or quasi-uniformly distributed in the M-dimensional subspace; and constructing a codeword of rank 2 using the M-dimensional vectors M w. ff as C greed
  • the step of constructing a codeword of rank 2 by using the "M-dimensional vectors M - M ⁇ " further includes: obtaining the arbitrary vectors w implicates and ⁇ , the steps of obtaining the arbitrary vectors w implicates and ⁇ include:
  • Storing or real-time computing obtains a plurality of vector sets ⁇ of M dimensions, and the chord distances of any two code words of the vector set ⁇ are greater than or equal to the threshold threshold; the vector is obtained by using the method of Grassmannian theory The set ⁇ is line-compressed; any two different vectors are taken as the sum ⁇ from the compressed vector set ⁇ .
  • the present invention also provides a device for constructing a channel information quantization codebook of a multi-antenna system, comprising: a codeword construction module, which is configured to: construct a channel matrix according to a correlation characteristic of a channel, an antenna polarization characteristic of a transmitting end and a receiving end a channel codeword of rank 1 and a channel codeword of rank 2 form a channel information quantization codebook, and the channel information quantization codebook includes a channel codeword of rank 1 of a channel matrix matched with dual polarization and a rank of 2 channel codewords.
  • a codeword construction module which is configured to: construct a channel matrix according to a correlation characteristic of a channel, an antenna polarization characteristic of a transmitting end and a receiving end a channel codeword of rank 1 and a channel codeword of rank 2 form a channel information quantization codebook
  • the channel information quantization codebook includes a channel codeword of rank 1 of a channel matrix matched with dual polarization and a rank of 2 channel
  • the constructing device further includes: an antenna polarization characteristic detecting module, configured to: detect and acquire an antenna polarization characteristic of the receiving end and the transmitting end, and send the antenna polarization characteristic to the code word constructing module.
  • an antenna polarization characteristic detecting module configured to: detect and acquire an antenna polarization characteristic of the receiving end and the transmitting end, and send the antenna polarization characteristic to the code word constructing module.
  • the codeword construction module includes a vector selection unit, a correlation codeword generation unit, and an unrelated codeword generation unit;
  • the vector selection unit is configured to: select a vector for performing codeword construction
  • the associated codeword generating unit is configured to: generate a codeword using the selected vector for the relevant channel;
  • the uncorrelated codeword generating unit is configured to: generate a codeword using the selected vector for the uncorrelated channel.
  • the codeword construction module is configured to: construct a channel codeword of rank 2 for the relevant channel as follows:
  • Vector selection unit selects ⁇ "M-dimensional vectors Where ⁇ is the number of related codewords to be generated in the codebook, and ⁇ is half of the number of transmitting antennas at the transmitting end;
  • the antenna polarization characteristic detecting module is configured to: detect that the antenna polarization characteristic of the transmitting end is
  • the antenna polarization characteristic at the receiving end is 1 antenna vertical polarization
  • the codeword construction module is set to: Construct a channel codeword with a rank of 1 for the relevant channel as follows: Vector selection unit selection ⁇ "M-dimensional vector Where ⁇ is the number of related codewords to be generated in the codebook, and ⁇ is half of the number of transmitting antennas at the transmitting end;
  • the correlation codeword generating unit constructs a codeword of rank 1 using ⁇ " ⁇ dimension vector M w. ff as
  • w is the ⁇ M-dimensional vector ⁇ ⁇ ⁇ any vector t antenna polarization characteristic detection module is set to: detect the antenna polarization characteristic of the transmitting end is
  • the antenna polarization characteristic at the receiving end is 45-degree polarization of one antenna
  • the codeword construction module is set to: construct a channel codeword of rank 1 for the relevant channel as follows: Vector selection unit selection ⁇ " 2M-dimensional vectors M - W where ⁇ " is the number of relevant codewords to be generated in the codebook, and M is half the number of transmitting antennas at the transmitting end;
  • the correlation codeword generating unit constructs a codeword of rank 1 by using "2M-dimensional vectors M w. ff as Where: wrada is any vector of the 2M-dimensional vector ⁇ ⁇ ⁇ , ⁇ ⁇ , is a real number, a vector consisting of two elements preceding the 2 ⁇ code word and a vector consisting of the following elements, one difference
  • the fixed phase ⁇ e i,p , the vector consisting of the first element is ⁇ critique multiplied by a constant coefficient / resort, the vector consisting of the last element is ⁇ multiplied by a constant coefficient /iller.
  • the vector selection unit is further configured to: when constructing the associated channel codeword, selecting the arbitrary vector to meet the following characteristics:
  • Cloth or evenly distributed within - ⁇ .
  • the antenna polarization characteristic detection module is set to: detect the antenna polarization characteristic of the transmitting end as
  • the antenna polarization characteristic of the receiving end is when the multiple antennas are vertically polarized
  • the codeword construction module is set to: construct a channel codeword of rank 2 for the uncorrelated channel as follows:
  • the vector selection unit selects ⁇ "M-dimensional vectors M - M ⁇ , where ⁇ is the number of related code words to be generated in the codebook, and ⁇ is half the number of transmitting antennas at the transmitting end, ⁇ ...! ⁇ in the dimension space Uniformly distributed or quasi-uniformly distributed;
  • the irrelevant codeword generating unit constructs a codeword of rank 2 by using the "" dimensional vector M w. ff as
  • Any one of the vector t antenna polarization characteristic detecting modules is configured to: detect the antenna polarization characteristic of the transmitting end as
  • the polarization characteristics of the antenna at the receiving end are when multiple antennas are vertically polarized
  • the 4 4 code word construction module is configured to: construct a channel codeword of rank 2 for an uncorrelated channel as follows:
  • the vector selection unit selects 2 AT M-dimensional vectors ⁇ ⁇ ⁇ ⁇ w ff and ⁇ ⁇ ⁇ ⁇ 1 ⁇ 4 ⁇ , where K is the number of associated codewords to be generated in the codebook, and ⁇ is half the number of transmit antennas at the transmitting end, ⁇ ...! ⁇ Hook distribution or quasi-homogeneous distribution in the Weiweizi space;
  • the chord distance is greater than or equal to the threshold threshold, where the chord distance ⁇ , ⁇ ;) is defined as the sine of the angle between the two vectors.
  • the vector selection unit is configured to: when the arbitrary vector and ⁇ are selected, the storage or real-time calculation obtains a plurality of vector sets ⁇ of M dimensions, and the chord distances of any two code words of the vector set ⁇ are greater than or equal to the threshold width Value; the vector set ⁇ is linearly compressed by the method of Grassmannian theory; two different vectors are randomly taken from the compressed vector set ⁇ as w barn andday.
  • the method and device for constructing a channel information quantization codebook of the multi-antenna system of the present invention, and constructing a channel information quantization codebook matched with dual polarization can have higher channel information quantization feedback efficiency under the dual-polarization channel, And it still has better performance when used for single-polarized channels.
  • FIG. 1 is a schematic diagram of a construction apparatus of a channel information quantization codebook in a multi-antenna system. Preferred embodiment of the invention
  • the method and device for constructing a channel information information (CSI) quantization codebook of a MIMO system proposed by the present invention is applicable to a channel information quantization codebook structure when a channel matrix is a low rank (Rank), which is suitable for dual A configuration scheme for a polarized channel, and is compatible with a single-polarized channel.
  • CSI channel information information
  • the core idea of the present invention is to construct a channel information codebook with ranks 1 and 2 of a channel matrix according to the correlation characteristics of the channel, the antenna polarization characteristics of the transmitting end and the receiving end, and form a channel information quantization codebook, and the channel information quantization code
  • the channel codewords of ranks 1 and 2 of the channel matrix constructed for the relevant channel are included herein; and the channel information quantization codebook further includes channel codewords of ranks 1 and 2 of the channel matrix constructed for the uncorrelated channel.
  • the channel information quantization codebook may further include other possible codewords. The following describes the codeword construction for the relevant channel and the uncorrelated channel, respectively.
  • Correlation channel codeword construction method The specific steps of the method for constructing the channel information information (CSI) quantization codebook of the MIMO system of the present invention are as follows: According to the correlation characteristics of the channel, the antenna polarization characteristics of the transmitting end and the receiving end, constructing the rank of the channel matrix is 1 and The channel codeword of 2 forms a channel information quantization codebook, and the channel information quantization codebook includes channel codewords of ranks 1 and 2 of the channel matrix constructed for the relevant channel.
  • CSI channel information information
  • the specific characteristics of the channel are as follows:
  • the specific steps of constructing the channel codeword of the rank 2 of the relevant channel are as follows: Select an M-dimensional vector M - w. ff , where K is the correlation code to be generated in the codebook
  • M is half of the number of transmitting antennas at the transmitting end; in general, the number of transmitting antennas is even, we assume that the transmitting end is 2M transmitting antennas. At this time, we have to construct a codebook of 2M antennas.
  • the codeword of the Rank1 of the Rank1 of the Rank1 defined in the LTE physical layer protocol 3GPP 36.211 may be used as the W l ⁇
  • the channel's correlation matrix (defined as the channel matrix)
  • the rank * channel matrix has a form similar to a block diagonal matrix whose feature vectors are close to some of the above models. Therefore, the code words constructed by the above model have very good performance for quantifying channel information (channel feature vector).
  • Transmitter antenna Polarization is the most common scenario, and the receiving end is a laptop or mobile phone.
  • one antenna at the receiving end When placed directly, one antenna at the receiving end can be regarded as vertical polarization.
  • the correlation characteristic of the channel is a related channel
  • the polarization characteristic of the antenna at the transmitting end is - / + ⁇ polarization
  • the polarization characteristic of the antenna at the receiving end is 1 day.
  • the specific steps of constructing the channel codeword of the rank 1 of the relevant channel are as follows: Select an M-dimensional vector M - w. ff , where K is the number of relevant codewords to be generated in the codebook, M is half the number of transmitting antennas at the transmitting end; constructing a codeword of rank 1 using RW M-dimensional vectors M w. ff
  • w n is any one of the above M-dimensional vectors M ⁇ ⁇ .
  • Transmitter antenna - /+ Polarization is the most common scene, the receiving end is the mobile phone 45 degree placement
  • one antenna at the receiving end can be regarded as a 45-degree polarization.
  • the correlation characteristic of the channel is the relevant channel
  • the polarization characteristic of the antenna at the transmitting end is polarization
  • the polarization characteristic of the antenna at the receiving end is 1 antenna.
  • the specific steps of constructing the channel codeword of the rank 1 of the relevant channel are as follows: Select a 2M-dimensional vector M - M ⁇ , where ⁇ " is the number of relevant codewords to be generated in the codebook, ⁇ is half the number of transmitting antennas at the transmitting end; Constructing a codeword with rank 1 as RW using ⁇ 2M-dimensional vectors M w. ff
  • wrada is any vector of the 2M-dimensional vector ⁇ ⁇ , ⁇ ⁇ , is a real number, a vector consisting of two elements preceding the 2 ⁇ code word and a vector consisting of the following elements, one difference
  • the fixed phase ⁇ e i,p , the vector consisting of the first element is ⁇ critique multiplied by a constant coefficient / resort, the vector consisting of the last element is ⁇ multiplied by a constant coefficient /iller.
  • the feature vector approximation has characteristics matching it, and thus has good performance.
  • ⁇ ⁇ is close to 1 in the actual situation
  • any one of the "" vectors has the following characteristics:
  • Cloth or evenly distributed within - ⁇ .
  • a channel codeword of a channel matrix with a rank of 1 and 2 according to a correlation characteristic of the channel, an antenna polarization characteristic of the transmitting end and the receiving end, and forming a channel information quantization codebook, where the channel information quantization codebook includes an uncorrelated channel
  • the channel codewords of the constructed channel matrix are 1 and 2 channel codewords. Transmitter antenna polarization is the most common scenario, and the receiving end is a laptop or mobile phone.
  • the correlation characteristic of the channel is an uncorrelated channel
  • the polarization characteristic of the antenna at the transmitting end is polarization
  • the polarization characteristic of the antenna at the receiving end is polarization
  • the relevant characteristics of the channel are uncorrelated channels, and the polarization characteristics of the antenna at the transmitting end are - / + ⁇ polarization,
  • the antenna polarization characteristic of the receiving end is multiple antennas.
  • the rank of the uncorrelated channel is 2
  • w cauliflower and ⁇ are arbitrary vectors
  • the chord distances of whan and ⁇ are greater than or equal to the threshold width
  • the chord distance ⁇ , ⁇ is defined as and ⁇ : the sine of the angle between the two vectors .
  • the ⁇ ⁇ preclude any vector obtained by the following steps:! Stored or calculated in real time to obtain a plurality of vectors ⁇ dimensional set [Omega], [Omega] in the set of vectors of any two codewords chord distance is greater than or equal to
  • the threshold threshold is obtained by linearly compressing the vector set ⁇ by using the Grassmannian theory; and taking two different vectors as the sum from the compressed vector set ⁇ .
  • Method 1 When there are many vectors in ⁇ , for any i and j, This has the best performance.
  • Method 2 When there are many vectors in ⁇ , for any i and j, This has the best performance.
  • the model can match the approximate feature of the feature vector under the dual-polarization non-correlated channel under the current polarization angle of the transceiver, and has good quantization performance.
  • the receiving end After the receiving end itself generates or obtains the channel information codebook constructed according to the above method, the codeword matching the channel is searched from the codebook according to the channel condition, and the codeword information or the codeword serial number is fed back to the transmitting end, and the transmitting end Finding the same codebook according to the codeword information or the codeword serial number, precoding according to the found codeword, can greatly save resources, and at the same time, since the codebook includes the codeword matched with the dual polarization, it is also compatible with the monopole. In the case of the invention, the construction method of the invention has better compatibility.
  • the present invention also provides a channel device information (CSI) quantization codebook construction apparatus 10 for a MIMO system. As shown in FIG. 1, the method includes:
  • the channel detecting module 11 is configured to detect the relevant characteristics of the acquired channel, determine whether the channel is an associated channel or an uncorrelated channel, and send the relevant characteristic result of the channel to the codeword construction module 13;
  • the antenna polarization characteristic detecting module 12 is configured to detect the antenna polarization characteristics of the receiving end and the transmitting end, and send the antenna polarization characteristic to the code word constructing module 13;
  • the codeword construction module 13 forms a channel information quantized codebook by constructing a channel codeword with ranks 1 and 2 of the channel matrix according to the correlation characteristics of the channel, the antenna polarization characteristics of the transmitting end and the receiving end, and the channel information quantization codebook includes The channel codewords of ranks 1 and 2 of the channel matrix constructed for the relevant channel further include channel codewords of ranks 1 and 2 of the channel matrix constructed for the uncorrelated channel.
  • the codeword construction module 13 is further divided into: a vector selection unit 131, a correlation codeword generation unit 132, an unrelated codeword generation unit 133;
  • the vector selection unit 131 is configured to select a vector for performing codeword construction
  • the associated codeword generating unit 132 is configured to generate a codeword by using the selected vector for the relevant channel;
  • the uncorrelated codeword generating unit 133 is configured to generate a codeword by using the selected vector for the uncorrelated channel.
  • the device for constructing the channel information information (CSI) of the MIMO system may be located at the transmitting end or at the receiving end. When it is located at the transmitting end or the receiving end, the working mode may be completed once.
  • the code word, and the constructed code word is stored at the transmitting end or the receiving end, and can also be constructed in real time.
  • the constructing device may also be located on a computer device or server other than the transmitting end or the receiving end, and all code words are constructed once, and the complete codebook is stored to the transmitting end and the receiving end.
  • the apparatus for constructing a channel information quantization codebook of the above MIMO system wherein the correlation characteristic of the channel may be acquired by the channel detection module, or may be obtained by other means, such as parameter transmission.
  • the antenna polarization characteristics can be obtained by the antenna polarization characteristic detection module, or can be obtained by other means, for example, from a dedicated channel correlation module.
  • the following is a specific codeword construction scheme in combination with specific channel characteristics and antenna polarization characteristics of the receiving end and the transmitting end.
  • the channel detection module 11 detects that the correlation characteristic of the channel is obtained as a correlation channel, and the specific scheme of the codeword structure of the codeword construction module 13 for the rank 2 of the relevant channel is as follows:
  • the vector selecting unit 131 selects ⁇ M-dimensional vectors M - W , where K is the number of related code words to be generated in the codebook, and M is half of the number of transmitting antennas at the transmitting end;
  • the channel detecting module 11 detects that the correlation characteristic of the channel is the relevant channel, and the antenna polarization characteristic detecting module 12 detects that the antenna polarization characteristic of the transmitting end is polarization, and the antenna pole of the receiving end
  • the vector selection unit 131 selects the M-dimensional vectors M - W where K For the number of related codewords to be generated in the codebook, M is half of the number of transmit antennas at the transmitting end;
  • Correlation code word generation unit 132 using the ⁇ M-dimensional vector M w ff configured codeword rank 1 in which:.
  • W n is the ⁇ M-dimensional vector M ⁇ ⁇ t vector of any one of
  • the channel detecting module 11 detects that the correlation characteristic of the channel is the relevant channel, and the antenna polarization characteristic detecting module 12 detects that the antenna polarization characteristic of the transmitting end is - polarization, and the antenna pole of the receiving end
  • the vector selection unit 131 selects two 2M-dimensional vectors. Wherein, the number of related codewords to be generated in the codebook, and M is half of the number of transmitting antennas at the transmitting end;
  • the correlation codeword generating unit 132 constructs a codeword of rank 1 using K 2M-dimensional vectors MM ⁇
  • wrada is any vector of the 2M-dimensional vector ⁇ ⁇ , ⁇ ⁇ , is a real number, a vector consisting of two elements preceding the 2 ⁇ code word and a vector consisting of the following elements, one difference
  • the vector consisting of the last M elements is ⁇ critique multiplied by a constant coefficient of the codeword Rfstory with a rank of 1, when ⁇ is equal to 1,
  • the vector selection unit 131 selects any one of the vectors, and meets the following characteristics:
  • the channel detecting module 11 detects that the correlation characteristic of the channel is an uncorrelated channel
  • the antenna polarization characteristic detecting module 12 detects that the antenna polarization characteristic of the transmitting end is polarization, and the antenna of the receiving end
  • the vector selection unit 131 selects the M-dimensional vectors M - W , K is the number of related codewords to be generated in the codebook, M is half of the number of transmitting antennas at the transmitting end, and ⁇ is evenly distributed or quasi-uniformly distributed in the M-dimensional subspace;
  • the channel detecting module 11 detects that the correlation characteristic of the channel is an uncorrelated channel
  • the antenna polarization characteristic detecting module 12 detects that the antenna polarization characteristic of the transmitting end is polarization, and the antenna of the receiving end
  • the specific scheme of the channel codeword construction with the rank rank of 2 is as follows:
  • the vector selection unit 131 selects 2 M M-dimensional vectors ⁇ ...! ⁇ And! ⁇ ... 1 ⁇ 4 ⁇ , where K is the number of related codewords to be generated in the codebook, M is half of the number of transmitting antennas at the transmitting end, and ⁇ ⁇ ⁇ ⁇ w ff is evenly distributed or quasi-uniformly distributed in the M-dimensional subspace;
  • the irrelevant codeword generating unit 133 constructs a codeword of rank 2 using ⁇ M-dimensional vectors M w. ff
  • chord distance is greater than or equal to the threshold width, where the chord distance fl nie, ) is defined as willer and ⁇ : the sine of the angle between the two vectors.
  • the vector selection unit 131 selects the arbitrary vector ⁇
  • the storage or real-time calculation obtains a plurality of vector sets ⁇ of the M dimension, and the chord distances of any two codewords of the vector set ⁇ are greater than or equal to the threshold threshold; ⁇
  • the line set ⁇ is linearly compressed by the method of Grassmannian theory; arbitrarily take two different vectors from the compressed vector set ⁇ as w Helen and.
  • the compressed vector set ⁇ for Any i and j , ⁇ ⁇ .
  • the antenna is polarized to the most common scene, and the receiving end is a laptop or a mobile phone.
  • the receiving end polarization angle information may not be obtained, and it may change at any time: Therefore, the following hybrid method can also be used for performance tradeoffs.
  • Example 4 8 adaptive dual polarization related channel codeword construction in 8 antenna Rank2
  • the 32-dimensional two-dimensional large-diameter 4-dimensional vector "i" is w l - w 32 , as shown in Table 3.
  • the first column of W n is denoted by perhaps (:,1)
  • the second column of W n is denoted by perhaps (:,2)
  • the construction codeword is admir(:,1) attend(:,2)
  • N antennas in the 8 antenna Rank2 are adapted to the dual polarization uncorrelated channel codeword construction.
  • the method and device for constructing a channel information quantization codebook of the multi-antenna system of the present invention, and constructing a channel information quantization codebook matched with dual polarization can have higher channel information quantization feedback efficiency under the dual-polarization channel, And it still has better performance when used for single-polarized channels.

