METHOD AND APPARATUS FOR DETERMINING PRECODING MATRIX INDICATOR, USER EQUIPMENT, AND BASE STATION

Abstract

A precoding matrix indicator (PMI) is determined for a user equipment or a base station, where the PMI corresponds to a precoding matrix W, and the precoding matrix W satisfies a first condition, a second condition, or a third condition; and the PMI is sent to a base station. The precoding matrix indicator can effectively control a beam, especially a beam shape and a beam orientation, in a horizontal direction and a perpendicular direction.

Claims

1. A method for determining a precoding matrix indicator (PMI), comprising: receiving, by a user equipment, a reference signal sent by a base station; and determining, by the user equipment, the PMI according to the reference signal, wherein the PMI corresponds to a precoding matrix W, and the precoding matrix W satisfies: the precoding matrix W comprises one or more column vectors of a block diagonal matrix W.sub.1, or the precoding matrix W is obtained by performing weighted combination on one or more column vectors of a block diagonal matrix W.sub.1, wherein W.sub.1=diag {X.sub.1, . . . ,X.sub.N.sub.B}, and N.sub.B1, wherein at least one block matrix X is a Kronecker product of a matrix A and a matrix B, X=A .Math. B, and X {X.sub.1,X.sub.2, . . . ,X.sub.N.sub.B}; the matrix A or the matrix B is a product of a matrix D and a matrix V; the matrix D is a diagonal matrix, D=.Math.diag {u.sub.1,u.sub.2, . . . ,u.sub.n,u.sub.n*,u.sub.n1*, . . . ,u.sub.1*}, is a complex factor, a complex number u.sub.i* is a conjugate complex number of a complex number u.sub.i, i=1, . . . ,n, and n is a quantity of row s of the matrix A or the matrix B; and the matrix V is a constant modulus matrix; wherein in the third condition, the precoding matrix W satisfies W=W.sub.1W.sub.2, wherein the matrix W.sub.2 is used to select one or more column vectors of the matrix W.sub.1; or is used to perform weighted combination on one or more column vectors of the W.sub.1 to obtain the precoding matrix W; and sending, by the user equipment, the PMI to the base station.

2. The method according to claim 1, wherein phases of diagonal elements u.sub.1,u.sub.2, . . . ,u.sub.n of the matrix D form an arithmetic progression.

3. The method according to claim 1, wherein the matrix V comprises at least one of a column vector 1 and one or more column vectors v.sub.m, the column vector 1 is a column vector whose elements are all 1, and any column vector v.sub.m is in the form v=[v.sub.1 v.sub.2 L v.sub.n v.sub.n v.sub.n1 L v.sub.1].sup.T, wherein an element is v.sub.i=v.sub.i,v.sub.i=1, and i=1, . . . ,n.

4. A method for determining a precoding matrix indicator, comprising: transmitting, by a base station, a reference signal to a user equipment; receiving, by the base station, a precoding matrix indicator PMI sent by the user equipment; determining, by the base station, a corresponding precoding matrix W according to the PMI, wherein the precoding matrix W satisfies: the precoding matrix W comprises one or more column vectors of a block diagonal matrix W.sub.1, or the precoding matrix W is obtained by performing weighted combination on one or more column vectors of a block diagonal matrix W.sub.1, wherein W.sub.1=diag {X.sub.1,X.sub.2, . . . ,X.sub.N.sub.B}, and N.sub.B24 1, wherein at least one block matrix X is a Kronecker product of a matrix A and a matrix B, X=A .Math. B, and X {X.sub.1,X.sub.2, . . . ,X.sub.N.sub.B}; the matrix A or the matrix B is a product of a matrix D and a matrix V, the matrix D is a diagonal matrix, i=1, . . . , n, and n is a quantity of rows of the matrix A or the matrix B, wherein the matrix D is a diagonal matrix, D=.Math.diag {u.sub.1,u.sub.2, . . . ,u.sub.n,u.sub.n*,u.sub.n18, . . . ,u.sub.1*}, is a complex factor, a complex number u.sub.i* is a conjugate complex number of a complex number u.sub.i, and n is determined by a quantity of antenna ports; and the matrix V is a constant modulus matrix; wherein in the second condition or the third condition, the precoding matrix W satisfies W=W.sub.1W.sub.2, wherein the matrix W.sub.2 is used to select one or more column vectors of the matrix W.sub.1; or is used to perform weighted combination on one or more column vectors of the W.sub.1 to obtain the precoding matrix W.

5. The method according to claim 4, wherein phases of diagonal elements u.sub.1,u.sub.2, . . . ,u.sub.n of the matrix D form an arithmetic progression.

6. The method according to claim 4, wherein the matrix V comprises at least one of a column vector 1 and one or more column vectors v.sub.m, the column vector 1 is a column vector whose elements are all 1, and any column vector v.sub.m is in the form v=[v.sub.1 v.sub.2 L v.sub.n v.sub.n v.sub.n1 L v.sub.1].sup.T, wherein an element is v.sub.i=v.sub.i,v.sub.i=1, and i=1, . . . ,n.

7. User equipment, comprising: a receiver, configured to receive a reference signal sent by a base station; a first processor, configured to determine a precoding matrix indicator PMI, wherein the PMI corresponds to a precoding matrix W, and the precoding matrix W satisfies the precoding matrix W comprises one or more column vectors of a block diagonal matrix W.sub.1, or the precoding matrix W is obtained by performing weighted combination on one or more column vectors of a block diagonal matrix W.sub.1, wherein W.sub.1=diag {X.sub.1, . . . ,X.sub.N.sub.B}, and N.sub.B1, wherein at least one block matrix X is a Kronecker product of a matrix A and a matrix B, X=A .Math. B, and X {X.sub.1,X.sub.2, . . . ,X.sub.N.sub.B}; the matrix A or the matrix B is a product of a matrix D and a matrix V; the matrix D is a diagonal matrix, D=.Math.diag {u.sub.1,u.sub.2, . . . ,u.sub.n,u.sub.n*,u.sub.n1*, . . . ,u.sub.1*}, is a complex factor, a complex number u.sub.i* is a conjugate complex number of a complex number u.sub.i, i=1, . . . .n, and n is a quantity of rows of the matrix A or the matrix B ; and the matrix V is a constant modulus matrix; wherein in the second condition or the third condition, the precoding matrix W satisfies W=w.sub.1W.sub.2, wherein the matrix W.sub.2 is used to select one or more column vectors of the matrix W.sub.1; or is used to perform weighted combination on one or more column vectors of the W.sub.1 to obtain the precoding matrix W; and a first transmitter, configured to send the PMI to the base station to be used by the base station.

8. The user equipment according to claim 7, wherein phases of diagonal elements u.sub.1,u.sub.2, . . . ,u.sub.n of the matrix D form an arithmetic progression.

9. The user equipment according to claim 7, wherein the matrix V comprises at least one of a column vector 1 and one or more column vectors v.sub.m, the column vector 1 is a column vector whose elements are all 1, and any column vector v.sub.m is in the form v=[v.sub.1 v.sub.2 L v.sub.n v.sub.n v.sub.n1 L v.sub.1].sup.T, wherein an element is v.sub.i=v.sub.i,v.sub.i=1, and i=1, . . . ,n.

10. Abase station, comprising: a transmitter, configured to transmit a reference signal to a user equipment; a receiver, configured to receive a precoding matrix indicator PMI sent by the user equipment; and a second processor, configured to: determine a corresponding precoding matrix W according to the PMI, wherein the precoding matrix W comprises one or more column vectors of a block diagonal matrix W.sub.1, or the precoding matrix W is obtained by performing weighted combination on one or more column vectors of a block diagonal matrix W.sub.1, wherein W.sub.1=diag {X.sub.1, . . . ,X.sub.N.sub.B}, and N.sub.B1, wherein at least one block matrix X is a Kronecker product of a matrix A and a matrix B, X=A .Math. B, and X {X.sub.1,X.sub.2, . . . ,X.sub.N.sub.B}; the matrix A or the matrix B is a product of a matrix D and a matrix V; the matrix D is a diagonal matrix, i=1, . . . ,n, and n is a quantity of rows of the matrix A or the matrix B, wherein the matrix D is a diagonal matrix, D=.Math.diag {u.sub.1,u.sub.2, . . . ,u.sub.n,u.sub.n*,u.sub.n1*, . . . ,u.sub.1*}, is a complex factor, a complex number u.sub.i* is a conjugate complex number of a complex number u.sub.i, and n is determined by a quantity of antenna ports; and the matrix V is a constant modulus matrix; wherein in the second condition or the third condition, the precoding matrix W satisfies W=W.sub.1W.sub.2, wherein the matrix W.sub.2 is used to select one or more column vectors of the matrix W.sub.1; or is used to perform weighted combination on one or more column vectors of the W.sub.1 to obtain the precoding matrix W.

11. The base station according to claim 10, wherein phases of diagonal elements u.sub.1,u.sub.2, . . . ,u.sub.n of the matrix D form an arithmetic progression.

12. The base station according to claim 10, wherein the matrix V comprises at least one of a column vector 1 and one or more column vectors v.sub.m, the column vector 1 is a column vector whose elements are all 1, and any column vector v.sub.m is in the form v=[v.sub.1 v.sub.2 L v.sub.n v.sub.n v.sub.n1 L v.sub.1].sup.T, wherein an element is v.sub.i=v.sub.i,v.sub.i=1, and i=1, . . . ,n.

13. The method according to claim 1, wherein the precoding matrix W is used to control a beam shape and a beam orientation, in a horizontal direction and a perpendicular direction.

Description

BRIEF DESCRIPTION OF DRAWINGS

[0206] To describe the technical solutions in the embodiments of the present invention more clearly, the following briefly introduces the accompanying drawings required for describing the embodiments. Apparently, the accompanying drawings in the following description show some embodiments of the present invention, and a person of ordinary skill in the art may still derive other drawings from these accompanying drawings without creative efforts.

[0207] FIG. 1 is a first schematic flowchart of a method for determining a precoding matrix indicator according to an embodiment of the present invention;

[0208] FIG. 2 is a second schematic flowchart of a method for determining a precoding matrix indicator according to an embodiment of the present invention;

[0209] FIG. 3 is a first schematic flowchart of a specific embodiment of the present invention;

[0210] FIG. 4 is a second schematic flowchart of a specific embodiment of the present invention;

[0211] FIG. 5 is a third schematic flowchart of a specific embodiment of the present invention;

[0212] FIG. 6 is a fourth schematic flowchart of a specific embodiment of the present invention;

[0213] FIG. 7 is a fifth schematic flowchart of a specific embodiment of the present invention;

[0214] FIG. 8 is a sixth schematic flowchart of a specific embodiment of the present invention;

[0215] FIG. 9 is a first schematic structural diagram of an apparatus for determining a precoding matrix indicator according to an embodiment of the present invention;

[0216] FIG. 10 is a second schematic structural diagram of an apparatus for determining a precoding matrix indicator according to an embodiment of the present invention;

[0217] FIG. 11 is a third schematic structural diagram of an apparatus for determining a precoding matrix indicator according to an embodiment of the present invention;

[0218] FIG. 12 is a fourth schematic structural diagram of an apparatus for determining a precoding matrix indicator according to an embodiment of the present invention;

[0219] FIG. 13 is a fifth schematic structural diagram of an apparatus for determining a precoding matrix indicator according to an embodiment of the present invention;

[0220] FIG. 14 is a sixth schematic structural diagram of an apparatus for determining a precoding matrix indicator according to an embodiment of the present invention;

[0221] FIG. 15 is a first schematic structural diagram of user equipment according to an embodiment of the present invention;

[0222] FIG. 16 is a first schematic structural diagram of a base station according to an embodiment of the present invention;

[0223] FIG. 17 is a second schematic structural diagram of user equipment according to an embodiment of the present invention;

[0224] FIG. 18 is a second schematic structural diagram of a base station according to an embodiment of the present invention;

[0225] FIG. 19 is a third schematic structural diagram of user equipment according to an embodiment of the present invention; and

[0226] FIG. 20 is a third schematic structural diagram of a base station according to an embodiment of the present invention.

DESCRIPTION OF EMBODIMENTS

[0227] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the following clearly describes the technical solutions in the embodiments of the present invention with reference to the accompanying drawings in the embodiments of the present invention. Apparently, the described embodiments are some but not all of the embodiments of the present invention. All other embodiments obtained by a person of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0228] An embodiment of the present invention provides a method for determining a precoding matrix indicator. FIG. 1 is a first schematic flowchart of the method for determining a precoding matrix indicator according to this embodiment of the present invention, and as shown in FIG. 1, the method includes the following steps:

[0229] Step 101: User equipment determines a precoding matrix indicator PMI, where the PMI corresponds to a precoding matrix W, and the precoding matrix W satisfies a first condition, a second condition, or a third condition.

[0230] Step 102: The user equipment sends the PMI to a base station.

[0231] The first condition is that the precoding matrix W satisfies W=DV; the second condition is that the precoding matrix W includes one or more column vectors of a block diagonal matrix W.sub.1, or the precoding matrix W is obtained by performing weighted combination on one or more column vectors of a block diagonal matrix W.sub.1, where W.sub.1=diag {X.sub.1, . . . ,X.sub.N.sub.B}, and N.sub.B1, where at least one block matrix X is a product X=DV of a matrix D and a matrix V, and X {X.sub.1,X.sub.2, . . . ,X.sub.N.sub.B}; and the third condition is that the precoding matrix W includes one or more column vectors of a block diagonal matrix W.sub.1, or the precoding matrix W is obtained by performing weighted combination on one or more column vectors of a block diagonal matrix W.sub.1, where W.sub.1=diag {X.sub.1, . . . ,X.sub.N.sub.B}, and N.sub.B1, where at least one block matrix X is a Kronecker product of a matrix A and a matrix B, X=A .Math. B, and X {X.sub.1, X.sub.2, . . . ,X.sub.N.sub.B}; the matrix A or the matrix B is a product of a matrix D and a matrix V; the matrix D is a diagonal matrix, i=1, . . . ,n, and n is a quantity of rows of the matrix A or the matrix B, where [0232] the matrix D is a diagonal matrix, D=.Math.diag {u.sub.1,u.sub.2, . . . ,u.sub.n, u.sub.n*, u.sub.n1*, . . . ,u.sub.1*}, is a complex factor, a complex number u.sub.i* is a conjugate complex number of a complex number u.sub.i, and n is determined by a quantity of antenna ports; and the matrix V is a constant modulus matrix.

[0233] In this embodiment, for detailed descriptions about three cases in which the precoding matrix W separately satisfies the first condition, the second condition, and the third condition, reference may be made to the following specific embodiments.

[0234] In addition, the method, provided in this embodiment of the present invention, for determining a precoding matrix indicator may further include: receiving a reference signal sent by the base station, and selecting, from a codebook according to the reference signal, the precoding matrix W corresponding to the PMI. Then, the determining a precoding matrix indicator PMI in the foregoing step 101 is specifically: determining the PMI according to the reference signal or the precoding matrix W.

[0235] Alternatively, the method, provided in this embodiment of the present invention, for determining a precoding matrix indicator may further include: receiving a reference signal sent by the base station. Then, the determining a precoding matrix indicator PMI in the foregoing step 101 is specifically: determining the precoding matrix indicator PMI according to the reference signal. In addition, after the determining the PMI according to the reference signal, the method further includes: determining the precoding matrix W according to the reference signal or the precoding matrix indicator PMI.

[0236] Corresponding to the embodiment, shown in FIG. 1, of the method on a user equipment side, the present invention further provides a method for determining a precoding matrix indicator on a base station side. FIG. 2 is a second schematic flowchart of the method for determining a precoding matrix indicator according to an embodiment of the present invention, and as shown in FIG. 2, the method includes the following steps:

[0237] Step 201: Receive a precoding matrix indicator PMI sent by user equipment.

[0238] Step 202: Determine a corresponding precoding matrix W according to the PMI, where the precoding matrix W satisfies a first condition, a second condition, or a third condition.

[0239] The first condition is that the precoding matrix W satisfies W=DV; [0240] the second condition is that the precoding matrix W includes one or more column vectors of a block diagonal matrix W.sub.1, or the precoding matrix W is obtained by performing weighted combination on one or more column vectors of a block diagonal matrix W.sub.1, where W.sub.1=diag {X.sub.1, . . . ,X.sub.N.sub.B} and N.sub.B1, where at least one block matrix X is a product X=DV of a matrix D and a matrix V, and X {X.sub.1,X.sub.2, . . . ,X.sub.N.sub.B}; and [0241] the third condition is that the precoding matrix W includes one or more column vectors of a block diagonal matrix W.sub.1, or the precoding matrix W is obtained by performing weighted combination on one or more column vectors of a block diagonal matrix W.sub.1, where W.sub.1=diag {X.sub.1, . . . ,X.sub.N.sub.B}, and N.sub.B1, where at least one block matrix X is a Kronecker product of a matrix A and a matrix B, X=A .Math. B, and X {X.sub.1X.sub.2, . . . ,X.sub.N.sub.B}; the matrix A or the matrix B is a product of a matrix D and a matrix V; the matrix D is a diagonal matrix, i=1, . . . ,n, and n is a quantity of rows of the matrix A or the matrix B, where [0242] the matrix D is a diagonal matrix, D=.Math.diag {u.sub.1,u.sub.2, . . . ,u.sub.n, u.sub.n*, u.sub.n1*, . . . ,u.sub.1*}, is a complex factor, a complex number u.sub.i* is a conjugate complex number of a complex number u.sub.i, and n is determined by a quantity of antenna ports; and the matrix V is a constant modulus matrix.

[0243] In the foregoing embodiment, for detailed descriptions about three cases in which the precoding matrix W separately satisfies the first condition, the second condition, and the third condition, reference may be made to the following specific embodiments. In addition, the determining a corresponding precoding matrix W according to the PMI in the foregoing step 202 may be specifically: selecting the corresponding precoding matrix W from a codebook according to the PMI.

[0244] FIG. 3 is a first schematic flowchart of a specific embodiment. This embodiment provides a method for determining a precoding matrix indicator that is executed on a user equipment side when a precoding matrix satisfies a first condition. As shown in FIG. 3, the method includes:

[0245] Step 301: User equipment receives a reference signal sent by a base station.

[0246] Specifically, the reference signal sent by the base station in this step may include: a channel state information reference signal (CSI RS), a demodulation reference signal (DM RS), or a cell-specific reference signal (CRS). The user equipment UE may acquire the reference signal by receiving a reference signal resource configuration notified by an eNB, or obtain a resource configuration of the reference signal according to a cell identity (cellID) and obtain the reference signal in a corresponding resource or subframe, where the eNB notification may be higher-layer signaling such as radio resource control (RRC) signaling, or dynamic signaling such as downlink control information (DCI). The higher-layer signaling is sent to the user equipment by using a physical downlink shared channel (PDSCH). The DCI may be sent to the user equipment by using a physical downlink control channel (PDCCH) or an enhanced PDCCH (ePDCCH).

[0247] Step 302: The user equipment selects, based on the reference signal, a precoding matrix from a codebook.

[0248] In this embodiment of the present invention, the codebook is a set of an available precoding matrix. The codebook and the PMI may be stored in a device in a table lookup manner; or a device may obtain the corresponding PMI by means of calculation by using a preset formula or algorithm according to the codebook, or obtain the corresponding codebook by means of calculation according to the PMI.

[0249] In an optional implementation manner of this embodiment, at least one precoding matrix W included in the codebook is a product of a matrix D and a matrix V ; and has the structure shown in formula (1):

W=DV(1) [0250] where the matrix D is a diagonal matrix, and satisfies:

D=.Math.diag{u.sub.1,u.sub.2, . . . ,u.sub.n,u.sub.n*,u.sub.n1*, . . . ,u.sub.1*}(2) [0251] where is a complex factor, and a real part or an imaginary part of the complex factor may be 0; a complex number u.sub.i* is a conjugate complex number of a complex number u.sub.i, and i=1, . . . ,n; and the matrix V is a constant modulus matrix, for example, elements of the matrix V maybe 1 or j.

[0252] It should be noted that, the so-called constant modulus matrix refers to a matrix whose elements have a same module or amplitude. It should be understood that, a constant modulus matrix generally is a non-diagonal matrix, or certainly, may be a diagonal matrix, for example, the constant modulus matrix is a matrix whose elements are 0.

[0253] In another optional implementation manner of this embodiment, the matrix V includes a column vector 1 and/or at least one column vector v, where the column vector 1 is a column vector whose elements are all 1, and the column vector v is:

v=[v.sub.1 v.sub.2 L v.sub.n v.sub.n v.sub.n1 L v.sub.1].sup.T, (3) [0254] where ( ).sup.T represents transposition of a matrix or vector, an element is v.sub.i=v.sub.i, i=1, . . . ,n, and v.sub.i=1, that is, a value of v.sub.i is +1 or 1. In an exemplary implementation manner, the matrix V is formed only by the column vector 1 and/or the at least one column vector v. That is, in the matrix V, except the included column vector 1, the other column vectors are column vectors v. Further, preferably, when the matrix V includes multiple vectors v, the multiple vectors v.sub.s are different; in this case, better orthogonality can be provided, thereby avoiding occurrence of strong interference.

