Apparatus and method for generating codebook in wireless communication system
An apparatus and method for generating a codebook in a wireless communication system are disclosed. The codebook generation method includes determining one or more dominant singular vectors in a channel matrix for antennas and setting each of the dominant singular vectors as a random non-zero vector, generating a first codebook having codewords, a minimum distance between the codewords being maximized, using the random non-zero vector in a region that includes unit norm vectors each having a Euclidean distance to each of the dominant singular vectors, equal to or less than a predetermined value, generating a second codebook corresponding to a unitary matrix that rotates the random non-zero vector toward the dominant singular vectors, and generating a final codebook using the first and second codebooks.
1. A method for generating a codebook in a wireless communication system, the method comprising:
determining one or more dominant singular vectors in a channel matrix of each antenna; and
generating codewords by sampling the codewords in a region that includes unit norm vectors, each norm vector having a Euclidean distance equal to or less than a predetermined Euclidean distance value t each of the dominant singular vectors,
wherein the codewords are generated through sampling of the codewords of the region by multiplying a rotation matrix being a diagonal matrix in front of each of the dominant singular vectors, and a minimum distance between the generated codewords is maximized, and
wherein the predetermined Euclidean distance value is adjusted according to a channel variation with respect to one of the dominant singular vectors.
2. The method according to claim 1 , wherein a phase rotation of the rotation matrix is one of four phase sets given by [Equation A] or a combination of the four phase sets,
Rot
=
{
0
,
±
π
8
}
or
{
0
,
±
π
16
}
or
{
0
,
±
π
32
}
or
{
0
,
±
π
n
}
,
[
Equation
A
]
where n is a real number.
3. The method according to claim 2 , wherein a codebook being a set of the generated codewords is expressed as [Equation B],
W
=
{
c
|
c
=
[
1
0
0
0
0
ⅇ
j
θ
1
0
0
0
0
ⅇ
j
θ
2
0
0
0
0
ⅇ
j
θ
3
]
[
v
1
v
2
…
v
n
]
,
θ
i
∈
Rot
}
[
Equation
B
]
where n is the number of the dominant singular vectors and k is the number of transmission antennas.
4. A method for generating a codebook in a wireless communication system, the method comprising:
determining one or more dominant singular vectors of in the channel matrix of each antenna;
determining a random non-zero vector;
generating a first codebook corresponding to a unitary matrix that rotates the random non-zero vector toward the one or more dominant singular vectors of a region;
generating a second codebook whose codewords are sampled from the region that includes unit norm vectors, each unit vector having Euclidean distance to the random non-zero vector is equal to or less than a predetermined value; and
generating a final codebook using the first and second codebooks.
5. The method according to claim 4 , wherein the final codebook is a product between the first codebook and the second codebook.
6. The method according to claim 5 wherein the second codebook includes a plurality of codebooks generated based on a plurality of Euclidean distances.
7. The method according to claim 6 , wherein the final codebook is expressed as [Equation C],
W=W 1 ( i ) W 2 k ( j ) [Equation C]
where W denotes the final codebook, W 1 (i) denotes the first codebook, W 2 k (j) denotes the second codebook, i denotes an index of a codeword in the first codebook, j denotes an index of a codeword in the second codebook, and k denotes an index of each of second codebooks generated using Euclidean distances to the one or more dominant singular vectors.
8. A method for generating a codebook in a wireless communication system, the method comprising:
determining one or more dominant singular vectors in the channel matrix of each antenna;
determining a random non-zero vector;
generating a first codebook by quantizing the one or more dominant singular vectors;
generating a second codebook by sampling codewords for a region that includes unit norm vectors, each unit norm vector having a Euclidean distance to the random non-zero vector is equal to or less than a predetermined value, so that a minimum distance between the codewords is maximized, and by subtracting the random non-zero vector from the sampled codewords; and
generating a final codebook using the first and second codebooks.
9. The method according to claim 8 , wherein the final codebook is expressed as sum of the first codebook and the second codebook.
10. The method according to claim 9 wherein the second codebook includes a plurality of codebooks generated based on a plurality of Euclidean distances.
11. The method according to claim 10 , wherein the final codebook is expressed as [Equation D],
W=W 1 ( i ) W 2 k ( j ) [Equation C]
where W denotes the final codebook, W 1 (i) denotes the first codebook, W 2 k (j) denotes the second codebook, i denotes an index of a codeword in the first codebook, j denotes an index of a codeword in the second codebook, and k denotes an index of each of second codebooks generated using Euclidean distances to the dominant singular vectors.
12. The method according to claim 9 , wherein the final codebook is expressed as [Equation E],
W=W 1 ( i )+ε k W 2 k ( j ) [Equation C]
where W denotes the final codebook, W 1 (i) denotes the first codebook, W n k (j) denotes the second codebook, i denotes an index of a codeword in the first codebook, j denotes an index of a codeword in the second codebook, and ε k denotes Euclidean distances to the dominant singular vectors.
13. A method for generating a codebook to support multi-cell communication at a base station (BS) in a multi-cell environment, the method comprising:
determining a serving cell dominant singular vector using a covariance matrix corresponding to channel measurement information between the base station and a user equipment (UE) within the serving cell of the base station and determining a neighboring cell dominant singular vector using a covariance matrix of an interference channel between the base station and a user equipment within the neighboring cell;
determining a modified dominant singular vector spaced from the serving cell dominant singular vector by a first predetermined value or less and spaced from the neighboring cell dominant singular vector by a second predetermined value or larger; and
generating codewords by sampling the codewords for a region that includes unit norm vectors each having a Euclidean distance to the modified dominant singular vector, equal to or less than a third predetermined value,
wherein the codewords are generated through sampling of the codewords in the region by multiplying a rotation matrix being a diagonal matrix in front of the modified dominant singular vector, and a minimum distance between the codewords is maximized.
14. The method according to claim 13 , further comprising:
transmitting a modified codebook or neighboring cell codebook generated by using the modified dominant singular vector to the user equipment served by the base station.
15. The method according to claim 13 , further comprising:
receiving information about the neighboring cell dominant singular vector from the user equipment within the neighboring cell or a base station of the user equipment within the neighboring cell.
16. The method according to claim 13 , wherein the serving cell, neighboring cell and modified dominant singular vectors are quantized values.
17. A method for generating a codebook at a user equipment (UE) in a multi-cell environment, the method comprising:
receiving one or more dominant singular vectors corresponding to long-term channel information from a base station (BS); and
generating codewords corresponding to short-term channel information by sampling the codewords for a region that includes unit norm vectors, each norm vector having a Euclidean distance to the received dominant singular vectors, equal to or less than a predetermined Euclidean distance value,
wherein the codewords are generated through sampling of the codewords in the region by multiplying a rotation matrix being a diagonal matrix in front of the dominant singular vector, and a minimum distance between the codewords is maximized, and
wherein the predetermined Euclidean distance value is adjusted according to a channel variation with respect to one of the dominant singular vectors.
18. The method according to claim 17 , wherein the received dominant singular vectors is determined by the base station based on a serving cell dominant singular vector of a channel between the base station and a user equipment within the serving cell and a neighboring cell dominant singular vector of a channel between a user equipment within the neighboring cell and the base station, and is spaced from the serving cell dominant singular vector by a first predetermined value or less and spaced from the neighboring cell dominant singular vector by a second predetermined value or larger.