A method of determining precoding matrix and corresponding communication
methods and devices
Technical field
The invention relates to multi-antenna technology, particularly relates
to precoding method in the multi-antenna technology.
Background of the art
According to the agreed way forward for Rel-10 feedback [1], a precoder for
a subband is composed of two matrices, one of which targets wideband and/or
long-term channel properties, denoted by W , and the other targets
frequency-selective and/or short-term channel properties, denoted by W2 .The
eNB determines the W via long-term wideband feedback, and determines the
W2 via short-term narrowband feedback.
For closely-spaced multi-antenna, the precoding scheme could take the
antenna-related characteristic brought by the closely spacing into
consideration, so as to simplify the design of the precoding scheme. The
industry has discussed about the precoding codebook of the closely-spaced
cross-polarized CLA. In Rl-103026, Samsung has proposed their two-stage
feedback approach with W a M r matrix and W2 a MxM square matrix; and
in Rl-101742, Ericsson has proposed their two-stage feedback approach with
W a x 2 matrix and W2 a 2xr matrix, wherein is the amount of the antennas
and r is the amount of the data flows. The performance of both above approaches
under the U- I O scenario, such as data throughput, is not satisfying.
Summary of the invention
The invention proposes a precoding scheme based on a new code book, and
this scheme has better data throughput.
According to a first aspect of the invention, it is proposed a method for
determining precoding matrix for sub-band precoding, wherein the
transmitter has two set of cross-polarized antennas, the channel vectors
of the two set of antennas with different polarization direction are
and H 2 respectively, these antennas are closely-spaced, is the amounts
of the transmitting antennas, r is the amount of the data flows, the channel
vectors and H 2 satisfy the following complex relationship:
Wherein is the amplitude of the complex relationship between the two
channel vector, and is the angle of the complex relationship between
the two channel vector;
the method comprises the following steps:
a . determining, according to long-term channel related information, an
optimized first matrix W with a dimension of MxM, the first matrix
corresponding to wideband and/or long-term channel properties, and the first
matrix W , is selected from first codebook below:
Wherein, R is a proximity of the transmitting space correlation matrix of
_ ~ -j(M 2 -\)-
H , j — 1 ,6 ,...,6 J is a dominant eigenvector of the transmitting
space correlation matrix of , proximated by DFT vector, is the angle
difference between the two neighboring item of the DFT vector;
b . multiplying the optimized first matrix W with each second matrix W 2
with a dimension of Mxr in a second codebook, so as to obtain a plurality
of candidate precoding matrices , wherein the second matrix W 2 corresponds
to frequency-selective and/or short-term channel properties and is selected
from second codebook below:
w =(w w ,...w/)
wherein, W 2' is the i-th column of W 2 , letting n = — 1) /2 + 1, the n-th
element of W 2' is 1 , and the (n+M/2)-th element is ' when mod(i,2)=l,
is 1/ ~ when mod ( ,2)=0 and other elements are 0 ;
c . selecting an optimized one from the plurality of candidate precoding
matrices for precoding the data to be transmitted based on predefined rules,
according to the short-term channel properties.
According to a second aspect of the invention, it is proposed a method,
in a user equipment, for feeding precoding matrix for sub-band precoding
back to an eNodeB, comprising the steps of :
- determining said optimized first matrix W and precoding matrix,
by using a method according to the aforesaid first aspect of the invention;
- determining an optimized second matrix W 2 corresponding to said
optimized precoding matrix;
- providing, for the eNodeB, identifications of said optimized first
matrix W and said second matrix W 2 .
Correspondingly, the invention also proposes a device, in a user equipment,
for feeding precoding matrix for sub-band precoding back to an eNodeB,
comprising :
- a determining means, for determining said optimized first matrix
W and precoding matrix, by using a method according to the aforesaid
first aspect of the invention, and determining an optimized second matrix
W 2 corresponding to said optimized precoding matrix;
- a sender, for providing, for the eNodeB, identifications of said
optimized first matrix W and said second matrix W 2 .