[0255] In another optional implementation manner of this embodiment, the codebook includes at least: a precoding matrix W.sub.i and a precoding matrix W.sub.j, where the W.sub.i and the W.sub.j satisfy the formula (4):

W.sub.i=D(i,j)W.sub.i(4) [0256] where the matrix D(i,j) is a diagonal matrix; optionally, phases of diagonal elements .sub.1,.sub.2, . . . ,.sub.n of the matrix D(i,j) form an arithmetic progression, and the matrix has the structure shown in the formula (5):

D(i,j)=.sub.(i,j)diag{.sub.1,.sub.2, . . . ,.sub.n,.sub.n*,.sub.n1*, . . . ,.sub.1*}(5) [0257] where .sub.(i,j) is a complex factor, and a real part or an imaginary part of the complex factor may be 0; and a complex number .sub.m* is a conjugate complex number of a complex number .sub.m, m=1, . . . ,n, and n is determined by a quantity of antenna ports.

[0258] For the foregoing two precoding matrices in the codebook, the user equipment may select different precoding matrices from the codebook at different time points according to a preset rule or randomly, that is, the user equipment may determine a first precoding matrix indicator PMI at a time point, where the PMI corresponds to the precoding matrix W.sub.i in the codebook, and send the first PMI to the base station; and determine a second precoding matrix indicator PMI at another time point, where the PMI corresponds to the precoding matrix W.sub.j in the codebook, and send the second PMI to the base station.

[0259] Corresponding to the case in which the foregoing user equipment sends the first PMI or the second PMI at different time points, on a base station side, the base station may also receive, at a time point, the first precoding matrix indicator PMI sent by the user equipment, and select the corresponding precoding matrix W.sub.i from the codebook according to the first PMI; and receive, at another time point, the second precoding matrix indicator PMI sent by the user equipment, and select the corresponding precoding matrix W.sub.j from the codebook according to the second PMI.

[0260] Optionally, the codebook includes at least: a precoding matrix W.sub.i and a precoding matrix W.sub.k, where the W.sub.i and W.sub.k satisfy the formula (6):

D.sub.i.sup.1W.sub.i=D.sub.k.sup.1W.sub.k=V(6) [0261] where the matrix V includes a column vector 1 and/or at least one column vector v, the column vector 1 is a column vector whose elements are all 1, and the column vector v has the structure shown in the formula (3); and the matrix D.sub.i and the matrix D.sub.k both are diagonal matrices, and have the structure shown in the formula (7):

D.sub.m=.sub.m.Math.diag{u.sub.m,1,u.sub.m,2, . . . ,u.sub.m,n,u.sub.m,n*,u.sub.m,n1*, . . . ,u.sub.m,1*},m=i,k(7) [0262] where .sub.m is a complex factor, and a real part or an imaginary part of the complex factor may be 0; a complex number u.sub.m,l* is a conjugate complex number of a complex number u.sub.m,l, m=i,k, l=1, . . . ,n, and n is determined by a quantity of antenna ports; and optionally, phases of diagonal elements u.sub.m,1,u.sub.m,2, . . . ,u.sub.m,n of the matrix D.sub.m form an arithmetic progression.

[0263] For the foregoing two precoding matrices in the codebook, the user equipment may select different precoding matrices from the codebook at different time points according to a preset rule or randomly, that is, the user equipment may determine a first precoding matrix indicator PMI at a time point, where the PMI corresponds to the precoding matrix W.sub.i in the codebook, and send the first PMI to the base station; and determine a second precoding matrix indicator PMI at another time point, where the PMI corresponds to the precoding matrix W.sub.k in the codebook, and send the second PMI to the base station.

[0264] Corresponding to the case in which the foregoing user equipment sends the first PMI or the second PMI at different time points, on a base station side, the base station may also receive, at a time point, the first precoding matrix indicator PMI sent by the user equipment, and select the corresponding precoding matrix W.sub.i from the codebook according to the first PMI; and receive, at another time point, the second precoding matrix indicator PMI sent by the user equipment, and select the corresponding precoding matrix W.sub.k from the codebook according to the second PMI.

[0265] It should be pointed out that diagonal elements of the foregoing diagonal matrix may have same amplitude. In this case, the structure of the foregoing precoding matrix allows that transmit antennas corresponding to rows of the precoding matrix have symmetric transmit powers based on actual considerations, and in this case, the foregoing codebook can still control a beam orientation by using a symmetric property of the powers of the transmit antennas, and meanwhile ensure orthogonality between transmission layers.

[0266] Step 303: The user equipment sends a precoding matrix indicator PMI to the base station, where the PMI corresponds to the selected precoding matrix.

[0267] In the foregoing embodiment of the present invention, user equipment selects, based on a reference signal, a precoding matrix from a codebook, and sends a precoding matrix indicator PMI. A precoding matrix W included in the codebook is a product of a matrix D and a matrix V. The D is a diagonal matrix and satisfies D=.Math.diag {u.sub.1,u.sub.2, . . . ,u.sub.n,u.sub.n*,u.sub.n1*, . . . ,u.sub.1*}, where u.sub.1,u.sub.2, . . . ,u.sub.n,u.sub.n*u.sub.n1*, . . . ,u.sub.1* forms a conjugate and symmetric sequence, which avoids constant modulus restrictions or a limit that antennas perform transmission by using equal powers, and can effectively control a beam shape and a beam orientation.

[0268] Further, the matrix V includes a column vector 1 and/or at least one column vector v=[v.sub.1 v.sub.2 L v.sub.n v.sub.n v.sub.n1 L v.sub.1].sup.T, so that column vectors of the precoding matrix are orthogonal to each other, which can effectively reduce inter-layer interference, thereby greatly improving performance of MIMO, especially MU-MIMO. Therefore, the foregoing method for determining a precoding matrix can fully use a degree of freedom of controlling a beam shape and a beam orientation of an antenna system, and meanwhile reduce inter-layer interference of MIMO transmission as much as possible, thereby improving precision of CSI feedback, and a system throughput.

[0269] Using n=5 as an example, diagonal elements shown in the formula (2) may be:

[00001]$\begin{array}{cc}\left[u_1,u_2,u_3,u_4,u_5\right]=\left[e^{j.Math.\frac{}{2}},{\left(\frac{7}{6}\right)}^{\frac{1}{2}}.Math.e^{j.Math.\frac{}{3}},{\left(\frac{8}{6}\right)}^{\frac{1}{2}}.Math.e^{j.Math.\frac{}{4}},{\left(\frac{9}{6}\right)}^{\frac{1}{2}}.Math.e^{j.Math.\frac{}{6}},{\left(\frac{10}{6}\right)}^{\frac{1}{2}}\right]& \left(8\right)\end{array}$ [0270] correspondingly, the following formula is satisfied:

[00002]$\begin{array}{cc}\left[u_5^{*},u_4^{*},u_3^{*},u_2^{*},u_1^{*}\right]=\left[{\left(\frac{10}{6}\right)}^{\frac{1}{2}},{\left(\frac{9}{6}\right)}^{\frac{1}{2}}.Math.e^{-j.Math.\frac{}{6}},{\left(\frac{8}{6}\right)}^{\frac{1}{2}}.Math.e^{-j.Math.\frac{}{4}},{\left(\frac{7}{6}\right)}^{\frac{1}{2}}.Math.e^{-j.Math.\frac{}{3}},e^{-j.Math.\frac{}{2}}\right]& \left(9\right)\end{array}$

[0271] Alternatively, diagonal elements shown in the formula (2) may be:

[00003]$\begin{array}{cc}\left[u_1,u_2,u_3,u_4,u_5\right]=\left[e^{j.Math.\frac{}{2}},e^{j.Math.\frac{}{3}},e^{j.Math.\frac{}{4}},e^{j.Math.\frac{}{6}},1\right]& \left(10\right)\end{array}$ [0272] correspondingly, the following formula is satisfied:

[00004]$\begin{array}{cc}\left[u_5^{*},u_4^{*},u_3^{*},u_2^{*},u_1^{*}\right]=\left[1,e^{-j.Math.\frac{}{6}},e^{-j.Math.\frac{}{4}},e^{-j.Math.\frac{}{3}},e^{-j.Math.\frac{}{2}}\right]& \left(11\right)\end{array}$

[0273] Correspondingly, the column vector v may be:

v=[1 1 1 1 1 1 1 1 1 1].sup.T(12)

[0274] Optionally, as another embodiment, in the matrix D, phases of the diagonal elements u.sub.1,u.sub.2,. . . . ,u.sub.n form an arithmetic progression, and phases of the diagonal elements u.sub.n*,u.sub.n1*, . . . ,u.sub.1* form an arithmetic progression.

[0275] Using n=5 as an example, diagonal elements shown in the formula (5) may be:

[00005]$\begin{array}{cc}\left[{}_1,{}_2,{}_3,{}_4,{}_5\right]=\left[e^{j.Math.\frac{}{2}},{\left(\frac{7}{6}\right)}^{\frac{1}{2}}.Math.e^{j.Math.\frac{}{3}},{\left(\frac{8}{6}\right)}^{\frac{1}{2}}.Math.e^{j.Math.\frac{}{4}},{\left(\frac{9}{6}\right)}^{\frac{1}{2}}.Math.e^{j.Math.\frac{}{6}},{\left(\frac{10}{6}\right)}^{\frac{1}{2}}\right]& \left(8.Math.a\right)\end{array}$ [0276] correspondingly, the following formula is satisfied:

[00006]$\begin{array}{cc}\left[{}_5^{*},{}_4^{*},{}_3^{*},{}_2^{*},{}_1^{*}\right]=\left[{\left(\frac{10}{6}\right)}^{\frac{1}{2}},{\left(\frac{9}{6}\right)}^{\frac{1}{2}}.Math.e^{-j.Math.\frac{}{6}},{\left(\frac{8}{6}\right)}^{\frac{1}{2}}.Math.e^{-j.Math.\frac{}{4}},{\left(\frac{7}{6}\right)}^{\frac{1}{2}}.Math.e^{-j.Math.\frac{}{3}},e^{-j.Math.\frac{}{2}}\right]& \left(9.Math.a\right)\end{array}$

[0277] Alternatively, diagonal elements shown in the formula (5) may be:

[00007]$\begin{array}{cc}\left[{}_1,{}_2,{}_3,{}_4,{}_5\right]=\left[e^{j.Math.\frac{}{2}},e^{j.Math.\frac{}{3}},e^{j.Math.\frac{}{4}},e^{j.Math.\frac{}{6}},1\right]& \left(10.Math.a\right)\end{array}$ [0278] correspondingly, the following formula is satisfied:

[00008]$\begin{array}{cc}\left[{}_5^{*},{}_4^{*},{}_3^{*},{}_2^{*},{}_1^{*}\right]=\left[1,e^{-j.Math.\frac{}{6}},e^{-j.Math.\frac{}{4}},e^{-j.Math.\frac{}{3}},e^{-j.Math.\frac{}{2}}\right]& \left(11\right)\end{array}$

[0279] Using n=4 as an example, diagonal elements shown in the formula (2) may be:

[00009]$\begin{array}{cc}\left[u_1,u_2,u_3,u_4\right]=\left[e^{j.Math.\frac{7.Math.}{12}},{\left(\frac{5}{4}\right)}^{\frac{1}{2}}.Math.e^{j.Math.\frac{5.Math.}{12}},{\left(\frac{6}{4}\right)}^{\frac{1}{2}}.Math.e^{j.Math.\frac{3.Math.}{12}},{\left(\frac{7}{4}\right)}^{\frac{1}{2}}.Math.e^{j.Math.\frac{}{12}}\right]& \left(13\right)\end{array}$ [0280] correspondingly, the following formula is satisfied:

[00010]$\begin{array}{cc}\left[u_4^{*},u_3^{*},u_2^{*},u_1^{*}\right]=\left[{\left(\frac{7}{4}\right)}^{\frac{1}{2}}.Math.e^{-j.Math.\frac{}{12}},{\left(\frac{6}{4}\right)}^{\frac{1}{2}}.Math.e^{-j.Math.\frac{3.Math.}{12}},{\left(\frac{5}{4}\right)}^{\frac{1}{2}}.Math.e^{-j.Math.\frac{5.Math.}{12}},e^{-j.Math.\frac{7.Math.}{12}}\right]& \left(14\right)\end{array}$

[0281] Alternatively, diagonal elements shown in the formula (2) are:

[00011]$\begin{array}{cc}\left[u_1,u_2,u_3,u_4\right]=\left[e^{j.Math.\frac{7.Math.}{12}},e^{j.Math.\frac{5.Math.}{12}},e^{j.Math.\frac{3.Math.}{12}},e^{j.Math.\frac{}{12}}\right]& \left(15\right)\end{array}$ [0282] correspondingly, the following formula is satisfied:

[00012]$\begin{array}{cc}\left[u_4^{*},u_3^{*},u_2^{*},u_1^{*}\right]=\left[e^{-j.Math.\frac{}{12}},e^{-j.Math.\frac{3.Math.}{12}},e^{-j.Math.\frac{5.Math.}{12}},e^{-j.Math.\frac{7.Math.}{12}}\right]& \left(16\right)\end{array}$

[0283] In the foregoing two formulas, a phase of the diagonal elements u.sub.1,u.sub.2, . . . ,u.sub.n and a phase progression of the diagonal elements u.sub.n*, u.sub.n1*, . . . ,u.sub.1* respectively form arithmetic progressions whose common differences respectively are /6 and +/6.

[0284] Using n=4 as an example, diagonal elements shown in the formula (5) may be:

[00013]$\begin{array}{cc}\left[{}_1,{}_2,{}_3,{}_4\right]=\left[e^{j.Math.\frac{7.Math.}{32}},{\left(\frac{5}{4}\right)}^{\frac{1}{2}}.Math.e^{j.Math.\frac{5.Math.}{32}},{\left(\frac{6}{4}\right)}^{\frac{1}{2}}.Math.e^{j.Math.\frac{3.Math.}{32}},{\left(\frac{7}{4}\right)}^{\frac{1}{2}}.Math.e^{j.Math.\frac{}{32}}\right]& \left(13.Math.a\right)\end{array}$ [0285] correspondingly, the following formula is satisfied:

[00014]$\begin{array}{cc}\left[{}_4^{*},{}_3^{*},{}_2^{*},{}_1^{*}\right]=\left[{\left(\frac{7}{4}\right)}^{\frac{1}{2}}.Math.e^{-j.Math.\frac{}{32}},{\left(\frac{6}{4}\right)}^{\frac{1}{2}}.Math.e^{-j.Math.\frac{3.Math.}{32}},{\left(\frac{5}{4}\right)}^{\frac{1}{2}}.Math.e^{-j.Math.\frac{5.Math.}{32}},e^{-j.Math.\frac{7.Math.}{32}}\right]& \left(14.Math.a\right)\end{array}$

[0286] Alternatively, diagonal elements shown in the formula (5) are:

[00015]$\begin{array}{cc}\left[{}_1,{}_2,{}_3,{}_4\right]=\left[e^{j.Math.\frac{7.Math.}{32}},e^{j.Math.\frac{5.Math.}{32}},e^{j.Math.\frac{3.Math.}{32}},e^{j.Math.\frac{}{32}}\right]& \left(15.Math.a\right)\end{array}$ [0287] correspondingly, the following formula is satisfied:

[00016]$\begin{array}{cc}\left[{}_4^{*},{}_3^{*},{}_2^{*},{}_1^{*}\right]=\left[e^{-j.Math.\frac{}{32}},e^{-j.Math.\frac{3.Math.}{32}},e^{-j.Math.\frac{5.Math.}{32}},e^{-j.Math.\frac{7.Math.}{32}}\right]& \left(16.Math.a\right)\end{array}$

[0288] In the foregoing two formulas, a phase of the diagonal elements u.sub.1,u.sub.2, . . . ,u.sub.n and a phase progression of the diagonal elements u.sub.n*,u.sub.n1*, . . . ,u.sub.1* respectively form arithmetic progressions whose common differences respectively are /16 and /16.

[0289] Using n=4 as an example, diagonal elements of the diagonal matrices shown in the formula (7) may be respectively:

[00017]$\begin{array}{cc}.Math.\left[{}_{i,1},{}_{i,2},{}_{i,3},{}_{i,4}\right]=\left[e^{j.Math.\frac{7.Math.}{16}},{\left(\frac{5}{4}\right)}^{\frac{1}{2}}.Math.e^{j.Math.\frac{5.Math.}{16}},{\left(\frac{6}{4}\right)}^{\frac{1}{2}}.Math.e^{j.Math.\frac{3.Math.}{16}},{\left(\frac{7}{4}\right)}^{\frac{1}{2}}.Math.e^{j.Math.\frac{}{16}}\right]& \left(17\right)\\ \left[{}_{k,1},{}_{k,2},{}_{k,3},{}_{k,4}\right]=\left[e^{j.Math.\frac{7.Math.}{8}},{\left(\frac{5}{4}\right)}^{\frac{1}{2}}.Math.e^{j.Math.\frac{5.Math.}{8}},{\left(\frac{6}{4}\right)}^{\frac{1}{2}}.Math.e^{j.Math.\frac{3.Math.}{8}},{\left(\frac{7}{4}\right)}^{\frac{1}{2}}.Math.e^{j.Math.\frac{}{8}}\right]& \left(18\right)\end{array}$

[0290] Alternatively, diagonal elements of the diagonal matrices shown in the formula (7) may be respectively:

[00018]$\begin{array}{cc}\left[{}_{i,1},{}_{i,2},{}_{i,3},{}_{i,4}\right]=\left[e^{j.Math.\frac{7.Math.}{16}},e^{j.Math.\frac{5.Math.}{16}},e^{j.Math.\frac{3.Math.}{16}},e^{j.Math.\frac{}{16}}\right]& \left(19\right)\\ \left[{}_{k,1},{}_{k,2},{}_{k,3},{}_{k,4}\right]=\left[e^{j.Math.\frac{7.Math.}{8}},e^{j.Math.\frac{5.Math.}{8}},e^{j.Math.\frac{3.Math.}{8}},e^{j.Math.\frac{}{8}}\right]& \left(20\right)\end{array}$

[0291] In this embodiment, in the matrix D, the phases of the diagonal elements u.sub.1,u.sub.2, . . . ,u.sub.n form an arithmetic progression, and the phases of the diagonal elements u.sub.n*,u.sub.n1*, . . . ,u.sub.1* form an arithmetic progression, which may match with an array structure of an antenna port, for example, a common uniform linear array or cross polarization array, where in the former array, array elements or antennas are arranged at a same spacing, and in the latter array, co-polarized antennas or array elements are arranged at a same spacing. Therefore, the phases in the arithmetic progression can improve precoding performance by using a property of the foregoing array structure.

[0292] Optionally, as another embodiment, the column vector v of the matrix V may be a column vector of a matrix [H.sup.T H.sup.T].sup.T, where a matrix H is a Hadamard matrix.

[0293] Using n =4 as an example, the column vector v is a column vector of the matrix [H.sup.T H.sup.T].sup.T, where

[00019]$\begin{array}{cc}H=\left[\begin{array}{cccc}1& 1& 1& 1\\ 1& -1& 1& -1\\ 1& 1& -1& -1\\ 1& -1& -1& 1\end{array}\right]& \left(21\right)\end{array}$

[0294] In this case, the column vector v may be:

v=[1 1 1 1 1 1 1 1].sup.T(22)

[0295] or,

v=[1 1 1 1 1 1 1 1].sup.T(23)

[0296] or,

v=[1 1 1 1 1 1 1 1].sup.T(24)

[0297] In this embodiment, the column vector v is a column vector of the matrix [H.sup.T H.sup.T].sup.T, and satisfies the property of the formula (3), and column vectors of the [H.sup.T H.sup.T].sup.T are orthogonal to each other, so that the obtained column vectors are orthogonal to each other, thereby reducing inter-layer interference that is generated when the precoding matrix is used for MIMO transmission.

[0298] In the foregoing embodiment of the present invention, the selecting, by the user equipment, a precoding matrix from a codebook according to the reference signal may be specifically: obtaining, by the user equipment based on the reference signal, channel estimation; and selecting, based on a predefined criterion according to the channel estimation, the precoding matrix from the codebook, where the foregoing predefined criterion may be a channel capacity maximization criterion, a throughput maximization criterion, or a cosine distance minimization criterion.

