Method of transmitting feedback information in a MIMO communication system and device thereof
View Patent ↗The present invention provides a method of providing feedback information in a user equipment of a communication system based on MIMO, comprising: determining a first latest measured long-term covariance matrix according to a downlink channel matrix model; transmitting the quantization of a first latest estimated long-term covariance matrix corresponding to the first latest measured long-term covariance matrix to the base station; determining a second latest measured long-term covariance matrix according to a downlink channel matrix model; determining the variation information of a second latest estimated long-term covariance matrix with respect to an estimated long-term covariance matrix of the last-time reset point; and transmitting the quantization of the variation information to the base station. With such a solution, overhead during the feedback of long-term covariance matrix is greatly reduced in MIMO scenario.
1. A method of providing feedback information to a base station in a user equipment of a communication system based on MIMO, the method comprising:
determining a first latest measured long-term covariance matrix according to a downlink channel matrix model;
transmitting a quantization of a first latest estimated long-term covariance matrix corresponding to the first latest measured long-term covariance matrix to the base station;
determining a second latest measured long-term covariance matrix according to a downlink channel matrix model;
determining variation information of a second latest estimated long-term covariance matrix with respect to an estimated long-term covariance matrix of a last-time reset point; wherein, the second latest estimated long-term covariance matrix is derived according to the variation information and the estimated long-term covariance matrix of the last-time reset point approximates the second latest measured long-term covariance matrix; and
transmitting a quantization of the variation information to the base station.
2. The method according to claim 1 , wherein the determining a second latest measured long-term covariance matrix is repeated for a plurality of times.
3. The method according to claim 2 , wherein the number of repetitions of the determining a second latest measured long-term covariance matrix is a predetermined threshold.
4. The method according to claim 1 , wherein determining a first latest measured long-term covariance matrix comprises:
determining the first latest measured long-term covariance matrix R (t) according to the following formula,
R
_
(
t
)
=
(
1
-
α
)
R
_
(
t
-
1
)
+
α
1
S
∑
f
∈
S
H
(
t
,
f
)
H
H
(
t
,
f
)
wherein, S denotes a set of subcarriers in the whole bands of the channel, |S |denotes cardinality of S, f denotes a subcarrier in the S, t denotes a subframe, H (t,f) denotes a downlink channel matrix based on the t and f,H (t,f) H denotes a conjugate transpose of the H (t,f), and α is an averaging factor.
5. The method according to claim 1 , wherein, the quantization of the variation information in determining a second latest measured long-term covariance matrix is index information of a codeword to which the variation information corresponds in a variation information codebook, and the quantization of the first latest estimated long-term covariance matrix in determining a first latest measured long-term covariance matrix is the index of a codeword to which the first latest estimated long-term covariance matrix corresponds in a long-term covariance matrix codebook.
6. The method according to claim 1 , wherein the determining a second latest measured long-term covariance matrix comprises:
determining the variation information according to the following formula,
W
=
arg
max
trace
(
R
_
(
t
+
nT
)
R
_
^
(
t
+
nT
)
H
)
R
_
(
t
+
nT
)
F
R
_
^
(
t
+
nT
)
F
,
{circumflex over (R)} ( t+nT )= W {circumflex over (R)} ( t ) W H
wherein, W denotes a codeword satisfying the above formula in the variation information codebook, W H denotes a conjugate transpose matrix of the codeword W, the t denotes the last-time reset point, T denotes a period of a long-term feedback, the t+nT denotes occurrence point of the second latest measurement, W denotes variation information of the estimated long-term covariance matrix {circumflex over (R)} (t+nT) in the t+nT with respect to the estimated long-term covariance matrix {circumflex over (R)} (t) in the t, ∥(•)∥ F denotes operation of normalization.
7. A method of providing feedback information to a base station in a user equipment of a communication system based on MIMO, the method comprising:
repeating method of claim 1 .
8. A method of processing feedback information from a user equipment in a base station of a communication system based on MIMO, the method comprising:
receiving the feedback information from the user equipment; and
determining a latest estimated long-term covariance matrix according to the feedback information, wherein when the feedback information comprises index information of a codeword to which variation information corresponds in a variation information codebook, determining a latest estimated long-term covariance matrix comprises at least:
determining a corresponding codeword in a variation information codebook according to index information, and
determining the latest estimated long-term covariance matrix according to the codeword to which the index information corresponds and an estimated long-term covariance matrix of the last-time reset point, wherein the variation information is used for reflecting the variation of the latest estimated long-term covariance matrix with respect to the estimated long-term covariance matrix of the last-time reset point.