According to the third aspect of the invention, it is proposed a method,
in an eNodeB, for precoding data, comprising:
- receiving identifications of an optimized first matrix W and second
matrix W 2 , fed back by a user equipment;
- determining, from the first codebook and the second code book in a
method according to the aforesaid first aspect of the invention, the
optimized first matrix W and second matrix W 2 , according to the
identifications;
- multiplying the optimized first matrix W and second matrix W 2,
and obtaining optimized precoding matrix;
- precoding data to be transmitted by using the optimized precoding
matrix, so as to transmit the data to the user equipment.
Correspondingly, the invention also proposes a device, in an eNodeB, for
precoding data, comprising:
- a receiver, for receiving identifications of an optimized first matrix
W and second matrix W 2, fed back by a user equipment;
- a inquiring means, for determining, from the first codebook and the
second code book in a method according to the aforesaid first aspect of
the invention, the optimized first matrix W and second matrix W 2, from
the identifications;
- a calculating means, for multiplying the optimized first matrix W
and second matrix W , and obtaining optimized precoding matrix;
- a precoder, for precoding data to be transmitted by using the optimized
precoding matrix, so as to transmit the data to the user equipment.
The above and other features will be elucidated in the following detailed
embodiments, or become obvious from the following detailed embodiments.
Detailed embodiments
Firstly, the design of codebook according to the invention will be elucidated
as following.
For two groups of cross- polarized antennas, by observation, the inventors
find that there exists a fast-varying complex vector relationship between
the two antenna groups, as described by the following formula:
Wherein is the channel vector representing one antenna group on one
polarization and H 2 is the channel vector representing the other antenna group
on the other polarization, is the amplitude of the complex relationship
between the two channel vector, and is the angle of the complex relationship
between the two channel vectors . can be treated as the long-term wideband
static channel. This observation agrees with the theoretical analysis on
cross-polarized antennas in reference document ( (L . Jiang, L . Thiele, and
V . Jungnickel, "On the Modelling of Polarized MIMO Channel ," 13th European
Wireless Conference, Paris, France, Apr. 2007).
For closely-spaced antennas, the dominant eigenvector in , , can be
approximated by a DFT vector:
= [1,£ -j0 -j(MI2-\)0- ' ',...,£ ' ] (2 )
For single layer (rank) (the amount of data flows is 1), the precodi
matrix for sub-band precoding is described as the following formula
Wherein
And the precoding matrix W 2 =[l,0 , ,0]T with a dimension of Mxl .
For closely-spaced cross-polarized CLA antennas, W can be treated as
a channel correlation matrix of the two groups of cross-polarized antennas,
and R can be treated as the channel correlation matrix of the group of
antennas corresponding to .
For two layers (the amount of data flows is 2), W doesn't change, and
W2 is determined according to the following formula:
Generally, for the case that the amount of data flows is r ,
W = W2
1,W ,...W ) wherein, W2' is the i-th column of W2 , letting
n = |_(i-l) / 2 + l , the n-th element of W2' is 1 , and the (n+M/2)-th element
is ' when mod( ,2) = l , is 1/ ~' when mod(j,2) = 0 and other elements are
0.
For example, in an embodiment, for transmitting antennas and maximum
2 layers, the 16 codewords 2 in the 4-bit second codebook W2 are determined
by the formula in the following table 1:
wherein ¾ =0.5 ,1= 1,2 /2 , 2 = 0 , =
The 256 codewords 1 in the 8-bit first codebook are determined by the
formula in the following table 2 :
Table 2
Serial No. Codeword
i = 0,...,2 - l
m n
re r M I - j(.MI2-V n
re
and R „ =
re
M/2-1 - ( /2- l)¾ re
m [i/2 J , and n = mod(j,26)
wherein , ¾ 0.25,O — 0 . 2 , — 2 r = 1 and
= /32, = 0,...,2 - l a
The selected values of the amplitude and are used for covering the
amplitude of the complex relationship falling into a certain range, and
the selected values of the angle difference are used for covering
the cases that the angle difference falls into difference location ranged
from 0-2n, in order to evenly and completely provide a plurality of candidate
complex relationships for selecting the nearest first matrix and second
matrix according to the channel status information. It should be noted,
in case that the bit number of the first codebook and second codebook,
namely the amount of the first matrices and the second matrices, change,
the values of the amplitude and angle difference can be adjusted
correspondingly, so as to evenly and completely provide a plurality of
candidate complex relationships.