[0299] In addition, in this embodiment of the present invention, the selecting, based on the reference signal, a precoding matrix from a codebook may include: [0300] selecting the precoding matrix from a codebook subset according to the reference signal, where the foregoing codebook subset is a predefined codebook subset, or a codebook subset reported to the base station, or a codebook subset reported to the base station, and returned and confirmed by the base station; the foregoing predefined codebook subset may be predefined in a protocol and is known by the user equipment and the base station in the system; and the codebook subset reported to the base station may be a codebook subset that is determined by the user equipment and is recently (recently) reported to the base station. In this embodiment, codebook subsets are set, for different application scenarios, in the codebook; and therefore, selecting a precoding matrix based on a codebook subset can effectively reduce feedback overheads and the implementation complexity.

[0301] Further, the codebook subset in the foregoing embodiment of the present invention may include a set of the precoding matrix W=DV; where the matrix D belongs to a subset of a universal set of the matrix D, or the matrix V belongs to a subset of a universal set of the matrix V.

[0302] It should be pointed out that diagonal elements of the foregoing diagonal matrix may have same amplitude. In this case, the structure of the foregoing precoding matrix allows that transmit antennas corresponding to rows of the precoding matrix have symmetric transmit powers based on actual considerations, and in this case, the foregoing codebook can still control a beam orientation by using a symmetric property of the powers of the transmit antennas, and meanwhile ensure orthogonality between transmission layers.

[0303] It should be understood that the precoding matrix in the codebook or the codebook subset may be pre-stored in the user equipment and the base station, or may be calculated by the user equipment and the base station according to the structure of the foregoing precoding matrix, or may be acquired from a network device, which is not limited in the present invention.

[0304] In step 303 shown in the foregoing FIG. 3, the precoding matrix indicator sent to the base station may include one or more indexes. Specifically, the codebook or the codebook subset usually is a set of one or more precoding matrices, where one precoding matrix indicator corresponds to one precoding matrix. Different precoding matrix indicators correspond to different precoding matrices in the codebook or the codebook subset, and in this embodiment, the sent precoding matrix indicator corresponds to the selected precoding matrix.

[0305] Specifically, the foregoing precoding matrix indicator PMI may include only one index, that is, one index directly indicates one precoding matrix, or the foregoing precoding matrix indicator may include two indexes, namely a first index PMI1 and a second index PMI2, where the first index PMI1 and the second index PMI2 jointly indicate the precoding matrix. In addition, the first index PMI1 corresponds to the matrix D. and the second index PMI2 corresponds to the matrix V. In an implementation manner of this implementation manner, for precoding matrices W indicated by two PMIs that have different first indexes PMI1 and a same second index PMI2, corresponding matrices D are different, and corresponding matrices V are the same. Optionally, for precoding matrices W indicated by two PMIs that have a same first index PMI1 and different second indexes PMI2, corresponding matrices D are the same, and corresponding matrices V are different.

[0306] Optionally, the foregoing first index PMI1 and second index PMI2 may have different time-domain granularities or frequency-domain granularities, that is, the PMI1 and the PMI2 separately represent channel characteristics of different periods or bandwidths, or are obtained based on different subframe periods or subbands.

[0307] Optionally, as another embodiment, the user equipment sends the first index PMI1 and the second index PMI2 to the base station by using different time periods, for example, the PMI1 may have a longer subframe period than the PMI2.

[0308] In addition, in step 303 in the foregoing embodiment of the present invention, the precoding matrix indicator PMI may be sent to the base station by using a physical uplink control channel (PUCCH) or a physical uplink shared channel (PUSCH).

[0309] The precoding matrix W in this embodiment may be a precoding matrix obtained by means of row or column transposition, for example, different antenna numbers correspondingly cause row transposition of the precoding matrix. In addition, the structure of the foregoing precoding matrix W not only may be used for antenna configuration in a horizontal direction in an AAS base station, but also may be used for antenna configuration in a perpendicular direction.

[0310] FIG. 4 is a second schematic flowchart of a specific embodiment. As shown in FIG. 4, the embodiment includes the following steps:

[0311] Step 401: A base station sends a reference signal to user equipment.

[0312] Specifically, the reference signal sent by the base station in this step may include a CSI RS, a DM RS, or a CRS. The user equipment UE may acquire the reference signal by receiving an eNB notification, or may obtain, based on a cell identity ID, a resource configuration of the reference signal and obtain the reference signal in a corresponding resource or subframe, where the eNB notification may be higher-layer signaling such as RRC signaling, or dynamic signaling such as DCI.

[0313] Step 402: The base station receives a precoding matrix indicator PMI sent by the user equipment.

[0314] Specifically, the precoding matrix indicator PMI corresponds to a precoding matrix that is selected from a codebook by the user equipment based on the reference signal.

[0315] Step 403: The base station determines a precoding matrix from a codebook according to the PMI, where the codebook includes one precoding matrix W that is a product of a matrix D and a matrix V and has the structure shown in the formula (25):

W=DV(25) [0316] where the matrix D is a diagonal matrix, and satisfies:

D=.Math.diag{u.sub.1,u.sub.2, . . . ,u.sub.n,u.sub.n1*, . . . ,u.sub.1*}(26) [0317] where is a complex factor, a complex number u.sub.1* is a conjugate complex number of a complex number u.sub.i, and i=1, . . . ,n; and the matrix V is a constant modulus matrix, for example, elements of the matrix V maybe 1 or j.

[0318] In another optional implementation manner of this embodiment, the matrix V includes a column vector 1 and/or at least one column vector v, where the column vector 1 is a column vector whose elements are all 1, and the column vector v is:

v=[v.sub.1 v.sub.2 L v.sub.n v.sub.n v.sub.n1 L v.sub.1].sup.T(27) [0319] where ( ).sup.T represents transposition of a matrix or vector, an element is v.sub.i=v.sub.i,, i=1, . . . , n, and v.sub.i=1, that is, a value of v.sub.i is +1 or 1. In an exemplary implementation manner, the matrix V is formed only by the column vector 1 and/or the at least one column vector v, that is, in the matrix V, except the included column vector 1, the other column vectors are all column vectors v.

[0320] In another optional implementation manner of this embodiment, the codebook includes multiple precoding matrices, and the multiple precoding matrices include a precoding matrix W.sub.i and a precoding matrix W.sub.j, where the W.sub.i and the W.sub.j satisfy the formula (4), and the matrix D(i,j) is a diagonal matrix and has the structure shown in the formula (5).

[0321] Optionally, the multiple precoding matrices include a precoding matrix W.sub.i and a precoding matrix W.sub.k, where the W.sub.i and the W.sub.k satisfy the formula (6), the matrix v includes a column vector 1 and/or at least one column vector v, the column vector 1 is a column vector whose elements are all 1, and the column vector v has the structure shown in the formula (27). The matrix D.sub.i and the matrix D.sub.k both are diagonal matrices, and have the structure shown in the formula (7).

[0322] It should be pointed out that diagonal elements of the foregoing diagonal matrix may have same amplitude. In this case, the structure of the foregoing precoding matrix allows that transmit antennas corresponding to rows of the precoding matrix have symmetric transmit powers based on actual considerations, and in this case, the foregoing codebook can still control a beam orientation by using a symmetric property of the powers of the transmit antennas, and meanwhile ensure orthogonality between transmission layers.

[0323] In the foregoing embodiment of the present invention, a base station receives a precoding matrix indicator PMI sent by user equipment, and determines a precoding matrix from a codebook according to the PMI. A precoding matrix W included in the codebook is a product of a matrix D and a matrix V. The D is a diagonal matrix and satisfies D=.Math.diag {u.sub.1,u.sub.2, . . . ,u.sub.n,u.sub.n*,u.sub.n1*, . . . ,u.sub.1*}, where u.sub.1,u.sub.2, . . . ,u.sub.nu.sub.n*,u.sub.n1,* . . . ,u.sub.1* forms a conjugate md symmetric sequence, which avoids constant modulus restrictions or a limit that antennas perform transmission by using equal powers, and can effectively control a beam shape and a beam orientation.

[0324] Further, the matrix V includes a column vector 1 and/or at least one column vector v=[v.sub.1 v.sub.2 L v.sub.n v.sub.n v.sub.n1 L v.sub.1].sup.T, so that column vectors of the precoding matrix are orthogonal to each other, which can effectively reduce inter-layer interference, thereby greatly improving performance of MIMO, especially MU-MIMO. Therefore, the foregoing method for determining a precoding matrix can fully use a degree of freedom of controlling a beam shape and a beam orientation of an antenna system, and meanwhile reduce inter-layer interference of MIMO transmission as much as possible, thereby improving precision of CSI feedback, and a system throughput.

[0325] Specifically, the base station may obtain the precoding matrix from the codebook according to the received PMI, where the codebook is the same as the codebook used by the user equipment. Further, the base station may further precode to-be-sent data according to the obtained precoding matrix.

[0326] Using n=5 as an example, the foregoing diagonal elements u.sub.1,u.sub.2, . . . ,u.sub.n, diagonal elements u.sub.n*,u.sub.n1*, . . . ,u.sub.1*, and column vector v in this embodiment may be respectively shown in the formula (8) to the formula (12).

[0327] Using n=5 as an example, the diagonal elements .sub.1,.sub.2,L,.sub.n and the diagonal elements .sub.n*,.sub.n1*,L,.sub.1* in the formula (5) used in this embodiment may be respectively shown in the formula (8a) to the formula (11a).

[0328] Optionally, as another embodiment, in the matrix D, phases of the diagonal elements u.sub.1,u.sub.2, . . . ,u.sub.n form an arithmetic progression, and phases of the diagonal elements u.sub.n*,u.sub.n1*, . . . ,u.sub.1* form an arithmetic progression. Using n=4 as an example, the diagonal elements u.sub.1,u.sub.2, . . . ,u.sub.n and the diagonal elements u.sub.n*,u.sub.n1*, . . . ,u.sub.1* may be respectively shown in the formula (13) to the formula (16).

[0329] Using n=4 as an example, the diagonal elements .sub.1,.sub.2,L,.sub.n and the diagonal elements .sub.n*,.sub.n1*,L,.sub.1* in the formula (5) used in this embodiment may be respectively shown in the formula (13a) to the formula (16a).

[0330] Using n=4 as an example, the diagonal elements of the diagonal matrices shown in the formula (7) used in this embodiment may be respectively shown in the formula (17) to the formula (20).

[0331] In this embodiment, in the matrix D, the phases of the diagonal elements u.sub.1,u.sub.2, . . . ,u.sub.n form an arithmetic progression, and the phases of the diagonal elements u.sub.n*,u.sub.n1*, . . . ,u.sub.1* form an arithmetic progression, which may match with an array structure of an antenna port, for example, a common uniform linear array or cross polarization array, where in the former array, array elements or antennas are arranged at a same spacing, and in the latter array, co-polarized antennas or array elements are arranged at a same spacing. Therefore, the phases in the arithmetic progression can improve precoding performance by using a property of the foregoing array structure.

[0332] Optionally, as another embodiment, the column vector v of the matrix V may be a column vector of a matrix [H.sup.T H.sup.T].sup.T, where a matrix H is a Hadamard matrix. Using n=4 as an example, the Hadamard matrix and the column vector v may be respectively shown in the formula (21) to the formula (24).

[0333] In this embodiment, the column vector v is a column vector of the matrix [H.sup.T H.sup.T].sup.T, and satisfies the property of the formula (27), and column vectors of the [H.sup.T H.sup.T].sup.T are orthogonal to each other, so that the obtained column vectors are orthogonal to each other, thereby reducing inter-layer interference that is generated when the precoding matrix is used for MIMO transmission.

[0334] In this embodiment of the present invention, the determining a precoding matrix from a codebook according to the PMI includes: [0335] determining the precoding matrix from a codebook subset according to the PMI, where the codebook subset is a predefined codebook subset, or a codebook subset reported to the base station, or a codebook subset reported to the base station, and returned and confirmed by the base station. The codebook subset may be a set of the precoding matrix W=DV, where the matrix D or the matrix V is a subset of a candidate matrix thereof.

[0336] It should be understood that the precoding matrix in the codebook or the codebook subset may be pre-stored in the user equipment and the base station, or may be calculated by the user equipment and the base station according to the structure of the foregoing precoding matrix, or may be acquired from a network device, which is not limited in the present invention.

[0337] The foregoing precoding matrix indicator PMI may include only one index, that is, one index directly indicates one precoding matrix, or the foregoing precoding matrix indicator may include two indexes, namely a first index PMI1 and a second index PMI2, where the first index PMI1 and the second index PMI2 jointly indicate the precoding matrix. In addition, the first index PMI1 corresponds to the matrix D. and the second index PMI2 corresponds to the matrix V. In an implementation manner of this implementation manner, for precoding matrices W indicated by two PMIs that have different first indexes PMI1 and a same second index PMI2, corresponding matrices D are different, and corresponding matrices V are the same. Optionally, for precoding matrices W indicated by two PMIs that have a same first index PMI1 and different second indexes PMI2, corresponding matrices D are the same, and corresponding matrices V are different.

[0338] Optionally, the foregoing first index PMI1 and second index PMI2 may have different time-domain granularities or frequency-domain granularities, that is, the PMI1 and the PMI2 separately represent channel characteristics of different periods or bandwidths, or are obtained based on different subframe periods or subbands.

[0339] Optionally, as another embodiment, the base station may receive, by using different time periods, the first index PMI1 and the second index PMI2 that are sent by the user equipment, for example, the PMI1 may have a longer subframe period than the PMI2. In addition, the base station may receive, by using a PUCCH or a PUSCH, the precoding matrix indicator PMI sent by the user equipment.

[0340] FIG. 5 is a third schematic flowchart of a specific embodiment of the present invention. This embodiment provides a method for determining a precoding matrix indicator that is executed on a user equipment side when a precoding matrix satisfies a first condition. As shown in FIG. 5, the method includes the following steps:

[0341] Step 501: User equipment receives a reference signal sent by a base station.

[0342] Similar to step 301 in the embodiment shown in FIG. 3, in this step, the reference signal sent by the base station may include a CSI RS, a DM RS, or a CRS. The user equipment UE may acquire the reference signal by receiving an eNB notification, or may obtain, based on a cell identity ID, a resource configuration of the reference signal and obtain the reference signal in a corresponding resource or subframe, where the eNB notification may be higher-layer signaling such as RRC signaling, or dynamic signaling such as DCI.

[0343] Step 502: The user equipment selects, based on the reference signal, a precoding matrix from a codebook, where a precoding matrix W included in the codebook is a product of a matrix W.sub.1 and a matrix W.sub.2, where

W=W.sub.1W.sub.2(28)

[0344] The matrix W.sub.1 is a block diagonal matrix,

W.sub.1=diag{X.sub.1, . . . ,X.sub.N.sub.B},N.sub.B1(29) [0345] where at least one block matrix X is a product of a matrix D and a matrix V, and X {X.sub.1,X.sub.2, . . . ,X.sub.N.sub.B}, that is, the block matrix X has the structure shown in the formula (30):

X=DV(30 )

[0346] The matrix D is a diagonal matrix,

D=.Math.diag{u.sub.1,u.sub.2, . . . ,u.sub.nU*,u.sub.n1*, . . . ,u.sub.1*}(31) [0347] where is a complex factor, a complex number u.sub.i* is a conjugate complex number of a complex number u.sub.i, and i=1, . . . ,n; and the matrix V includes a column vector 1 and/or at least one column vector v, where the column vector 1 is a column vector whose elements are all 1, and the column vector v is:

V=[v.sub.1v.sub.2Lv.sub.nv.sub.nv.sub.n1Lv.sub.1].sup.T(32) [0348] where ( ).sup.T represents transposition of a matrix or vector, an element is v.sub.i=v.sub.i, i=1, . . . ,n, and v.sub.i=1, that is, a value of v.sub.i is +1 or 1. The matrix W.sub.2 is used to select one or more column vectors of the matrix W.sub.1, or is used to perform weighted combination on one or more column vectors of the W.sub.1 to obtain the precoding matrix W.

[0349] In another optional implementation manner of this embodiment, the block matrix X may be multiple different matrices, including a matrix P.sub.i and a matrix P.sub.j, where the P.sub.i and the P.sub.j satisfy the formula (33):

P.sub.i=D(i,j)P.sub.j(33) [0350] where the matrix D(i,j) is a diagonal matrix, and has the structure shown in the formula (5); and optionally, phases of diagonal elements .sub.1,.sub.2, . . . ,.sub.n of the matrix D(i,j) form an arithmetic progression.

[0351] Optionally, the block matrix X may be multiple different matrices, including a matrix P.sub.i and a matrix P.sub.k, where the P.sub.i and the P.sub.k satisfy the formula (34):

D.sub.i.sup.1P.sub.i=D.sub.k.sup.1P.sub.k=V(34) [0352] where the matrix V includes a column vector 1 and/or at least one column vector v, the column vector 1 is a column vector whose elements are all 1, and the column vector v has the structure shown in the formula (32); and the matrix D.sub.i and the matrix D.sub.k both are diagonal matrices, and have the structure shown in the formula (35):

D.sub.m=.Math.diag{u.sub.m,1,u.sub.m,2, . . . ,u.sub.m,n,u.sub.m,n*,u.sub.m,n1*, . . . ,u.sub.m,1*},m=i,k(35) [0353] where .sub.m is a complex factor, and a real part or an imaginary part of the complex factor may be 0; a complex number u.sub.m,l* is a conjugate complex number of a complex number u.sub.m,l, m=i,k, l=1, . . . , n, and n is determined by a quantity of antenna ports; and optionally, phases of diagonal elements u.sub.m,2,u.sub.m,2, . . . ,u.sub.m,n of the matrix D.sub.m form an arithmetic progression.

[0354] The user equipment may select different precoding matrices at different time points according to a preset rule or randomly, that is, the user equipment may determine a first precoding matrix indicator PMI at a time point, where the PMI corresponds to the precoding matrix P.sub.i in the codebook, and send the first PMI to the base station; and determine a second precoding matrix indicator PMI at another time point, where the PMI corresponds to the precoding matrix D.sub.k in the codebook, and send the second PMI to the base station.

[0355] Corresponding to the case in which the foregoing user equipment sends the first PMI or the second PMI at different time points, on a base station side, the base station may also receive, at a time point, the first precoding matrix indicator PMI sent by the user equipment, and select the corresponding precoding matrix P.sub.i from the codebook according to the first PMI; and receive, at another time point, the second precoding matrix indicator PMI sent by the user equipment, and select the corresponding precoding matrix D.sub.k from the codebook according to the second PMI.

[0356] It should be pointed out that diagonal elements of the foregoing diagonal matrix may have same amplitude. In this case, the structure of the foregoing precoding matrix allows that transmit antennas corresponding to rows of the precoding matrix have symmetric transmit powers based on actual considerations, and in this case, the foregoing codebook can still control a beam orientation by using a symmetric property of the powers of the transmit antennas, and meanwhile ensure orthogonality between transmission layers.

[0357] Step 503: The user equipment sends a precoding matrix indicator PMI to the base station, where the PMI corresponds to the selected precoding matrix.

[0358] In the foregoing embodiment of the present invention, user equipment selects, based on a reference signal, a precoding matrix from a codebook, and sends a precoding matrix indicator PMI. A precoding matrix W included in the codebook is a product of a matrix W.sub.1 and a matrix W.sub.2, where the matrix W.sub.1 is a block diagonal matrix, W.sub.1=diag {X.sub.1, . . . ,X.sub.N.sub.B}, and N.sub.B1, where at least one block matrix X is a product of a matrix D and a matrix V, X=DV, and X {X.sub.1,X.sub.2, . . . ,X.sub.N.sub.B}. The D is a diagonal matrix and satisfies D=.Math.diag {u.sub.1,u.sub.2, . . . ,u.sub.n,u.sub.n*,u.sub.n1*, . . . ,u.sub.1*}, where u.sub.1,u.sub.2, . . . ,u.sub.n,u.sub.n*,u.sub.n1*, . . . ,u.sub.1forms a conjugate md symmetric sequence, which avoids constant modulus restrictions or a limit that antennas perform transmission by using equal powers, and can effectively control a beam shape and a beam orientation. The matrix V includes a column vector 1 and/or at least one column vector v=[v.sub.1 v.sub.2 L v.sub.n v.sub.n v.sub.n1 L v.sub.1].sup.T, so that column vectors of the precoding matrix are orthogonal to each other, which can effectively reduce inter-layer interference, thereby greatly improving performance of MIMO, especially MU-MIMO. Therefore, the foregoing method for determining a precoding matrix can fully use a degree of freedom of controlling a beam shape and a beam orientation of an antenna system, and meanwhile reduce inter-layer interference of MIMO transmission as much as possible, thereby improving precision of CSI feedback, and a system throughput.

