Uplink power allocation method, uplink power allocation device and uplink power allocation system
An uplink power allocation method, an uplink power allocation device and an uplink power allocation system are provided. The uplink power allocation method includes: determining users to be paired, based on locations of the users within the cell; determining Patten Division Multiple Access (PDMA) pattern matrixes, allocating different PDMA pattern vector groups for the users respectively, and obtaining a power control pattern matrix based on the PDMA pattern vector groups; and determining power control factors of the users in each pair based on the power control pattern matrix, and determining an uplink transmission power of each user based on the power control factors.
1. An uplink power control method, applied to a base station forming at least one cell, wherein the method comprises:
determining users to be paired, based on locations of the users within the cell;
determining Pattern Division Multiple Access (PDMA) pattern matrixes, allocating different PDMA pattern vector groups for the users respectively, and obtaining a power control pattern matrix based on the PDMA pattern vector groups; and
determining power control factors of the users in each pair based on the power control pattern matrix, and determining an uplink transmission power of each user based on the power control factors;
wherein the determining PDMA pattern matrixes, allocating different PDMA pattern vector groups for the users respectively, and obtaining a power control pattern matrix based on the PDMA pattern vector groups comprises:
selecting a PDMA pattern matrix H PDMA (M*N) for an uplink based on a total number of the users within the cell and total resources of a system, wherein M is a number of rows of the PDMA pattern matrix H PDMA (M*N) and N is a number of columns of the PDMA pattern matrix H PDMA (M*N);
calculating the PDMA pattern vector group for each user based on the PDMA pattern matrix H PDMA (M*N) and a power control factor matrix
[
α
11
1
L
α
1
N
1
1
L
α
11
K
L
α
1
N
K
K
α
21
1
L
α
2
N
1
1
L
α
21
K
L
α
2
N
K
K
M
O
M
L
M
O
M
α
M
1
1
L
α
MN
1
1
L
α
M
1
K
L
α
MN
K
K
]
,
to generate the power control pattern matrix
A
=
[
α
11
1
L
α
1
N
1
1
L
α
11
K
L
α
1
N
K
K
α
21
1
L
α
2
N
1
1
L
α
21
K
L
α
2
N
K
K
M
O
M
L
M
O
M
α
M
1
1
L
α
MN
1
1
L
α
M
1
K
L
α
MN
K
K
]
e
H
PDMA
=
[
A
1
,
A
2
,
L
,
A
K
]
,
wherein
K represents a number of uplink paired user, N 1 , N 2 , . . . , N k represent respectively the numbers of columns of the PDMA pattern matrix H PDMA (M*N) occupied by the users 1 , 2 , . . . , K, wherein N=N 1 +N 2 +L+N K , A 1 , A 2 , . . . , A K represent respectively power control pattern vector groups for the users 1 , 2 , . . . , K, and □ represents a dot matrixing;
wherein the determining power control factors of the users in each pair based on the power control pattern matrix comprises:
selecting an uplink power control optimization mode;
adjusting a current combination of the PDMA pattern vector group and the power control factor of each user, to enable a Proportional Fair (PF) weighted sum throughput to satisfy the formula
A
opt
=
argmax
A
{
∑
k
=
1
K
T
k
(
SINR
k
,
A
)
·
(
1
-
BLER
k
(
SINR
k
,
A
)
)
R
_
k
(
t
)
}
and ∥A∥ 1 =1 , and to maximize the PF weighted sum throughput per power consumption unit in the case that the users are paired, wherein
T k (SINR k,A ) represents a transmission rate of a user k in the case that the power control pattern matrix of the user k is A, calculated based on transmission bit numbers mapped from a Signal to Interference plus Noise Ratio (SINR) of the user k and resources occupied by the user k, a value of the SINR of the user k is related to the power control factors in the PDMA pattern matrix, BLER k (SINR k,A ) represents an estimated Block Error Rate (BLER) in the case that the power control pattern matrix of the user k is A, calculated based on a mapping of the SINR and the BLER, R k (t) represents a historical average rate of the user k at a time point t, P t (∥A k ∥ 1 ) represents total uplink transmission power in the case that the power control pattern matrix of the user k is A and P t (∥A k ∥ 1 ) is smaller than or equal to maximum uplink transmission power P max of a user terminal, and ∥·∥ 1 represents summing up matrix elements;
the historical average rate is calculated based on the formula R k (t)=α·R k (t)+(1−α)· R k (t−1), wherein R k (t) represents the historical average rate of the user k at a current time point t, R k (t−1) represents the historical average rate of the user k at a previous time point t−1, R k (t) represents a current rate of the user k, α represents a forgetting factor, and the PF weighted sum throughput is related to the current rate of the user k in the case that α=1.