9. The method according to claim 8 , wherein, when the feedback information comprises the index information of a codeword to which the latest estimated long-term covariance matrix corresponds in a long-term covariance matrix codebook, determining a latest estimated long-term covariance matrix comprises:
determining the latest estimated long-term covariance matrix from the long-term covariance matrix codebook according to the index information.
10. The method according to claim 8 , wherein, the step of determining the latest estimated long-term covariance matrix in determining a latest estimated long-term covariance matrix further comprises:
determining the latest estimated long-term covariance matrix according to the following formula,
{circumflex over (R)} ( t+nT )= W {circumflex over (R)} ( t ) W H
wherein t denotes the last-time reset point, n denotes a positive integer, T denotes a period of a long-term feedback, {circumflex over (R)} (t) denotes the latest estimated long-term covariance matrix, {circumflex over (R)} (t) denotes the estimated long-term covariance matrix of the last-time reset point, W denotes a codeword to which the variation information corresponds in a variation information codebook, and W H denotes a conjugate transpose matrix of the codeword W.
11. A device for providing feedback information to a base station in a user equipment of a communication system based on MIMO, the device comprising:
a first determining unit, configured to determine a first latest measured long-term covariance matrix according to a downlink channel matrix;
a first transmitting unit, configured to transmit a quantization of a first latest estimated long-term covariance matrix corresponding to the first latest measured long-term covariance matrix to the base station;
a second determining unit, configured to determine a second latest measured long-term covariance matrix according to a downlink channel matrix model;
a variation information determining unit, configured to determine variation information of a second latest estimated long-term covariance matrix with respect to an estimated long-term covariance matrix of a last-time reset point; wherein, the second latest estimated long-term covariance matrix derived according to the variation information and the estimated long-term covariance matrix of the last-time reset point approximates the second latest measured long-term covariance matrix; and
a variation information transmitting unit, configured to transmit the quantization of the variation information to the base station.
12. The device according to claim 11 , wherein, the quantization of the variation information in the variation information transmitting unit is index information of a codeword to which the variation information corresponds in a variation information codebook, and the quantization of the first latest estimated long-term covariance matrix in the first transmitting unit of latest measured long-term covariance matrix is the index of a codeword to which the first latest estimated long-term covariance matrix in a long-term covariance matrix codebook, and the variation information determining unit is configured to determine the variation information according to the following formula,
W
=
arg
max
trace
(
R
_
(
t
+
nT
)
R
_
^
(
t
+
nT
)
H
)
R
_
(
t
+
nT
)
F
R
_
^
(
t
+
nT
)
F
,
{circumflex over (R)} ( t+nT )= W {circumflex over (R)} ( t ) W H
wherein, W denotes the codeword satisfying the above formula in the variation information codebook, W H denotes a conjugate transpose matrix of the codeword W, t denotes the last-time reset point, T denotes the period of a long-term feedback, t+nT denotes occurrence point of the second latest measurement, W denotes the variation information of the estimated long-term covariance matrix {circumflex over (R)} (t+nT) in t+nT with respect to the estimated long-term covariance matrix {circumflex over (R)} (t) in t, and ∥(•)∥ F denotes operation of normalization.
13. A device for processing feedback information from a user equipment in a base station of a communication system based on MIMO, the device comprising:
a feedback information receiving unit, configured to receive the feedback information from the user equipment;
a determining unit of latest estimated long-term covariance matrix, configured to determine a latest estimated long-term covariance matrix according to the feedback information, wherein when the feedback information comprises index information of a codeword to which variation information corresponds in a variation information codebook, determining a latest estimated long-term covariance matrix comprises at least:
determining a corresponding codeword in a variation information codebook according to the index information, and
determining the latest estimated long-term covariance matrix according to the codeword to which the index information corresponds and an estimated long-term covariance matrix of the last-time reset point, wherein the variation information is used for reflecting the variation of the latest estimated long-term covariance matrix with respect to the estimated long-term covariance matrix of the last-time reset point.
14. The device according to claim 13 ,
wherein, when the feedback information comprises the index information of a codeword to which the latest estimated long-term covariance matrix corresponds in a long-term covariance matrix codebook, the determining unit of latest estimated long-term covariance matrix is further configured to determine the latest estimated long-term covariance matrix from the long-term covariance matrix codebook according to the index information.