[0359] Specifically, using n=5 as an example, the foregoing diagonal elements u.sub.1,u.sub.2, . . . ,u.sub.n, diagonal elements u.sub.n*,u.sub.n1*, . . . ,u.sub.1*, and column vector v may be respectively shown in the formula (8) to the formula (12).

[0360] Using n=5 as an example, the diagonal elements .sub.1,.sub.2, L,.sub.n and the diagonal elements .sub.n*,.sub.n1*,L,.sub.1* in the formula (5) used in this embodiment may be respectively shown in the formula (8a) to the formula (11a).

[0361] Optionally, as another embodiment, in the matrix D, phases of the diagonal elements u.sub.1,u.sub.2, . . . ,u.sub.n form an arithmetic progression, and phases of the diagonal elements u.sub.n*,u.sub.n1*, . . . ,u.sub.1* form an arithmetic progression. Using n=4 as an example, the diagonal elements u.sub.1,u.sub.2, . . . ,u.sub.n and the diagonal elements u.sub.n*,u.sub.n1*, . . . ,u.sub.1* may be respectively shown in the formula (13) to the formula (16).

[0362] Using n=4 as an example, the diagonal elements .sub.1,.sub.2,L,.sub.n and the diagonal elements .sub.n*,.sub.n1*,L,.sub.1* in the formula (5) used in this embodiment may be respectively shown in the formula (13a) to the formula (16a).

[0363] Using n=4 as an example, the diagonal elements of the diagonal matrices in the formula (35) used in this embodiment may be respectively shown in the formula (17) to the formula (20).

[0364] In this embodiment, in the matrix D, the phases of the diagonal elements u.sub.1,u.sub.2, . . . ,u.sub.n form an arithmetic progression, and the phases of the diagonal elements u.sub.n*,u.sub.n1*, . . . ,u.sub.1* form an arithmetic progression, which may match with an array structure of an antenna port, for example, a common uniform linear array or cross polarization array, where in the former array, array elements or antennas are arranged at a same spacing, and in the latter array, co-polarized antennas or array elements are arranged at a same spacing. Therefore, the phases in the arithmetic progression can improve precoding performance by using a property of the foregoing array structure.

[0365] Optionally, as another embodiment, the column vector v of the matrix V may be a column vector of a matrix [H.sup.T H.sup.T].sup.T, where a matrix H is a Hadamard matrix. Using n=4 as an example, the Hadamard matrix and the column vector v may be respectively shown in the formula (21) to the formula (24).

[0366] In this embodiment, the column vector v is a column vector of the matrix [H.sup.T H.sup.T].sup.T, and satisfies the property of the formula (32), and column vectors of the [H.sup.T H.sup.T].sup.T are orthogonal to each other, so that the obtained column vectors are orthogonal to each other, thereby reducing inter-layer interference that is generated when the precoding matrix is used for MIMO transmission.

[0367] In the foregoing embodiment of the present invention, the matrix W.sub.2 is used to select a column vector of the matrix W.sub.1, or is used to perform weighted combination on a column vector of the W.sub.1 to form the matrix W.

[0368] Using W.sub.1=diag {X.sub.1,X.sub.2} as an example, where each of a block matrix X.sub.1 and a block matrix X.sub.2 has four columns, the W.sub.2 may be the following matrix:

[00020]$\begin{array}{cc}{W}_2\left\{\frac{1}{\sqrt{2}}\left[\begin{array}{c}{e}_i\\ e_i\end{array}\right],\frac{1}{\sqrt{2}}\left[\begin{array}{c}{e}_i\\ {\mathrm{je}}_i\end{array}\right],\frac{1}{\sqrt{2}}\left[\begin{array}{c}{e}_i\\ -e_i\end{array}\right],\frac{1}{\sqrt{2}}\left[\begin{array}{c}{e}_i\\ -{\mathrm{je}}_i\end{array}\right]|i=1,2,3,4\right\}& \left(36\right)\end{array}$ [0369] where e.sub.i, i=1,2,3,4 represents a 41 selection vector, and except that the i.sup.th element is 1, other elements are all 0.

[0370] Using W.sub.1=diag {X.sub.1,X.sub.2} as an example, where each of a block matrix X.sub.1 and a block matrix X.sub.2 has eight columns, the W.sub.2 may be the following matrix:

[00021]$\begin{array}{cc}{W}_2\left\{\frac{1}{\sqrt{2}}\left[\begin{array}{c}Y\\ Y\end{array}\right],\frac{1}{\sqrt{2}}\left[\begin{array}{c}Y\\ \mathrm{jY}\end{array}\right],\frac{1}{\sqrt{2}}\left[\begin{array}{c}Y\\ -Y\end{array}\right],\frac{1}{\sqrt{2}}\left[\begin{array}{c}Y\\ -\mathrm{jY}\end{array}\right]\right\}& \left(37\right)\end{array}$
Y{e.sub.1,e.sub.2,e.sub.3,e.sub.4,e.sub.5,e.sub.6,e.sub.7,e.sub.8}(38) [0371] or,

[00022]$\begin{array}{cc}{W}_2\left\{\frac{1}{2}\left[\begin{array}{cc}{Y}_1& Y_2\\ Y_1& -Y_2\end{array}\right],\frac{1}{2}\left[\begin{array}{cc}{Y}_1& Y_2\\ {\mathrm{jY}}_1& -{\mathrm{jY}}_2\end{array}\right]\right\}& \left(39\right)\end{array}$
(Y.sub.1Y.sub.2){(e.sub.1e.sub.1),(e.sub.2,e.sub.2),(e.sub.3,e.sub.3),(e.sub.4,e.sub.4,),(e.sub.1,e.sub.2),(e.sub.2,e.sub.3),(e.sub.1,e.sub.4),(e.sub.2,e.sub.4)}(40) [0372] where e.sub.n, n=1,2,L,8 represents an 81 selection vector, and except that the n.sup.th element is 1, other elements are all 0.

[0373] Using W.sub.1=diag {X.sub.1,X.sub.2,X.sub.3,X.sub.4} as an example, where each block matrix of X.sub.1,X.sub.2,X.sub.3,X.sub.4 has four columns, the W.sub.2 may be the following matrix:

[00023]$\begin{array}{cc}{W}_2\left\{\frac{1}{2}\left[\begin{array}{c}{e}_i\\ e^{j.Math..Math.}.Math.e_i\\ e^{j.Math..Math.}.Math.e_k\\ e^{j\left(+\right)}.Math.e_k\end{array}\right],\frac{1}{2}\left[\begin{array}{c}{e}_i\\ -e^{j.Math..Math.}.Math.e_i\\ e^{j.Math..Math.}.Math.e_k\\ -e^{j\left(+\right)}.Math.e_k\end{array}\right],\frac{1}{2}\left[\begin{array}{c}{e}_i\\ {\mathrm{je}}^{j.Math..Math.}.Math.e_i\\ e^{j.Math..Math.}.Math.e_k\\ {\mathrm{je}}^{j\left(+\right)}.Math.e_k\end{array}\right],\frac{1}{2}\left[\begin{array}{c}{e}_i\\ -{\mathrm{je}}^{j.Math..Math.}.Math.e_i\\ e^{j.Math..Math.}.Math.e_k\\ -{\mathrm{je}}^{j\left(+\right)}.Math.e_k\end{array}\right]|\begin{array}{c}i=1,2,3,4;\\ k=1,2,3,4\end{array}\right\}& \left(41\right)\end{array}$ [0374] where e.sub.i, i=1,2,3,4 represents a 41 selection vector, and except that the i.sup.th element is 1, other elements are all 0, where and are phases, for example,

[00024]$=\frac{m.Math..Math.}{32},m=0,1,2,3,\mathrm{.Math.}.Math..Math.\mathrm{and}.Math..Math.=\frac{n.Math..Math.}{32},n=0,1,2,3,\mathrm{.Math.}.Math..$

[0375] Further, the block matrix is X.sub.1=X.sub.2,X.sub.3=X.sub.4 or X.sub.1=X.sub.2=X.sub.3=X.sub.4.

[0376] Using W.sub.1=diag {X.sub.1,X.sub.2,X.sub.3,X.sub.4} as an example where each block matrix of X.sub.1,X.sub.2,X.sub.3,X.sub.4 has eight columns, the W.sub.2 may be the following matrix:

[00025]$\begin{array}{cc}{W}_2\left\{\frac{1}{2}\left[\begin{array}{c}{Y}_1\\ e^{j.Math..Math.}.Math.Y_1\\ e^{j.Math..Math.}.Math.Y_2\\ e^{j\left(+\right)}.Math.Y_2\end{array}\right],\frac{1}{2}\left[\begin{array}{c}{Y}_1\\ -e^{j.Math..Math.}.Math.Y_1\\ e^{j.Math..Math.}.Math.Y_2\\ -e^{j\left(+\right)}.Math.Y_2\end{array}\right],\frac{1}{2}\left[\begin{array}{c}{Y}_1\\ {\mathrm{je}}^{j.Math..Math.}.Math.Y_1\\ e^{j.Math..Math.}.Math.Y_2\\ {\mathrm{je}}^{j\left(+\right)}.Math.Y_2\end{array}\right],\frac{1}{2}\left[\begin{array}{c}{Y}_1\\ -{\mathrm{je}}^{j.Math..Math.}.Math.Y_1\\ e^{j.Math..Math.}.Math.Y_2\\ -{\mathrm{je}}^{j\left(+\right)}.Math.Y_2\end{array}\right]\right\}& \left(42\right)\end{array}$
Y.sub.1{e.sub.1,e.sub.2,e.sub.3,e.sub.4,e.sub.5,e.sub.6,e.sub.7,e.sub.8}, Y.sub.2{e.sub.1,e.sub.2,e.sub.3,e.sub.4,e.sub.5,e.sub.6,e.sub.7,e.sub.8}(43)

[0377] or,

[00026]$.Math.\left(44\right)$$W_2\left\{\begin{array}{c}\frac{1}{2.Math.\sqrt{2}}\left[\begin{array}{cc}{Y}_{1,1}& Y_{1,2}\\ e^{j.Math..Math.}.Math.Y_{1,1}& e^{j.Math..Math.}.Math.Y_{1,2}\\ e^{j.Math..Math.}.Math.Y_{2,1}& e^{j.Math..Math.}.Math.Y_{2,2}\\ e^{j\left(+\right)}.Math.Y_{2,1}& e^{j\left(+\right)}.Math.Y_{2,2}\end{array}\right],\frac{1}{2.Math.\sqrt{2}}\left[\begin{array}{cc}{Y}_{1,1}& Y_{1,2}\\ e^{j.Math..Math.}.Math.Y_{1,1}& -e^{j.Math..Math.}.Math.Y_{1,2}\\ e^{j.Math..Math.}.Math.Y_{2,1}& e^{j.Math..Math.}.Math.Y_{2,2}\\ e^{j\left(+\right)}.Math.Y_{2,1}& -e^{j\left(+\right)}.Math.Y_{2,2}\end{array}\right],\\ \frac{1}{2.Math.\sqrt{2}}\left[\begin{array}{cc}{Y}_{1,1}& Y_{1,2}\\ -e^{j.Math..Math.}.Math.Y_{1,1}& e^{j.Math..Math.}.Math.Y_{1,2}\\ e^{j.Math..Math.}.Math.Y_{2,1}& e^{j.Math..Math.}.Math.Y_{2,2}\\ e^{j\left(+\right)}.Math.Y_{2,1}& -e^{j\left(+\right)}.Math.Y_{2,2}\end{array}\right],\frac{1}{2.Math.\sqrt{2}}\left[\begin{array}{cc}{Y}_{1,1}& Y_{1,2}\\ -e^{j.Math..Math.}.Math.Y_{1,1}& e^{j.Math..Math.}.Math.Y_{1,2}\\ e^{j.Math..Math.}.Math.Y_{2,1}& e^{j.Math..Math.}.Math.Y_{2,2}\\ -e^{j\left(+\right)}.Math.Y_{2,1}& e^{j\left(+\right)}.Math.Y_{2,2}\end{array}\right]\end{array}\right\}$$.Math.\left(45\right)$$.Math.\left(Y_{m,1},Y_{m,2}\right)\left\{\begin{array}{c}\left(e_1,e_1\right),\left(e_2,e_2\right),\left(e_3,e_3\right),\left(e_4,e_4\right),\\ \left(e_1,e_2\right),\left(e_2,e_3\right),\left(e_1,e_4\right),\left(e_2,e_4\right)\end{array}\right\},m=1,2$ [0378] where e.sub.n,n=1,2,L,8 represents an 81 selection vector, and except that the n.sup.th element is 1, other elements are all 0, where and are phases, for example,

[00027]$=\frac{m.Math..Math.}{32},m=0,1,2,3,\mathrm{.Math.}.Math..Math.\mathrm{and}.Math..Math.=\frac{n.Math..Math.}{32},n=0,1,2,3,\mathrm{.Math.}.Math..$

[0379] Further, the block matrix is X.sub.1=X.sub.2,X.sub.3=X.sub.4 or X.sub.1=X.sub.2=X.sub.3=X.sub.4.

[0380] In the foregoing embodiment of the present invention, the selecting, by the user equipment based on the reference signal, a precoding matrix from a codebook may be specifically: obtaining, by the user equipment based on the reference signal, channel estimation; and selecting, based on a predefined criterion according to the channel estimation, the precoding matrix from the codebook, where the foregoing predefined criterion may be a channel capacity maximization criterion, a throughput maximization criterion, or a cosine distance minimization criterion.

[0381] In addition, in the foregoing embodiment of the present invention, the selecting, based on the reference signal, a precoding matrix from a codebook may include: selecting the precoding matrix from a codebook subset according to the reference signal, where the foregoing codebook subset is a predefined codebook subset, or a codebook subset reported to the base station, or a codebook subset reported to the base station, and returned and confirmed by the base station; the foregoing predefined codebook subset may be predefined in a protocol and is known by the user equipment and the base station in the system; and the codebook subset reported to the base station may be a codebook subset that is recently determined by the user equipment and reported to the base station. In this embodiment, codebook subsets are set, for different application scenarios, in the codebook; and therefore, selecting a precoding matrix based on a codebook subset can effectively reduce feedback overheads and the implementation complexity.

[0382] Further, the codebook subset in the foregoing embodiment of the present invention may include: [0383] a set or a precoding matrix W=V.sub.1W.sub.2, where W.sub.1=diag {X.sub.1, . . . ,X.sub.N.sub.B}, at least one block matrix X is a product of a matrix D and a matrix V, X {X.sub.1,X.sub.2, . . . ,X.sub.N.sub.B}, and X=DV; and the matrix D, the matrix V, or the matrix W.sub.2 is a subset of a candidate matrix thereof.

[0384] It should be pointed out that diagonal elements of the foregoing diagonal matrix may have same amplitude. In this case, the structure of the foregoing precoding matrix allows that transmit antennas corresponding to rows of the precoding matrix have symmetric transmit powers based on actual considerations, and in this case, the foregoing codebook can still control a beam orientation by using a symmetric property of the powers of the transmit antennas, and meanwhile ensure orthogonality between transmission layers.

[0385] It should be understood that the precoding matrix in the codebook or the codebook subset may be pre-stored in the user equipment and the base station, or may be calculated by the user equipment and the base station according to the structure of the foregoing precoding matrix, or may be obtained from a network device, which is not limited in the present invention.

[0386] Further, in the foregoing precoding matrix, a block matrix X.sub.k and a block matrix X.sub.l,kl may be different, or may be the same. In a case in which there are multiple X.sub.k s that are the same as the X.sub.l,kl, for example, the X.sub.k and the X.sub.l,kl that are the same exist in pairs, feedback overheads can be further reduced. Multiple block matrices X.sub.i in the foregoing matrix W.sub.1 may respectively correspond to antenna port groups having different polarizations or at different positions, so that the foregoing precoding matrix matches with multiple antenna deployments or configurations.

[0387] In step 503 shown in the foregoing FIG. 5, the precoding matrix indicator sent to the base station may include one or more indexes. Specifically, the codebook or the codebook subset usually is a set of one or more precoding matrices, where one precoding matrix indicator corresponds to one precoding matrix. Different precoding matrix indicators correspond to different precoding matrices in the codebook or the codebook subset, and in this embodiment, the sent precoding matrix indicator corresponds to the selected precoding matrix.

[0388] Specifically, the foregoing precoding matrix indicator PMI may include only one index, that is, one index directly indicates one precoding matrix, or the foregoing precoding matrix indicator may include two indexes, namely a first index PMI1 and a second index PMI2, where the first index PMI1 and the second index PMI2 jointly indicate the precoding matrix. In addition, the first index PMI1 is used for indicating the matrix W.sub.1, and the second index PMI2 is used for indicating the matrix W.sub.2. The foregoing first index PMI1 and second index PMI2 may have different time-domain granularities or frequency-domain granularities, that is, the PMI1 and the PMI2 separately represent channel characteristics of different periods or bandwidths, or are obtained based on different subframe periods or subbands.

[0389] Optionally, the foregoing PMI may also include three indexes, where the three indexes are respectively used for indicating the matrix D, the matrix V, and the matrix W.sub.2.

[0390] Optionally, as another embodiment, the user equipment sends the first index PMI1 and the second index PMI2 to the base station by using different time periods, for example, the PMI1 may have a longer subframe period than the PMI2.

[0391] In addition, in step 503 in the foregoing embodiment of the present invention, the precoding matrix indicator information PMI may be sent to the base station by using a PUCCH or a PUSCH.

[0392] The precoding matrix W in this embodiment may be a precoding matrix obtained by means of row or column transposition, for example, different antenna numbers correspondingly cause row transposition of the precoding matrix. In addition, the structure of the foregoing precoding matrix W not only may be used for antenna configuration in a horizontal direction in an AAS base station, but also may be used for antenna configuration in a perpendicular direction.

[0393] In this embodiment of the present invention, user equipment selects, based on a reference signal, a precoding matrix from a codebook, and sends a precoding matrix indicator PMI, where the PMI corresponds to the selected precoding matrix. A precoding matrix W included in the codebook is a product of a matrix W.sub.1 and a matrix W.sub.2, where the W.sub.1 is a block diagonal matrix, and W.sub.1=diag {X.sub.1, . . . ,X.sub.N.sub.B}, where at least one block matrix X is a product of a matrix D and a matrix v, X=DV, and X {X.sub.1,X.sub.2, . . . ,X.sub.N.sub.B}. The D is a diagonal matrix and satisfies D=.Math.diag {u.sub.1,u.sub.2, . . . ,u.sub.n,u.sub.n*,u.sub.n1*, . . . ,u.sub.1*}, where u.sub.1,u.sub.2, . . . ,u.sub.n,u.sub.n*,u.sub.n1*, . . . ,u.sub.1* forms a conjugate and symmetric sequence, which avoids constant modulus restrictions or a limit that antennas perform transmission by using equal powers, and can effectively control a beam shape and a beam orientation. The matrix V includes a column vector 1 and at least one column vector v=[v.sub.1 v.sub.2 L v.sub.n v.sub.n v.sub.n1 L v.sub.1].sup.T, so that column vectors of the precoding matrix are orthogonal to each other, which can effectively reduce inter-layer interference, thereby greatly improving performance of MIMO, especially MU-MIMO. Multiple block matrices X.sub.i may respectively correspond to antenna port groups having different polarizations or at different positions, so that the foregoing precoding matrix matches with multiple antenna deployments or configurations. Therefore, the foregoing method for determining a precoding matrix can fully use a degree of freedom of controlling a beam shape and a beam orientation in a horizontal and/or perpendicular direction of an active antenna system, and meanwhile reduce inter-layer interference during transmission as much as possible, thereby improving precision of CSI feedback, and a system throughput.

[0394] Corresponding to the embodiment shown in FIG. 5, an embodiment of the present invention further provides a method for determining a precoding matrix indicator that is executed on a base station side when a precoding matrix satisfies a first condition. FIG. 6 is a fourth schematic flowchart of a specific embodiment of the present invention. As shown in FIG. 6, the embodiment includes the following steps:

[0395] Step 601: A base station sends a reference signal to user equipment.

[0396] The foregoing sent reference signal may include multiple forms, and for details, reference may be made to step 501 in the embodiment shown in FIG. 5.

[0397] Step 602: The base station receives a precoding matrix indicator PMI sent by the user equipment.

[0398] Specifically, the precoding matrix indicator PMI corresponds to a precoding matrix that is selected from a codebook by the user equipment based on the reference signal.