2. The method according to claim 1 , wherein the determining power control factors of the users in each pair based on the power control pattern matrix comprises:
selecting an uplink power control simplification mode;
determining whether a first simplification mode or a second simplification mode for determining the power control pattern matrix is used, based on a ratio of the uplink transmission power of the user to an uplink transmission power of an Orthogonal Frequency Division Multiple Access (OFDMA) user, wherein
in the first simplification mode, the uplink transmission power of the user is equal to the uplink transmission power of the OFDMA user, a smallest unit of resources occupied by the user is one PDMA pattern vector group and one PDMA pattern vector group is permitted to occupy a plurality of Physical Resource Blocks (PRBs), and a smallest unit of resources occupied by the OFDMA user is one PRB;
in the second simplification mode, the uplink transmission power of the user is a multiple of a number of rows of the PDMA pattern vector group of the uplink transmission power of the OFDMA user, and the power control pattern matrix enables a transmission power of the smallest unit of the resources occupied by the PDMA user to be equal to a multiple of the number of rows of the PDMA pattern vector group of a transmission power of the smallest unit of the resources occupied by the OFDMA user.
3. The method according to claim 1 , wherein subsequent to the determining users needing to be paired, the method further comprises:
determining uplink transmission power of each user per PRB unit, through an uplink power control.
4. An uplink power control device, comprising a processor, a memory and a transceiver, wherein the processor is configured to read a program stored in the memory to:
determine users to be paired, based on locations of the users within the cell;
determine Pattern Division Multiple Access (PDMA) pattern matrixes, allocate different PDMA pattern vector groups for the users respectively, and obtain a power control pattern matrix based on the PDMA pattern vector groups; and
determine power control factors of the users, and determine an uplink transmission power of each user:
the transceiver is configured to receive and transmit data;
the processor is configured to perform a general processing; and
the memory is configured to store data for the operation of the processor;
wherein the processor is further configured to read a program stored in the memory to:
determine a PDMA pattern matrix H PDMA (M*N) based on a total number of the users within the cell and total resources of a system, wherein M is a number of rows of the PDMA pattern matrix H PDMA (M*N) and N is a number of columns of the PDMA pattern matrix H PDMA (M*N);
calculate the PDMA pattern vector group for each user based on the PDMA pattern matrix H PDMA (M*N) and a power control factor matrix
[
α
11
1
L
α
1
N
1
1
L
α
11
K
L
α
1
N
K
K
α
21
1
L
α
2
N
1
1
L
α
21
K
L
α
2
N
K
K
M
O
M
L
M
O
M
α
M
1
1
L
α
MN
1
1
L
α
M
1
K
L
α
MN
K
K
]
,
to generate the power control pattern matrix
A
=
[
α
11
1
L
α
1
N
1
1
L
α
11
K
L
α
1
N
K
K
α
21
1
L
α
2
N
1
1
L
α
21
K
L
α
2
N
K
K
M
O
M
L
M
O
M
α
M
1
1
L
α
MN
1
1
L
α
M
1
K
L
α
MN
K
K
]
e
H
PDMA
=
[
A
1
,
A
2
,
L
,
A
K
]
,
wherein
K represents a number of uplink paired user, N 1 , N 2 , . . . , N k represent respectively the numbers of columns of the PDMA pattern matrix H PDMA (M*N) occupied by the users 1 , 2 , . . . , K, wherein N=N 1 +N 2 +L+N K , A 1 , A 2 , . . . , A K represent respectively power control pattern vector groups for the users 1 , 2 , . . . , K, and □ represents a dot matrixing;