Landscapes

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

Abstract

The invention discloses a method and device for constructing the channel information quantization codebook of a multi-antenna system. According to the correlation characteristic of the channel and the antenna polarization characteristic of the transmitting terminal and the receiving terminal, constructing channel code words whose ranks of channel matrix are 1 or 2 to form the channel information quantization codebook. The channel information quantization codebook includes channel code words whose ranks of channel matrix are matching with the dual polarization and are 1 or 2. Applying the invention, the quantization codebook satisfying the channel information feedback request of the dual polarization channel can be constructed and the well matched channel characteristic for the channel of the dual polarization antenna in the multi-antenna system can be obtained.

Description

多天线系统中信道信息量化码本的构造方法及装置  Method and device for constructing channel information quantization codebook in multi-antenna system

技术领域 Technical field

本发明涉及无线通信技术领域, 尤其涉及多天线系统中信道信息量化码 本的构造方法及装置。  The present invention relates to the field of wireless communication technologies, and in particular, to a method and an apparatus for constructing a channel information quantization codebook in a multi-antenna system.

背景技术 Background technique

在无线通信中, 如果发送端和接收端都使用多根天线, 可釆取空间复用 的方式来获取更高的速率。相对于一般的空间复用方法, 一种增强的技术是: 接收端反馈给发送端信道信息, 发送端根据从接收端获得的信道信息使用一 些发射预编码技术, 极大地提高传输性能。 简单的利用方法即直接使用信道 特征矢量信息进行预编码, 主要用于单用户多输入多输出 (Multiple-Input Multiple-Out-put, MIMO )中, 也有其它一些更优但更复杂的方法, 主要用于 多用户 MIMO中。  In wireless communication, if multiple antennas are used by both the transmitting end and the receiving end, spatial multiplexing can be used to obtain a higher rate. Compared with the general spatial multiplexing method, an enhanced technology is: the receiving end feeds back the channel information of the transmitting end, and the transmitting end uses some transmitting precoding techniques according to the channel information obtained from the receiving end, thereby greatly improving the transmission performance. The simple method of using the channel feature vector information for precoding is mainly used in Single-Input Multiple-Out-put (MIMO), and there are other better but more complicated methods. Used in multi-user MIMO.

在发送端我们定义层的概念, 在同一时频资源上, 每层可以传输不同的 数据符号, 层数等于信道矩阵的秩(Rank ) 。 如果在发送端能够准确的知道 完整信道信息, 那么利用获取的信道矩阵(Channel State Information, CSI ) 信息, 对层上数据进行线性或非线性的预编码, 使得到达接收端时用户接收 数据的信噪比最大, 层间干扰和用户间干扰最小。  At the transmitting end, we define the concept of layers. On the same time-frequency resource, each layer can transmit different data symbols, and the number of layers is equal to the rank of the channel matrix (Rank). If the complete channel information can be accurately known at the transmitting end, the obtained channel matrix information (CSI) information is used to perform linear or non-linear precoding on the layer data, so that the user receives the data when receiving the receiving end. The noise ratio is the largest, and the inter-layer interference and inter-user interference are minimal.

如果能够准确的获取该 CSI信息, 那么就可以做到最优的预编码。 然而, 信道信息(CSI )往往只有在接收端才能直接准确的获取, 而在发送端想要获 取 CSI—般只能通过接收端向发送端进行 CSI信息的反馈。 一个重要的问题 是如何有效的量化反馈 CSI的信息。 在目前主流的标准中, 系统提供给 CSI 信息的反馈容量都是比较有限的, 由于反馈整个信道信息的反馈量是十分大 的。 因此主流的反馈方法都是基于码本的量化方式。  If the CSI information can be accurately obtained, optimal precoding can be achieved. However, the channel information (CSI) can only be obtained directly and accurately at the receiving end, and the CSI needs to obtain the CSI information from the receiving end through the receiving end. An important question is how to effectively quantify the feedback CSI information. In the current mainstream standards, the feedback capacity provided by the system to CSI information is relatively limited, because the feedback amount of feedback on the entire channel information is very large. Therefore, the mainstream feedback methods are based on the quantification of codebooks.

基于码本的信道信息量化反馈, 其基本原理如下: 假设有限反馈信道容 量为 B bps/Hz, 那么可用的码字的个数为 w = 2S个,信道矩阵的特征矢量空间 经过量化构成码本空间 9 = , F2… } ,发送端与接收端共同保存或实时产生 此码本(收发端相同) 。 对每次信道实现 H, 接收端根据一定准则从码本空 间 中选择一个与信道最匹配的码字 并将码字序号 i反馈回发送端。发送 端根据此码字序号 i找到对应的预编码码字 获得信道信息,主要为信道的 特征矢量信息。 The basic principle of channel information based channel information quantization feedback is as follows: Assuming that the limited feedback channel capacity is B bps/Hz, then the number of available codewords is w = 2S , and the feature vector space of the channel matrix is quantized to form a codebook. Space 9 = , F 2 ... }, the sender and receiver are saved together or generated in real time This codebook (same as the transceiver). For each channel implementation H, the receiving end selects a codeword that best matches the channel from the codebook space according to a certain criterion and feeds back the codeword sequence number i back to the transmitting end. The transmitting end finds the corresponding precoding codeword according to the codeword sequence number i to obtain channel information, which is mainly feature vector information of the channel.

一般来说 91可以进一步的被划分为多个 Rank对应的码本,每个 Rank下 会对应多个码字来量化该 Rank下的信道特征矢量构成的预编码矩阵。由于信 道的 Rank和非零特征矢量个数是相等的, 因此, 一般来说 Rank为 N时的码 字都会有 N列。 所以我们可以把码本 91按 Rank分为多个子码本。  In general, 91 may be further divided into codebooks corresponding to multiple ranks, and each rank corresponding to multiple codewords to quantize the precoding matrix formed by the channel feature vectors under the rank. Since the number of ranks and non-zero feature vectors of the channel are equal, in general, the codeword with the rank N will have N columns. Therefore, we can divide the codebook 91 into multiple subcodebooks according to the Rank.

在能够完全准确获取信道信息 CSI情况下, 根据 CSI进行预编码的性能 是最好的, 由于反馈开销的限制, 我们往往釆用基于码本的信道信息量化反 馈。 因此反馈方法中最重要的一部分内容一码本的设计就很重要。 在实际的 通信系统中,用户的信道矩阵的秩绝大多数的情况下都集中在 Rank为 1和 2, 此时用于信道信息量化反馈的 Rankl码本和 Rank2码本性能对整个系统传输 性能就很关键。 如表 1所列, 显示了码本 ^按 Rank划分的示意结构。  In the case of fully accurate acquisition of channel information CSI, the performance of precoding according to CSI is the best. Due to the limitation of feedback overhead, we often use codebook based channel information to quantify feedback. Therefore, the most important part of the feedback method is the design of a codebook. In the actual communication system, the majority of the user's channel matrix ranks in Rank 1 and 2, and the Rank1 codebook and Rank2 codebook performance for channel information quantization feedback is transmitted to the entire system. It is very important. As shown in Table 1, the codebook is shown as a schematic structure divided by Rank.

表 1  Table 1

Figure imgf000003_0001
Figure imgf000003_0001

现有的用于信道信息量化反馈的码本主要是考虑发送端单极化天线应用 的信道, 在这种单极化天线配置情况下再以相关性来区别信道, 使用适应强 相关信道和独立信道特征的码字。  The existing codebook for channel information quantization feedback mainly considers the channel applied by the single-polarized antenna at the transmitting end. In this single-polarized antenna configuration, the channel is distinguished by correlation, and the strong correlation channel is used and independent. The code word of the channel feature.

( 1 )现有码本设计的一个方向是其中一些码字考虑完全独立不相关的信 道, 即信道中每个元素 (每一个收发天线对之间的信道被表示为一个信道矩 阵中的信道元素)都是独立不相关 (i.i.d ) 的, 此时 Rank为"的码本中不相 关信道码字的设计的方法为: 寻找多个"列的矩阵( "为 1时退化到矢量) , 使得这多个矩阵(即子空 间)在全空间内均匀分布, 主要的手段是通过格拉斯曼( Grassmannian )线压 缩或子空间压缩的方法, 来找到全空间内均勾分布的多个码字矩阵。 (1) One direction of the existing codebook design is that some of the codewords consider completely independent and uncorrelated channels, ie each element in the channel (the channel between each pair of transceiver antennas is represented as a channel element in a channel matrix) ) are independent and irrelevant (iid), at this time Rank is "the codebook is not in the same The method of designing the channel codeword is as follows: Find multiple "columns of columns ("degenerate to vector when 1", so that the multiple matrices (ie subspaces) are evenly distributed in the whole space, the main means is to pass the Gera Grassmannian line compression or subspace compression method to find multiple codeword matrices in a uniform space.

( 2 )现有码本设计的另外一个方向是考虑较小天线间距的单极化阵列天 线场景, 信道具有较强的相关性, 此时信道不是独立不相关的, 其特征矢量 表现出一种固定的模型, 例如: 8 天线时, 单极化天线强相关信道的特征矢 量为 [1 e e … , 其中 表示不同的相位值。 这种模型与离散傅 立叶变换( Discrete Fourier transform, DFT )矩阵的列矢量有相同的模型, 因 此, 选取 DFT矩阵的列作为相关信道的码字。 (2) Another direction of the existing codebook design is a single-polarized array antenna scenario with smaller antenna spacing. The channel has strong correlation. At this time, the channel is not independent and irrelevant, and its feature vector shows a kind. For a fixed model, for example: 8 antenna, the eigenvector of the strongly correlated channel of the single-polarized antenna is [1 e e ... , which represents different phase values. This model has the same model as the column vector of the Discrete Fourier Transform (DFT) matrix. Therefore, the column of the DFT matrix is selected as the codeword of the relevant channel.

现有的反馈装置和码本构造思想被应用在各种主流标准中,如 3GPP LTE 就釆用了该装置和方法。 但是现有的用于信道信息量化反馈的码本设计主要 是针对单极化天线信道的相关情况和不相关情况考虑的, 而双极化天线由于 在实际应用中具有更广泛的应用前景, 逐渐成为主流的应用, 尤其是天线数 大于等于 4时, 但是双极化信道表现出的一些特征往往是非常复杂的, 例如 双极化的相关信道, 仅仅在同一个极化方向上表现出强相关性, 而在极化方 向之间则不具备强相关性, 双极化的不相关信道, 在同一个极化方向上, 表 现出独立的性质, 在极化方向之间表现出的关系却并不是独立的。  Existing feedback devices and codebook construction ideas are applied in various mainstream standards, such as 3GPP LTE. However, the existing codebook design for channel information quantization feedback is mainly considered for the related situation and irrelevant situation of the single-polarized antenna channel, and the dual-polarized antenna has a wider application prospect in practical applications, and gradually Become a mainstream application, especially when the number of antennas is greater than or equal to 4, but some characteristics of dual-polarized channels are often very complex, such as dual-polarized correlated channels, showing strong correlation only in the same polarization direction. Sex, but there is no strong correlation between the polarization directions. The dual-polarized uncorrelated channels exhibit independent properties in the same polarization direction, and the relationship between the polarization directions is Not independent.

目前,现有的码本构造技术都是针对单极化天线的应用场景进行考虑的, 上述现有技术中针对单极化天线时相关信道和非相关信道的码本设计方法不 能在双极化天线的信道中很好的匹配信道特性, 在双极化信道中表现出的性 能较差。 因而, 目前就需要一种很好的针对双极化信道的信道信息反馈的技 术和码本构造方案的出现。  At present, the existing codebook construction techniques are considered for the application scenario of a single-polarized antenna. In the above prior art, the codebook design method for the correlated channel and the uncorrelated channel for the single-polarized antenna cannot be dual-polarized. The channel characteristics of the antenna are well matched, and the performance exhibited in the dual-polarized channel is poor. Therefore, there is a need for a good technique and codebook construction scheme for channel information feedback for dual-polarized channels.

