Granted Patent B2

US 12,633,987 · App. 18/516,860 · Granted May 19, 2026

Inventors: Md. Saifur Rahman (Plano, TX); Eko Onggosanusi (Coppell, TX); Gilwon Lee (McKinney, TX)

Assignee: Samsung Electronics Co., Ltd.

H04B7/0626H04W24/10

Abstract

Claims (101)

1 . A user equipment (UE) comprising:

a transceiver configured to receive information about a channel state information (CSI) report, the report based on at least two components: a set of basis entities and a set of coefficients; and

a processor operably coupled to the transceiver, the processor, based on the information, configured to determine the set of basis entities and the set of coefficients,

wherein the transceiver is further configured to transmit the CSI report including at least one indicator indicating at least one of the set of basis entities and the set of coefficients,

wherein the set of basis entities is based on X dimensions,

wherein the set of coefficients is not based on the X dimensions and is from a set of convolutional-based coefficients, and

wherein X is associated with the CSI report and X≥1.

2 . The UE of claim 1 , wherein the set of convolutional-based coefficients is based on a machine-learning (ML) or artificial intelligence (AI) based algorithm.

3 . The UE of claim 2 , wherein:

the ML or AI based algorithm is based on a training using a predetermined set of coefficients from a codebook,

the information indicates a type of the set of coefficients, the type indicated from among (a) a convolutional-based coefficients, which are learnt or adaptive, and (b) codebook-based coefficients, which are fixed or predetermined, and

the processor is further configured to:

identify the type of the set of coefficients; and

determine the CSI report based on the identified type of the set of coefficients.

4 . The UE of claim 1 , wherein:

each basis entity in the set of basis entities is a vector, and

the vector is at least one of the following vector types:

separate for each dimension of the X dimensions, and

joint across at least two of the X dimensions.

5 . The UE of claim 1 , wherein:

each basis entity in the set of basis entities is a matrix, and

the matrix is at least one of the following matrix types:

a Kronecker product of at least two vectors, and

joint across at least two of the X dimensions.

6 . The UE of claim 1 , wherein:

X=3,

a first of the X=3 dimensions is associated with a spatial-domain (SD), a second of the X=3 dimensions is associated with a frequency-domain (FD), and a third of the X=3 dimensions is associated with a time-domain (TD),

P is a number of SD units, N 3 is a number of FD units, and N 4 is a number of TD units,

a value of P is based on properties associated with antenna ports, the properties being at least one of: (i) antenna structure, (ii) antenna polarization, (iii) antenna geometry, including shape or size, and (iv) locations, including co-located or non-co-located, and

P≥1, N 3 ≥1, and N 4 ≥1.

7 . The UE of claim 6 , wherein:

the set of basis entities is based on a codebook,

the codebook comprises vectors or matrices, and

each of the vectors or matrices has a length or size, respectively, that depends on at least one of the value of P, a value of N 3 , and a value of N 4 .

8 . The UE of claim 6 , wherein:

P=sN 1 N 2 ,

N 1 and N 2 are a number of antenna ports associated with two dimensions, respectively, of a two-dimensional planar antenna array, and

s∈{1,2} is a number of polarizations associated with each location in the two-dimensional planar antenna array.

9 . The UE of claim 1 , wherein:

the transceiver is further configured to receive information about K≥1 antenna port groups (AGs),

the processor is further configured to:

measure the K AGs, and

determine the CSI report based on the measurement,

the K AGs are measured via at least one of the X dimensions, and

each of the K AGs includes at least one non-zero power (NZP) CSI-RS port.

10 . A base station (BS) comprising:

a processor; and

a transceiver operably coupled with the processor, the transceiver configured to:

transmit information about a channel state information (CSI) report, the report based on at least two components: a set of basis entities and a set of coefficients; and

receive the CSI report including at least one indicator indicating at least one of the set of basis entities and the set of coefficients,

wherein the set of basis entities is based on X dimensions,

wherein the set of coefficients is not based on the X dimensions and is from a set of convolutional-based coefficients, and

wherein X is associated with the CSI report and X≥1.

11 . The BS of claim 10 , wherein the set of convolutional-based coefficients is based on a machine-learning (ML) or artificial intelligence (AI) based algorithm.

12 . The BS of claim 11 , wherein:

the ML or AI based algorithm is based on a training using a predetermined set of coefficients from a codebook,

the information indicates a type of the set of coefficients, the type indicated from among (a) a convolutional-based coefficients, which are learnt or adaptive, and (b) codebook-based coefficients, which are fixed or predetermined, and

the CSI report is based on the indicated type of the set of coefficients.

13 . The BS of claim 10 , wherein:

each basis entity in the set of basis entities is a vector, and

the vector is at least one of the following vector types:

separate for each dimension of the X dimensions, and

joint across at least two of the X dimensions.

14 . The BS of claim 10 , wherein:

each basis entity in the set of basis entities is a matrix, and

the matrix is at least one of the following matrix types:

a Kronecker product of at least two vectors, and

joint across at least two of the X dimensions.

15 . The BS of claim 10 , wherein:

X=3,

a first of the X=3 dimensions is associated with a spatial-domain (SD), a second of the X=3 dimensions is associated with a frequency-domain (FD), and a third of the X=3 dimensions is associated with a time-domain (TD),

P is a number of SD units, N 3 is a number of FD units, and N 4 is a number of TD units,

a value of P is based on properties associated with antenna ports, the properties being at least one of: (i) antenna structure, (ii) antenna polarization, (iii) antenna geometry including a shape or size, and (iv) locations including co-located or non-co-located, and

P≥1, N 3 ≥1, and N 4 ≥1.

16 . The BS of claim 15 , wherein:

the set of basis entities is based on a codebook,

the codebook comprises vectors or matrices, and

each of the vectors or matrices has a length or size, respectively, that depends on at least one of the value of P, a value of N 3 , and a value of N 4 .

17 . The BS of claim 15 , wherein:

P=sN 1 N 2 ,

N 1 and N 2 are a number of antenna ports associated with two dimensions, respectively, of a two-dimensional planar antenna array, and

s∈{1,2} is a number of polarizations associated with each location in the two-dimensional planar antenna array.

18 . The BS of claim 10 , wherein:

the transceiver is further configured to transmit information about K≥1 antenna port groups (AGs),

the CSI report is based on measurement of the K AGs,

the K AGs are measured via at least one of the X dimensions, and

each of the K AGs includes at least one non-zero power (NZP) CSI-RS port.

19 . A method performed by a user equipment (UE), the method comprising:

receiving information about a channel state information (CSI) report, the report based on at least two components: a set of basis entities and a set of coefficients;

based on the information, determining the set of basis entities and the set of coefficients; and

transmitting the CSI report including at least one indicator indicating at least one of the set of basis entities and the set of coefficients,

wherein the set of basis entities is based on X dimensions,

wherein the set of coefficients is not based on the X dimensions and is from a set of convolutional-based coefficients, and

wherein X is associated with the CSI report and X≥1.

20 . The method of claim 19 , wherein:

the set of convolutional-based coefficients is based on a machine-learning (ML) or artificial intelligence (AI) based algorithm,

the ML or AI based algorithm is based on a training using a predetermined set of coefficients from a codebook,

the information indicates a type of the set of coefficients, the type indicated from among (a) a convolutional-based coefficients, which are learnt or adaptive, and (b) codebook-based coefficients, which are fixed or predetermined, and

the method further comprises:

identifying the type of the set of coefficients; and

determining the CSI report based on the identified type of the set of coefficients.

Assignments (1)

Continuity (2)

References Cited (17)