Rate control technique for layered architectures with multiple transmit and receive antennas
View Patent ↗In a wireless communication system using multiple antennas at the transmitter and multiple antennas at the receiver, with a sufficiently large number of antennas and/or diversity order, the same performance that is obtainable when employing instantaneous rate indication feedback can be attained without any such short-term feedback by adjusting the data rate of each transmit antenna according to a prescribed function. The prescribed function does not depend on the short-term state of the channel, but instead only depends on its long-term statistics, which may be fed back from the receiver to the transmitter at relatively lengthy intervals.
1. A transmitter of a multiple-input multiple-output (MIMO) system for transmitting a plurality of data substreams derived from a data stream, comprising:
means for obtaining a long term parameter; and
a rate controller that develops a rate for each transmit data substream as function of at least said long term parameter and no short term parameters;
wherein said transmitter is a wireless transmitter and each of said substreams is transmitted by a separate antenna over a channel to a receiver having a plurality of receive antennas, and wherein said rate controller develops each rate for each of said transmit data substream using a derivative of asymptotic capacity of said system.
2. The invention as defined in claim 1 wherein said long term parameter is a signal-to-noise ratio (SNR).
3. The invention as defined in claim 1 wherein said long term parameter is supplied by said receiver coupled via a channel to said transmitter.
4. The invention as defined in claim 1 wherein said long term parameter is developed at said transmitter.
5. The invention as defined in claim 1 wherein said long term parameter is characteristic of a channel between said transmitter and said receiver.
6. The invention as defined in claim 1 wherein said rate controller determines an approximation of the optimal rate for each transmit data substream.
7. A transmitter of a multiple-input multiple-output (MIMO) system for transmitting a plurality of data substreams derived from a data stream, comprising
means for obtaining a long term parameter; and
a rate controller that develops a rate for each transmit data substream as function of at least said long term parameter and no short term parameters;
wherein said rate controller determines an approximation of the optimal rate for each transmit data substream and said optimal rate for each transmit data substream is given by
R
n
=
log
2
(
1
+
h
n
H
(
H
n
H
n
H
+
N
σ
2
P
I
)
-
1
h
n
)
where
n is a data substream index;
h n =[h I,n . . . h M,n ] T is the complex M-dimensional channel vector for transmit substream, with h m,n being the channel transfer coefficient between transmit substream n and receiver branch m;
X T indicates the matrix transpose operation for matrix X;
H n =[h n+1 h n+2 . . . h N ], an M-by-(N-n) matrix;
P represents the total radiated power;
X H indicates the Hermitian transpose operation for matrix X;
I the M-by-M identity matrix; and
σ 2 is the noise variance.
8. The invention as defined in claim 1 wherein said rate controller determines said rate for each transmit data substream also as a function of fixed parameters.
9. The invention as defined in claim 1 wherein said rate controller determines said rate for each transmit antenna also as a function of the ratio of transmit data substreams and receive antennas β, where
β
=
N
M
,
N is the number transmit antennas and M is the number of receive antennas.
10. A transmitter of a multiple-input multiple-output (MIMO) system for transmitting a plurality of data substreams derived from a data stream, comprising:
means for obtaining a long term parameter; and
a rate controller that develops a rate for each transmit data substream as function of at least said long term parameter and no short term parameters;
wherein said data substreams are transmitted via a channel impaired by additive white Gaussian noise (AWGN), and said rate controller determines said rates by computing
R
(
b
)
=
log
2
(
1
+
SNR
β
-
1
4
[
1
+
SNR
β
(
1
+
β
-
b
)
2
-
1
+
SNR
β
(
1
-
β
-
b
)
2
]
2
)
where
β
=
N
M
,
N is the number transmit data substreams,
M is the number of receive branches,
SNR is the long-term average signal-to-noise ratio, and
h is n/M, where n is the index of the transmit data substreams.
11. A wireless transmitter of a multiple-input multiple-output (MIMO) system for transmitting wirelessly a plurality of data substreams derived from a data stream, each of said data substreams being transmitted via an antenna of said wireless transmitter, said data substreams being for receipt at a receiver having a plurality of receive antennas, said transmitter comprising
means for obtaining a long term parameter; and
a rate controller that develops a rate for each transmit data substream as function of at least said long term parameter and no short term parameters;
said data substreams being transmitted via a channel impaired by additive white Gaussian noise (AWGN), and said rate controller determines said rates by computing
R
(
b
)
=
log
2
(
1
+
SNR
β
-
1
4
[
1
+
SNR
β
(
1
+
β
-
b
)
2
-
1
+
SNR
β
(
1
-
β
-
b
)
2
]
2
)
where
β
=
N
M
,
N is the number transmit antennas,
M is the number of receive antennas,
SNR is the long-term average signal-to-noise ratio, and
h is n/M, where n is the index of the transmit antennas.
12. A method for use in transmitter of a multiple-input multiple-output (MIMO) system that is transmitting a plurality of data substreams derived from a data stream, the method comprising the steps of:
obtaining a long term parameter; and
determining a rate for each transmit data substream as function of at least said long term parameter and no short term parameters;
wherein said data substreams are transmitted via a channel impaired by additive white Gaussian noise (AWGN), and said rate controller determines said rates by computing
R
(
b
)
=
log
2
(
1
+
SNR
β
-
1
4
[
1
+
SNR
β
(
1
+
β
-
b
)
2
-
1
+
SNR
β
(
1
-
β
-
b
)
2
]
2
)
where
β
=
N
M
,
N is the number transmit data substreams,
M is the number of receive branches,
SNR is the long-term average signal-to-noise ratio, and
h is n/M, where n is the index of the transmit data substreams.