发明内容 Summary of the invention

本发明所要解决的技术问题在于, 提供一种多天线系统中信道信息量化 码本的构造方法及装置, 用于构造满足双极化信道的信道信息反馈要求的量 化码本, 在多天线系统中对双极化天线的信道获得良好的匹配信道特性。 为了解决上述问题, 本发明提出了一种多天线系统中信道信息量化码本 的构造方法, 包括: The technical problem to be solved by the present invention is to provide a method and a device for constructing a channel information quantization codebook in a multi-antenna system, which are used for constructing a quantized codebook that satisfies the channel information feedback requirement of a dual-polarized channel, in a multi-antenna system. Good matching channel characteristics are obtained for the channels of the dual-polarized antenna. In order to solve the above problems, the present invention provides a method for constructing a channel information quantization codebook in a multi-antenna system, including:

根据信道的相关特性、 发送端和接收端的天线极化特性, 构造信道矩阵 的秩为 1的信道码字和秩为 2的信道码字, 形成信道信息量化码本, 所述信 道信息量化码本包括针对相关信道构造的与双极化匹配的信道矩阵的秩为 1 的信道码字和秩为 2的信道码字。 所述信道的相关特性为相关信道, 该构造相关信道的秩为 2的信道码字 的步骤包括:  Constructing a channel codeword of rank 1 of the channel matrix and a channel codeword of rank 2 according to relevant characteristics of the channel, antenna polarization characteristics of the transmitting end and the receiving end, forming a channel information quantization codebook, and the channel information quantization codebook A channel codeword of rank 1 and a channel codeword of rank 2 of a channel matrix matched to the dual polarization for the correlation channel configuration are included. The correlation characteristic of the channel is an associated channel, and the step of constructing a channel codeword of rank 2 of the associated channel includes:

选取 个 M维矢量 M -w.ff, 其中, K为码本中需要产生的相关码字个 数, M为发送端的发射天线数目的一半; 以及 Selecting an M-dimensional vector M - w. ff , where K is the number of related code words to be generated in the codebook, and M is half of the number of transmitting antennas at the transmitting end;

利用 ^个 M维矢量 M -M^构造秩为 2的码字为 0^=^^ Un2], 其中, Constructing a codeword with a rank of 2 using ^M-dimensional vectors M -M^ is 0^=^^ U n2 ], where

= αΛ +*Α , "Ac 为常数, w„为所述^"

Figure imgf000005_0003
= c u, +d I 个 M维矢量 ^…! ^中任意一个矢量。 所述秩为 2的码字 CV„ = [Unl Un2 ]包括: = αΛ + * Α , "Ac is a constant, w„ is the ^"
Figure imgf000005_0003
= cu, +d I M-dimensional vectors ^...! Any vector in ^. The codeword CV = [U nl U n2 ] of rank 2 includes:

CW„ = CW„ =

或(W„Or (W„

Figure imgf000005_0001
所述信道的相关特性为相关信道, 发送端的天线极化特性为 极
Figure imgf000005_0001
The relevant characteristics of the channel are related channels, and the polarization characteristics of the antenna at the transmitting end are extreme

4 4 化, 接收端的天线极化特性为 1根天线垂直极化时, 该构造相关信道的秩为 1的信道码字的步骤包括: 选取 个 M维矢量 M -w.ff, 其中, K为码本中需要产生的相关码字个 数, M为发送端的发射天线数目的一半; 以及 利用 ^个 M维矢量 M w.ff构造秩为 1的码字为 RW„ =4, when the antenna polarization characteristic of the receiving end is 1 antenna vertical polarization, the step of constructing the channel codeword of the rank 1 of the relevant channel includes: selecting an M-dimensional vector M - w. ff , where K is The number of associated codewords to be generated in the codebook, M is half of the number of transmit antennas at the transmitting end; and the codeword with rank 1 is constructed using ^M-dimensional vectors M w. ff as RW„ =

Figure imgf000005_0002
wn为所述 ^个 M维矢量 M · · 中任意一个矢量 t 所述信道的相关特性为相关信道, 发送端的天线极化特性为 -^/+^极
Figure imgf000005_0002
w n is the correlation coefficient of the channel of any one of the M-dimensional vectors M · · is the relevant channel, and the antenna polarization characteristic of the transmitting end is -^/+^

4 4 化, 接收端的天线极化特性为 1 根天线 45度极化时, 该构造相关信道的秩 为 1的信道码字的步骤包括: 选取 个 2M维矢量

Figure imgf000006_0001
其中, ^"为码本中需要产生的相关码字 个数, Μ为发送端的发射天线数目的一半; 以及 利用 ^个 2Μ维矢量 M w.ff构造秩为 1的码字为 RW„4 4, when the antenna polarization characteristic of the receiving end is 45-degree polarization of one antenna, the step of constructing the channel codeword of the rank 1 of the relevant channel includes: selecting a 2M-dimensional vector
Figure imgf000006_0001
Where ^" is the number of associated codewords to be generated in the codebook, Μ is half of the number of transmit antennas at the transmitting end; and the codeword of rank 1 is constructed using ^2Μ-dimensional vectors M w. ff

Figure imgf000006_0002
其中: w„为所述 ^个 2M维矢量 Μ · · 中任意一个矢量, μη , 为实数, 2Μ维的码字的前面 Μ个元素组成的矢量和后面 Μ个元素组成的矢量,相差 一个固定的相位^ ei,p= , 前 Μ个元素组成的矢量为 ^„乘以一个常系数 /„, 后 Μ个元素组成的矢量为 ^乘以一个常系数/„。 所述秩为 1的码字 RW„ = , 在 , ^等于 1时, 为 n =
Figure imgf000006_0002
Where: w„ is any vector of the 2M-dimensional vector Μ · · , μ η , is a real number, a vector consisting of two elements preceding the 2 Μ code word and a vector consisting of the following elements, one difference The fixed phase ^ e i,p = , the vector consisting of the first element is ^„ multiplied by a constant coefficient / „, the vector consisting of the last element is ^ multiplied by a constant coefficient / „. The codeword RW „ = of rank 1 is n = when ^ is equal to 1.

Figure imgf000006_0003
上述的相关信道码字构造中, 所述任意一个矢量 符合以下特征:
Figure imgf000006_0003
In the above related channel codeword structure, any one of the vectors meets the following characteristics:

1€ =τχ\ d … 其中, τ为一个复常数, 为一个复常数值, θ · · ·θκ 〜 i 1€ =τχ\ d ... where τ is a complex constant and is a complex constant value, θ · · · θ κ ~ i

2 2  twenty two

布, 或在 - 〜 内均匀分布 t Cloth, or - the uniform distribution t ~

3 3  3 3

本发明还提供另一种多天线系统中信道信息量化码本的构造方法,包括: 根据信道的相关特性、 发送端和接收端的天线极化特性, 构造信道矩阵 的秩为 1的信道码字和秩为 2的信道码字, 形成信道信息量化码本, 所述信 道信息量化码本包括针对不相关信道构造的与双极化匹配的信道矩阵的秩为 1的信道码字和秩为 2的信道码字, The present invention also provides a method for constructing a channel information quantization codebook in another multi-antenna system, comprising: constructing a channel codeword with a rank of 1 of a channel matrix according to a correlation characteristic of a channel, an antenna polarization characteristic of a transmitting end and a receiving end; a channel codeword of rank 2, forming a channel information quantization codebook, wherein the channel information quantization codebook includes a rank of a channel matrix matched to the dual polarization for the uncorrelated channel structure a channel codeword of 1 and a channel codeword of rank 2,

所述信道的相关特性为不相关信道,发送端的天线极化特性为 - /+ 极  The correlation characteristic of the channel is an uncorrelated channel, and the antenna polarization characteristic of the transmitting end is - / + pole

4 4 化, 接收端的天线极化特性为多根天线垂直极化时, 该构造不相关信道的秩 为 2的信道码字的步骤包括:  In the case where the antenna polarization characteristic of the receiving end is vertical polarization of a plurality of antennas, the steps of constructing the channel codeword of the rank 2 of the uncorrelated channel include:

选取 个 M维矢量 M -w.ff , 其中, K为码本中需要产生的相关码字个 数, M为发送端的发射天线数目的一半, ^…!^在 M维子空间内均匀分布 或准均匀分布; 以及 利用 ^个 M 维矢量 M w.ff构造秩为 2 的码字为 Cf „Select an M-dimensional vector M - w. ff , where K is the number of related code words to be generated in the codebook, and M is half the number of transmitting antennas at the transmitting end, ^...! ^ uniformly distributed or quasi-uniformly distributed in the M-dimensional subspace; and constructing a codeword of rank 2 using the M-dimensional vectors M w. ff as Cf „

Figure imgf000007_0001
Figure imgf000007_0001

CW„ = , 其中: 为所述 ^个 M维矢量 Μ ·· 中任意一个矢量 w„ - w„ 所述信道的相关特性为不相关信道,发送端的天线极化特性为 - /+ 极 化, 接收端的天线极化特性为多根天线 垂直极化时, 该构造不相关信 CW„ = , where: is any vector of the M-dimensional vectors Μ ·· w w w w w w w w w w w w w w w w w w w w w w w w w w w w w w w w w w w w w w w w w w w w w w w w w w w w w w w w w w w w w w w w w w w w w w w w w w w w w w When the antenna polarization characteristic at the receiving end is vertical polarization of multiple antennas, the configuration is uncorrelated.

4 4  4 4

道的秩为 2的信道码字的步骤包括: 选取 2ΛΓ个 M维矢量 ^…!^和!^… ¼^ , 其中, AT为码本中需要产生 的相关码字个数, M为发送端的发射天线数目的一半, ^…!^在 M维子空 间内均匀分布或准均匀分布; 以及 利用 ^个 M 维矢量 M w.ff构造秩为 2 的码字为 C „The steps of the channel codeword with rank 2 of the track include: selecting 2 M M-dimensional vectors ^...! ^And! ^... 1⁄4^ , where AT is the number of associated codewords to be generated in the codebook, and M is half the number of transmit antennas at the transmitting end, ^...! ^ uniformly distributed or quasi-uniformly distributed in the M-dimensional subspace; and constructing a codeword of rank 2 using the M-dimensional vectors M w. ff as C „

Figure imgf000007_0002
Figure imgf000007_0002

Wn W„ W n W„

cw„ = , 其中: 和^为任意的矢量, w„和^的弦距离大于或等 w„ -w 于门限阔值,其中弦距离 ^^^,^ 定义为 和^:两个矢量之间夹角的求正 弦值。 所述利用 ^"个 M维矢量 M - M^构造秩为 2的码字的步骤还包括: 获得 所述任意矢量 w„和^ , 所述获得所述任意矢量 w„和^的步骤包括: Cw„ = , where: and ^ are arbitrary vectors, the chord distances of w„ and ^ are greater than or equal to w„ -w at the threshold threshold, where the chord distance ^^^,^ is defined as and ^: between the two vectors Find the sine of the angle. The step of constructing a codeword of rank 2 by using the "M-dimensional vectors M - M^" further includes: obtaining the arbitrary vectors w „ and ^, the steps of obtaining the arbitrary vectors w „ and ^ include:

存储或实时计算获得 M维的多个矢量集合 Ω , 矢量集合 Ω中任意 2个 码字的弦距离都大于或等于所述门限阔值; 通过釆用格拉斯曼 Grassmannian 理论的方法对所述矢量集合 Ω进行线压缩;从压缩后的矢量集合 Ω中任意取 2 个不同的矢量作为 和^。  Storing or real-time computing obtains a plurality of vector sets Ω of M dimensions, and the chord distances of any two code words of the vector set Ω are greater than or equal to the threshold threshold; the vector is obtained by using the method of Grassmannian theory The set Ω is line-compressed; any two different vectors are taken as the sum ^ from the compressed vector set Ω.

从压缩后的矢量集合 Ω中任意选取 w„和^的步骤包括: 从压缩后的矢量集合 Ω中任意选取 M ^和 时, 对于任意的 i 和 j , ; 或者,从压缩后的矢量集合 Ω中任意选取 和 时,若 Ω为 多天线系统中 M天线非双极化信道秩为 1的码本, 对于任意的 i和 j , Wj = Wj 或^ = ^ , 但不能同时相等。 The steps of arbitrarily selecting w„ and ^ from the compressed vector set Ω include: arbitrarily selecting M^ from the compressed vector set Ω, for any i and j, or alternatively, from the compressed vector set Ω If Ω is arbitrarily selected, if Ω is a codebook with a M antenna non-dual-polarized channel rank of 1 in a multi-antenna system, Wj = Wj or ^ = ^ for any i and j, but not equal at the same time.

本发明还提供一种多天线系统的信道信息量化码本的构造装置, 包括: 码字构造模块, 其设置为: 根据信道的相关特性、 发送端和接收端的天 线极化特性, 构造信道矩阵的秩为 1的信道码字和秩为 2的信道码字, 形成 信道信息量化码本, 所述信道信息量化码本包括与双极化匹配的信道矩阵的 秩为 1的信道码字和秩为 2的信道码字。 The present invention also provides a device for constructing a channel information quantization codebook of a multi-antenna system, comprising: a codeword construction module, which is configured to: construct a channel matrix according to a correlation characteristic of a channel, an antenna polarization characteristic of a transmitting end and a receiving end a channel codeword of rank 1 and a channel codeword of rank 2 form a channel information quantization codebook, and the channel information quantization codebook includes a channel codeword of rank 1 of a channel matrix matched with dual polarization and a rank of 2 channel codewords.

所述构造装置还包括: 天线极化特性检测模块, 其设置为: 检测获取接 收端和发送端的天线极化特性, 将天线极化特性发送给码字构造模块。  The constructing device further includes: an antenna polarization characteristic detecting module, configured to: detect and acquire an antenna polarization characteristic of the receiving end and the transmitting end, and send the antenna polarization characteristic to the code word constructing module.

所述码字构造模块包括矢量选择单元、 相关码字生成单元和不相关码字 生成单元;  The codeword construction module includes a vector selection unit, a correlation codeword generation unit, and an unrelated codeword generation unit;

所述矢量选择单元设置为: 选择进行码字构造的矢量;  The vector selection unit is configured to: select a vector for performing codeword construction;

所述相关码字生成单元设置为: 针对相关信道利用所选择的矢量生成码 字;  The associated codeword generating unit is configured to: generate a codeword using the selected vector for the relevant channel;

所述不相关码字生成单元设置为: 针对不相关信道利用所选择的矢量生 成码字。 所述码字构造模块是设置为: 按如下方式构造针对相关信道的秩为 2的 信道码字: The uncorrelated codeword generating unit is configured to: generate a codeword using the selected vector for the uncorrelated channel. The codeword construction module is configured to: construct a channel codeword of rank 2 for the relevant channel as follows:

矢量选择单元选取^ "个 M维矢量

Figure imgf000009_0001
其中, Κ为码本中需要产生 的相关码字个数, Μ为发送端的发射天线数目的一半; Vector selection unit selects ^ "M-dimensional vectors
Figure imgf000009_0001
Where Κ is the number of related codewords to be generated in the codebook, and Μ is half of the number of transmitting antennas at the transmitting end;

相关码字生成单元利用 ^"个 Μ 维矢量 M w.ff构造秩为 2 的码字为 cwn = [unl u , 其中, The correlation codeword generating unit constructs a codeword of rank 2 using c" Μ dimension vector M w. ff as cw n = [u nl u , where

= αΛ +*Α , "Ac 为常数, w„为所述^"

Figure imgf000009_0004
= c u, +d u 个 M维矢量 ^…! ^中任意一个矢量。 w„ o 所述秩为 2 的码字 O„=| 或 = αΛ + * Α , "Ac is a constant, w„ is the ^"
Figure imgf000009_0004
= cu, +du M-dimensional vectors ^...! Any vector in ^. w„ o The codeword with rank 2 is O„=| or

0 w„ 或

Figure imgf000009_0002
天线极化特性检测模块是设置为:检测发送端的天线极化特性为 0 w„ or
Figure imgf000009_0002
The antenna polarization characteristic detecting module is configured to: detect that the antenna polarization characteristic of the transmitting end is

4 4 极化, 接收端的天线极化特性为 1根天线垂直极化时, 以及 码字构造模块是设置为: 按如下方式构造针对相关信道的秩为 1的信道 码字: 矢量选择单元选取^ "个 M维矢量

Figure imgf000009_0003
其中, Κ为码本中需要产生 的相关码字个数, Μ为发送端的发射天线数目的一半; 4 4 Polarization, the antenna polarization characteristic at the receiving end is 1 antenna vertical polarization, and the codeword construction module is set to: Construct a channel codeword with a rank of 1 for the relevant channel as follows: Vector selection unit selection ^ "M-dimensional vector
Figure imgf000009_0003
Where Κ is the number of related codewords to be generated in the codebook, and Μ is half of the number of transmitting antennas at the transmitting end;

相关码字生成单元利用 ^"个 Μ 维矢量 M w.ff构造秩为 1 的码字为 The correlation codeword generating unit constructs a codeword of rank 1 using ^" Μ dimension vector M w. ff as

, 其中: w„为所述 ^个 M维矢量 Μ ·· 中任意一个矢量 t 天线极化特性检测模块是设置为:检测发送端的天线极化特性为 , where: w is the ^ M-dimensional vector Μ · · any vector t antenna polarization characteristic detection module is set to: detect the antenna polarization characteristic of the transmitting end is

4 4 极化, 接收端的天线极化特性为 1 根天线 45度极化时, 以及 码字构造模块是设置为: 按如下方式构造针对相关信道的秩为 1的信道 码字: 矢量选择单元选取^ "个 2M维矢量 M - W 其中, ^"为码本中需要产 生的相关码字个数, M为发送端的发射天线数目的一半; 4 4 Polarization, the antenna polarization characteristic at the receiving end is 45-degree polarization of one antenna, and the codeword construction module is set to: construct a channel codeword of rank 1 for the relevant channel as follows: Vector selection unit selection ^ " 2M-dimensional vectors M - W where ^" is the number of relevant codewords to be generated in the codebook, and M is half the number of transmitting antennas at the transmitting end;

相关码字生成单元利用 ^"个 2M维矢量 M w.ff构造秩为 1的码字为

Figure imgf000010_0001
其中: w„为所述 ^个 2M维矢量 Μ · · 中任意一个矢量, μη , 为实数, 2Μ维的码字的前面 Μ个元素组成的矢量和后面 Μ个元素组成的矢量,相差 一个固定的相位^ ei,p = , 前 Μ个元素组成的矢量为 ^„乘以一个常系数 /„, 后 Μ个元素组成的矢量为 ^乘以一个常系数/„。 所述秩为 1的码字 RW„ = , 在 , ^等于 1时, 为 η =The correlation codeword generating unit constructs a codeword of rank 1 by using "2M-dimensional vectors M w. ff as
Figure imgf000010_0001
Where: w„ is any vector of the 2M-dimensional vector Μ · · , μ η , is a real number, a vector consisting of two elements preceding the 2 Μ code word and a vector consisting of the following elements, one difference The fixed phase ^ e i,p = , the vector consisting of the first element is ^„ multiplied by a constant coefficient / „, the vector consisting of the last element is ^ multiplied by a constant coefficient / „. The codeword RW „ = of rank 1 is η = when ^ is equal to 1.

Figure imgf000010_0002
所述矢量选择单元还设置为: 在相关信道码字构造时, 选取所述任意- 个矢量 符合以下特征:
Figure imgf000010_0002
The vector selection unit is further configured to: when constructing the associated channel codeword, selecting the arbitrary vector to meet the following characteristics:

1€ =τχ\ d … 其中, τ为一个复常数, 为一个复常数值, · · · 在— 〜 内均匀分 1€ =τχ\ d ... where τ is a complex constant and is a complex constant value, · · · evenly within -

2 2  twenty two

布, 或在- 〜 内均匀分布。 Cloth, or evenly distributed within -~.