wherein the processor is further configured to read a program stored in the memory to:
select an uplink power control optimization mode;
adjust a current combination of the PDMA pattern vector group and the power control factor of each user, to enable a Proportional Fair (PF) weighted sum throughput to satisfy the formula
A
opt
=
argmax
A
{
∑
k
=
1
K
T
k
(
SINR
k
,
A
)
·
(
1
-
BLER
k
(
SINR
k
,
A
)
)
R
_
k
(
t
)
}
and ∥A∥ 1 =1, and to maximize the PF weighted sum throughput per power consumption unit in the case that the users are paired, wherein
T k (SINR k,A ) represents a transmission rate of a user k in the case that the power control pattern matrix of the user k is A, calculated based on transmission bit numbers mapped from a Signal to Interference plus Noise Ratio (SINR) of the user k and resources occupied by the user k, a value of the SINR of the user k is related to the power control factors in the PDMA pattern matrix, BLER k (SINR k,A ) represents an estimated Block Error Rate (BLER) in the case that the power control pattern matrix of the user k is A, calculated based on a mapping of the SINR and the BLER, R k (t) represents a historical average rate of the user k at a time point t, P t (∥A k ∥ 1 ) represents total uplink transmission power in the case that the power control pattern matrix of the user k is A and P t (∥A k ∥ 1 ) is smaller than or equal to maximum uplink transmission power P max of a user terminal, and ∥·∥ 1 represents summing up matrix elements;
the historical average rate is calculated based on the formula R k (t)=α·R k (t)+(1−α)· R k (t−1), wherein R k (t) represents the historical average rate of the user k at a current time point t, R k (t−1) represents the historical average rate of the user k at a previous time point t−1, R k (t) represents a current rate of the user k, α represents a forgetting factor, and the PF weighted sum throughput is related to the current rate of the user k in the case that α=1.
5. The device according to claim 4 , wherein the processor is further configured to read a program stored in the memory to select an uplink power control simplification mode, and determine whether a first simplification mode or a second simplification mode for determining the power control pattern matrix is used, based on a ratio of the uplink transmission power of the user to an uplink transmission power of an Orthogonal Frequency Division Multiple Access (OFDMA) user, wherein
in the first simplification mode, the uplink transmission power of the user is equal to the uplink transmission power of the OFDMA user, a smallest unit of resources occupied by the user is one PDMA pattern vector group and one PDMA pattern vector group is permitted to occupy a plurality of Physical Resource Blocks (PRBs), and a smallest unit of resources occupied by the OFDMA user is one PRB;
in the second simplification mode, the uplink transmission power of the user is a multiple of a number of rows of the PDMA pattern vector group of the uplink transmission power of the OFDMA user, and the power control pattern matrix enables a transmission power of the smallest unit of the resources occupied by the PDMA user to be equal to a multiple of the number of rows of the PDMA pattern vector group of a transmission power of the smallest unit of the resources occupied by the OFDMA user.
6. The device according to claim 4 , wherein the processor is further configured to read a program stored in the memory to determine uplink transmission power of each user per PRB unit, through an uplink power control.