[0399] Step 603: The base station determines a precoding matrix from a codebook according to the PMI, where a precoding matrix W included in the codebook is a product of a matrix W.sub.1 and a matrix W.sub.2, where

W=W.sub.1W.sub.2(46)

[0400] The matrix W.sub.1 is a block diagonal matrix,

W.sub.1=diag{X.sub.1, . . . ,X.sub.N.sub.B},N.sub.B1(47) [0401] where at least one block matrix X is a product of a matrix D and a matrix V, and X {X.sub.1,X.sub.2, . . . ,X.sub.N.sub.B}, that is the block matrix X has the structure shown in the formula (48):

X=DV(48)

[0402] The matrix D is a diagonal matrix,

D=.Math.diag{u.sub.1,u.sub.2, . . .,u.sub.n,u.sub.n*,u.sub.n1*, . . . ,u.sub.1*}(49) [0403] where is a complex factor, a complex number u.sub.i* is a conjugate complex number of a complex number u.sub.i, and i=1, . . . ,n; and the matrix V includes a column vector 1 and/or at least one column vector v, where the column vector 1 is a column vector whose elements are all 1, and the column vector v is:

v=[v.sub.1v.sub.2Lv.sub.nv.sub.nv.sub.n1Lv.sub.1].sup.T(50) [0404] where ( ).sup.T represents transposition of a matrix or vector, an element is v.sub.i=v.sub.i, i=1, . . . , n, and v.sub.i=1, that is, a value of v.sub.i is +1 or 1. The matrix W.sub.2 is used to select one or more column vectors of the matrix W.sub.1, or is used to perform weighted combination on one or more column vectors of the W.sub.1 to obtain the precoding matrix W.

[0405] In another optional implementation manner of this embodiment, the block matrix X may be multiple different matrices, including a matrix P.sub.i and a matrix P.sub.j, where the P.sub.i and the P.sub.j satisfy the formula (51):

P.sub.i=D(i,j)P.sub.j(51) [0406] where the matrix D(i,j) is a diagonal matrix, and has the structure shown in the formula (5); and optionally, phases of diagonal elements .sub.1,.sub.2, . . . ,.sub.n of the matrix D(i,j) form an arithmetic progression.

[0407] The user equipment may select different precoding matrices at different time points according to a preset rule or randomly, that is, the user equipment may determine a first precoding matrix indicator PMI at a time point, where the PMI corresponds to the precoding matrix P.sub.i in the codebook, and send the first PMI to the base station; and determine a second precoding matrix indicator PMI at another time point, where the PMI corresponds to a precoding matrix D.sub.k in the codebook, and send the second PMI to the base station.

[0408] Corresponding to the case in which the foregoing user equipment sends the first PMI or the second PMI at different time points, on a base station side, the base station may also receive, at a time point, the first precoding matrix indicator PMI sent by the user equipment, and select the corresponding precoding matrix P.sub.i from the codebook according to the first PMI; and receive, at another time point, the second precoding matrix indicator PMI sent by the user equipment, and select the corresponding precoding matrix D.sub.k from the codebook according to the second PMI.

[0409] Optionally, the block matrix X may be multiple different matrices, including a matrix P.sub.i and a matrix P.sub.k, where the P.sub.i and the P.sub.k satisfy the formula (52):

D.sub.i.sup.1P.sub.i=D.sub.k.sup.1P.sub.k=V(52) [0410] where the matrix V includes a column vector 1 and/or at least one column vector v, the column vector 1 is a column vector whose elements are all 1, and the column vector v has the structure shown in the formula (50); and the matrix D.sub.i and the matrix D.sub.k both are diagonal matrices, and have the structure shown in the formula (53):

D.sub.m=.sub.m.Math.diag{u.sub.m,1,u.sub.m,2, . . . ,u.sub.m,n,u.sub.m,n*,u.sub.m,n1*, . . . ,u.sub.m,1*}m=i,k(53) [0411] where .sub.m is a complex factor, and a real part or an imaginary part of the complex factor may be 0; a complex number u.sub.m,l* is a conjugate complex number of a complex number u.sub.m,l, m=i,k, l=1, . . . ,n, and n is determined by a quantity of antenna ports; and optionally, phases of diagonal elements u.sub.m,1,u.sub.m,2, . . . ,u.sub.m,n of the matrix D.sub.m form an arithmetic progression.

[0412] It should be pointed out that diagonal elements of the foregoing diagonal matrix may have same amplitude. In this case, the structure of the foregoing precoding matrix allows that transmit antennas corresponding to rows of the precoding matrix have symmetric transmit powers based on actual considerations, and in this case, the foregoing codebook can still control a beam orientation by using a symmetric property of the powers of the transmit antennas, and meanwhile ensure orthogonality between transmission layers.

[0413] In the foregoing embodiment of the present invention, a base station receives a precoding matrix indicator PMI sent by user equipment, and determines a precoding matrix from a codebook according to the PMI. A precoding matrix W included in the codebook is a product of a matrix W.sub.1 and a matrix W.sub.2, where the matrix W.sub.1 is a block diagonal matrix, W.sub.1=diag {X.sub.1, . . . ,X.sub.N.sub.B}, and N.sub.B1, where at least one block matrix X is a product of a matrix D and a matrix V, X {X.sub.1,X.sub.2, . . . ,X.sub.N.sub.B}, and X=DV. The D is a diagonal matrix and satisfies D=.Math.diag {u.sub.1,u.sub.2, . . . ,u.sub.n,u.sub.n*,u.sub.n1*, . . . ,u.sub.1*}, where u.sub.1,u.sub.2, . . . ,u.sub.n,u.sub.n*,u.sub.n1*, . . . ,u.sub.1* forms a conjugate and symmetric sequence, which avoids constant modulus restrictions, and can effectively control a beam shape and a beam orientation. The matrix V includes a column vector 1 and at least one column vector v=[v.sub.1 v.sub.2 L v.sub.n v.sub.n v.sub.n1 L v.sub.1].sup.T, so that column vectors of the precoding matrix are orthogonal to each other, which can effectively reduce inter-layer interference, thereby greatly improving performance of MIMO, especially MU-MIMO. Therefore, the foregoing method for determining a precoding matrix can fully use a degree of freedom of controlling a beam shape and a beam orientation of an antenna system, and meanwhile reduce inter-layer interference of MIMO transmission as much as possible, thereby improving precision of CSI feedback, and a system throughput.

[0414] Specifically, the base station may obtain the precoding matrix from the codebook according to the received PMI, where the codebook is the same as the codebook used by the user equipment. Further, the base station may further precode to-be-sent data according to the obtained precoding matrix.

[0415] Using n=5 as an example, the foregoing diagonal elements u.sub.1,u.sub.2, . . . ,u.sub.n, diagonal elements u.sub.n*,u.sub.n1*, . . . ,u.sub.1*, and column vector v may be respectively shown in the formula (8) to the formula (12).

[0416] Using n=5 as an example, the diagonal elements .sub.1,.sub.2,L,.sub.n and the diagonal elements .sub.n*,.sub.n1*,L,.sub.1* in the formula (5) used in this embodiment may be respectively shown in the formula (8a) to the formula (11a).

[0417] Optionally, as another embodiment, in the matrix D, phases of the diagonal elements u.sub.1,u.sub.2, . . . ,u.sub.n form an arithmetic progression, and phases of the diagonal elements u.sub.n*,u.sub.n1*, . . . ,u.sub.1* form an arithmetic progression. Using n=4 as an example, the diagonal elements u.sub.1,u.sub.2, . . . ,u.sub.n and the diagonal elements u.sub.n*,u.sub.n1*, . . . ,u.sub.1* may be respectively shown in the formula (13) to the formula (16).

[0418] Using n=4 as an example, the diagonal elements .sub.1,.sub.2,L,.sub.n and the diagonal elements .sub.n*,.sub.n1*,L,.sub.1* in the formula (5) used in this embodiment may be respectively shown in the formula (13a) to the formula (16a).

[0419] Using n=4 as an example, the diagonal elements of the diagonal matrices shown in the formula (53) used in this embodiment may be respectively shown in the formula (17) to the formula (20).

[0420] In this embodiment, in the matrix D, the phases of the diagonal elements u.sub.1,u.sub.2, . . . ,u.sub.n form an arithmetic progression, and the phases of the diagonal elements u.sub.n*,u.sub.n1*, . . . ,u.sub.1* form an arithmetic progression, which may match with an array structure of an antenna port, for example, a common uniform linear array or cross polarization array, where in the former array, array elements or antennas are arranged at a same spacing, and in the latter array, co-polarized antennas or array elements are arranged at a same spacing. Therefore, the phases in the arithmetic progression can improve precoding performance by using a property of the foregoing array structure.

[0421] Optionally, as another embodiment, the column vector v of the matrix V may be a column vector of a matrix [H.sup.T H.sup.T].sup.T, where a matrix H is a Hadamard matrix. Using n=4 as an example, the Hadamard matrix and the column vector v may be respectively shown in the formula (21) to the formula (24).

[0422] In this embodiment, the column vector v is a column vector of the matrix [H.sup.T H.sup.T].sup.T, and satisfies the property of the formula (50), and column vectors of the [H.sup.T H.sup.T].sup.T are orthogonal to each other, so that the obtained column vectors are orthogonal to each other, thereby reducing inter-layer interference that is generated when the precoding matrix is used for MIMO transmission. In the foregoing embodiment of the present invention, the matrix W.sub.2 is used to select a column vector of the matrix W.sub.1, or is used to perform weighted combination on a column vector of the W.sub.1 to form the matrix W.

[0423] Using W.sub.1=diag {X.sub.1,X.sub.2} an example, where each of a block matrix X.sub.1 and a block matrix X.sub.2 has four columns, the W.sub.2 may be the matrix shown in the formula (36). Using W.sub.1=diag {X.sub.1,X.sub.2} as an example, where each of a block matrix X.sub.1 and a block matrix X.sub.2 has eight columns, the W.sub.2 may be the matrix shown in the formula (37) to the formula (40).

[0424] Using W.sub.1=diag {X.sub.1,X.sub.2,X.sub.3,X.sub.4} an example, where each block matrix of X.sub.1,X.sub.2,X.sub.3,X.sub.4 has four columns, the W.sub.2 may be the matrix shown in the formula (41). Further, the block matrix is X.sub.1=X.sub.2,X.sub.3=X.sub.4 or X.sub.1=X.sub.2=X.sub.3=X.sub.4.

[0425] Using W.sub.1=diag {X.sub.1,X.sub.2,X.sub.3,X.sub.4} an example, where each block matrix of X.sub.1,X.sub.2,X.sub.3,X.sub.4 has eight columns, the W.sub.2 may be the matrix shown in the formula (42) to the formula (45). Further, the block matrix is X.sub.1=X.sub.2,X.sub.3=X.sub.4 or X.sub.1=X.sub.2=X.sub.3=X.sub.4.

[0426] In the foregoing embodiment of the present invention, the determining, a precoding matrix from a codebook according to the PMI may include: [0427] determining the precoding matrix from a codebook subset according to the PMI, where the foregoing codebook subset is a predefined codebook subset, or a codebook subset reported to the base station, or a codebook subset reported to the base station, and returned and confirmed by the base station.

[0428] The foregoing codebook subset may include a set of the precoding matrix W=W.sub.1W.sub.2, where W.sub.1=diag {X.sub.1, . . . ,X.sub.N.sub.B}, X.sub.k=DV, and the matrix D, the matrix V, or the matrix W.sub.2 is a subset of a candidate matrix thereof.

[0429] It should be understood that the precoding matrix in the codebook or the codebook subset may be pre-stored in the user equipment and the base station, or may be calculated by the user equipment and the base station according to the structure of the foregoing precoding matrix, which is not limited in the present invention.

[0430] In addition, the foregoing precoding matrix indicator PMI may include only one index, that is, one index directly indicates one precoding matrix, or the foregoing precoding matrix indicator may include two indexes, namely a first index PMI1 and a second index PMI2, where the first index PMI1 and the second index PMI2 jointly indicate the precoding matrix. In addition, the first index PMI1 is used for indicating the matrix W.sub.1. and the second index PMI2 is used for indicating the matrix W.sub.2. The foregoing first index PMI1 and second index PMI2 may have different time-domain granularities or frequency-domain granularities, that is, the PMI1 and the PMI2 separately represent channel characteristics of different periods or bandwidths, or are obtained based on different subframe periods or subbands.

[0431] Optionally, the foregoing PMI may also include three indexes, where the three indexes are respectively used for indicating the matrix D, the matrix V, and the matrix W.sub.2.

[0432] Optionally, as another embodiment, the base station receives, by using different time periods, the first index PMI1 and the second index PMI2 that are sent by the user equipment, for example, the PMI1 may have a longer subframe period than the PMI2.

[0433] Specifically, the foregoing base station may receive, by using a PUCCH or a PUSCH, the precoding matrix indicator PMI sent by the user equipment UE.

[0434] The precoding matrix W m this embodiment of the present invention may be a precoding matrix obtained by means of row or column transposition, for example, different antenna numbers correspondingly cause row transposition of the precoding matrix. The precoding matrix provided in the foregoing embodiment of the present invention not only may be used for antenna configuration in a horizontal direction in an AAS base station, but also may be used for antenna configuration in a perpendicular direction.

[0435] FIG. 7 is a fifth schematic flowchart of a specific embodiment of the present invention. This embodiment provides a method for determining a precoding matrix indicator that is executed on a user equipment side when a precoding matrix satisfies a third condition. As shown in FIG. 7, the method includes the following steps:

[0436] Step 701: User equipment receives a reference signal sent by a base station.

[0437] Specifically, in this step, the user equipment may receive the reference signal in multiple manners, and the reference signal sent by the base station may include a CSI RS, a DM RS, or a CRS. The user equipment UE may acquire the reference signal by using a received eNB notification, or may obtain, based on a cell identity ID, a resource configuration of the reference signal and obtain the reference signal in a corresponding resource or subframe, where the eNB notification may be higher-layer signaling such as RRC signaling, or dynamic signaling such as DCI.

[0438] Step 702: The user equipment selects, based on the reference signal, a precoding matrix from a codebook, where a precoding matrix W included in the codebook is a product of a matrix W.sub.1 and a matrix W.sub.2, where

W=W.sub.1W.sub.2(54)

[0439] The matrix W.sub.1 is a block diagonal matrix,

W.sub.1=diag{X.sub.1, . . . X.sub.N.sub.B},N.sub.B1(55) [0440] where at least one block matrix X is a Kronecker product of a matrix A and a matrix B, and X {X.sub.1,X.sub.2, . . . ,X.sub.N.sub.B}, that is, the block matrix X has the structure shown in the formula (56):

X=A.Math.B(56) [0441] where the matrix A or the matrix B is a product of a matrix D and a matrix V,

A=DV(57)

[0442] or

B=DV(58)

[0443] The matrix D is a diagonal matrix, and satisfies:

D=.Math.diag{u.sub.1,u.sub.2, . . . ,u.sub.n,u.sub.N*,u.sub.n1*, . . . ,u.sub.1*}(59) [0444] where is a complex factor, a complex number u.sub.i* is a conjugate complex number of a complex number .sub.i, and i=1, . . . ,n; and the matrix V includes a column vector 1 and/or at least one column vector v, where the column vector 1 is a column vector whose elements are all 1, and the column vector v is:

v=[v.sub.1v.sub.2lv.sub.nv.sub.nv.sub.n1Lv.sub.1].sup.T(60) [0445] where an element is v.sub.i=v.sub.i, i=1, . . . ,n, and v.sub.i=1, that is, a value of v.sub.i is +1 or 1. The matrix W.sub.2 is used to select one or more column vectors of the matrix W.sub.1, or is used to perform weighted combination on one or more column vectors of the W.sub.1 to obtain the precoding matrix W.

[0446] In another optional implementation manner of this embodiment, the matrix A or the matrix B may be multiple different matrices, including a matrix P.sub.i and a matrix P.sub.j, where the P.sub.i and the P.sub.j satisfy the formula (61):

P.sub.i=D9i,j)P.sub.j(61) [0447] where the matrix D(i,j) is a diagonal matrix, and has the structure shown in the formula (5); and optionally, phases of diagonal elements .sub.1,.sub.2, . . . ,.sub.n of the matrix D(i,j) form an arithmetic progression.

[0448] Optionally, the matrix A or the matrix B may be multiple different matrices, including a matrix P.sub.i and a matrix P.sub.k, where the P.sub.j and the P.sub.k satisfy the formula (62):

D.sub.i.sup.1P.sub.i=D.sub.k.sup.1P.sub.k=V(62) [0449] where the matrix V includes a column vector 1 and/or at least one column vector v, the column vector 1 is a column vector whose elements are all 1, and the column vector v has the structure shown in the formula (60); and the matrix D.sub.i and the matrix D.sub.k both are diagonal matrices, and have the structure shown in the formula (63):

D.sub.m=.sub.m.Math.diag{u.sub.m,1,u.sub.m,2, . . . ,u.sub.m,n,u.sub.m,n*,u.sub.m,n1*, . . . ,u.sub.m,1*},m=i,k(63) [0450] where .sub.m is a complex factor, and a real part or an imaginary part of the complex factor may be 0; a complex number u.sub.m,l* is a conjugate complex number of a complex number u.sub.m,l, m=i,k, l=1, . . . ,n, and n is determined by a quantity of antenna ports; and optionally, phases of diagonal elements u.sub.m,1,u.sub.m,2, . . . ,u.sub.m,n of the matrix D.sub.m form an arithmetic progression.

[0451] It should be pointed out that diagonal elements of the foregoing diagonal matrix may have same amplitude. In this case, the structure of the foregoing precoding matrix allows that transmit antennas corresponding to rows of the precoding matrix have symmetric transmit powers based on actual considerations, and in this case, the foregoing codebook can still control a beam orientation by using a symmetric property of the powers of the transmit antennas, and meanwhile ensure orthogonality between transmission layers.

[0452] Step 703: The user equipment sends a precoding matrix indicator PMI to the base station, where the PMI corresponds to the selected precoding matrix.

[0453] In the foregoing embodiment of the present invention, user equipment selects, based on a reference signal, a precoding matrix from a codebook, and sends a precoding matrix indicator PMI. A precoding matrix W included in the codebook is a product of a matrix W.sub.1 and a matrix W.sub.2, where the matrix W is a block diagonal matrix, W.sub.1=diag {X.sub.1, . . . ,X.sub.N.sub.B}, and N.sub.B1, where at least one block matrix X is a Kronecker product of a matrix A and a matrix B . X=A .Math. B, and X {X.sub.1,X.sub.2, . . . ,X.sub.N.sub.B}; and the matrix A or the matrix B is a product of a matrix D and a matrix V, where A=DV or B=DV. The D is a diagonal matrix and satisfies D=.Math.diag {u.sub.1,u.sub.2, . . . ,u.sub.n,u.sub.n*,u.sub.n1*, . . . ,u.sub.1*}, where u.sub.1,u.sub.2, . . . ,u.sub.n,u.sub.n*,u.sub.n1*, . . . ,u.sub.1* forms a conjugate and symmetric sequence, which avoids constant modulus restrictions or a limit that antennas perform transmission by using equal powers, and can effectively control a beam shape and a beam orientation. The matrix V includes a column vector 1 and at least one column vector v=[v.sub.1 v.sub.2 L v.sub.n v.sub.n v.sub.n1 L v.sub.1].sup.T, so that column vectors of the precoding matrix are orthogonal to each other, which can effectively reduce inter-layer interference, thereby greatly improving performance of MIMO, especially MU-MIMO. Therefore, the foregoing method for determining a precoding matrix can fully use a degree of freedom of controlling a beam shape and a beam orientation in horizontal and perpendicular directions of an active antenna system, and meanwhile reduce inter-layer interference during transmission as much as possible, thereby improving precision of CSI feedback, and a system throughput.

[0454] Optionally, the base station may further obtain the precoding matrix according to the received PMI.

[0455] Specifically, the base station may obtain the precoding matrix from the codebook according to the received PMI, where the codebook is the same as the codebook used by the user equipment.

[0456] Specifically, when the matrix A is the matrix having the structure shown in the formula (57), the matrix B may be the matrix having the structure shown in the formula (58). In addition, the matrix B may also be a discrete fourier transformation (DFT) matrix, a Householder matrix, a Hadamard matrix, or a precoding matrix in a two-antenna codebook, a four-antenna codebook, or an eight-antenna codebook in an LTE R10 system.

[0457] Specifically, when the matrix B is the matrix having the structure shown in the formula (57), the matrix A may be the matrix having the structure shown in the formula (58). In addition, the matrix A may also be a DFT matrix, a Householder matrix, a Hadamard matrix, or a precoding matrix in a two-antenna codebook, a four-antenna codebook, or an eight-antenna codebook in an LTE R10 system.

[0458] Specifically, using n=5 as an example, the foregoing diagonal elements u.sub.1,u.sub.2, . . . ,u.sub.n, diagonal elements u.sub.n*,u.sub.n1*, . . . ,u.sub.1*, and column vector v may be respectively shown in the formula (8) to the formula (12).

[0459] Using n=5 as an example, the diagonal elements .sub.1,.sub.2,L,.sub.n and the diagonal elements .sub.n*,.sub.n1*,L,.sub.1* in the formula (5) used in this embodiment may be respectively shown in the formula (8a) to the formula (11a).