3 3 天线极化特性检测模块是设置为:检测发送端的天线极化特性为  3 3 The antenna polarization characteristic detection module is set to: detect the antenna polarization characteristic of the transmitting end as

4 4 极化, 接收端的天线极化特性为多根天线垂直极化时, 以及 码字构造模块是设置为: 按如下方式构造针对不相关信道的秩为 2的信 道码字: 矢量选择单元选取^ "个 M维矢量 M -M^, 其中, Κ为码本中需要产生 的相关码字个数, Μ为发送端的发射天线数目的一半, ^…!^在 Μ维子空 间内均匀分布或准均匀分布; 4 4 Polarization, the antenna polarization characteristic of the receiving end is when the multiple antennas are vertically polarized, and the codeword construction module is set to: construct a channel codeword of rank 2 for the uncorrelated channel as follows: The vector selection unit selects ^ "M-dimensional vectors M - M^, where Κ is the number of related code words to be generated in the codebook, and Μ is half the number of transmitting antennas at the transmitting end, ^...!^ in the dimension space Uniformly distributed or quasi-uniformly distributed;

不相关码字生成单元利用 ^"个 Μ维矢量 M w.ff构造秩为 2 的码字为 The irrelevant codeword generating unit constructs a codeword of rank 2 by using the "" dimensional vector M w. ff as

,其中: „为所述 ^个 M维矢量 Μ ··ι ^中

Figure imgf000011_0001
, where: „ is the ^ M-dimensional vector Μ ··ι ^
Figure imgf000011_0001

任意一个矢量 t 天线极化特性检测模块是设置为:检测发送端的天线极化特性为 Any one of the vector t antenna polarization characteristic detecting modules is configured to: detect the antenna polarization characteristic of the transmitting end as

4 4 极化, 接收端的天线极化特性为多根天线 垂直极化时, 以及  4 4 polarization, the polarization characteristics of the antenna at the receiving end are when multiple antennas are vertically polarized, and

4 4 码字构造模块是设置为: 按如下方式构造针对不相关信道的秩为 2的信 道码字: 矢量选择单元选取 2 AT个 M维矢量 ^ · · · wff和 ^ · · · ¼^ , 其中, K为码 本中需要产生的相关码字个数,Μ为发送端的发射天线数目的一半, ^…!^ 在 Μ维子空间内均勾分布或准均勾分布; The 4 4 code word construction module is configured to: construct a channel codeword of rank 2 for an uncorrelated channel as follows: The vector selection unit selects 2 AT M-dimensional vectors ^ · · · w ff and ^ · · · 1⁄4^ , where K is the number of associated codewords to be generated in the codebook, and Μ is half the number of transmit antennas at the transmitting end, ^...! ^ Hook distribution or quasi-homogeneous distribution in the Weiweizi space;

不相关码字生成单元利用 ^"个 Μ维矢量 M w.ff构造秩为 2 的码字为 CW„ = ,其中: w„和^为任意的矢量,Unrelated ^ codeword generating means uses the "one-dimensional vectors Μ M w ff 2 codeword rank configured as CW." =, Where: w "^ and arbitrary vector,

Figure imgf000011_0002
Figure imgf000011_0002

弦距离大于或等于门限阔值,其中弦距离 ^^ ,^;)定义为 和^:两个矢量 之间夹角的求正弦值。 矢量选择单元是设置为: 选取所述任意矢量 和^时, 存储或实时计算 获得 M维的多个矢量集合 Ω , 矢量集合 Ω中任意 2个码字的弦距离都大于或 等于所述门限阔值; 通过釆用格拉斯曼 Grassmannian理论的方法对所述矢量 集合 Ω进行线压缩; 从压缩后的矢量集合 Ω中任意取 2个不同的矢量作为 w„ 和 „。 所述矢量选择单元还设置为: 所述压缩后的矢量集合 Ω中, 对于任意的 i 和 j , 或者, 所述压缩后的矢量集合 Ω中, Ω为多天线系统 中 M天线非双极化信道的秩为 1的码本,对于任意的 i和 j , w; 或^ = ^ , 但不能同时相等。 The chord distance is greater than or equal to the threshold threshold, where the chord distance ^^ , ^;) is defined as the sine of the angle between the two vectors. The vector selection unit is configured to: when the arbitrary vector and ^ are selected, the storage or real-time calculation obtains a plurality of vector sets Ω of M dimensions, and the chord distances of any two code words of the vector set Ω are greater than or equal to the threshold width Value; the vector set Ω is linearly compressed by the method of Grassmannian theory; two different vectors are randomly taken from the compressed vector set Ω as w „ and „. The vector selection unit is further configured to: in the compressed vector set Ω, for any i and j, or in the compressed vector set Ω, Ω is a non-dual polarization of the M antenna in the multi-antenna system A codebook with a rank of 1 for the channel, for any i and j, w; or ^ = ^, but not equal at the same time.

本发明的多天线系统的信道信息量化码本的构造方法及装置, 构造出的 与双极化匹配的信道信息量化码本, 能够在双极化信道下具有较高的信道信 息量化反馈效率, 且用于单极化信道时仍然具有较好的性能。 附图概述 The method and device for constructing a channel information quantization codebook of the multi-antenna system of the present invention, and constructing a channel information quantization codebook matched with dual polarization, can have higher channel information quantization feedback efficiency under the dual-polarization channel, And it still has better performance when used for single-polarized channels. BRIEF abstract

图 1是多天线系统中信道信息量化码本的构造装置的示意图。 本发明的较佳实施方式  1 is a schematic diagram of a construction apparatus of a channel information quantization codebook in a multi-antenna system. Preferred embodiment of the invention

为使本发明的目的、 技术方案和优点更加清楚, 以下结合附图对本发明 作进一步地详细说明。  In order to make the objects, technical solutions and advantages of the present invention more apparent, the present invention will be further described in detail with reference to the accompanying drawings.

本发明提出的 MIMO系统的信道信息 ( Channel State Information, CSI ) 量化码本的构造方法及装置, 针对一种信道矩阵为低秩(Rank ) 时的信道信 息量化码本构造, 可实现适于双极化信道的构造方案, 并且兼容单极化信道。 本发明的核心思想在于, 根据信道的相关特性、 发送端和接收端的天线极化 特性, 构造信道矩阵的秩为 1和 2的信道码字, 形成信道信息量化码本, 所 述信道信息量化码本中包括针对相关信道构造的信道矩阵的秩为 1和 2的信 道码字; 所述信道信息量化码本中还包括针对不相关信道构造的信道矩阵的 秩为 1和 2的信道码字。 进一步的, 对于相关特性较模糊的, 所述信道信息 量化码本中还可包括其它可能的码字。 下面分别针对相关信道和不相关信道 的码字构造进行说明。  The method and device for constructing a channel information information (CSI) quantization codebook of a MIMO system proposed by the present invention is applicable to a channel information quantization codebook structure when a channel matrix is a low rank (Rank), which is suitable for dual A configuration scheme for a polarized channel, and is compatible with a single-polarized channel. The core idea of the present invention is to construct a channel information codebook with ranks 1 and 2 of a channel matrix according to the correlation characteristics of the channel, the antenna polarization characteristics of the transmitting end and the receiving end, and form a channel information quantization codebook, and the channel information quantization code The channel codewords of ranks 1 and 2 of the channel matrix constructed for the relevant channel are included herein; and the channel information quantization codebook further includes channel codewords of ranks 1 and 2 of the channel matrix constructed for the uncorrelated channel. Further, for the related feature is ambiguous, the channel information quantization codebook may further include other possible codewords. The following describes the codeword construction for the relevant channel and the uncorrelated channel, respectively.

相关信道码字构造方法, 本发明的 MIMO系统的信道信息( Channel State Information, CSI )量化 码本的构造方法的具体步骤如下: 根据信道的相关特性、 发送端和接收端的天线极化特性, 构造信道矩阵 的秩为 1和 2的信道码字, 形成信道信息量化码本, 所述信道信息量化码本 中包括针对相关信道构造的信道矩阵的秩为 1和 2的信道码字。 Correlation channel codeword construction method, The specific steps of the method for constructing the channel information information (CSI) quantization codebook of the MIMO system of the present invention are as follows: According to the correlation characteristics of the channel, the antenna polarization characteristics of the transmitting end and the receiving end, constructing the rank of the channel matrix is 1 and The channel codeword of 2 forms a channel information quantization codebook, and the channel information quantization codebook includes channel codewords of ranks 1 and 2 of the channel matrix constructed for the relevant channel.

所述信道的相关特性为相关信道, 该相关信道的秩为 2的信道码字构造 的具体步骤如下: 选取 个 M维矢量 M -w.ff, 其中, K为码本中需要产生的相关码字个 数, M为发送端的发射天线数目的一半; 一般来说, 发射天线数都是偶数, 我们假设发送端为 2M根发射天线, 此时, 我们要构造 2M根天线的码本。 The specific characteristics of the channel are as follows: The specific steps of constructing the channel codeword of the rank 2 of the relevant channel are as follows: Select an M-dimensional vector M - w. ff , where K is the correlation code to be generated in the codebook The number of words, M is half of the number of transmitting antennas at the transmitting end; in general, the number of transmitting antennas is even, we assume that the transmitting end is 2M transmitting antennas. At this time, we have to construct a codebook of 2M antennas.

利用 ^个 M维矢量 M -M^构造秩为 2的码字为 0^=^^ Un2], 其中, Constructing a codeword with a rank of 2 using ^M-dimensional vectors M -M^ is 0^=^^ U n2 ], where

Unl =anUl +bnu2 , a ,c 为常数, 为所述 f

Figure imgf000013_0002
Un2 =cnux +dnu2 个 M维矢量 ^…! ^中任意一个矢量。 使用该方法,可以很好的在双极化天线的同一个极化方向 M根天线上 成较好的波束, 匹配其天线之间固定相位差的特征。 如果需要码字数较少, 且为了发送端的处理复杂度低, 且兼容性好, 也 可以釆用 LTE物理层协议 3GPP 36.211中定义的 Rankl的 Index为 0-7的码 字, 作为 W l^ 所述秩 2的码字 C „ U nl = a nUl + b n u 2 , a , c is a constant, for the f
Figure imgf000013_0002
U n2 =c n u x +d n u 2 M-dimensional vectors ^...! Any vector in ^. Using this method, a good beam can be well formed on the M antennas in the same polarization direction of the dual-polarized antenna, and the characteristics of the fixed phase difference between the antennas are matched. If the number of codewords required is small, and the processing complexity of the transmitting end is low, and the compatibility is good, the codeword of the Rank1 of the Rank1 of the Rank1 defined in the LTE physical layer protocol 3GPP 36.211 may be used as the W l^ The code word C of rank 2

CW„ = CW„ =

或(W„Or (W„

Figure imgf000013_0001
在相关信道下, 往往都是存在一条直射径的信道, 此时该直射径所占功 率经常超过 80%, 对于这样的信道, 其信道的相关矩阵 (定义为信道矩阵的转 秩 *信道矩阵)有着类似块对角矩阵的形式, 其特征矢量接近上述的一些模型。 因此用上述的模型构造出的码字对信道信息 (信道特征矢量)量化, 有着非 常好的性能。
Figure imgf000013_0001
Under the relevant channel, there is always a channel with a direct path. At this time, the direct path takes up more than 80% of the power. For such a channel, the channel's correlation matrix (defined as the channel matrix) The rank * channel matrix) has a form similar to a block diagonal matrix whose feature vectors are close to some of the above models. Therefore, the code words constructed by the above model have very good performance for quantifying channel information (channel feature vector).

发送端天线 极化为最常见的场景,接收端为笔记本电脑或手机竖 Transmitter antenna Polarization is the most common scenario, and the receiving end is a laptop or mobile phone.

4 4  4 4

直放置时, 接收端 1根天线可看成垂直极化。 所述信道的相关特性为相关信 道, 发送端的天线极化特性为- /+^极化, 接收端的天线极化特性为 1根天 When placed directly, one antenna at the receiving end can be regarded as vertical polarization. The correlation characteristic of the channel is a related channel, the polarization characteristic of the antenna at the transmitting end is - / + ^ polarization, and the polarization characteristic of the antenna at the receiving end is 1 day.

4 4  4 4

线垂直极化时, 该相关信道的秩为 1的信道码字构造的具体步骤如下: 选取 个 M维矢量 M - w.ff , 其中, K为码本中需要产生的相关码字个 数, M为发送端的发射天线数目的一半; 利用 ^个 M维矢量 M w.ff构造秩为 1的码字为 RW„When the line is vertically polarized, the specific steps of constructing the channel codeword of the rank 1 of the relevant channel are as follows: Select an M-dimensional vector M - w. ff , where K is the number of relevant codewords to be generated in the codebook, M is half the number of transmitting antennas at the transmitting end; constructing a codeword of rank 1 using RW M-dimensional vectors M w. ff

Figure imgf000014_0001
wn为所述 ^个 M维矢量 M · · 中任意一个矢量。 此时, 信道的 2个极化方向上有相同的预编码, 比较吻合收发端该极化 配置下信道的特征。
Figure imgf000014_0001
w n is any one of the above M-dimensional vectors M · ·. At this time, the same precoding is performed in the two polarization directions of the channel, and the characteristics of the channel in the polarization configuration of the transceiver end are matched.

发送端天线 - /+ 极化为最常见的场景, 接收端为手机 45度放置接打 Transmitter antenna - /+ Polarization is the most common scene, the receiving end is the mobile phone 45 degree placement

4 4  4 4

电话时,接收端 1根天线可看成 45度极化。所述信道的相关特性为相关信道, 发送端的天线极化特性为 极化, 接收端的天线极化特性为 1 根天线 When the phone is on, one antenna at the receiving end can be regarded as a 45-degree polarization. The correlation characteristic of the channel is the relevant channel, the polarization characteristic of the antenna at the transmitting end is polarization, and the polarization characteristic of the antenna at the receiving end is 1 antenna.

4 4  4 4

45度极化时, 该相关信道的秩为 1的信道码字构造的具体步骤如下: 选取 个 2M维矢量 M - M^ , 其中, ^"为码本中需要产生的相关码字 个数, Μ为发送端的发射天线数目的一半; 利用 ^个 2M维矢量 M w.ff构造秩为 1的码字为 RW„When the 45-degree polarization is performed, the specific steps of constructing the channel codeword of the rank 1 of the relevant channel are as follows: Select a 2M-dimensional vector M - M^ , where ^" is the number of relevant codewords to be generated in the codebook, Μ is half the number of transmitting antennas at the transmitting end; Constructing a codeword with rank 1 as RW using ^2M-dimensional vectors M w. ff

Figure imgf000015_0001
其中: w„为所述 ^个 2M维矢量 Μ ·· 中任意一个矢量, μη, 为实数, 2Μ维的码字的前面 Μ个元素组成的矢量和后面 Μ个元素组成的矢量,相差 一个固定的相位^ ei,p = , 前 Μ个元素组成的矢量为 ^„乘以一个常系数 /„, 后 Μ个元素组成的矢量为 ^乘以一个常系数/„。 此时, 特征矢量近似的有与其匹配的特征, 因此具有艮好的性能。 根据统计概率情况, μη, 在实际情况中都接近于 1, 所述秩为 1的码 字鮮„ =
Figure imgf000015_0001
Where: w„ is any vector of the 2M-dimensional vector Μ ··, μ η , is a real number, a vector consisting of two elements preceding the 2 Μ code word and a vector consisting of the following elements, one difference The fixed phase ^ e i,p = , the vector consisting of the first element is ^„ multiplied by a constant coefficient / „, the vector consisting of the last element is ^ multiplied by a constant coefficient / „. At this time, the feature vector approximation has characteristics matching it, and thus has good performance. According to the statistical probability situation, μ η is close to 1 in the actual situation, and the codeword with rank 1 is fresh „ =

Figure imgf000015_0002
上述的相关信道码字构造中, 所述^ "个矢量中的任意一个矢量 , 符合 以下特征:
Figure imgf000015_0002
In the above-mentioned related channel codeword structure, any one of the "" vectors has the following characteristics:

1€ =τχ\ d … 其中, τ为一个复常数, 为一个复常数值, ··· 在— 〜 内均匀 1€ =τχ\ d ... where τ is a complex constant, which is a complex constant value, ··· is uniform in -

2 2  twenty two

布, 或在- 〜 内均匀分布。 Cloth, or evenly distributed within -~.

不相关信道码字构造方法 本发明的多天线( ΜΙΜΟ ) 系统的信道信息 ( Channel State Information, CSI)量化码本的构造方法的具体步骤如下: Unrelated Channel Codeword Construction Method The specific steps of the method for constructing the channel information information (CSI) quantization codebook of the multi-antenna (ΜΙΜΟ) system of the present invention are as follows:

根据信道的相关特性、 发送端和接收端的天线极化特性, 构造信道矩阵 的秩为 1和 2的信道码字, 形成信道信息量化码本, 所述信道信息量化码本 中包括针对不相关信道构造的信道矩阵的秩为 1和 2的信道码字。 发送端天线 极化为最常见的场景,接收端为笔记本电脑或手机竖 Configuring a channel codeword of a channel matrix with a rank of 1 and 2 according to a correlation characteristic of the channel, an antenna polarization characteristic of the transmitting end and the receiving end, and forming a channel information quantization codebook, where the channel information quantization codebook includes an uncorrelated channel The channel codewords of the constructed channel matrix are 1 and 2 channel codewords. Transmitter antenna polarization is the most common scenario, and the receiving end is a laptop or mobile phone.