7. An uplink power allocation system, comprising a base station and a receiver, wherein the base station comprises an uplink power allocation device, the uplink power allocation device comprises a processor, a memory and a transceiver, wherein the processor is configured to read a program stored in the memory to:
determine different kinds of users to be paired, based on locations of the users within the cell;
determine Pattern Division Multiple Access (PDMA) pattern matrixes, allocate different PDMA pattern vector groups for the users respectively, and obtain a power control pattern matrix based on the PDMA pattern vector groups; and
determine power control factors of the users, and determine an uplink transmission power of each user;
the transceiver is configured to receive and transmit data;
the processor is configured to perform a general processing; and
the memory is configured to store data for the operation of the processor;
wherein the receiver comprises a processor, a memory and a transceiver, wherein the processor of the receiver is configured to read a program stored in the memory of the receiver to cancel the users demodulated correctly through a successive interference cancellation and detect the other users;
the transceiver of the receiver is configured to receive and transmit data;
the processor of the receiver is configured to perform a general processing; and
the memory of the receiver is configured to store data for the operation of the processor of the receiver;
wherein the processor is further configured to read a program stored in the memory to:
determine a PDMA pattern matrix H PDMA (M*N) based on a total number of the users within the cell and total resources of a system, wherein M is a number of rows of the PDMA pattern matrix H PDMA (M*N) and N is a number of columns of the PDMA pattern matrix H PDMA (M*N);
calculate the PDMA pattern vector group for each user based on the PDMA pattern matrix H PDMA (M*N) a power control factor matrix
[
α
11
1
L
α
1
N
1
1
L
α
11
K
L
α
1
N
K
K
α
21
1
L
α
2
N
1
1
L
α
21
K
L
α
2
N
K
K
M
O
M
L
M
O
M
α
M
1
1
L
α
MN
1
1
L
α
M
1
K
L
α
MN
K
K
]
,
to generate the power control pattern matrix
A
=
[
α
1
1
1
L
α
1
N
1
1
L
α
11
K
L
α
1
N
K
K
α
2
1
1
L
α
2
N
1
1
L
α
2
1
K
L
α
2
N
K
K
M
O
M
L
M
O
M
α
M
1
1
L
α
MN
1
1
L
α
M
1
K
L
α
M
N
K
K
]
e
H
P
D
M
A
=
[
A
1
,
A
2
,
L
,
A
K
]
,
wherein
K represents a number of uplink paired user, N 1 , N 2 , . . . , N k represent respectively the numbers of columns of the PDMA pattern matrix H PDMA (M*N) occupied by the users 1 , 2 , . . . , K, wherein N=N 1 +N 2 +L+N K , A 1 , A 2 , . . . , A K represent respectively power control pattern vector groups for the users 1 , 2 , . . . , K, and □ represents a dot matrixing;
wherein the processor is further configured to read a program stored in the memory to:
select an uplink power control optimization mode;
adjust a current combination of the PDMA pattern vector group and the power control factor of each user, to enable a Proportional Fair (PF) weighted sum throughput to satisfy the formula
A
opt
=
argmax
A
{
∑
k
=
1
K
T
k
(
S
I
N
R
k
,
A
)
·
(
1
-
BLER
k
(
S
I
N
R
k
,
A
)
)
R
¯
k
(
t
)
}
and ∥A∥ 1 =1, and to maximize the PF weighted sum throughput per power consumption unit in the case that the users are paired, wherein
T k (SINR k,A ) represents a transmission rate of a user k in the case that the power control pattern matrix of the user k is A, calculated based on transmission bit numbers mapped from a Signal to Interference plus Noise Ratio (SINR) of the user k and resources occupied by the user k, a value of the SINR of the user k is related to the power control factors in the PDMA pattern matrix, BLER k (SINR k,A ) represents an estimated Block Error Rate (BLER) in the case that the power control pattern matrix of the user k is A, calculated based on a mapping of the SINR and the BLER, R k (t) represents a historical average rate of the user k at a time point t, P t (∥A k ∥ 1 ) represents total uplink transmission power in the case that the power control pattern matrix of the user k is A and P t (∥A k ∥ 1 ) is smaller than or equal to maximum uplink transmission power P max of a user terminal, and ∥·∥ 1 represents summing up matrix elements;
the historical average rate is calculated based on the formula R k (t)=α·R k (t)+(1−α)· R k (t−1), wherein R k (t) represents the historical average rate of the user k at a current time point t, R k (t−1) represents the historical average rate of the user k at a previous time point t−1, R k (t) represents a current rate of the user k, α represents a forgetting factor, and the PF weighted sum throughput is related to the current rate of the user k in the case that α=1.