The codewords determined according to the invention is described by the
above embodiment. It should be noted that the invention is not limited
by this set of codeword, and those skilled in the art could determine other
suitable codeword based on the disclosure in the invention.
The following part will describe the embodiment of carrying out precoding
communication based on the codewords determined by the invention.
Firstly, the user equipment (UE) measures long-term channel properties,
and selects, from the first codebook such as the codebook in table 2 , an optimized
first matrix W with a dimension of Mx , according to the long-term channel
properties. It is well known for those skilled in the art that how to select
a corresponding optimized codeword according to the long-term channel
properties, and the description will not give unnecessary detail.
After that, to multiply the first matrix W with each second matrix W 2
with a dimension of Mxr in a second codebook, so as to obtain a plurality
of candidate precoding matrices .The second codebook is such as the codebook
as shown in table 1 .
Then, the UE selects an optimized precoding matrix from the plurality of
candidate precoding matrices based on predefined rules, according to the
measured short-term channel properties. The predefined rules are such as
to select a candidate precoding matrix enabling a maximum channel capacity .
It should be noted that other rules are also applicable, and the description
will not give unnecessary details.
Then, the UE can determine an optimized second matrix W 2 corresponding
to the optimized precoding matrix.
At last, the UE provides, for the eNodeB, identifications of the optimized
first matrix W and the second matrix W 2 . The identifications are such
as the serial numbers of the matrix in the codebook.
On the side of the eNodeB, it receives identifications of the optimized
first matrix W and second matrix W 2, fed back by user equipments.
Then, the eNodeB determines, from the first codebook and the second code
book, the optimized first matrix W and second matrix W , according to
the identifications.
After that, the eNodeB multiplies the optimized first matrix W and second
matrix W , and obtains an optimized precoding matrix for sub-band
precoding .
At last, the eNodeB precodes data to be transmitted by using the optimized
precoding matrix, so as to transmit the data to the user equipment.
The inventor simulates the performance of the codebook proposed in the
invention, and compares it with the simulation results of the schemes
proposed in the prior art. The following table 3 shows the assumed wireless
network environment, based on which the simulation is carried out
Table 3
The simulation result is shown by the following table 4
Table 4
Average throughput Cell edge
Rl-101742 2.8 (100%) 0.08 (100%)
Rl-103026 2.3 (82.1%) 0.10 (125%)
The invention 3.0 (107.1%) 0.11 (137.5%)
It can be seen that the invention achieve better performance than the art.
Those ordinary skilled in the art could understand and realize modifications
to the disclosed embodiments, through studying the description, drawings
and appended claims. The word "comprising" does not exclude the presence
of elements or steps not listed in a claim or in the description. The word
"a" or "an" preceding an element does not exclude the presence of a plurality
of such elements. In the practice of present invention, several technical
features in the claim can be embodied by one component. In the claims,
any reference signs placed between parentheses shall not be construed as
limiting the claim.