4 4  4 4

直放置时, 接收端多根天线可看成垂直水平极化。 所述信道的相关特性为不 相关信道, 发送端的天线极化特性为 极化, 接收端的天线极化特性为 When placed directly, multiple antennas at the receiving end can be considered as vertical horizontal polarization. The correlation characteristic of the channel is an uncorrelated channel, the polarization characteristic of the antenna at the transmitting end is polarization, and the polarization characteristic of the antenna at the receiving end is

4 4  4 4

多根天线垂直极化时, 该不相关信道的秩为 2的信道码字构造的具体步骤如 下: When multiple antennas are vertically polarized, the specific steps of constructing the channel codeword of rank 2 of the uncorrelated channel are as follows:

选取 AT个 M维矢量 M W.ff , ^…!^在 M维子空间内均勾分布或准均 匀分布。 其中, ^"为码本中需要产生的相关码字个数, M为发送端的发射天 线数目的一半, ^…!^在 M维子空间内均匀分布或准均匀分布; 利用 ^个 M 维矢量 M w.ff构造秩为 2 的码字为 Cf „Select AT M-dimensional vectors M W. ff , ^...! ^ Hook or quasi-uniform distribution in the M-dimensional subspace. Where ^" is the number of relevant codewords to be generated in the codebook, M is half of the number of transmit antennas at the transmitting end, ^...!^ is evenly distributed or quasi-uniformly distributed in the M-dimensional subspace; using ^M-dimensional vectors M w. ff constructs a codeword with a rank of 2 as Cf „

Figure imgf000016_0001
cw^ = , 其中: 为所述 ^个 M维矢量 Μ · · 中任意一个矢量 w„ - w„ 使用该模型能很好的匹配当前收发端极化角度条件下, 双极化非相关信 道下特征矢量的近似特征, 具有艮好的量化性能。
Figure imgf000016_0001
Cw^ = , where: is any vector of the ^ M-dimensional vectors Μ · · w„ - w„ Using this model can match the current polarization angle of the transceiver at the same time, under the dual-polarization non-correlated channel The approximate feature of the feature vector has good quantization performance.

在发送端天线 极化, 接收端天线多根天线为 极化时, 所 When the antenna at the transmitting end is polarized and the antenna at the receiving end is polarized,

4 4 4 4 述信道的相关特性为不相关信道, 发送端的天线极化特性为- /+^极化, 接  4 4 4 4 The relevant characteristics of the channel are uncorrelated channels, and the polarization characteristics of the antenna at the transmitting end are - / + ^ polarization,

4 4 收端的天线极化特性为多根天线 垂直极化时, 该不相关信道的秩为 2  4 4 The antenna polarization characteristic of the receiving end is multiple antennas. When vertically polarized, the rank of the uncorrelated channel is 2

4 4  4 4

的信道码字构造的具体步骤如下: 选取 2 ΛΓ个 M维矢量 ^…!^和!^… ¼^ , 其中, AT为码本中需要产生 的相关码字个数, M为发送端的发射天线数目的一半, ^…!^在 M维子空 间内均匀分布或准均匀分布; 所述矢量中任意的 ^和 应该有较大弦距离, 其中弦距离 fl „, )的定义为这两个矢量之间夹角的求正弦值。 此时比较符合收发端当前极化角度下, 有一定程度的极化泄露时, 双极 化信道特征矢量的近似特征。 利用 ^个 M 维矢量 M w.ff构造秩为 2 的码字为 C „ =

Figure imgf000017_0001
The specific steps of the channel codeword construction are as follows: Select 2 M M-dimensional vectors ^...! ^And! ^... 1⁄4^ , where AT is the number of associated codewords to be generated in the codebook, and M is half the number of transmit antennas at the transmitting end, ^...! ^ uniformly distributed or quasi-uniformly distributed in the M-dimensional subspace; any ^ in the vector should have a large chord distance, The chord distance fl „, ) is defined as the sine value of the angle between the two vectors. At this time, the double-polarized channel eigenvector is obtained when there is a certain degree of polarization leakage under the current polarization angle of the transceiver. Approximate characteristics of the construct. Use the M-dimensional vectors M w. ff to construct a codeword with a rank of 2 as C „ =
Figure imgf000017_0001

-w„ w wn w„ -w„ ww n w„

, 其中: w„和^为任意的矢量, w„和^的弦距离大于或等 于门限阔值,其中弦距离 ^^^,^ 定义为 和^:两个矢量之间夹角的求正 弦值。 为了减少系统存储量, 所述任意矢量^ !和 ^釆用以下步骤获得: 存储或实时计算获得 Μ维的多个矢量集合 Ω, 矢量集合 Ω中任意 2个 码字的弦距离都大于或等于所述门限阔值;通过釆用格拉斯曼( Grassmannian ) 理论的方法对所述矢量集合 Ω进行线压缩;从压缩后的矢量集合 Ω中任意取 2 个不同的矢量作为 和^。 , where: w„ and ^ are arbitrary vectors, the chord distances of w„ and ^ are greater than or equal to the threshold width, where the chord distance ^^^,^ is defined as and ^: the sine of the angle between the two vectors . To reduce the amount of system memory, and the ^ ^ preclude any vector obtained by the following steps:! Stored or calculated in real time to obtain a plurality of vectors Μ dimensional set [Omega], [Omega] in the set of vectors of any two codewords chord distance is greater than or equal to The threshold threshold is obtained by linearly compressing the vector set Ω by using the Grassmannian theory; and taking two different vectors as the sum from the compressed vector set Ω.

从 Ω中任意取 2个不同的矢量作为 w„和^, 对于不同的 n, w„和 的 取法不同。 方法 1: 在 Ω中矢量较多时, 对于任意的 i和 j,

Figure imgf000017_0002
此时具有最好 的性能。 方法 2: Take two different vectors from Ω as w„ and ^, for different n, w„ and the difference is different. Method 1: When there are many vectors in Ω, for any i and j,
Figure imgf000017_0002
This has the best performance. Method 2:

Ω可以为系统中 M天线非双极化信道 Rankl码本, 此时兼容性很好, 不 用增加新的存储, 相对于方法 1 此时可以允许 或^ = 但不能同时 相等。 所述 和^可以为正交的, 此时, 任意的 lMk,^)=i , 此时比较符 合收发端当前极化角度下, 有很小的极化泄露时, 双极化信道特征矢量的近 似特征。 使用该模型能很好的匹配当前收发端极化角度条件下, 双极化非相关信 道下特征矢量的近似特征, 具有艮好的量化性能。 接收端自己产生或者从别处获取根据上述方法构造的信道信息码本后, 根据信道情况从码本中查找到与信道匹配的码字, 将码字信息或码字序号反 馈给发送端, 发送端根据码字信息或码字序号查找相同的码本, 按查找到的 码字进行预编码, 可大大节省资源, 同时, 由于码本中包含了与双极化匹配 的码字, 还兼容单极化的情况, 因而本发明的构造方法具有更好的兼容性。 Ω can be the non-dual-polarized channel Rank1 codebook of the M antenna in the system. At this time, the compatibility is very good, and no new storage is needed. Compared with the method 1, it can be allowed or ^ = but not equal at the same time. The sum and ^ can be orthogonal, in this case, any lM k, ^) = i, the comparator Approximate characteristics of the dual-polarized channel eigenvectors when there is little polarization leakage at the current polarization angle of the transceiver. The model can match the approximate feature of the feature vector under the dual-polarization non-correlated channel under the current polarization angle of the transceiver, and has good quantization performance. After the receiving end itself generates or obtains the channel information codebook constructed according to the above method, the codeword matching the channel is searched from the codebook according to the channel condition, and the codeword information or the codeword serial number is fed back to the transmitting end, and the transmitting end Finding the same codebook according to the codeword information or the codeword serial number, precoding according to the found codeword, can greatly save resources, and at the same time, since the codebook includes the codeword matched with the dual polarization, it is also compatible with the monopole. In the case of the invention, the construction method of the invention has better compatibility.

本发明还提供一种 MIMO系统的信道信息 (Channel State Information, CSI )量化码本的构造装置 10, 如图 1所示, 包括: The present invention also provides a channel device information (CSI) quantization codebook construction apparatus 10 for a MIMO system. As shown in FIG. 1, the method includes:

信道检测模块 11 , 用于检测获取信道的相关特性, 判断信道为相关信道 还是不相关信道, 将信道的相关特性结果发送给码字构造模块 13;  The channel detecting module 11 is configured to detect the relevant characteristics of the acquired channel, determine whether the channel is an associated channel or an uncorrelated channel, and send the relevant characteristic result of the channel to the codeword construction module 13;

天线极化特性检测模块 12, 用于检测获取接收端和发送端的天线极化特 性, 将天线极化特性发送给码字构造模块 13;  The antenna polarization characteristic detecting module 12 is configured to detect the antenna polarization characteristics of the receiving end and the transmitting end, and send the antenna polarization characteristic to the code word constructing module 13;

码字构造模块 13 , 根据信道的相关特性、 发送端和接收端的天线极化特 性, 构造信道矩阵的秩为 1和 2的信道码字形成信道信息量化码本, 所述信 道信息量化码本包括针对相关信道构造的信道矩阵的秩为 1和 2的信道码字, 还包括针对不相关信道构造的信道矩阵的秩为 1和 2的信道码字。  The codeword construction module 13 forms a channel information quantized codebook by constructing a channel codeword with ranks 1 and 2 of the channel matrix according to the correlation characteristics of the channel, the antenna polarization characteristics of the transmitting end and the receiving end, and the channel information quantization codebook includes The channel codewords of ranks 1 and 2 of the channel matrix constructed for the relevant channel further include channel codewords of ranks 1 and 2 of the channel matrix constructed for the uncorrelated channel.

所述码字构造模块 13进一步可分为: 矢量选择单元 131、 相关码字生成 单元 132、 不相关码字生成单元 133;  The codeword construction module 13 is further divided into: a vector selection unit 131, a correlation codeword generation unit 132, an unrelated codeword generation unit 133;

所述矢量选择单元 131 , 用于选择进行码字构造的矢量;  The vector selection unit 131 is configured to select a vector for performing codeword construction;

所述相关码字生成单元 132 , 用于针对相关信道利用所选择的矢量生成 码字;  The associated codeword generating unit 132 is configured to generate a codeword by using the selected vector for the relevant channel;

所述不相关码字生成单元 133 , 用于针对不相关信道利用所选择的矢量 生成码字。 上述的 MIMO系统的信道信息( Channel State Information, CSI )量化码 本的构造装置, 可以位于发送端, 也可以位于接收端, 当位于发送端或接收 端时, 其工作方式可以为一次构造完成所有码字, 而将构造好的码字存储在 发送端或接收端, 也可以为实时的进行构造。 该构造装置, 也可以位于发送 端或接收端之外的计算机装置或服务器上, 一次构造完成所有码字, 而将完 整的码本存储到发送端和接收端。 上述的 MIMO系统的信道信息量化码本的构造装置, 其中的信道的相关 特性可以由信道检测模块获取, 也可以通过其它方式, 例如参数传递来获得。 天线极化特性可以由天线极化特性检测模块来获得,也可釆用其它方式获得, 例如从专门的信道相关模块来获得。 The uncorrelated codeword generating unit 133 is configured to generate a codeword by using the selected vector for the uncorrelated channel. The device for constructing the channel information information (CSI) of the MIMO system may be located at the transmitting end or at the receiving end. When it is located at the transmitting end or the receiving end, the working mode may be completed once. The code word, and the constructed code word is stored at the transmitting end or the receiving end, and can also be constructed in real time. The constructing device may also be located on a computer device or server other than the transmitting end or the receiving end, and all code words are constructed once, and the complete codebook is stored to the transmitting end and the receiving end. The apparatus for constructing a channel information quantization codebook of the above MIMO system, wherein the correlation characteristic of the channel may be acquired by the channel detection module, or may be obtained by other means, such as parameter transmission. The antenna polarization characteristics can be obtained by the antenna polarization characteristic detection module, or can be obtained by other means, for example, from a dedicated channel correlation module.

下面结合具体的信道特性, 及接收端和发送端的天线极化特性进行具体 码字构造的方案。 The following is a specific codeword construction scheme in combination with specific channel characteristics and antenna polarization characteristics of the receiving end and the transmitting end.

信道检测模块 11检测获得所述信道的相关特性为相关信道,码字构造模 块 13针对该相关信道的秩为 2的信道码字构造的具体方案如下:  The channel detection module 11 detects that the correlation characteristic of the channel is obtained as a correlation channel, and the specific scheme of the codeword structure of the codeword construction module 13 for the rank 2 of the relevant channel is as follows:

矢量选择单元 131选取 ^个 M维矢量 M - W 其中, K为码本中需要 产生的相关码字个数, M为发送端的发射天线数目的一半;  The vector selecting unit 131 selects ^ M-dimensional vectors M - W , where K is the number of related code words to be generated in the codebook, and M is half of the number of transmitting antennas at the transmitting end;

相关码字生成单元 132利用 ^个 M维矢量 M w.ff构造秩为 2的码字为 cwn = [unl u , 其中, The correlation codeword generating unit 132 constructs a codeword of rank 2 by using the M-dimensional vectors M w. ff as cw n = [u nl u , where

= αΛ + *Α , "A c 为常数, w„为所述^"

Figure imgf000019_0002
= c u, + d u 个 M维矢量 ^…! ^中任意一个矢量。 = αΛ + * Α , "A c is a constant, w„ is the ^^"
Figure imgf000019_0002
= cu, + du M-dimensional vectors ^...! Any vector in ^.

Figure imgf000019_0001
信道检测模块 11检测获得所述信道的相关特性为相关信道,天线极化特 性检测模块 12检测发送端的天线极化特性为 极化, 接收端的天线极
Figure imgf000019_0001
The channel detecting module 11 detects that the correlation characteristic of the channel is the relevant channel, and the antenna polarization characteristic detecting module 12 detects that the antenna polarization characteristic of the transmitting end is polarization, and the antenna pole of the receiving end

4 4  4 4

化特性为 1根天线垂直极化时,码字构造模块 13针对该相关信道的秩为 1的 信道码字构造的具体方案如下: 矢量选择单元 131选取 ^个 M维矢量 M -W 其中, K为码本中需要 产生的相关码字个数, M为发送端的发射天线数目的一半; When the characteristics of the antenna are vertically polarized, the specific scheme of the codeword structure of the codeword construction module 13 for the rank 1 of the relevant channel is as follows: The vector selection unit 131 selects the M-dimensional vectors M - W where K For the number of related codewords to be generated in the codebook, M is half of the number of transmit antennas at the transmitting end;

相关码字生成单元 132利用 ^个 M维矢量 M w.ff构造秩为 1的码字为 其中: wn为所述 ^个 M维矢量 M · · 中任意一个矢量 t Correlation code word generation unit 132 using the ^ M-dimensional vector M w ff configured codeword rank 1 in which:. W n is the ^ M-dimensional vector M · · t vector of any one of

信道检测模块 11检测获得所述信道的相关特性为相关信道,天线极化特 性检测模块 12检测发送端的天线极化特性为- 极化, 接收端的天线极 The channel detecting module 11 detects that the correlation characteristic of the channel is the relevant channel, and the antenna polarization characteristic detecting module 12 detects that the antenna polarization characteristic of the transmitting end is - polarization, and the antenna pole of the receiving end

4 4  4 4

化特性为 1 根天线 45度极化时, 码字构造模块 13针对该相关信道的秩为 1 的信道码字构造的具体方案如下: 矢量选择单元 131选取 ^个 2M维矢量

Figure imgf000020_0001
其中, 为码本中需 要产生的相关码字个数, M为发送端的发射天线数目的一半; When the characteristic is a 45-degree polarization of one antenna, the specific scheme of the codeword structure of the codeword construction module 13 for the rank 1 of the relevant channel is as follows: The vector selection unit 131 selects two 2M-dimensional vectors.
Figure imgf000020_0001
Wherein, the number of related codewords to be generated in the codebook, and M is half of the number of transmitting antennas at the transmitting end;

相关码字生成单元 132利用 K个 2M维矢量 M M^构造秩为 1的码字为

Figure imgf000020_0002
其中: w„为所述 ^个 2M维矢量 Μ ·· 中任意一个矢量, μη, 为实数, 2Μ维的码字的前面 Μ个元素组成的矢量和后面 Μ个元素组成的矢量,相差 一个固定的相位^ ei,p = , 前 Μ个元素组成的矢量为 ^„乘以一个常系数 /„, 后 M个元素组成的矢量为 ^„乘以一个常系数 所述秩为 1的码字 Rf „ , 在 , ^等于 1时, 为The correlation codeword generating unit 132 constructs a codeword of rank 1 using K 2M-dimensional vectors MM^
Figure imgf000020_0002
Where: w„ is any vector of the 2M-dimensional vector Μ ··, μ η , is a real number, a vector consisting of two elements preceding the 2 Μ code word and a vector consisting of the following elements, one difference The fixed phase ^ e i,p = , the vector of the first element is ^„ multiplied by a constant coefficient / „, The vector consisting of the last M elements is ^„ multiplied by a constant coefficient of the codeword Rf „ with a rank of 1, when ^ is equal to 1,