[0460] Optionally, as another embodiment, in the matrix D, phases of the diagonal elements u.sub.1,u.sub.2, . . . ,u.sub.n form m arithmetic progression, and phases of the diagonal elements u.sub.n*,u.sub.n1*, . . . ,u.sub.1* form an arithmetic progression. Using n=4 as an example, the diagonal elements u.sub.1,u.sub.2, . . . ,u.sub.n and the diagonal elements u.sub.n*,u.sub.n1*, . . . ,u.sub.1* may be respectively shown in the formula (13) to the formula (16).

[0461] Using n=4 as an example, the diagonal elements .sub.1,.sub.2,L,.sub.n and the diagonal elements .sub.n*,.sub.n1*,L,.sub.1* in the formula (5) used in this embodiment may be respectively shown in the formula (13a) to the formula (16a).

[0462] Using n=4 as an example, the diagonal elements of the diagonal matrices in the formula (63) used in this embodiment may be respectively shown in the formula (17) to the formula (20).

[0463] In this embodiment, in the matrix D the phases of the diagonal elements u.sub.1,u.sub.2, . . . ,u.sub.n form an arithmetic progression, and the phases of the diagonal elements u.sub.n*,u.sub.n1*, . . . ,u.sub.1* form an arithmetic progression, which may match with an array structure of an antenna port, for example, a common uniform linear array or cross polarization array, where in the former array, array elements or antennas are arranged at a same spacing, and in the latter array, co-polarized antennas or array elements are arranged at a same spacing. Therefore, the phases in the arithmetic progression can improve precoding performance by using a property of the foregoing array structure.

[0464] Optionally, as another embodiment, the column vector v of the matrix V may be a column vector of a matrix [H.sup.T H.sup.T].sup.T, where a matrix H is a Hadamard matrix. Using n=4 as an example, the Hadamard matrix and the column vector v may be respectively shown in the formula (21) to the formula (24).

[0465] In this embodiment, the column vector v is a column vector of the matrix [H.sup.T H.sup.T].sup.T, and satisfies the property of the formula (60), and column vectors of the [H.sup.T H.sup.T].sup.T are orthogonal to each other, so that the obtained column vectors are orthogonal to each other, thereby reducing inter-layer interference that is generated when the precoding matrix is used for MIMO transmission.

[0466] In the foregoing embodiment of the present invention, the matrix W.sub.2 is used to select a column vector of the matrix W.sub.1, or is used to perform weighted combination on a column vector of the W.sub.1 to form the matrix W.

[0467] Using W.sub.1=diag {X.sub.1,X.sub.2} as an example, where each of a block matrix X.sub.1 and a block matrix X.sub.2 has four columns, the W.sub.2 may be the matrix shown in the formula (36). Using W.sub.1=diag {X.sub.1,X.sub.2} as an example where each of a block matrix X.sub.1 and a block matrix X.sub.2 has eight columns, the W.sub.2 may be the matrix shown in the formula (37) to the formula (40).

[0468] Using W.sub.1=diag {X.sub.1,X.sub.2,X.sub.3,X.sub.4} as an example where each block matrix of X.sub.1,X.sub.2,X.sub.3,X.sub.4 has four columns, the W.sub.2 may be the matrix shown in the formula (41). Further, the block matrix is X.sub.1=X.sub.2,X.sub.3=X.sub.4 or X.sub.1=X.sub.2=X.sub.3=X.sub.4.

[0469] Using W.sub.1=diag {X.sub.1,X.sub.2,X.sub.3,X.sub.4} as an example, where each block matrix of X.sub.1,X.sub.2,X.sub.3,X.sub.4 has eight columns, the W.sub.2 may be the matrix shown in the formula (42) to the formula (45). Further, the block matrix is X.sub.1=X.sub.2,X.sub.3=X.sub.4 or X.sub.1=X.sub.2=X.sub.3=X.sub.4.

[0470] In the foregoing embodiment of the present invention, the selecting, by the user equipment, a precoding matrix from a codebook according to the reference signal may be specifically: obtaining, by the user equipment based on the reference signal, channel estimation; and selecting, based on a predefined criterion according to the channel estimation, the precoding matrix from the codebook, where the foregoing predefined criterion may be a channel capacity maximization criterion, a throughput maximization criterion, or a cosine distance minimization criterion.

[0471] In addition, in the foregoing embodiment of the present invention, the selecting a precoding matrix from a codebook according to the reference signal may include: [0472] selecting the precoding matrix from a codebook subset according to the reference signal, where the foregoing codebook subset is a predefined codebook subset, or a codebook subset reported to the base station, or a codebook subset reported to the base station, and returned and confirmed by the base station. In this embodiment, codebook subsets are set, for different application scenarios, in the codebook; and therefore, selecting a precoding matrix based on a codebook subset can effectively reduce feedback overheads and the implementation complexity.

[0473] Specifically, the codebook subset in the foregoing embodiment of the present invention may include: [0474] a set of a precoding matrix W=W.sub.1W.sub.2, where W.sub.1=diag {X.sub.1, . . . ,X.sub.N.sub.B}, at least one block matrix is X=A .Math. B, and X {X.sub.1,X.sub.2, . . . ,X.sub.N.sub.B}, where A=DV or B=DV, and the matrix D, the matrix V, or the matrix W.sub.2 is a subset of a candidate matrix thereof.

[0475] It should be pointed out that diagonal elements of the foregoing diagonal matrix may have same amplitude. In this case, the structure of the foregoing precoding matrix allows that transmit antennas corresponding to rows of the precoding matrix have symmetric transmit powers based on actual considerations, and in this case, the foregoing codebook can still control a beam orientation by using a symmetric property of the powers of the transmit antennas, and meanwhile ensure orthogonality between transmission layers.

[0476] It should be understood that the precoding matrix in the codebook or the codebook subset may be pre-stored in the user equipment and the base station, or may be calculated by the user equipment and the base station according to the structure of the foregoing precoding matrix, which is not limited in the present invention.

[0477] Further, in the foregoing precoding matrix, a block matrix X.sub.k and a block matrix X.sub.l,kl may different, or may be the same. In a case in which there are multiple X.sub.k s that are the same as the X.sub.l,kl, for example, the X.sub.k and the X.sub.l,kl that are the same exist in pairs, feedback overheads can be further reduced. Multiple block matrices X.sub.i in the foregoing matrix W.sub.1 may respectively correspond to antenna port groups having different polarizations or at different positions, so that the foregoing precoding matrix matches with multiple antenna deployments or configurations.

[0478] In step 703 shown in the foregoing FIG. 7, the precoding matrix indicator sent to the base station may include one or more indexes. Specifically, the codebook or the codebook subset usually is a set of one or more precoding matrices, where one precoding matrix indicator corresponds to one precoding matrix Different precoding matrix indicators correspond to different precoding matrices in the codebook or the codebook subset, and in this embodiment, the sent precoding matrix indicator corresponds to the selected precoding matrix.

[0479] Specifically, the foregoing precoding matrix indicator PMI may include only one index, that is, one index directly indicates one precoding matrix, or the foregoing precoding matrix indicator may include two indexes, namely a first index PMI1 and a second index PMI2, where the first index PMI1 and the second index PMI2 jointly indicate the precoding matrix. In addition, the first index PMI1 is used for indicating the matrix W.sub.1. and the second index PMI2 is used for indicating the matrix W.sub.2. The foregoing first index PMI1 and second index PMI2 may have different time-domain granularities or frequency-domain granularities, that is, the PMI1 and the PMI2 separately represent channel characteristics of different periods or bandwidths, or are obtained based on different subframe periods or subbands.

[0480] Optionally, the foregoing PMI may also include three indexes, where the three indexes

[0481] are respectively used for indicating the matrix D, the matrix V, and the matrix W.sub.2.

[0482] Optionally, as another embodiment, the user equipment sends the first index PMI1 and the second index PMI2 to the base station by using different time periods, for example, the PMI1 may have a longer subframe period than the PMI2.

[0483] In addition, in step 703 in the foregoing embodiment of the present invention, the precoding matrix indicator information PMI may be sent to the base station by using a PUCCH or a PUSCH.

[0484] The precoding matrix W in this embodiment may be a precoding matrix obtained by means of row or column transposition, for example, different antenna numbers correspondingly cause row transposition of the precoding matrix. In addition, the structure of the foregoing precoding matrix W not only may be used for antenna configuration in a horizontal direction in an AAS base station, but also may be used for antenna configuration in a perpendicular direction.

[0485] In this embodiment of the present invention, user equipment selects, based on a reference signal, a precoding matrix from a codebook, and sends a precoding matrix indicator PMI, where the PMI corresponds to the selected precoding matrix. A precoding matrix W included in the codebook is a product of a matrix W.sub.1 and a matrix W.sub.2, where the W.sub.1 is a block diagonal matrix, W.sub.1=diag {X.sub.1, . . . ,X.sub.N.sub.B}, and N.sub.B1, where at least one block matrix X is a Kronecker product of a matrix A and a matrix B , X=A .Math. B, and X {X.sub.1,X.sub.2, . . . ,X.sub.N.sub.B}; and the matrix A or the matrix B is a product of a matrix D and a matrix V. The D is a diagonal matrix and satisfies D=.Math.diag {u.sub.1,u.sub.2, . . . ,u.sub.n,u.sub.n*,u.sub.n*, . . . ,u.sub.1*}, where u.sub.1,u.sub.2, . . . ,u.sub.n,u.sub.n*,u.sub.n1*, . . . ,u.sub.1* forms a conjugate and symmetric sequence, which avoids constant modulus restrictions or a limit that antennas perform transmission by using equal powers, and can effectively control a beam shape and a beam orientation. The matrix V includes a column vector 1 and at least one column vector v=[v.sub.1 v.sub.2 L v.sub.n v.sub.n v.sub.n1 L v.sub.1].sup.T, so that column vectors of the precoding matrix are orthogonal to each other, which can effectively reduce inter-layer interference, thereby greatly improving performance of MIMO, especially MU-MIMO. Multiple block matrices X.sub.i may respectively correspond to antenna port groups having different polarizations or at different positions, so that the foregoing precoding matrix matches with multiple antenna deployments or configurations. Meanwhile, the matrix A or the matrix B may separately quantize a beam in a horizontal direction and a perpendicular direction. Therefore, the foregoing method for determining a precoding matrix can fully use a degree of freedom of controlling a beam shape and a beam orientation in horizontal and perpendicular directions of an active antenna system, and meanwhile reduce inter-layer interference during transmission as much as possible, thereby improving precision of CSI feedback, and a system throughput.

[0486] Corresponding to the embodiment shown in FIG. 7, the present invention further provides a method executed on a base station side. FIG. 8 is a sixth schematic flowchart of a specific embodiment of the present invention. As shown in FIG. 8, the embodiment includes the following steps:

[0487] Step 801: A base station sends a reference signal to user equipment.

[0488] Specifically, for a manner in which the base station sends the reference signal, reference may be made to step 701 shown in FIG. 7.

[0489] Step 802: Receive a precoding matrix indicator PMI sent by the user equipment.

[0490] Specifically, the precoding matrix indicator PMI corresponds to a precoding matrix that is selected from a codebook by the user equipment based on the reference signal.

[0491] Step 803: Determine a precoding matrix from a codebook according to the PMI, where a precoding matrix W included in the codebook is a product of a matrix W.sub.1 and a matrix W.sub.2, where

W=W.sub.1W.sub.2(64)

[0492] The matrix W.sub.1 is a block diagonal matrix,

W.sub.1=diag{X.sub.1, . . . ,X.sub.N.sub.B},N.sub.B1(65) [0493] where at least one block matrix X is a Kronecker product of a matrix A and a matrix B, and X {X.sub.1,X.sub.2, . . . ,X.sub.N.sub.B}, that is, the block matrix X has the structure shown in the formula (66):

X=A.Math.B(66) [0494] where the matrix A or the matrix B is a product of a matrix D and a matrix V,

A=DV(67)

[0495] or,

B=DV(68) [0496] where the matrix D is a diagonal matrix, and satisfies:

D=.Math.diag{u.sub.1,u.sub.2, . . . ,u.sub.n,u.sub.n*,u.sub.n1*, . . . ,u.sub.1*}(69) [0497] where is a complex factor, a complex number u.sub.i* is a conjugate complex number of a complex number u.sub.i, and i=1, . . . ,n; and the matrix V includes a column vector 1 and at least one column vector v, where the column vector 1 is a column vector whose elements are all 1, and the column vector v is:

v=[v.sub.1v.sub.2Lv.sub.nv.sub.nv.sub.n1Lv.sub.1].sup.T(70) [0498] where an element is v.sub.i=v.sub.i, i=1, . . . ,n, and v.sub.i=1, that is, a value of v.sub.i is +1 or 1. The matrix W.sub.2 is used to select one or more column vectors of the matrix W.sub.1, or is used to perform weighted combination on one or more column vectors of the W.sub.1 to obtain the precoding matrix W.

[0499] In another optional implementation manner of this embodiment, the matrix A or the matrix B may be multiple different matrices, including a matrix P.sub.i and a matrix P.sub.j, where the P.sub.i and the P.sub.j satisfy the formula (71):

P.sub.iD(i,j)P.sub.j(71) [0500] where the matrix D(i,j) is a diagonal matrix, and has the structure shown in the formula (5); and optionally, phases of diagonal elements .sub.1,.sub.2, . . . ,.sub.n of the matrix D(i,j) form an arithmetic progression.

[0501] Optionally, the matrix A or the matrix B may be multiple different matrices, including a matrix P.sub.i and a matrix P.sub.k, where the P.sub.i and the P.sub.k satisfy the formula (72):

D.sub.i.sup.1P.sub.i=D.sub.k.sup.1P.sub.k=V(72) [0502] where the matrix V includes a column vector 1 and/or at least one column vector v, the column vector 1 is a column vector whose elements are all 1, and the column vector v has the structure shown in the formula (60); and the matrix D.sub.i and the matrix D.sub.k both are diagonal matrices, and have the structure shown in the formula (73):

D.sub.m=.sub.m.Math.diag{u.sub.m,1,u.sub.m,2, . . . ,u.sub.m,n,u.sub.m,n*,u.sub.m,n1*, . . . ,u.sub.m,1*},m=i,k(73) [0503] where .sub.m is a complex factor, and a real part or an imaginary part of the complex factor may be 0; a complex number u.sub.m,l* is a conjugate complex number of a complex number u.sub.m,l, m=i,k, l=1, . . . ,n, and n is determined by a quantity of antenna ports; and optionally, phases of diagonal elements u.sub.m,1,u.sub.m,2, . . . ,u.sub.m,n of the matrix D.sub.m form an arithmetic progression.

[0504] It should be pointed out that diagonal elements of the foregoing diagonal matrix may have same amplitude. In this case, the structure of the foregoing precoding matrix allows that transmit antennas corresponding to rows of the precoding matrix have symmetric transmit powers based on actual considerations, and in this case, the foregoing codebook can still control a beam orientation by using a symmetric property of the powers of the transmit antennas, and meanwhile ensure orthogonality between transmission layers.

[0505] In the foregoing embodiment of the present invention, a base station receives a precoding matrix indicator PMI sent by user equipment, and determines a precoding matrix from a codebook according to the PMI. A precoding matrix W included in the codebook is a product of a matrix W.sub.1 and a matrix W.sub.2, where the matrix W.sub.1 is a block diagonal matrix, W.sub.1=diag {X.sub.1, . . . ,X.sub.N.sub.B}, and N.sub.B1, where at least one block matrix X is a Kronecker product of a matrix A and a matrix B, X=A .Math. B, and X {X.sub.1,X.sub.2, . . . ,X.sub.N.sub.B}; and the matrix A or the matrix B is a product of a matrix D and a matrix V, where A=DV or B=DV. The D is a diagonal matrix and satisfies D=.Math.diag {u.sub.1,u.sub.2, . . . ,u.sub.n,u.sub.n*,u.sub.n1*, . . . ,u.sub.1*}, where u.sub.1,u.sub.2, . . . ,u.sub.n,u.sub.n*,u.sub.n1, . . . ,u.sub.1* forms a conjugate and symmetric sequence, which avoids constant modulus restrictions or a limit that antennas perform transmission by using equal powers, and can effectively control a beam shape and a beam orientation. The matrix V includes a column vector 1 and at least one column vector v=[v.sub.1 v.sub.2 L v.sub.n v.sub.n v.sub.n1 L v.sub.1].sup.T, so that column vectors of the precoding matrix are orthogonal to each other, which can effectively reduce inter-layer interference, thereby greatly improving performance of MIMO, especially MU-MIMO. Therefore, the foregoing method for determining a precoding matrix can fully use a degree of freedom of controlling a beam shape and a beam orientation in horizontal and perpendicular directions of an active antenna system, and meanwhile reduce inter-layer interference during transmission as much as possible, thereby improving precision of CSI feedback, and a system throughput.

[0506] Specifically, the base station may obtain the precoding matrix from the codebook according to the received PMI, where the codebook is the same as the codebook used by the user equipment. Further, the base station may further precode to-be-sent data according to the obtained precoding matrix.

[0507] Specifically, when the matrix A is the matrix having the structure shown in the formula (67), the matrix B may be the matrix having the structure shown in the formula (68). The matrix B may also be a DFT matrix, a Householder matrix, a Hadamard matrix, or a precoding matrix in a two-antenna codebook, a four-antenna codebook, or an eight-antenna codebook in an LTE R10 system.

[0508] Specifically, when the matrix B is the matrix having the structure shown in the formula (68), the matrix A may be the matrix having the structure shown in the formula (67). The matrix A may also be a DFT matrix, a Householder matrix, a Hadamard matrix, or a precoding matrix in a two-antenna codebook, a four-antenna codebook, or an eight-antenna codebook in an LTE R10 system.

[0509] Specifically, using n=5 as an example, the foregoing diagonal elements u.sub.1u.sub.2, . . . ,u.sub.n, diagonal elements u.sub.n*,u.sub.n1*, . . . ,u.sub.1*, and column vector v may be respectively shown in the formula (8) to the formula (12).

[0510] Using n=5 as an example, the diagonal elements .sub.1,.sub.2,L,.sub.n and the diagonal elements .sub.n*,.sub.n1*,L,.sub.1* in the formula (5) used in this embodiment may be respectively shown in the formula (8a) to the formula (11a).

[0511] Optionally, as another embodiment, in the matrix D, phases of the diagonal elements u.sub.1,u.sub.2, . . . ,u.sub.n form an arithmetic progression, and phases of the diagonal elements u.sub.n*,u.sub.n1*, . . . ,u.sub.1* form an arithmetic progression.

[0512] Using n4 as an example, the diagonal elements u.sub.1,u.sub.2, . . . ,u.sub.n and the diagonal elements u.sub.n*,u.sub.n1, . . . ,u.sub.1* may be respectively shown in the formula (13) to the formula (16).

[0513] Using n=4 as an example, the diagonal elements .sub.1,.sub.2,L,.sub.n and the diagonal elements .sub.n*,.sub.n1*,L.sub.1* in the formula (5) used in this embodiment may be respectively shown in the formula (13a) to the formula (16a).

[0514] Using n=4 as an example, the diagonal elements of the diagonal matrices in the formula (73) used in this embodiment may be respectively shown in the formula (17) to the formula (20).

[0515] In this embodiment, in the matrix D, the phases of the diagonal elements u.sub.1,u.sub.2, . . . ,u.sub.n form an arithmetic progression, and the phases of the diagonal elements u.sub.n*,u.sub.n1*, . . . ,u.sub.1* form an arithmetic progression, which may match with an array structure of an antenna port, for example, a common uniform linear array or cross polarization array, where in the former array, array elements or antennas are arranged at a same spacing, and in the latter array, co-polarized antennas or array elements are arranged at a same spacing. Therefore, the phases in the arithmetic progression can improve precoding performance by using a property of the foregoing array structure.

[0516] Optionally, as another embodiment, the column vector v of the matrix V may be a column vector of a matrix [H.sup.T H.sup.T].sup.T, where a matrix H is a Hadamard matrix. Using n=4 as an example, the Hadamard matrix and the column vector v may be respectively shown in the formula (21) to the formula (24).

[0517] In this embodiment, the column vector v is a column vector of the matrix [H.sup.T H.sup.T].sup.T, and satisfies the property of the formula (70), and column vectors of the [H.sup.T H.sup.T].sup.T are orthogonal to each other, so that the obtained column vectors are orthogonal to each other, thereby reducing inter-layer interference that is generated when the precoding matrix is used for MIMO transmission.