What is claimed is:
1 .A method for determining precoding matrix for sub-band precoding, wherein
the transmitter has two sets of cross-polarized antennas, the channel
vectors of the two sets of antennas with different polarization direction
are and H 2 respectively ,these antennas are closely spaced, is the amount
of the transmitting antennas, r is the amount of the data flows, the channel
vectors and H 2 of two sets of antennas satisfy the following complex
relationship :
Wherein is the amplitude of the complex relationship between the two
channel vectors, and is the angle of the complex relationship between
the two channel vectors;
the method comprises the following steps:
a . determining, according to long-term channel related information, an
optimized first matrix W with a dimension of MxM, the first matrix
corresponding to wideband and/or long-term channel properties, and the first
matrix W , is selected from a first codebook below:
wherein, R is a proximity of the transmitting space correlation matrix of
H , j is a dominant eigenvector of the transmitting
space correlation matrix of , proximated by DFT vector, is the angle
difference between the two neighboring items of the DFT vector;
b . multiplying the optimized first matrix W with each second matrix W 2
with a dimension of Mxr in a second codebook, so as to obtain a plurality
of candidate precoding matrices , wherein the second matrix W 2 corresponds
to frequency-selective and/or short-term channel properties and is selected
from second codebook below:
w = (w w ,...w )
r
wherein, W2 is the i-th column of W2 , letting n = — 1) /2 + 1, the n-th
element of W2' is 1 , and the (n+M/2)-th element is ' when mod (i,2)=l,
is 1/ ~ when mod (i,2)=0 and other elements are 0 ;
c . selecting an optimized one from the plurality of candidate precoding
matrices for precoding the data to be transmitted based on predefined rules,
according to the short-term channel properties.
2 . A method as claimed in claim 1 , wherein, for the second codebook W2,
when the amount of data flows is 1 ,
W2 = [1,0, ^,0]
and when the amount of data flows is 2 ,
3 . A method as claimed in claim 2 , wherein, for transmitting antennas
and maximum 2 ranks, the codeword in the 4-bit second codebook W2 are
determined by the following formula:
wherein '= 0 ...,2 4 - l III = i l l 2 n = m od( ,22)
the codeword 8-bit first codebook iare determined by the following
formula :
and
4 . A method as claimed in claim 1 , wherein the mode of the correlation
factor of the closely-spaced antennas is greater than a predetermined
threshold, or the distance between the antennas is equal to or less than
half of the wavelength of the signal.
5 .A method, in a user equipment, for feeding precoding matrix for sub-band
precoding back to an eNodeB, comprising the steps of :
- determining said optimized first matrix W and precoding matrix,
by using a method as claimed in any one of claims 1 to 4 ;
- determining an optimized second matrix W 2 corresponding to said
optimized precoding matrix;
- providing, for the eNodeB, identifications of said optimized first
matrix W and said second matrix W 2 .
6 . A method, in an eNodeB, for precoding data, comprising:
- receiving identifications of an optimized first matrix W and an
optimized second matrix W 2 , fed back by a user equipment;
- determining, from the first codebook and the second code book in a
method according to any one of claims 1 to 4 , the optimized first matrix
W and second matrix W 2 , according to the identifications;
- multiplying the optimized first matrix W and second matrix W 2 ,
and obtaining an optimized precoding matrix;
- precoding data to be transmitted by using the optimized precoding
matrix, so as to transmit the data to the user equipment.
7.A device, in a user equipment, for feeding precoding matrix for sub-band
precoding back to an eNodeB, comprising:
- a determining means, for determining an optimized first matrix W
and a precoding matrix, by using a method as claimed in any one of claims
1 to 4 , and determining an optimized second matrix W 2 corresponding to
said optimized precoding matrix;
- a sender, for providing, for the eNodeB, identifications of said
optimized first matrix W and said second matrix W 2 .
8 . A device, in an eNodeB, for precoding data, comprsising:
- a receiver, for receiving identifications of an optimized first matrix
W and an optimized second matrix W 2, fed back by a user equipment;
- a inquiring means, for determining, from the first codebook and the
second code book in a method according to any one of claims 1 to 4 , the
optimized first matrix W and second matrix W 2, from the identifications ;
- a calculating means, for multiplying the optimized first matrix W
and second matrix W , and obtaining an optimized precoding matrix;
- a precoder, for precoding data to be transmitted by using the optimized
precoding matrix, so as to transmit the data to the user equipment.