Figure imgf000021_0001
Figure imgf000021_0001

所述矢量选择单元 131选取所述任意一个矢量 , 符合以下特征: The vector selection unit 131 selects any one of the vectors, and meets the following characteristics:

1€

Figure imgf000021_0002
έθ" " … 其中, τ为一个复常数, 为一个复常数值, Α··· 在 布, 或在 - 〜 内均匀分布 t 1€
Figure imgf000021_0002
έ θ "1θ" ... where, [tau] is a complex constant, a normalization value, Α ··· cloth, or - the uniform distribution t ~

信道检测模块 11检测获得所述信道的相关特性为不相关信道,天线极化 特性检测模块 12检测发送端的天线极化特性为 极化, 接收端的天线 The channel detecting module 11 detects that the correlation characteristic of the channel is an uncorrelated channel, and the antenna polarization characteristic detecting module 12 detects that the antenna polarization characteristic of the transmitting end is polarization, and the antenna of the receiving end

4 4  4 4

极化特性为多根天线垂直极化时,码字构造模块 13针对该不相关信道的秩为 2的信道码字构造的具体方案如下: 矢量选择单元 131选取 ^个 M维矢量 M -W 其中, K为码本中需要 产生的相关码字个数, M为发送端的发射天线数目的一半, Ή ··Ή^在 M维 子空间内均匀分布或准均匀分布; When the polarization characteristic is that the plurality of antennas are vertically polarized, the specific scheme of the codeword structure of the codeword construction module 13 for the rank 2 of the uncorrelated channel is as follows: The vector selection unit 131 selects the M-dimensional vectors M - W , K is the number of related codewords to be generated in the codebook, M is half of the number of transmitting antennas at the transmitting end, and Ή··Ή^ is evenly distributed or quasi-uniformly distributed in the M-dimensional subspace;

不相关码字生成单元 133利用 ^个 Μ维矢量 M w.ff构造秩为 2的码字 为 CW„ = ,其中: w„为所述 ^个 M维矢量 Μ ··The irrelevant codeword generating unit 133 constructs a codeword of rank 2 into CW „ = using ^ Μ dimension vectors M w. ff , where: w „ is the ^ M-dimensional vectors Μ ··

Figure imgf000021_0003
Figure imgf000021_0003

中任意一 Any one of

信道检测模块 11检测获得所述信道的相关特性为不相关信道,天线极化 特性检测模块 12检测发送端的天线极化特性为 极化, 接收端的天线 The channel detecting module 11 detects that the correlation characteristic of the channel is an uncorrelated channel, and the antenna polarization characteristic detecting module 12 detects that the antenna polarization characteristic of the transmitting end is polarization, and the antenna of the receiving end

4 4 极化特性为多根天线- /+ 垂直极化时, 码字构造模块 13针对该不相关信 4 4 When the polarization characteristic is multiple antennas - / + vertical polarization, the code word construction module 13 is for the irrelevant signal

4 4  4 4

道的秩为 2的信道码字构造的具体方案如下: 矢量选择单元 131选取 2 ΛΓ个 M维矢量 ^…!^和!^… ¼^ , 其中, K 为码本中需要产生的相关码字个数, M为发送端的发射天线数目的一半, ^ · · · wff在 M维子空间内均匀分布或准均匀分布; The specific scheme of the channel codeword construction with the rank rank of 2 is as follows: The vector selection unit 131 selects 2 M M-dimensional vectors ^...! ^And! ^... 1⁄4^ , where K is the number of related codewords to be generated in the codebook, M is half of the number of transmitting antennas at the transmitting end, and ^ · · · w ff is evenly distributed or quasi-uniformly distributed in the M-dimensional subspace;

不相关码字生成单元 133利用 ^个 M维矢量 M w.ff构造秩为 2的码字 The irrelevant codeword generating unit 133 constructs a codeword of rank 2 using ^ M-dimensional vectors M w. ff

,其中: 和 为任意的矢量, 和^

Figure imgf000022_0002
, where: and for any vector, and ^
Figure imgf000022_0002

的弦距离大于或等于门限阔值,其中弦距离 fl „, )定义为 w„和^:两个矢 量之间夹角的求正弦值。 矢量选择单元 131选取所述任意矢量 ^和 时,存储或实时计算获得 M 维的多个矢量集合 Ω , 矢量集合 Ω中任意 2个码字的弦距离都大于或等于所 述门限阔值; 通过釆用格拉斯曼 Grassmannian理论的方法对所述矢量集合 Ω 进行线压缩; 从压缩后的矢量集合 Ω中任意取 2个不同的矢量作为 w„和 。 所述压缩后的矢量集合 Ω中, 对于任意的 i和 j , ^ ^ 。 或者, 所 述压缩后的矢量集合 Ω中, Ω为多天线系统中 M天线非双极化信道的秩为 1 的码本, 对于任意的 i和 j , ^. = 或^ = ^ , 但不能同时相等。 The chord distance is greater than or equal to the threshold width, where the chord distance fl „, ) is defined as w „ and ^: the sine of the angle between the two vectors. When the vector selection unit 131 selects the arbitrary vector ^, the storage or real-time calculation obtains a plurality of vector sets Ω of the M dimension, and the chord distances of any two codewords of the vector set Ω are greater than or equal to the threshold threshold;线The line set Ω is linearly compressed by the method of Grassmannian theory; arbitrarily take two different vectors from the compressed vector set Ω as w „ and. The compressed vector set Ω, for Any i and j , ^ ^. Or, in the compressed vector set Ω, Ω is a codebook of rank 1 of the M antenna non-dual-polarized channel in the multi-antenna system, for any i and j, ^ . = or ^ = ^ , but not equal at the same time.

实施例 1 Example 1

8天线 Rank2中 8个适应双极化相关信道码字构造 ^为如下表 2所示 的 8个 4维矢量: 表 2

Figure imgf000022_0001
M2 =[l j -l -7] w6 = [1 qi —j -qo] 8 adaptive double-polarized correlation channel codeword constructions in 8 antennas Rank2 are 8 4-dimensional vectors as shown in Table 2 below: Table 2
Figure imgf000022_0001
M 2 =[lj -l -7] w 6 = [1 qi —j -qo]

w3 =[l -1 1 -l] u1 = [l q2 j q>] w4 =[! -J -1 7'] w8 =[l ^3 —j ql] 其中: q0= (1+j) /V2; q1= (-1+j)

Figure imgf000023_0001
q3= (1-j) l l 构造码字
Figure imgf000023_0002
w 3 =[l -1 1 -l] u 1 = [l q2 j q>] w 4 =[! -J -1 7'] w 8 =[l ^3 —j ql] where: q0= (1+ j) /V2; q1= (-1+j)
Figure imgf000023_0001
Q3= (1-j) ll Construct codeword
Figure imgf000023_0002

cw„ = n = 2,4,6,8

Figure imgf000023_0003
Cw„ = n = 2,4,6,8
Figure imgf000023_0003

上述这种混合模式也是可以的 , 实施例 2 The above mixing mode is also possible, Embodiment 2

8天线 Rank2中 N个适应双极化相关信道码字构造 选取 16个 4维矢量

Figure imgf000023_0004
对于任意的 n
Figure imgf000023_0005
θη = -y + η = \〜Ν Selecting 16 4D Vectors for N Adapted Dual Polarization Correlated Channel Codeword Structures in 8 Antenna Rank2
Figure imgf000023_0004
For any n
Figure imgf000023_0005
θ η = -y + η = \~Ν

构造码字 cw— n = l ~ N Construct codeword Cw— n = l ~ N

u„ —u  u„ —u

实施例 3 Example 3

8天线 Rankl中 N个适应双极化相关信道码字构造  8 antennas in Rank1 N adaptive dual polarization correlation channel codeword construction

选取 N个 4维矢量 n , 方法同实施例 1或实施例 2 发送端天线 极化为最常见的场景,接收端为笔记本电脑或手机竖  Select N four-dimensional vectors n, and the method is the same as that of Embodiment 1 or Embodiment 2. The antenna is polarized to the most common scene, and the receiving end is a laptop or a mobile phone.

4 4  4 4

直放置时。 此时码字构造为 n = \〜 N 如果接收端为 - /+ 极化 When placed straight. At this point the codeword is constructed as n = \~ N if the receiver is - /+ polarized

4 4 〜 N 4 4 ~ N

Figure imgf000024_0001
在一些情况下可能不能获得接收端极化角度信息, 且其可能随时改变: 因此, 也可以使用下面的混合方法进行性能折衷
Figure imgf000024_0001
In some cases, the receiving end polarization angle information may not be obtained, and it may change at any time: Therefore, the following hybrid method can also be used for performance tradeoffs.

n为奇数时

Figure imgf000024_0002
n为偶数时
Figure imgf000024_0003
When n is an odd number
Figure imgf000024_0002
When n is even
Figure imgf000024_0003

实施例 4 8天线 Rank2中 8个适应双极化相关信道码字构造 Example 4 8 adaptive dual polarization related channel codeword construction in 8 antenna Rank2

选取 8个 4维矢量集合, 如 LTE中定义的 4天线 Rankl的后 8个码字记 为 ~ w8 使用下面的方式构造 Rank2的码字

Figure imgf000025_0001
Select 8 4-dimensional vector sets. The last 8 code words of the 4-antenna Rank1 defined in LTE are recorded as ~ w 8. The following steps are used to construct the codeword of Rank2.
Figure imgf000025_0001

实施例 5 Example 5

8天线 Rank2中 16个适应双极化不相关信道码字构造 8 antennas in Rank 2 16 adaptive dual-polarization uncorrelated channel codeword construction

选取 8个两两弦距离较大的 4维矢量集合,如 LTE中定义的 4天线 Rankl 的后 8个码字记为 。 使用下面的方式构造 Rank2的码字  Select a 4-dimensional vector set with a large distance between two pairs of two strings. For example, the last 8 code words of the 4-antenna Rank1 defined in LTE are recorded as . Use the following method to construct the codeword of Rank2

CW CW

CW5 CW 5

Figure imgf000025_0002
实施例 6
Figure imgf000025_0002
Example 6

8天线 Rank2中 16个适应双极化不相关信道码字构造选取 32个两两弦 巨离较大的 4维矢量 "i己为 wl - w32 , 如表 3所示。 表 3 16 antennas in the 8 antenna Rank2 are adapted to the dual-polarization uncorrelated channel codeword structure. The 32-dimensional two-dimensional large-diameter 4-dimensional vector "i" is w l - w 32 , as shown in Table 3. Table 3

Figure imgf000025_0003
5 -0. 5000 0. 5000i 0. 5000 -0. 5000i
Figure imgf000025_0003
5 -0. 5000 0. 5000i 0. 5000 -0. 5000i

6 0. 5000 0. 5000 0. 5000 0. 5000 6 0. 5000 0. 5000 0. 5000 0. 5000

7 0. 5000 0. 5000i 0. 5000 0. 5000i  7 0. 5000 0. 5000i 0. 5000 0. 5000i

8 0. 5000 0. 5000 0. 5000 -0. 5000  8 0. 5000 0. 5000 0. 5000 -0. 5000

9 0. 5000 0. 5000i -0. 5000 0. 5000i  9 0. 5000 0. 5000i -0. 5000 0. 5000i

10 0. 5000 -0. 5000 0. 5000 0. 5000  10 0. 5000 -0. 5000 0. 5000 0. 5000

11 0. 5000 -0. 5000i -0. 5000 -0. 5000i 11 0. 5000 -0. 5000i -0. 5000 -0. 5000i

12 0. 5000 0. 3536 + 0. 3536i 0. 5000i -0. 3536 + 0. 3536i12 0. 5000 0. 3536 + 0. 3536i 0. 5000i -0. 3536 + 0. 3536i

13 0. 5000 -0. 3536 + 0. 3536i -0. 5000i 0. 3536 + 0. 3536i13 0. 5000 -0. 3536 + 0. 3536i -0. 5000i 0. 3536 + 0. 3536i

14 0. 5000 -0. 3536 - 0. 3536i 0. 5000i 0. 3536 - 0. 3536i14 0. 5000 -0. 3536 - 0. 3536i 0. 5000i 0. 3536 - 0. 3536i

15 0. 5000 0. 3536 - 0. 3536i -0. 5000i -0. 3536 - 0. 3536i15 0. 5000 0. 3536 - 0. 3536i -0. 5000i -0. 3536 - 0. 3536i

16 0. 5000 -0. 4619 - 0. 1913i 0. 3536 + 0. 3536i -0. 1913 - 0. 4619i16 0. 5000 -0. 4619 - 0. 1913i 0. 3536 + 0. 3536i -0. 1913 - 0. 4619i

17 0. 3117 0. 6025 + 0. 1995i -0. 4030 - 0. 4903i - 0. 1122 - 0. 2908i17 0. 3117 0. 6025 + 0. 1995i -0. 4030 - 0. 4903i - 0. 1122 - 0. 2908i

18 0. 3117 -0. 6025 - 0. 1995i - 0. 1122 - 0. 2908i 0. 4030 + 0. 4903i18 0. 3117 -0. 6025 - 0. 1995i - 0. 1122 - 0. 2908i 0. 4030 + 0. 4903i

19 0. 3058 0. 1901 - 0. 6052i 0. 1195 + 0. 2866i 0. 4884 - 0. 4111 i19 0. 3058 0. 1901 - 0. 6052i 0. 1195 + 0. 2866i 0. 4884 - 0. 4111 i

20 0. 5000 -0. 1913 + 0. 4619i -0. 3536 - 0. 3536i 0. 4619 - 0. 1913i20 0. 5000 -0. 1913 + 0. 4619i -0. 3536 - 0. 3536i 0. 4619 - 0. 1913i

21 0. 5000 0. 1913 - 0. 4619i -0. 3536 - 0. 3536i -0. 4619 + 0. 1913i21 0. 5000 0. 1913 - 0. 4619i -0. 3536 - 0. 3536i -0. 4619 + 0. 1913i

0. 5000 0. 4619 + 0. 1913i 0. 3536 + 0. 3536i 0. 1913 + 0. 4619i0. 5000 0. 4619 + 0. 1913i 0. 3536 + 0. 3536i 0. 1913 + 0. 4619i

22 twenty two

0. 3082 0. 0104 + 0. 3151 i 0. 4077 + 0. 4887i -0. 4783 + 0. 4145i 0. 3082 0. 0104 + 0. 3151 i 0. 4077 + 0. 4887i -0. 4783 + 0. 4145i

23 twenty three

0. 3117 0. 3573 - 0. 2452i 0. 6025 - 0. 1995i - 0. 1578 + 0. 5360i 0. 3117 0. 3573 - 0. 2452i 0. 6025 - 0. 1995i - 0. 1578 + 0. 5360i

24 twenty four

0. 3117 0. 2452 + 0. 3573i -0. 6025 + 0. 1995i 0. 5360 + 0. 1578i 0. 3117 0. 2452 + 0. 3573i -0. 6025 + 0. 1995i 0. 5360 + 0. 1578i

25 25

0. 3082 - 0. 3666 + 0. 2426i 0. 6092 - 0. 1842i 0. 1615 - 0. 5298i 0. 3082 - 0. 3666 + 0. 2426i 0. 6092 - 0. 1842i 0. 1615 - 0. 5298i

26 26

0. 3117 -0. 2452 - 0. 3573i -0. 6025 + 0. 1995i - 0. 5360 - 0. 1578i 0. 3117 -0. 2452 - 0. 3573i -0. 6025 + 0. 1995i - 0. 5360 - 0. 1578i

27 27

0. 3117 0. 4260 + 0. 0793i 0. 1995 + 0. 6025i 0. 2674 + 0. 4906i 0. 3117 0. 4260 + 0. 0793i 0. 1995 + 0. 6025i 0. 2674 + 0. 4906i

28 28

0. 3117 -0. 0793 + 0. 4260i -0. 1995 - 0. 6025i 0. 4906 - 0. 2674i 0. 3117 -0. 0793 + 0. 4260i -0. 1995 - 0. 6025i 0. 4906 - 0. 2674i

29 29

30 0. 3117 -0. 4260 - 0. 0793i 0. 1995 + 0. 6025i -0. 2674 - 0. 4906i 30 0. 3117 -0. 4260 - 0. 0793i 0. 1995 + 0. 6025i -0. 2674 - 0. 4906i

31 0. 3117 0. 0793 - 0. 4260i -0. 1995 - 0. 6025i -0. 4906 + 0. 2674i31 0. 3117 0. 0793 - 0. 4260i -0. 1995 - 0. 6025i -0. 4906 + 0. 2674i

32 0. 5636 -0. 3332 - 0. 2672i 0. 1174 + 0. 55121 -0. 3308 - 0. 2702i 使用下面的方式构造 Rank2的码字 w w 32 0. 5636 -0. 3332 - 0. 2672i 0. 1174 + 0. 55121 -0. 3308 - 0. 2702i Use the following method to construct the code word of Rank2 w w

CW^ w = 1〜 16  CW^ w = 1~ 16

w„ - w 实施例 6  w„ - w Example 6

8天线 Rank2中 8个适应双极化不相关信道码字构造 选取 8个两两弦距离较大的 4*2的正交矩阵集合, 如 LTE中定义的 4天 线 Rank2的后 8个码字记为 H 8 adaptive double-polarization uncorrelated channel codeword construction in 8 antenna Rank2 Select a 4*2 orthogonal matrix set with a large distance between two pairs of two chords. For example, the last 8 code words of the 4 antenna Rank2 defined in LTE are recorded as H.