[0518] In the foregoing embodiment of the present invention, the matrix W.sub.2 is used to select a column vector of the matrix W.sub.1, or is used to perform weighted combination on a column vector of the W.sub.1 to form the matrix W. Using W.sub.1=diag {X.sub.1,X.sub.2} as an example where each of a block matrix X.sub.1 and a block matrix X.sub.2 has four columns, the W.sub.2 may be the matrix shown in the formula (36). Using W.sub.1=diag {X.sub.1,X.sub.2} as an example where each of a block matrix X.sub.1 and a block matrix X.sub.2 has eight columns, the W.sub.2 may be the matrix shown in the formula (37) to the formula (40).

[0519] Using W.sub.1=diag {X.sub.1,X.sub.2,X.sub.3,X.sub.4} as an example where each block matrix of X.sub.1,X.sub.2,X.sub.3,X.sub.4 has four columns, the W.sub.2 may be the matrix shown in the formula (41). Further, the block matrix is X.sub.1=X.sub.2,X.sub.3=X.sub.4 or X.sub.1=X.sub.2=X.sub.3=X.sub.4.

[0520] Using W.sub.1=diag {X.sub.1,X.sub.2,X.sub.3,X.sub.4} as an example where each block matrix of X.sub.1,X.sub.2, X.sub.3,X.sub.4 has eight columns, the W.sub.2 may be the matrix shown in the formula (42) to the formula (45). Further, the block matrix is X.sub.1=X.sub.2,X.sub.3=X.sub.4 or X.sub.1=X.sub.2=X.sub.3=X.sub.4.

[0521] In the foregoing embodiment of the present invention, the determining a precoding matrix from a codebook according to the PMI includes: [0522] determining the precoding matrix from the codebook according to the PMI, where the foregoing codebook subset is a predefined codebook subset, or a codebook subset reported to the base station, or a codebook subset reported to the base station, and returned and confirmed by the base station. In this embodiment, codebook subsets are set, for different application scenarios, in the codebook; and therefore, selecting a precoding matrix based on a codebook subset can effectively reduce feedback overheads and the implementation complexity.

[0523] Specifically, the codebook subset in the foregoing embodiment of the present invention may include: [0524] a set or a precoding matrix W=W.sub.1W.sub.2, where W.sub.1=diag {X.sub.1, . . . ,X.sub.N.sub.B}, at least one block matrix is X=A .Math. B, and X {X.sub.1,X.sub.2, . . . ,X.sub.N.sub.B}, where A=DV or B=DV, and the matrix D, the matrix V, or the matrix W.sub.2 is a subset of a candidate matrix thereof.

[0525] It should be pointed out that diagonal elements of the foregoing diagonal matrix may have same amplitude. In this case, the structure of the foregoing precoding matrix allows that transmit antennas corresponding to rows of the precoding matrix have symmetric transmit powers based on actual considerations, and in this case, the foregoing codebook can still control a beam orientation by using a symmetric property of the powers of the transmit antennas, and meanwhile ensure orthogonality between transmission layers.

[0526] It should be understood that the precoding matrix in the codebook or the codebook subset may be pre-stored in the user equipment and the base station, or may be calculated by the user equipment and the base station according to the structure of the foregoing precoding matrix, which is not limited in the present invention.

[0527] In addition, the precoding matrix indicator sent to the base station may include one or more indexes. Specifically, the codebook or the codebook subset usually is a set of one or more precoding matrices, where one precoding matrix indicator corresponds to one precoding matrix. Different precoding matrix indicators correspond to different precoding matrices in the codebook or the codebook subset, and in this embodiment, the sent precoding matrix indicator corresponds to the selected precoding matrix.

[0528] Specifically, the foregoing precoding matrix indicator PMI may include only one index, that is, one index directly indicates one precoding matrix, or the foregoing precoding matrix indicator may include two indexes, namely a first index PMI1 and a second index PMI2, where the first index PMI1 and the second index PMI2 jointly indicate the precoding matrix. In addition, the first index PMI1 is used for indicating the matrix W.sub.1, and the second index PMI2 is used for indicating the matrix W.sub.2. The foregoing first index PMI1 and second index PMI2 may have different time-domain granularities or frequency-domain granularities, that is, the PMI1 and the PMI2 separately represent channel characteristics of different periods or bandwidths, or are obtained based on different subframe periods or subbands.

[0529] Optionally, the foregoing PMI may also include three indexes, where the three indexes are respectively used for indicating the matrix D, the matrix V, and the matrix W.sub.2.

[0530] Optionally, as another embodiment, the base station receives, by using different time periods, the first index PMI1 and the second index PMI2 that are sent by the user equipment, for example, the PMI1 may have a longer subframe period than the PMI2.

[0531] Specifically, the foregoing base station may receive, by using a PUCCH or a PUSCH, the precoding matrix indicator PMI sent by the user equipment UE.

[0532] The precoding matrix W m this embodiment of the present invention may be a precoding matrix obtained by means of row or column transposition, for example, different antenna numbers correspondingly cause row transposition of the precoding matrix.

[0533] In the foregoing embodiment of the present invention, a base station sends a reference signal and receives a precoding matrix indicator PMI sent by user equipment, where the PMI corresponds to a precoding matrix that is selected from a codebook by the user equipment based on the reference signal. A precoding matrix W included in the codebook is a product of a matrix W.sub.1 and a matrix W.sub.2, where the W.sub.1 is a block diagonal matrix, W.sub.1=diag {W.sub.1, . . . ,X.sub.N.sub.B}, and N.sub.B1, where at least one block matrix X is a Kronecker product of a matrix A and a matrix B, X=A .Math. B, and X {X.sub.1,X.sub.2, . . . ,X.sub.N.sub.B}; and the matrix A or the matrix B is a product of a matrix D and a matrix V. The D is a diagonal matrix and satisfies D=.Math.diag {u.sub.1,u.sub.2, . . . ,u.sub.m,u.sub.n*,u.sub.n1*, . . . ,u.sub.1*}, where u.sub.1,u.sub.2, . . . ,u.sub.n,u.sub.n*,u.sub.n1*, . . . ,u.sub.1* forms a conjugate and symmetric sequence, which avoids constant modulus restrictions, and can effectively control a beam shape and a beam orientation. The matrix V includes a column vector 1 and at least one column vector v=[v.sub.1 v.sub.2 L v.sub.n v.sub.n v.sub.n1 L v.sub.1].sup.T, so that column vectors of the precoding matrix are orthogonal to each other, which can effectively reduce inter-layer interference, thereby greatly improving performance of MIMO, especially MU-MIMO. Therefore, the foregoing method for determining a precoding matrix can fully use a degree of freedom of controlling a beam shape and a beam orientation in horizontal and perpendicular directions of an active antenna system, and meanwhile reduce inter-layer interference during transmission as much as possible, thereby improving precision of CSI feedback, and a system throughput.

[0534] FIG. 9 is a first schematic structural diagram of an apparatus for determining a precoding matrix indicator according to an embodiment of the present invention. As shown in FIG. 9, the apparatus includes a first determining module 11 and a first sending module 12. The first determining module is configured to determine a precoding matrix indicator PMI, where the PMI corresponds to a precoding matrix and the precoding matrix W satisfies a first condition, a second condition, or a third condition; and [0535] the first sending module is configured to send the PMI to a base station, where [0536] the first condition is that the precoding matrix W satisfies W=DV, where the matrix D is a diagonal matrix, D=.Math.diag {u.sub.1,u.sub.2, . . . ,u.sub.n,u.sub.n*,u.sub.n1*, . . . ,u.sub.1*}, is a complex factor, a complex number u.sub.i* is a conjugate complex number of a complex number .sub.i, and n is determined by a quantity of antenna ports; and the matrix V is a constant modulus matrix; [0537] the second condition is that the precoding matrix W includes one or more column vectors of a block diagonal matrix W.sub.1, or the precoding matrix W is obtained by performing weighted combination on one or more column vectors of a block diagonal matrix W.sub.1, where W.sub.1=diag {X.sub.1, . . . ,X.sub.N.sub.B}, and N.sub.B1, where at least one block matrix X is a product X=DV of a matrix D and a matrix V, and X {X.sub.1,X.sub.2, . . . ,X.sub.N.sub.B}; the matrix D is a diagonal matrix, D=.Math.diag {u.sub.1,u.sub.2, . . . ,u.sub.n,u.sub.n*,u.sub.n1*, . . . ,u.sub.1*}, is a complex factor, a complex number u.sub.i* is a conjugate complex number of a complex number u.sub.i, and n is determined by a quantity of antenna ports; and the matrix V is a constant modulus matrix; and [0538] the third condition is that the precoding matrix W includes one or more column vectors of a block diagonal matrix W.sub.1, or the precoding matrix W is obtained by performing weighted combination on one or more column vectors of a block diagonal matrix W.sub.1, where W.sub.1=diag {X.sub.1, . . . ,X.sub.N.sub.B}, and N.sub.B1, where at least one block matrix X is a Kronecker product of a matrix A and a matrix B, X=A .Math. B, and X {X.sub.1,X.sub.2, . . . ,X.sub.N.sub.B}; the matrix A or the matrix B is a product of a matrix D and a matrix V; the matrix D is a diagonal matrix, D=.Math.diag {u.sub.1,u.sub.2, . . . ,u.sub.n,u.sub.n*,u.sub.n1*, . . . ,u.sub.1*}, is a complex factor, a complex number u.sub.i* is a conjugate complex number of a complex number u.sub.i, i=1, . . .,n, and n is a quantity of rows of the matrix A or the matrix B; and the matrix V is a constant modulus matrix.

[0539] In the second condition or the third condition, the precoding matrix W satisfies W=W.sub.1W.sub.2, where the matrix W.sub.2 is used to select one or more column vectors of the matrix W.sub.1; or is used to perform weighted combination on one or more column vectors of the W.sub.1 to obtain the precoding matrix W.

[0540] Phases of diagonal elements u.sub.1,u.sub.2, . . . ,u.sub.n of the foregoing matrix D form an arithmetic progression.

[0541] The foregoing matrix V includes a column vector 1 and/or at least one column vector v, the column vector 1 is a column vector whose elements are all 1, and the column vector v is v=[v.sub.1 v.sub.21 L v.sub.n v.sub.n v.sub.n1 L v.sub.1].sup.T, where an element is v.sub.i=v.sub.i,v.sub.i=1, and i=1, . . . ,n. The matrix V includes only the column vector 1 and the at least one column vector v, and when the matrix V includes multiple column vectors v, the multiple column vectors v are different. Optionally, the column vector v of the matrix V is a column vector of a matrix [H.sup.T H.sup.T].sup.T, where a matrix H is a Hadamard matrix.

[0542] In addition, the foregoing PMI includes a first index PMI1 and a second index PMI2, where [0543] when the precoding matrix W satisfies the first condition, the first index PMI1 corresponds to the matrix D, and the second index PMI2 corresponds to the matrix V; [0544] when the precoding matrix W satisfies the second condition, the first index PMI1 corresponds to the matrix W.sub.1, and the second index PMI2 corresponds to the matrix W.sub.2; or [0545] when the precoding matrix W satisfies the third condition, the first index PMI1 corresponds to the matrix W.sub.1, and the second index PMI2 corresponds to the matrix W.sub.2. Further, the foregoing first index PMI1 and second index PMI2 have different time-domain granularities or frequency-domain granularities; or the first index PMI1 and the second index PMI2 are sent to the base station by using different time periods.

[0546] In the foregoing embodiment of the present invention, the foregoing apparatus further includes: [0547] a first receiving module, configured to receive a reference signal sent by the base station, and select, from a codebook according to the reference signal, the precoding matrix W corresponding to the PMI.

[0548] The codebook includes a precoding matrix W.sub.i and a precoding matrix W.sub.j, and the two precoding matrices satisfy W.sub.i=D(i,j)W.sub.j, where D(i,j)=.sub.(i,j)diag {.sub.1,.sub.2, . . . ,.sub.n,.sub.n*,.sub.n1*, . . . ,.sub.1*}, .sub.(i,j) is a complex factor a complex number .sub.m* is a conjugate complex number of a complex number .sub.m, m=1, . . . ,n, and n is determined by a quantity of antenna ports. Optionally, phases of diagonal elements .sub.1,.sub.2, . . . ,.sub.n of the foregoing matrix D(i,j) form an arithmetic progression.

[0549] Alternatively, the foregoing codebook includes a precoding matrix w.sub.i and a precoding matrix W.sub.k, and the two precoding matrices satisfy D.sub.i.sup.1W.sub.i=D.sub.k.sup.1W.sub.k=V, where D.sub.m=.sub.m.Math.diag {u.sub.m,1,u.sub.m,2, . . . ,u.sub.m,n,u.sub.m,n*,u.sub.m,n1*, . . . ,u.sub.m,1*}, m=i,k, .sub.m is a complex factor, a complex number u.sub.m,l* is a conjugate complex number of a complex number u.sub.m,l, m=i,k, l=1, . . . ,n, and n is determined by a quantity of antenna ports.

[0550] Optionally, phases of diagonal elements u.sub.m,1,u.sub.m,2, . . . ,u.sub.m,n of the matrix D.sub.m form an arithmetic progression.

[0551] FIG. 10 is a second schematic structural diagram of an apparatus for determining a precoding matrix indicator according to an embodiment of the present invention. As shown in FIG. 10, the apparatus includes a second receiving module 21 and a second determining module 22. The second receiving module 21 is configured to receive a precoding matrix indicator PMI sent by user equipment; and the second determining module 22 is configured to determine a corresponding precoding matrix W according to the PMI, where the precoding matrix W satisfies a first condition, a second condition, or a third condition, where the first condition is that the precoding matrix W satisfies W=DV; the second condition is that the precoding matrix W includes one or more column vectors of a block diagonal matrix W.sub.1, or the precoding matrix W is obtained by performing weighted combination on one or more column vectors of a block diagonal matrix W.sub.1, where W.sub.1=diag {X.sub.1, . . . ,X.sub.N.sub.B}, and N.sub.B1, where at least one block matrix X is a product X=DV of a matrix D and a matrix V, and X {X.sub.1,X.sub.2, . . . ,X.sub.N.sub.B}; and the third condition is that the precoding matrix W includes one or more column vectors of a block diagonal matrix W.sub.1, or the precoding matrix W is obtained by performing weighted combination on one or more column vectors of a block diagonal matrix W.sub.1, where W.sub.1=diag {X.sub.1, . . . ,X.sub.N.sub.B}, and N.sub.B1, where at least one block matrix X is a Kronecker product of a matrix A and a matrix B, X=A .Math. B, and X {X.sub.1,X.sub.2, . . . ,X.sub.N.sub.B}; the matrix A or the matrix B is a product of a matrix D and a matrix V; the matrix D is a diagonal matrix, i=1, . . . ,n, and n is a quantity of rows of the matrix A or the matrix B, where [0552] the matrix D is a diagonal matrix, D=.Math.diag {u.sub.1,u.sub.2, . . . ,u.sub.n,u.sub.n*u.sub.n1*, . . . ,u.sub.1*}, is a complex factor, a complex number u.sub.i* is a conjugate complex number of a complex number u.sub.i, and n is determined by a quantity of antenna ports; and the matrix V is a constant modulus matrix.

[0553] In the foregoing embodiment of the present invention, in the second condition or the third condition, the precoding matrix W satisfies W=W.sub.1W.sub.2, where the matrix W.sub.2 is used to select one or more column vectors of the matrix W.sub.1; or is used to perform weighted combination on one or more column vectors of the W.sub.1 to obtain the precoding matrix W.

[0554] Optionally, phases of diagonal elements u.sub.1,u.sub.2, . . . ,u.sub.n of the foregoing matrix D form an arithmetic progression.

[0555] Further, the matrix V includes a column vector 1 and/or at least one column vector v, the column vector 1 is a column vector whose elements are all 1, and the column vector v is v=[v.sub.1 v.sub.2 L v.sub.n v.sub.n v.sub.n1 L v.sub.1].sup.T, where an element is v.sub.i=v.sub.i,v.sub.i=1, and i=1, . . . ,n. Optionally, the matrix V includes only the column vector 1 and the at least one column vector v, and when the matrix V includes multiple column vectors v, the multiple column vectors v are different.

[0556] In addition, the column vector v of the foregoing matrix V is a column vector of a matrix [H.sup.T H.sup.T].sup.T, where a matrix H is a Hadamard matrix.

[0557] In the foregoing embodiment of the present invention, the precoding matrix indicator PMI includes a first index PMI1 and a second index PMI2, where [0558] when the precoding matrix W satisfies the first condition, the first index PMI1 corresponds to the matrix D, and the second index PMI2 corresponds to the matrix V; [0559] when the precoding matrix W satisfies the second condition, the first index PMI1 corresponds to the matrix W.sub.1, and the second index PMI2 corresponds to the matrix W.sub.2; or [0560] when the precoding matrix W satisfies the third condition, the first index PMI1 corresponds to the matrix W.sub.1, and the second index PMI2 corresponds to the matrix W.sub.2.

[0561] Optionally, the first index PMI1 and the second index PMI2 have different time-domain granularities or frequency-domain granularities; or the first index PMI1 and the second index PMI2 are sent to the base station by using different time periods.

[0562] In addition, the foregoing determining a corresponding precoding matrix W according to the PMI includes: [0563] selecting the corresponding precoding matrix W from a codebook according to the PMI.

[0564] Specifically, the foregoing codebook includes a precoding matrix W.sub.i and a precoding matrix W.sub.j, and the two precoding matrices satisfy W.sub.i=D(i,j)W.sub.j, where D(i,j)=.sub.(i,j)diag {.sub.1,.sub.2, . . . ,.sub.n,.sub.n*,.sub.n1*, . . . ,.sub.1*}, .sub.(i,j) is a complex factor a complex number .sub.m* is a conjugate complex number of a complex number .sub.m, n=1, . . . ,n, and n is determined by a quantity of antenna ports.

[0565] Phases of diagonal elements .sub.1,.sub.2, . . . ,.sub.n of the matrix D(i,j) form an arithmetic progression.

[0566] Alternatively, the foregoing codebook includes a precoding matrix W.sub.i and a precoding matrix W.sub.k, and the two precoding matrices satisfy D.sub.i.sup.1W.sub.i=D.sub.k.sup.1W.sub.k=V, where D.sub.m=.sub.m.Math.diag {u.sub.m,1,u.sub.m,2, . . . ,u.sub.m,n,u.sub.m,n*,u.sub.m,n1*, . . . ,u.sub.m,1*}, m=i,k, .sub.m is a complex factor, a complex number u.sub.m,l* is a conjugate complex number of a complex number u.sub.m,l, m=i,k, l=1, . . . ,n, and n is determined by a quantity of antenna ports.

[0567] Phases of diagonal elements u.sub.m,1,u.sub.m,2, . . . ,u.sub.m,n of the foregoing matrix D.sub.m form an arithmetic progression.

[0568] FIG. 11 is a third schematic structural diagram of an apparatus for determining a precoding matrix indicator according to an embodiment of the present invention. As shown in FIG. 11, the apparatus includes a third determining module 31 and a second sending module 32. The third determining module 31 is configured to determine a first precoding matrix indicator PMI, where the PMI corresponds to a precoding matrix W.sub.i in a codebook; and the second sending module 32 is configured to send the first PMI to a base station, where the codebook includes at least: the precoding matrix W.sub.i and a precoding matrix W.sub.j, and the precoding matrix W.sub.i and the precoding matrix W.sub.j in the codebook satisfy W.sub.i=D(i,j)W.sub.j, where D(i,j)=.sub.(i,j)diag {.sub.1,.sub.2, . . . ,.sub.n,.sub.n*,.sub.n1*, . . . ,.sub.1*}, .sub.(i,j) is a complex factor a complex number .sub.m* is a conjugate complex number of a complex number .sub.m, m=1, . . . ,n, and n is determined by a quantity of antenna ports.

[0569] Optionally, phases of diagonal elements .sub.1,.sub.2, . . . ,.sub.n of the foregoing matrix D(i,j) form an arithmetic progression.

[0570] FIG. 12 is a fourth schematic structural diagram of an apparatus for determining a precoding matrix indicator according to an embodiment of the present invention. As shown in FIG. 12, the apparatus includes a third receiving module 41 and a fourth determining module 42. The third receiving module 41 is configured to receive a first precoding matrix indicator PMI sent by user equipment; and the fourth determining module 42 is configured to determine a corresponding precoding matrix W.sub.i from a codebook according to the first PMI, where the codebook includes at least: the precoding matrix W.sub.i and a precoding matrix W.sub.j, and the precoding matrix W.sub.i and the precoding matrix W.sub.j in the codebook satisfy W.sub.i=D(i,j)W.sub.j, where D(i,j)=.sub.(i,j).Math.diag {.sub.1,.sub.2, . . . ,.sub.n,.sub.m*,.sub.n1*, . . . ,.sub.1*}, .sub.(i,j) is a complex factor a complex number .sub.m* is a conjugate complex number of a complex number .sub.m, m=1, . . . ,n, and n is determined by a quantity of antenna ports.

[0571] Optionally, phases of diagonal elements .sub.1,.sub.2, . . . ,.sub.n of the foregoing matrix D(i,j) form an arithmetic progression.

[0572] FIG. 13 is a fifth schematic structural diagram of an apparatus for determining a precoding matrix indicator according to an embodiment of the present invention. As shown in FIG. 13, the apparatus includes a fifth determining module 51 and a third sending module 52. The fifth determining module 51 is configured to determine a first precoding matrix indicator PMI, where the first PMI corresponds to a precoding matrix W.sub.i in a codebook; and the third sending module 52 is configured to send the first PMI to a base station, where the codebook includes at least: the precoding matrix W.sub.i and a precoding matrix W.sub.j, and the precoding matrix W.sub.i and a precoding matrix W.sub.k in the codebook satisfy D.sub.i.sup.1W.sub.i=D.sub.k.sup.1W.sub.k=V, where D.sub.m=.sub.m.Math.diag {u.sub.m,1u.sub.m,2, . . . ,u.sub.m,nu.sub.m,n*,u.sub.m,n1*, . . . ,u.sub.m,1*}, m=i,k, .sub.m is a complex factor, a complex number u.sub.m,l* is a conjugate complex number of a complex number u.sub.m,l, m=i,k, l=1, . . . ,n, and n is determined by a quantity of antenna ports; and the matrix V is a constant modulus matrix.

[0573] Optionally, phases of diagonal elements u.sub.m,1,u.sub.m,2, . . . ,u.sub.m,n of the foregoing matrix D.sub.m form an arithmetic progression.

[0574] FIG. 14 is a sixth schematic structural diagram of an apparatus for determining a precoding matrix indicator according to an embodiment of the present invention.

[0575] As shown in FIG. 14, the apparatus includes a fourth receiving module 61 and a sixth determining module 62. The fourth receiving module 61 is configured to receive a first precoding matrix indicator PMI sent by user equipment; and the sixth determining module 62 is configured to determine a corresponding precoding matrix W.sub.i from a codebook according to the first PMI, where the codebook includes at least: the precoding matrix W.sub.i and a precoding matrix W.sub.j, and the precoding matrix W.sub.i and a precoding matrix W.sub.k in the codebook satisfy D.sub.i.sup.1W.sub.i=D.sub.k.sup.1W.sub.k=V, where D.sub.m=.sub.m.Math.diag {u.sub.m,1,u.sub.m,2, . . . ,u.sub.m,n,u.sub.m,n*,u.sub.m,n1*, . . . ,u.sub.m,1*}, m=i,k, .sub.m is a complex factor, a complex number u.sub.m,l* is a conjugate complex number of a complex number u.sub.m,l, m=i,k, l=1, . . . ,n, and n is determined by a quantity of antenna ports; and the matrix V is a constant modulus matrix.

[0576] Optionally, phases of diagonal elements u.sub.m,1,u.sub.m,2. . . ,u.sub.m,n of the foregoing matrix D.sub.m form an arithmetic progression.

[0577] FIG. 15 is a first schematic structural diagram of user equipment according to an embodiment of the present invention. As shown in FIG. 14, the user equipment includes a first processor 71 and a first transmitter 72. The first processor 71 is configured to determine a precoding matrix indicator PMI, where the PMI corresponds to a precoding matrix W, and the precoding matrix W satisfies a first condition, a second condition, or a third condition; and the first transmitter 72 is configured to send the PMI to a base station, where the first condition is that the precoding matrix W satisfies W=DV, where the matrix D is a diagonal matrix, D=.Math.diag {u.sub.1,u.sub.2, . . . ,u.sub.n,u.sub.n*,u.sub.n1*, . . . ,u.sub.1*}, is a complex factor, a complex number u.sub.i* is a conjugate complex number of a complex number u.sub.i, and n is determined by a quantity of antenna ports; and the matrix V is a constant modulus matrix; the second condition is that the precoding matrix W includes one or more column vectors of a block diagonal matrix W.sub.1, or the precoding matrix W is obtained by performing weighted combination on one or more column vectors of a block diagonal matrix W.sub.1, where W.sub.1=diag {X.sub.1, . . . ,X.sub.N.sub.B}, and N.sub.B1, where at least one block matrix X is a product X=DV of a matrix D and a matrix V, and X {X.sub.1,X.sub.2, . . . ,X.sub.N.sub.B}; the matrix D is a diagonal matrix, D=.Math.diag {u.sub.1,u.sub.2, . . . ,u.sub.n,u.sub.n*,u.sub.n1*, . . . ,u.sub.1*}, is a complex factor, a complex number u.sub.i* is a conjugate complex number of a complex number u.sub.i, and n is determined by a quantity of antenna ports; and the matrix V is a constant modulus matrix; and the third condition is that the precoding matrix W includes one or more column vectors of a block diagonal matrix W.sub.1, or the precoding matrix W is obtained by performing weighted combination on one or more column vectors of a block diagonal matrix W.sub.1, where W.sub.1=diag {X.sub.1, . . . ,X.sub.N.sub.B}, and N.sub.B1, where at least one block matrix X is a Kronecker product of a matrix A and a matrix B, X=A .Math. B, and X {X.sub.1,X.sub.2, . . . ,X.sub.N.sub.B}; the matrix A or the matrix B is a product of a matrix D and a matrix V; the matrix D is a diagonal matrix, D=.Math.diag {u.sub.1,u.sub.2, . . . ,u.sub.nu.sub.n*,u.sub.n1*, . . . ,u.sub.1*}, is a complex factor, a complex number u.sub.i* is a conjugate complex number of a complex number u.sub.i, i=1, . . . ,n, and n is a quantity of rows of the matrix A or the matrix B; and the matrix V is a constant modulus matrix.

[0578] In the foregoing embodiment of the present invention, in the second condition or the third condition, the precoding matrix W satisfies W=W.sub.1W.sub.2, where the matrix W.sub.2 is used to select one or more column vectors of the matrix W.sub.1; or is used to perform weighted combination on one or more column vectors of the W.sub.1 to obtain the precoding matrix W.

[0579] Optionally, phases of diagonal elements u.sub.1,u.sub.2, . . . ,u.sub.n of the foregoing matrix D form an arithmetic progression.

[0580] In addition, the matrix V includes a column vector 1 and/or at least one column vector v, the column vector 1 is a column vector whose elements are all 1, and the column vector v is v=[v.sub.1 v.sub.2 L v.sub.n v.sub.n v.sub.n1 L v.sub.1].sup.T, where and element is v.sub.i=v.sub.i,v.sub.1=1, and i=1, . . . ,n. Optionally, the matrix V includes only the column vector 1 and the at least one column vector v, and when the matrix V includes multiple column vectors v, the multiple column vectors v are different. The column vector v of the foregoing matrix V is a column vector of a matrix [H.sup.T H.sup.T].sup.T, where a matrix H is a Hadamard matrix.

[0581] In the foregoing embodiment of the present invention, the PMI includes a first index PMI1 and a second index PMI2, where [0582] when the precoding matrix W satisfies the first condition, the first index PMI1 corresponds to the matrix D, and the second index PMI2 corresponds to the matrix V; [0583] when the precoding matrix W satisfies the second condition, the first index PMI1 corresponds to the matrix W.sub.1, and the second index PMI2 corresponds to the matrix W.sub.2; or [0584] when the precoding matrix W satisfies the third condition, the first index PMI1 corresponds to the matrix W.sub.1, and the second index PMI2 corresponds to the matrix W.sub.2.

[0585] Optionally, the foregoing first index PMI1 and second index PMI2 have different time-domain granularities or frequency-domain granularities; or the first index PMI1 and the second index PMI2 are sent to the base station by using different time periods.

[0586] In the foregoing embodiment of the present invention, the user equipment further includes: [0587] a first receiver, configured to receive a reference signal sent by the base station, and select, from a codebook according to the reference signal, the precoding matrix W corresponding to the PMI.

[0588] The foregoing codebook includes the precoding matrix W.sub.1 and a precoding matrix W.sub.j, and the two precoding matrices satisfy W.sub.i=D(i,j)W.sub.j, where D(i,j)=.sub.(i,j)diag {.sub.1,.sub.2, . . . ,.sub.n,.sub.n*,.sub.n1*, . . . ,.sub.1*}, .sub.(i,j) is a complex factor a complex number .sub.n* is a conjugate complex number of a complex number .sub.m, m=1, . . . ,n, and n is determined by a quantity of antenna ports.

[0589] Optionally, phases of diagonal elements .sub.1,.sub.2, . . . ,.sub.n of the matrix D(i,j) form an arithmetic progression.

[0590] Alternatively, the foregoing codebook includes the precoding matrix W.sub.i and a precoding matrix W.sub.k, and the two precoding matrices satisfy D.sub.i.sup.1W.sub.i=D.sub.k.sup.1W.sub.k=V, where D.sub.m=.sub.m.Math.diag {u.sub.m,1,u.sub.m,2, . . . ,u.sub.m,n,u.sub.m,n*,u.sub.m,n1, . . . ,u.sub.m,1*}, m=i,k, .sub.m is a complex factor, a complex number u.sub.m,l* is a conjugate complex number of a complex number u.sub.m,l, m=i,k, l=1, . . . ,n, and n is determined by a quantity of antenna ports.

[0591] Optionally, phases of diagonal elements u.sub.m,1,u.sub.m,2, . . . ,u.sub.m,n of the foregoing matrix D.sub.m form an arithmetic progression.

[0592] FIG. 16 is a first schematic structural diagram of a base station according to an embodiment of the present invention. As shown in FIG. 16, the base station includes a second receiver 81 and a second processor 82. The second receiver 81 is configured to receive a precoding matrix indicator PMI sent by user equipment; and the second processor 82 is configured to determine a corresponding precoding matrix W according to the PMI, where the precoding matrix W satisfies a first condition, a second condition, or a third condition, where [0593] the first condition is that the precoding matrix W satisfies W=DV; [0594] the second condition is that the precoding matrix W includes one or more column vectors of a block diagonal matrix W.sub.1, or the precoding matrix W is obtained by performing weighted combination on one or more column vectors of a block diagonal matrix W.sub.1, where W.sub.1=diag {X.sub.1, . . . ,X.sub.N.sub.B}, and N.sub.B1, where at least one block matrix X is a product X=DV of a matrix D and a matrix V, and X {X.sub.1,X.sub.2, . . . ,X.sub.N.sub.B}; and [0595] the third condition is that the precoding matrix W includes one or more column vectors of a block diagonal matrix W.sub.1, or the precoding matrix W is obtained by performing weighted combination on one or more column vectors of a block diagonal matrix W.sub.1, where W.sub.1=diag {X.sub.1, . . . ,X.sub.N.sub.B}, and N.sub.B1, where at least one block matrix X is a Kronecker product of a matrix A and a matrix B, X=A .Math. B, and X {X.sub.1,X.sub.2, . . . ,X.sub.N.sub.B}; the matrix A or the matrix B is a product of a matrix D and a matrix V; the matrix D is a diagonal matrix, i=1, . . . , n, and n is a quantity of rows of the matrix A or the matrix B, where [0596] the matrix D is a diagonal matrix, D=.Math.diag {u.sub.1,u.sub.2, . . . ,u.sub.n,u.sub.n*,u.sub.n1*, . . . ,u.sub.1*}, is a complex factor, a complex number u.sub.i* is a conjugate complex number of a complex number u.sub.i, and n is determined by a quantity of antenna ports; and the matrix V is a constant modulus matrix.

[0597] In the foregoing embodiment of the present invention, in the second condition or the third condition, the precoding matrix W satisfies W=W.sub.1W.sub.2, where the matrix W.sub.2 is used to select one or more column vectors of the matrix W.sub.1; or is used to perform weighted combination on one or more column vectors of the W.sub.1 to obtain the precoding matrix W.

[0598] Optionally, phases of diagonal elements u.sub.1,u.sub.2, . . . ,u.sub.n of the matrix D form an arithmetic progression.

[0599] In the foregoing embodiment of the present invention, the matrix V includes a column vector 1 and/or at least one column vector v, the column vector 1 is a column vector whose elements are all 1, and the column vector v is v=[v.sub.1 v.sub.2 L v.sub.n v.sub.n v.sub.n1 L v.sub.1].sup.T, where an element is v.sub.i=v.sub.i,v.sub.i=1, and i=1, . . . ,n. Optionally, the matrix V includes only the column vector 1 and the at least one column vector v, and when the matrix V includes multiple column vectors v, the multiple column vectors v are different. In addition, the column vector v of the foregoing matrix V is a column vector of a matrix [H.sup.T H.sup.T].sup.T, where a matrix H is a Hadamard matrix.

[0600] In the foregoing embodiment of the present invention, the precoding matrix indicator PMI includes a first index PMI1 and a second index PMI2, where when the precoding matrix W satisfies the first condition, the first index PMI1 corresponds to the matrix D, and the second index PMI2 corresponds to the matrix V; when the precoding matrix W satisfies the second condition, the first index PMI1 corresponds to the matrix and the second index PMI2 corresponds to the matrix W.sub.2; or when the precoding matrix W satisfies the third condition, the first index PMI1 corresponds to the matrix W.sub.1, and the second index PMI2 corresponds to the matrix W.sub.2.

[0601] Optionally, the first index PMI1 and the second index PMI2 have different time-domain granularities or frequency-domain granularities; or the first index PMI1 and the second index PMI2 are sent to the base station by using different time periods.

[0602] In the foregoing embodiment of the present invention, the determining a corresponding precoding matrix W according to the PMI includes: [0603] selecting the corresponding precoding matrix W from a codebook according to the PMI.

[0604] In addition, the foregoing codebook includes a precoding matrix W.sub.j and a precoding matrix W.sub.j, and the two precoding matrices satisfy W.sub.i=D(i,j)W.sub.j, where D(i,j)=.sub.(i,j)diag {.sub.1,.sub.2, . . . ,.sub.n,.sub.n*,.sub.n1*, . . . ,.sub.1*}, .sub.(i,j) is a complex factor, a complex number .sub.m* is a conjugate complex number of a complex number .sub.m, m=1, . . . ,n, and n is determined by a quantity of antenna ports. Optionally, phases of diagonal elements .sub.1,.sub.2, . . . ,.sub.n of the matrix D(i,j) form an arithmetic progression.

[0605] Alternatively, the codebook includes a precoding matrix W.sub.i and a precoding matrix W.sub.k, and the two precoding matrices satisfy D.sub.i.sup.1W.sub.i=D.sub.k.sup.1W.sub.k=V, where D.sub.m=.sub.m.Math.diag {u.sub.m,1,u.sub.m,2, . . . ,u.sub.m,n,u.sub.m,n*,u.sub.m,n1*, . . . ,u.sub.m,1*}, m=i,k, .sub.m is a complex factor, a complex number .sub.m,l* is a conjugate complex number of a complex number u.sub.m,l, m=i,k, i=1, . . . ,n, and n is determined by a quantity of antenna ports. Optionally, phases of diagonal elements u.sub.m,1,u.sub.m,2, . . . ,u.sub.m,n of the foregoing matrix m form an arithmetic progression.

[0606] FIG. 17 is a second schematic structural diagram of user equipment according to an embodiment of the present invention. As shown in FIG. 17, the user equipment includes a third processor 73 and a second transmitter 74. The third processor 73 is configured to determine a first precoding matrix indicator PMI, where the PMI corresponds to a precoding matrix W.sub.i in a codebook; and the second transmitter 74 is configured to send the first PMI to a base station, where [0607] the codebook includes at least: the precoding matrix W.sub.i and a precoding matrix W.sub.j, and the precoding matrix W.sub.i and the precoding matrix W.sub.j in the codebook satisfy W.sub.i=D(i,j)W.sub.j, where D(i,j)=.sub.(i,j)diag {.sub.1,.sub.2,. . . ,.sub.n,.sub.n*,.sub.n1*, . . . ,.sub.1*}, .sub.(i,j) is a complex factor, a complex number .sub.m* is a conjugate complex number of a complex number .sub.m, m=1, . . . ,n, and n is determined by a quantity of antenna ports.

[0608] Optionally, phases of diagonal elements .sub.1,.sub.2, . . . ,.sub.n of the foregoing matrix D(i,j) form an arithmetic progression.

[0609] FIG. 18 is a second schematic structural diagram of a base station according to an embodiment of the present invention. As shown in FIG. 18, the base station includes a third receiver 81 and a fourth processor 82. The third receiver 81 included in the base station is configured to receive a first precoding matrix indicator PMI sent by user equipment; and the fourth processor 82 is configured to determine a corresponding precoding matrix W.sub.i from a codebook according to the first PMI, where the codebook includes at least: the precoding matrix W.sub.i and a precoding matrix W.sub.j, and the precoding matrix W.sub.i and the precoding matrix W.sub.j in the codebook satisfy W.sub.i=D(i,j)W.sub.j, where D(i,j)=.sub.(i,j)diag {.sub.1,.sub.2, . . . ,.sub.n,.sub.n*, .sub.n1*, . . . ,.sub.1*}, .sub.(i,j) is a complex factor, a complex number .sub.m* is a conjugate complex number of a complex number .sub.m, m=1, . . . ,n, and n is determined by a quantity of antenna ports.

[0610] Optionally, phases of diagonal elements .sub.1,.sub.2, . . . ,.sub.n of the foregoing matrix D(i,j) form an arithmetic progression.

[0611] FIG. 19 is a third schematic structural diagram of user equipment according to an embodiment of the present invention. As shown in FIG. 19, the user equipment includes a fifth processor 75 and a third transmitter 76. The fifth processor 75 is configured to determine a first precoding matrix indicator PMI, where the first PMI corresponds to a precoding matrix W.sub.i in a codebook; and the third transmitter 76 is configured to send the first PMI to a base station, where the codebook includes at least: the precoding matrix W.sub.i and a precoding matrix W.sub.j, and the precoding matrix W.sub.i and a precoding matrix W.sub.k in the codebook satisfy D.sub.i.sup.1W.sub.i=D.sub.k.sup.1W.sub.k=V, where D.sub.m=.sub.m.Math.diag {u.sub.m,1,u.sub.m,2, . . . ,u.sub.m,n,u.sub.m,n*,u.sub.m,n1*, . . . ,u.sub.m,1*}, m=i,l, .sub.m is a complex factor, a complex number u.sub.m,l* is a conjugate complex number of a complex number u.sub.m,l, m=i,k, l=1, . . . ,n, and n is determined by a quantity of antenna ports; and the matrix V is a constant modulus matrix.

[0612] Optionally, phases of diagonal elements u.sub.m,1,u.sub.m,2, . . . ,u.sub.m,n of the foregoing matrix D.sub.m form an arithmetic progression.

[0613] FIG. 20 is a third schematic structural diagram of a base station according to an embodiment of the present invention. As shown in FIG. 20, the base station includes a fourth receiver 85 and a sixth processor 86. The fourth receiver 85 is configured to receive a first precoding matrix indicator PMI sent by user equipment; and the sixth processor 86 is configured to determine a corresponding precoding matrix W.sub.i from a codebook according to the first PMI, where the codebook includes at least: the precoding matrix W.sub.i and a precoding matrix W.sub.j, and the precoding matrix W.sub.i and a precoding matrix W.sub.k in the codebook satisfy D.sub.i.sup.1W.sub.i=D.sub.k.sup.1W.sub.k=V, where D.sub.m=.sub.m.Math.diag {u.sub.m,1,u.sub.m,2, . . . ,u.sub.m,n,u.sub.m,n*,u.sub.m,n1*, . . . ,u.sub.m,1*}, m=i,k, .sub.m is a complex factor, a complex number u.sub.m,l* is a conjugate complex number of a complex number u.sub.m,l, m=i,k, l=1, . . . ,n, and n is determined by a quantity of antenna ports; and the matrix V is a constant modulus matrix.

[0614] Optionally, phases of diagonal elements u.sub.m,1,u.sub.m,2, . . . ,u.sub.m,n of the foregoing matrix D.sub.m form an arithmetic progression.

[0615] A person of ordinary skill in the art may understand that all or some of the steps of the method embodiments may be implemented by a program instructing relevant hardware. The program may be stored in a computer-readable storage medium. When the program runs, the steps of the method embodiments are performed. The foregoing storage medium includes: any medium that can store program code, such as a ROM, a RAM, a magnetic disk, or an optical disc.

[0616] Finally, it should be noted that the foregoing embodiments are merely intended for describing the technical solutions of the present invention, but not for limiting the present invention. Although the present invention is described in detail with reference to the foregoing embodiments, persons of ordinary skill in the art should understand that they may still make modifications to the technical solutions described in the foregoing embodiments or make equivalent replacements to some or all technical features thereof, without departing from the scope of the technical solutions of the embodiments of the present invention.