Wn的第一列记为 „ (:,1) , Wn的第二列记为 „ (:,2) 构造码字为 „(:,1) „(:,2) The first column of W n is denoted by „ (:,1) , and the second column of W n is denoted by „ (:,2) The construction codeword is „(:,1) „(:,2)

CW = w = 1 ~ 8 CW = w = 1 ~ 8

„(:,1) - Wn (:,2) „(:,1) - W n (:,2)

实施例 7 Example 7

8天线 Rank2中 N个适应双极化不相关信道码字构造。 选取 2N个两两弦距离较大的 4维矢量集合 ^〜 w2N。 从中选取 N个矢量 ^。 以 W„为基础矢量, 对 做正交化处理得到 使得 „+ΛίW„正交; 构造码字: CWn = = \〜 NN antennas in the 8 antenna Rank2 are adapted to the dual polarization uncorrelated channel codeword construction. Select a 2N vector set ^ 2 w 2N with a larger 2N two-chord distance. Select N vectors ^ from it. Based on the W „ vector, the orthogonalization is performed to make „ +Λί and W „ orthogonal; construct codeword: CW n = = \~ N

Figure imgf000027_0001
以上所述仅为本发明的实施例而已, 并不用于限制本发明, 对于本领域 的技术人员来说, 本发明可以有各种更改和变化。 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权利要求 范围之内。
Figure imgf000027_0001
The above is only the embodiment of the present invention, and is not intended to limit the present invention, and various modifications and changes can be made to the present invention. Any modifications, equivalents, improvements, etc., made within the spirit and scope of the invention are intended to be included within the scope of the appended claims.

工业实用性 Industrial applicability

本发明的多天线系统的信道信息量化码本的构造方法及装置, 构造出的 与双极化匹配的信道信息量化码本, 能够在双极化信道下具有较高的信道信 息量化反馈效率, 且用于单极化信道时仍然具有较好的性能。  The method and device for constructing a channel information quantization codebook of the multi-antenna system of the present invention, and constructing a channel information quantization codebook matched with dual polarization, can have higher channel information quantization feedback efficiency under the dual-polarization channel, And it still has better performance when used for single-polarized channels.

Claims

权 利 要 求 书 Claim 1、 一种多天线系统中信道信息量化码本的构造方法, 所述方法包括: 根据信道的相关特性、 发送端和接收端的天线极化特性, 构造信道矩阵 的秩为 1的信道码字和秩为 2的信道码字, 形成信道信息量化码本, 所述信 道信息量化码本包括针对相关信道构造的与双极化匹配的信道矩阵的秩为 1 的信道码字和秩为 2的信道码字。  A method for constructing a channel information quantization codebook in a multi-antenna system, the method comprising: constructing a channel codeword of a rank 1 of a channel matrix according to a correlation characteristic of a channel, an antenna polarization characteristic of a transmitting end and a receiving end; a channel codeword of rank 2, forming a channel information quantization codebook, the channel information quantization codebook including a channel codeword of rank 1 and a channel of rank 2 of a channel matrix matched with dual polarization for a correlation channel configuration Codeword. 2、 如权利要求 1所述的构造方法, 其中, 所述信道的相关特性为相关信 道, 该构造相关信道的秩为 2的信道码字的步骤包括: 2. The method according to claim 1, wherein the correlation characteristic of the channel is a correlation channel, and the step of constructing a channel codeword with a rank of 2 of the associated channel comprises: 选取 个 M维矢量 M -w.ff , 其中, K为码本中需要产生的相关码字个 数, M为发送端的发射天线数目的一半; 以及 Selecting an M-dimensional vector M - w. ff , where K is the number of related code words to be generated in the codebook, and M is half of the number of transmitting antennas at the transmitting end; 利用 ^个 M维矢量 M -M^构造秩为 2的码字为 0^ =^^ Un2], 其中, Constructing a codeword with a rank of 2 using ^M-dimensional vectors M -M^ is 0^ =^^ U n2 ], where = αΛ +*Α , "Ac 为常数, w„为所述^"
Figure imgf000028_0002
= c u, +d u 个 M维矢量 ^ ··· 中任意一个矢量-
= αΛ + * Α , "Ac is a constant, w„ is the ^"
Figure imgf000028_0002
= cu, +du M-dimensional vectors ^ ··· Any vector -
3、如权利要求 2所述的构造方法,其中,所述秩为 2的码字 CW„ The construction method according to claim 2, wherein said code word of rank 2 is CW 包括 Including
Figure imgf000028_0001
Figure imgf000028_0001
4、 如权利要求 1所述的构造方法, 其中, 所述信道的相关特性为相关信 道, 发送端的天线极化特性为 - /+ 极化, 接收端的天线极化特性为 1根天 4. The method according to claim 1, wherein the correlation characteristic of the channel is a correlation channel, the antenna polarization characteristic of the transmitting end is - / + polarization, and the antenna polarization characteristic of the receiving end is 1 day. 4 4  4 4 线垂直极化时, 该构造相关信道的秩为 1的信道码字的步骤包括: 选取 个 M维矢量 M -w.ff , 其中, K为码本中需要产生的相关码字个 数, M为发送端的发射天线数目的一半; 以及 利用 ^个 M维矢量 M w.ff构造秩为 1的码字为 RW„When the line is vertically polarized, the step of constructing the channel codeword of the rank 1 of the associated channel includes: selecting an M-dimensional vector M - w. ff , where K is the number of related codewords to be generated in the codebook, M Half of the number of transmit antennas at the transmitting end; Constructing a codeword with rank 1 as RW using ^ M-dimensional vectors M w. ff
Figure imgf000029_0001
wn为所述 ^个 M维矢量 M · · 中任意一个矢量 t 5、 如权利要求 1所述的构造方法, 其中, 所述信道的相关特性为相关信道, 发送端的天线极化特性为 极
Figure imgf000029_0001
^ w n is the M-dimensional vector of any one of M · · vector t 5, the construction method as claimed in claim 1, wherein the channel correlation property for the relevant channel, transmitting end antenna polarization characteristics of polar
4 4 化, 接收端的天线极化特性为 1 根天线 45度极化时, 该构造相关信道的秩 为 1的信道码字的步骤包括: 选取 个 2M维矢量
Figure imgf000029_0002
其中, ^"为码本中需要产生的相关码字 个数, Μ为发送端的发射天线数目的一半; 以及 利用 ^个 2Μ维矢量 M w.ff构造秩为 1的码字为 RW„ =
4 4, when the antenna polarization characteristic of the receiving end is 45-degree polarization of one antenna, the step of constructing the channel codeword of the rank 1 of the relevant channel includes: selecting a 2M-dimensional vector
Figure imgf000029_0002
Where ^" is the number of relevant codewords to be generated in the codebook, Μ is half of the number of transmit antennas at the transmitting end; and constructing a codeword with a rank of 1 using ^2Μ-dimensional vectors M w. ff is RW„ =
Figure imgf000029_0003
其中: w„为所述 ^"个 2M维矢量 Μ ·· 中任意一个矢量, μη, 为实数,
Figure imgf000029_0003
Where: w„ is any vector of the 2" dimensional vector Μ ··, μ η , is a real number,
2Μ维的码字的前面 Μ个元素组成的矢量和后面 Μ个元素组成的矢量,相差 一个固定的相位^ ei,p= , 前 Μ个元素组成的矢量为 ^„乘以一个常系数 /„, 后 Μ个元素组成的矢量为 ^乘以一个常系数/„。 The vector consisting of two elements in front of the two-dimensional codeword and the vector consisting of the following elements are separated by a fixed phase ^ e i,p = , and the vector consisting of the preceding elements is ^„ multiplied by a constant coefficient / „, the vector consisting of the last element is ^ multiplied by a constant coefficient / „. 6、 如权利要求 5 所述的构造方法, 其中, 所述秩为 1 的码字
Figure imgf000029_0004
6. The construction method according to claim 5, wherein the codeword of rank 1 is
Figure imgf000029_0004
7、 如权利要求 2至 6中任意一项所述的构造方法, 其中, 上述的相关信 道码字构造中, 所述任意一个矢量 W„, 符合以下特征: The construction method according to any one of claims 2 to 6, wherein in the above-described correlation channel codeword structure, the arbitrary one of the vectors W „ conforms to the following features: 1€
Figure imgf000029_0005
έθ" " … 其中, τ为一个复常数, 为一个复常数值, Α··· 在 布, 或在 - 〜 内均匀分布 t
1€
Figure imgf000029_0005
έ θ "1θ" ... where, [tau] is a complex constant, a normalization value, Α ··· cloth, or - the uniform distribution t ~
8、 一种多天线系统中信道信息量化码本的构造方法, 所述方法包括: 根据信道的相关特性、 发送端和接收端的天线极化特性, 构造信道矩阵 的秩为 1的信道码字和秩为 2的信道码字, 形成信道信息量化码本, 所述信 道信息量化码本包括针对不相关信道构造的与双极化匹配的信道矩阵的秩为 1的信道码字和秩为 2的信道码字。 A method for constructing a channel information quantization codebook in a multi-antenna system, the method comprising: constructing a channel codeword of a rank 1 of a channel matrix according to a correlation characteristic of a channel, an antenna polarization characteristic of a transmitting end and a receiving end; a channel codeword of rank 2, forming a channel information quantization codebook, the channel information quantization codebook comprising a channel codeword of rank 1 and a rank of 2 of a channel matrix matched with dual polarization for uncorrelated channel construction Channel codeword. 9、 如权利要求 8所述的构造方法, 其中, 所述信道的相关特性为不相关 信道, 发送端的天线极化特性为- /+ 极化, 接收端的天线极化特性为多根  9. The method according to claim 8, wherein the correlation characteristic of the channel is an uncorrelated channel, the antenna polarization characteristic of the transmitting end is - / + polarization, and the antenna polarization characteristic of the receiving end is multiple roots. 4 4  4 4 天线垂直极化时, 该构造不相关信道的秩为 2的信道码字的步骤包括: 选取 个 M维矢量 M - w.ff , 其中, K为码本中需要产生的相关码字个 数, M为发送端的发射天线数目的一半, ^…!^在 M维子空间内均匀分布 或准均匀分布; 以及 利用 ^个 M 维矢量 M w.ff构造秩为 2 的码字为 Cf „When the antenna is vertically polarized, the step of constructing the channel codeword of the rank 2 of the uncorrelated channel includes: selecting an M-dimensional vector M - w. ff , where K is the number of related codewords to be generated in the codebook, M is half the number of transmit antennas at the transmitting end, ^...! ^ uniformly distributed or quasi-uniformly distributed in the M-dimensional subspace; and constructing a codeword of rank 2 using the M-dimensional vectors M w. ff as Cf „
Figure imgf000030_0001
Figure imgf000030_0001
CW„ = , 其中: w„为所述 ^个 M维矢量 Μ · · 中任意一个矢量 w„ - w„ 10、 如权利要求 8所述的构造方法, 其中, CW „ = , where: w „ is the vector of any one of the M-dimensional vectors w · ·, the configuration method according to claim 8, wherein 所述信道的相关特性为不相关信道,发送端的天线极化特性为 - /+ 极  The correlation characteristic of the channel is an uncorrelated channel, and the antenna polarization characteristic of the transmitting end is - / + pole 4 4 化, 接收端的天线极化特性为多根天线 垂直极化时, 该构造不相关信  4 4 , the antenna polarization characteristic of the receiving end is multiple antennas. When the vertical polarization is used, the structure is uncorrelated. 4 4  4 4 道的秩为 2的信道码字的步骤包括: 选取 2 ΛΓ个 M维矢量 ^…!^和!^… ¼^ , 其中, AT为码本中需要产生 的相关码字个数, M为发送端的发射天线数目的一半, ^…!^在 M维子空 间内均匀分布或准均匀分布; 以及 利用 ^个 M 维矢量 M w.ff构造秩为 2 的码字为 C „The steps of the channel codeword with rank 2 of the track include: selecting 2 M M-dimensional vectors ^...! ^And! ^... 1⁄4^ , where AT is the number of associated codewords to be generated in the codebook, and M is half the number of transmit antennas at the transmitting end, ^...! ^ uniformly distributed or quasi-uniformly distributed in the M-dimensional subspace; and constructing a codeword of rank 2 using the M-dimensional vectors M w. ff as C „
Figure imgf000030_0002
, 其中: 和^为任意的矢量, W„和^的弦距离大于或等
Figure imgf000031_0001
Figure imgf000030_0002
, where: and ^ are arbitrary vectors, and the distance between the strings of W„ and ^ is greater than or equal to
Figure imgf000031_0001
于门限阔值,其中弦距离 ^^^,^ 定义为 和^:两个矢量之间夹角的求正 弦值。 At the threshold value, where the chord distance ^^^,^ is defined as and ^: the sine of the angle between the two vectors.
11、 如权利要求 10 所述的构造方法, 其中, 所述利用 ^个 Μ维矢量 M - w.ff构造秩为 2 的码字的步骤还包括: 获得所述任意矢量 ^和 , 所述 获得所述任意矢量 和^的步骤包括: The constructing method according to claim 10, wherein the step of constructing the codeword of rank 2 by using the two dimension vectors M - w. ff further comprises: obtaining the arbitrary vector ^ and the obtaining The steps of the arbitrary vector and ^ include: 存储或实时计算获得 M维的多个矢量集合 Ω , 矢量集合 Ω中任意 2个 码字的弦距离都大于或等于所述门限阔值;  Store or real-time calculation to obtain multiple vector sets Ω of M-dimensional, vector set Ω The chord distance of any two codewords is greater than or equal to the threshold threshold; 通过釆用格拉斯曼( Grassmannian )理论的方法对所述矢量集合 Ω进行线 压缩; 以及  Linearly compressing the vector set Ω by using the Grassmannian theory; and 从压缩后的矢量集合 Ω中任意取 2个不同的矢量作为 w„和^。  Take two different vectors from the compressed vector set Ω as w„ and ^. 12、 如权利要求 11所述的构造方法, 其中, 从压缩后的矢量集合 Ω中任 意选取 和^ 的步骤包括: 从压缩后的矢量集合 Ω中任意选取 M ^和 时, 对于任意的 i 和 j , wt≠ Wj≠ wt≠ Wj; 或者, 从压缩后的矢量集合 Ω中任意选取 ^和 ^时, 若 Ω为多天线系统 中 Μ天线非双极化信道秩为 1的码本, 对于任意的 i和 j , w; 或^ = ^ , 但不能同时相等。 12. The constructing method according to claim 11, wherein the step of arbitrarily selecting and selecting from the compressed vector set Ω comprises: arbitrarily selecting M^ from the compressed vector set Ω, for any i and j , w t ≠ Wj≠ w t ≠ Wj; or, if φ and ^ are randomly selected from the compressed vector set Ω, if Ω is a codebook of a non-dual-polarized channel rank of 1 in a multi-antenna system, For any i and j, w; or ^ = ^, but not equal at the same time. 13、 一种多天线系统的信道信息量化码本的构造装置, 所述装置包括: 码字构造模块, 其设置为: 根据信道的相关特性、 发送端和接收端的天 线极化特性, 构造信道矩阵的秩为 1的信道码字和秩为 2的信道码字, 形成 信道信息量化码本, 所述信道信息量化码本包括与双极化匹配的信道矩阵的 秩为 1的信道码字和秩为 2的信道码字。 13. A device for constructing a channel information quantization codebook of a multi-antenna system, the device comprising: a codeword construction module, configured to: construct a channel matrix according to a correlation characteristic of a channel, an antenna polarization characteristic of a transmitting end and a receiving end, a channel codeword of rank 1 and a channel codeword of rank 2, forming a channel information quantization codebook, the channel information quantization codebook including a channel codeword and rank of rank 1 of a channel matrix matched with dual polarization Is a channel codeword of 2. 14、 如权利要求 13所述的构造装置, 所述构造装置还包括: 天线极化特性检测模块, 其设置为: 检测获取接收端和发送端的天线极 化特性, 将天线极化特性发送给码字构造模块。 14. The constructing device of claim 13, the constructing device further comprising: The antenna polarization characteristic detecting module is configured to: detect and acquire antenna polarization characteristics of the receiving end and the transmitting end, and send the antenna polarization characteristic to the code word construction module. 15、 如权利要求 14所述的构造装置, 其中, 所述码字构造模块包括矢量 选择单元、 相关码字生成单元和不相关码字生成单元; 所述矢量选择单元设置为: 选择进行码字构造的矢量;  The constructing apparatus according to claim 14, wherein the codeword constructing module comprises a vector selecting unit, a correlated codeword generating unit, and an unrelated codeword generating unit; and the vector selecting unit is configured to: select a codeword Constructed vector 所述相关码字生成单元设置为: 针对相关信道利用所选择的矢量生成码 字;  The associated codeword generating unit is configured to: generate a codeword using the selected vector for the relevant channel; 所述不相关码字生成单元设置为: 针对不相关信道利用所选择的矢量生 成码字。  The uncorrelated codeword generating unit is configured to: generate a codeword using the selected vector for the uncorrelated channel. 16、 如权利要求 15所述的构造装置, 其中,  16. The construction device according to claim 15, wherein 所述码字构造模块是设置为: 按如下方式构造针对相关信道的秩为 2的 信道码字:  The codeword construction module is configured to: construct a channel codeword of rank 2 for the relevant channel as follows: 矢量选择单元选取^ "个 M维矢量
Figure imgf000032_0001
其中, Κ为码本中需要产生 的相关码字个数, Μ为发送端的发射天线数目的一半;
Vector selection unit selects ^ "M-dimensional vectors
Figure imgf000032_0001
Where Κ is the number of related codewords to be generated in the codebook, and Μ is half of the number of transmitting antennas at the transmitting end;
相关码字生成单元利用 ^"个 Μ 维矢量 M w.ff构造秩为 2 的码字为 cwn = [unl u , 其中, unl =\W" un2 =\ ° ], U =anul +b u2 ^ a b c d^ t, Wn为所述 K L0」 Lw"」 Un2 =cnU +dnu2 个 M维矢量 ^…! ^中任意一个矢量。 The correlation codeword generating unit constructs a codeword of rank 2 using ^" Μ dimension vector M w. ff as cw n = [u nl u , where u nl =\ W " u n2 =\ ° ], U = a n u l +b u 2 ^ abcd ^ t , Wn is the KL 0 "L w "" U n2 =c nU +d n u 2 M-dimensional vectors ^...! Any vector in ^.
17、 如权利要求 16 所述的构造装置, 其中, 所述秩为 2 的码字 cwn = [unl f/„2]包括: cwn =17. The constructing apparatus according to claim 16, wherein the codeword cw n = [u nl f / „ 2 ] of rank 2 comprises: cw n =
Figure imgf000032_0002
或 c „
Figure imgf000032_0002
Or c „
Figure imgf000032_0003
Figure imgf000032_0003
18、 如权利要求 15所述的构造装置, 其中 天线极化特性检测模块是设置为:检测发送端的天线极化特性为 The configuration device according to claim 15, wherein the antenna polarization characteristic detecting module is configured to: detect that the antenna polarization characteristic of the transmitting end is 4 4 极化, 接收端的天线极化特性为 1根天线垂直极化时; 以及 码字构造模块是设置为: 按如下方式构造针对相关信道的秩为 1的信道 码字: 矢量选择单元选取^ "个 M维矢量
Figure imgf000033_0001
其中, Κ为码本中需要产生 的相关码字个数, Μ为发送端的发射天线数目的一半;
4 4 Polarization, the antenna polarization characteristic at the receiving end is 1 antenna vertical polarization; and the codeword construction module is set to: construct a channel codeword with rank 1 for the relevant channel as follows: Vector selection unit selects ^ " M-dimensional vector
Figure imgf000033_0001
Where Κ is the number of related codewords to be generated in the codebook, and Μ is half of the number of transmitting antennas at the transmitting end;
相关码字生成单元利用 ^"个 Μ 维矢量 M w.ff构造秩为 1 的码字为 The correlation codeword generating unit constructs a codeword of rank 1 using ^" Μ dimension vector M w. ff as , 其中: w„为所述 ^个 M维矢量 Μ · · 中任意一个矢量 t , where: w„ is the ^ M-dimensional vector Μ · · any one of the vectors t
19、 如权利要求 15所述的构造装置, 其中, 天线极化特性检测模块是设置为:检测发送端的天线极化特性为 The configuration device according to claim 15, wherein the antenna polarization characteristic detecting module is configured to: detect that the antenna polarization characteristic of the transmitting end is 4 4 极化, 接收端的天线极化特性为 1 根天线 45度极化时; 以及 码字构造模块是设置为: 按如下方式构造针对相关信道的秩为 1的信道 码字: 矢量选择单元选取^ "个 2M维矢量 M - W 其中, ^"为码本中需要产 生的相关码字个数, M为发送端的发射天线数目的一半;  4 4 Polarization, the antenna polarization characteristic of the receiving end is 1 antenna 45 degree polarization; and the codeword construction module is set to: construct the channel codeword with rank 1 for the relevant channel as follows: Vector selection unit selection ^ "2M-dimensional vectors M - W where ^" is the number of associated codewords to be generated in the codebook, and M is half the number of transmit antennas at the transmitting end; 相关码字生成单元利用 ^"个 2M维矢量 M w.ff构造秩为 1的码字为
Figure imgf000033_0002
其中: w„为所述 ^个 2M维矢量 Μ · · 中任意一个矢量, μη , 为实数, 2Μ维的码字的前面 Μ个元素组成的矢量和后面 Μ个元素组成的矢量,相差 一个固定的相位^ ei,p = , 前 Μ个元素组成的矢量为 ^„乘以一个常系数 /„, 后 Μ个元素组成的矢量为 ^乘以一个常系数/„。
The correlation codeword generating unit constructs a codeword of rank 1 by using "2M-dimensional vectors M w. ff as
Figure imgf000033_0002
Where: w„ is any vector of the 2M-dimensional vector Μ · · , μ η , is a real number, a vector consisting of two elements preceding the 2 Μ code word and a vector consisting of the following elements, one difference The fixed phase ^ e i,p = , the vector consisting of the first element is ^„ multiplied by a constant coefficient / „, the vector consisting of the last element is ^ multiplied by a constant coefficient / „.
20、 如权利要求 19 所述的构造装置, 其中, 所述秩为 1 的码字
Figure imgf000034_0001
20. The constructing apparatus according to claim 19, wherein the codeword of rank 1 is
Figure imgf000034_0001
21、 如权利要求 15至 20中任一项所述的构造装置, 其中, 所述矢量选 择单元还设置为: 选取所述任意一个矢量 , 符合以下特征: The constructing apparatus according to any one of claims 15 to 20, wherein the vector selecting unit is further configured to: select any one of the vectors, and meet the following characteristics: 1€
Figure imgf000034_0002
έθ" " … 其中, τ为一个复常数, 为一个复常数值, Α · · · 在 布, 或在 - 〜 内均匀分布 t
1€
Figure imgf000034_0002
έ θ "1θ" ... where, [tau] is a complex constant, a normalization value, Α · · · cloth, or - ~ t uniform distribution
22、 如权利要求 15所述的构造装置, 其中, 天线极化特性检测模块是设置为:检测发送端的天线极化特性为 The configuration device according to claim 15, wherein the antenna polarization characteristic detecting module is configured to: detect that the antenna polarization characteristic of the transmitting end is 4 4 极化, 接收端的天线极化特性为多根天线垂直极化时; 以及 码字构造模块是设置为: 按如下方式构造针对不相关信道的秩为 2的信 道码字: 矢量选择单元选取^ "个 M维矢量
Figure imgf000034_0003
其中, Κ为码本中需要产生 的相关码字个数, Μ为发送端的发射天线数目的一半, ^…!^在 Μ维子空 间内均匀分布或准均匀分布;
4 4 polarization, the antenna polarization characteristic of the receiving end is when the multiple antennas are vertically polarized; and the codeword construction module is set to: construct a channel codeword of rank 2 for the uncorrelated channel as follows: Vector selection unit selection ^ "M-dimensional vector
Figure imgf000034_0003
Where Κ is the number of related codewords to be generated in the codebook, and Μ is half the number of transmitting antennas at the transmitting end, ^...! ^ evenly distributed or quasi-uniform distribution in the dimension space;
不相关码字生成单元利用 ^"个 Μ维矢量 M w.ff构造秩为 2 的码字为 The irrelevant codeword generating unit constructs a codeword of rank 2 by using the "" dimensional vector M w. ff as CW„ = ,其中: w„为所述 ^个 M维矢量 Μ · · ι ^中
Figure imgf000034_0004
CW„ = , where: w„ is the ^ M-dimensional vector Μ · · ι ^
Figure imgf000034_0004
任意一个矢量。 Any vector.
23、 如权利要求 15所述的构造装置, 其中, 天线极化特性检测模块是设置为:检测发送端的天线极化特性为 The configuration device according to claim 15, wherein the antenna polarization characteristic detecting module is configured to: detect that the antenna polarization characteristic of the transmitting end is 4 4 极化, 接收端的天线极化特性为多根天线 - /+ 垂直极化时; 以及  4 4 polarization, the antenna polarization characteristics at the receiving end are multiple antennas - / + when vertically polarized; 4 4 码字构造模块是设置为: 按如下方式构造针对不相关信道的秩为 2的信 道码字: 矢量选择单元选取 2 AT个 M维矢量 ^ · · · wff和 ^ · · · ¼^ , 其中, K为码 本中需要产生的相关码字个数,Μ为发送端的发射天线数目的一半, ^…!^ 在 Μ维子空间内均勾分布或准均勾分布; 4 4 The codeword construction module is configured to: construct a channel codeword of rank 2 for the uncorrelated channel as follows: The vector selection unit selects 2 AT M-dimensional vectors ^ · · · w ff and ^ · · · 1⁄4^ , where , K is the number of related codewords to be generated in the codebook, and Μ is half the number of transmitting antennas at the transmitting end, ^...! ^ Hook distribution or quasi-homogeneous distribution in the Weiweizi space; 不相关码字生成单元利用 ^"个 Μ维矢量 M w.ff构造秩为 2 的码字为 cwn=\ w" ^^或 c „= " w- ],其中: 和^:为任意的矢量, 和^:的The irrelevant codeword generating unit constructs a codeword of rank 2 using c" n dimension vectors M w. ff as cw n =\ w " ^^ or c „= " w - ], where: and ^: are arbitrary Vector, and ^: \_-wn v„」 - 《」 弦距离大于或等于门限阔值,其中弦距离 ^^ ,^;)定义为 和^:两个矢量 之间夹角的求正弦值。 \_-w n v„" - The string distance is greater than or equal to the threshold width, where the chord distance ^^ , ^;) is defined as the sine of the angle between the two vectors. 24、 如权利要求 23所述的构造装置, 其中, 矢量选择单元是设置为: 选取所述任意矢量 和^时, 存储或实时计算 获得 M维的多个矢量集合 Ω , 矢量集合 Ω中任意 2个码字的弦距离都大于或 等于所述门限阔值; 通过釆用格拉斯曼(Grassmannian)理论的方法对所述矢 量集合 Ω进行线压缩; 从压缩后的矢量集合 Ω中任意取 2个不同的矢量作为 The constructing apparatus according to claim 23, wherein the vector selecting unit is configured to: when the arbitrary vector sum is selected, store or real-time calculate to obtain a plurality of vector sets Ω of M dimensions, and any of the vector sets Ω The chord distances of the codewords are all greater than or equal to the threshold threshold; the vector set Ω is linearly compressed by the method of Grassmannian theory; and the arbitrary vector set Ω is taken from the compressed vector set Ω Different vectors as 25、如权利要求 24所述的构造装置,其中,所述矢量选择单元还设置为: 所述压缩后的矢量集合 Ω中, 对于任意的 i和 j, WiWj≠^≠^ ; 或者, 所述压缩后的矢量集合 Ω中, Ω为多天线系统中 Μ天线非双极化 信道的秩为 1的码本,对于任意的 i和 j, Wi= 或^ = 但不能同时相等。 The constructing apparatus according to claim 24, wherein said vector selecting unit is further configured to: in said compressed vector set Ω, for any i and j, WiWj ≠ ^ ≠ ^ ; In the compressed vector set Ω, Ω is a codebook of rank 1 of the Μ antenna non-dual-polarized channel in the multi-antenna system, and Wi = or ^ = but not equal at the same for any i and j.
PCT/CN2010/075994 2009-10-10 2010-08-13 Method and device for constructing channel information quantization codebook of multi-antenna system Ceased WO2011041965A1 (en)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN 200910181128 CN101667895B (en) 2009-10-10 2009-10-10 Construction method of channel information quantization codebook in multi-antenna system and device thereof
CN200910181128.1 2009-10-10

Publications (1)

Publication Number Publication Date
WO2011041965A1 true WO2011041965A1 (en) 2011-04-14

Family

ID=41804352

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2010/075994 Ceased WO2011041965A1 (en) 2009-10-10 2010-08-13 Method and device for constructing channel information quantization codebook of multi-antenna system

Country Status (2)

Country Link
CN (1) CN101667895B (en)
WO (1) WO2011041965A1 (en)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2014513482A (en) * 2011-04-28 2014-05-29 アルカテル−ルーセント Method and apparatus for generating codebooks and associated data in a network device
CN104541470A (en) * 2013-05-10 2015-04-22 上海贝尔股份有限公司 Multi-antenna information channel extraction codebook feedback method and device in MIMO system

Families Citing this family (11)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101667895B (en) * 2009-10-10 2013-12-18 中兴通讯股份有限公司南京分公司 Construction method of channel information quantization codebook in multi-antenna system and device thereof
CN101826941B (en) * 2010-04-02 2015-12-16 中兴通讯股份有限公司 The feedback method of channel quantitative information and device
CN101834707B (en) * 2010-04-04 2016-03-30 中兴通讯股份有限公司 A kind of obtain channel condition information method and corresponding communication system
CN101834706B (en) * 2010-04-05 2016-08-03 中兴通讯股份有限公司 The codebook quantification feedback method of a kind of channel information and system
CN101854236B (en) 2010-04-05 2015-04-01 中兴通讯股份有限公司 Method and system for feeding back channel information
CN104135348B (en) * 2010-05-04 2018-03-27 华为技术有限公司 Precoding processing method and user equipment
CN105075142B (en) * 2013-03-27 2019-05-24 上海诺基亚贝尔股份有限公司 Multi-antenna channel codebook feedback method and device in MIMO system
CN104753628B (en) * 2013-12-26 2019-12-10 中兴通讯股份有限公司 Channel information feedback method, system, base station and terminal
US9319249B2 (en) * 2014-08-27 2016-04-19 eTopus Technology Inc. Receiver for high speed communication channel
CN111327398B (en) * 2020-02-11 2021-06-22 北京邮电大学 Signal transmitting and receiving method and device of polarized multi-antenna serial number modulation system
CN111342931B (en) * 2020-02-11 2021-08-17 北京邮电大学 Encoding and decoding method and device for polarized multi-antenna generalized serial number modulation system

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101330479A (en) * 2007-06-20 2008-12-24 中兴通讯股份有限公司 Method for pre-encoding multi-input multi-output transmission and codebook encoding
WO2009022871A2 (en) * 2007-08-14 2009-02-19 Lg Electronics Inc. Method for feedback and method for configuring a codebook in multi-antenna system
KR20090054357A (en) * 2007-11-26 2009-05-29 엘지전자 주식회사 Codebook for Channel Adaptation and Precoding Method Using the Same
CN101667895A (en) * 2009-10-10 2010-03-10 中兴通讯股份有限公司 Construction method of channel information quantization codebook in multi-antenna system and device thereof

Family Cites Families (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101252381B (en) * 2008-04-01 2011-07-20 东南大学 Multi-aerial radio communication system high efficiency feedback method based on tree code book

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101330479A (en) * 2007-06-20 2008-12-24 中兴通讯股份有限公司 Method for pre-encoding multi-input multi-output transmission and codebook encoding
WO2009022871A2 (en) * 2007-08-14 2009-02-19 Lg Electronics Inc. Method for feedback and method for configuring a codebook in multi-antenna system
KR20090054357A (en) * 2007-11-26 2009-05-29 엘지전자 주식회사 Codebook for Channel Adaptation and Precoding Method Using the Same
CN101667895A (en) * 2009-10-10 2010-03-10 中兴通讯股份有限公司 Construction method of channel information quantization codebook in multi-antenna system and device thereof

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2014513482A (en) * 2011-04-28 2014-05-29 アルカテル−ルーセント Method and apparatus for generating codebooks and associated data in a network device
CN104541470A (en) * 2013-05-10 2015-04-22 上海贝尔股份有限公司 Multi-antenna information channel extraction codebook feedback method and device in MIMO system
CN104541470B (en) * 2013-05-10 2017-09-15 上海贝尔股份有限公司 Multi-antenna channel in mimo system extracts code book feedback method and device

Also Published As

Publication number Publication date
CN101667895A (en) 2010-03-10
CN101667895B (en) 2013-12-18

Similar Documents

Publication Publication Date Title
WO2011041965A1 (en) Method and device for constructing channel information quantization codebook of multi-antenna system
US9178590B2 (en) Channel information feedback method and wireless communication device using the same
JP5878646B2 (en) Method and apparatus for feeding back channel state information
WO2018127044A1 (en) Communication method, base station and terminal device
CN110086516A (en) The method and apparatus for being used to indicate the precoder parameter in cordless communication network
CN101854236B (en) Method and system for feeding back channel information
WO2011020383A1 (en) Method and apparatus for constructing pre-coding codebook in multiple-input multiple-output system
CN109474315A (en) A method and device for indicating and determining a precoding matrix
WO2015018030A1 (en) Method for determining precoding matrix indicator, receiving device and transmitting device
CN115053465A (en) Information transmission method and device
CN104753628B (en) Channel information feedback method, system, base station and terminal
CN111800172A (en) A communication method and device
CN107733476A (en) The feedback method and device of channel condition information
CN107181513A (en) The feedback method and device of a kind of channel condition information
TW201817181A (en) Apparatus and method for precoder selection in multiple-input multiple-output (MIMO) systems, manufacturing method of apparatus for precoder selection and method for constructing integrated circuit method having apparatus for precoder selection
WO2011095064A1 (en) Method and apparatus for transmitting/receiving information in closed loop multiple antenna system
CN102725991B (en) Feedback method and system of correlation matrix for antenna array
CN111435848A (en) Method and communication device for indicating and determining precoding vector
WO2020063804A1 (en) Enhanced type ii channel state information in mobile communications
CN107222249B (en) Method and device for acquiring channel state information
CN107046431B (en) Information transmission and receiving method and device
WO2017167157A1 (en) Method and device for processing channel information
WO2017107671A1 (en) Method for transmitting and receiving channel information and apparatus, terminal and base station therefor
WO2011137729A1 (en) Method and user equipment for pre-coding processing
CN114710188A (en) Transmission method and device

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: 10821574

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: 10821574

Country of ref document: EP

Kind code of ref document: A1