Granted Patent B2

US 12,603,693 · App. 19/261,157 · Granted Apr 14, 2026

Inventors: Foad Sohrabi (Murray Hill, NJ); Pavan Koteshwar Srinath (Massy, FR); Jinfeng Du (Murray Hill, NJ)

Assignee: Nokia Solutions and Networks Oy

H04B7/086

Abstract

Claims (36)

1 . An apparatus, comprising:

at least one processor; and

at least one memory storing instructions which, when executed by the at least one processor, cause the apparatus at least to:

receive at least one uplink pilot associated with a receiver, wherein the at least one uplink pilot is transmitted over at least one time-frequency unit, and wherein the receiver comprises at least one digital beamforming, DBF, -panel and at least one subarray-based beamforming, SABBF, -panel, the at least one DBF-panel having an associated DBF-channel and the at least one SABBF-panel having an associated SABBF-channel;

based on channel distributions of the DBF-channel and the SABFF-channel and further based on the at least one uplink pilot, determine at least one estimated strongest direction of a common covariance matrix associated with the at least one DBF-panel and the at least one SABBF-panel; and

determine a beamforming vector for at least one transmission phase or at least one reception phase based on the at least one estimated strongest direction of the common covariance matrix.

2 . The apparatus according to claim 1 , wherein the at least one estimated strongest direction of the common covariance matrix comprises an eigenvector corresponding to a strongest eigenvalue of the common covariance matrix.

3 . The apparatus according to claim 1 , wherein the instructions, when executed by the at least one processor, further cause the apparatus at least to:

determine a plurality of candidates for a strongest direction of the common covariance matrix; and

determine the at least one estimated strongest di-rection of the common covariance matrix further based on the plurality of candidates.

4 . The apparatus according to claim 3 , wherein the plurality of candidates for the strongest di-rection of the common covariance matrix comprises an over-sampled Discrete Fourier Transform, DFT, matrix.

5 . The apparatus according to claim 3 , wherein the instructions, when executed by the at least one processor, further cause the apparatus at least to:

based on the plurality of candidates and the at least one uplink pilot, determine a plurality of signal metrics, wherein each signal metric from the plurality of signal metrics is associated with one from the plurality of candidates; and

determine the at least one estimated strongest di-rection of the common covariance matrix further based on the determined plurality of signal metrics.

6 . The apparatus according to claim 5 , wherein the plurality of signal metrics is determined based at least partially on an inner product between the plurality of candidates and the at least one uplink pilot.

7 . The apparatus according to claim 1 , wherein the instructions, when executed by the at least one processor, further cause the apparatus at least to:

determine a common direction for more than one user a received uplink pilot is associated with; and

determine the beamforming vector further based on the determined common direction.

8 . The apparatus according to claim 1 , wherein the instructions, when executed by the at least one processor, further cause the apparatus at least to:

determine at least one user specific direction for at least one user the at least one uplink pilot is associated with; and

determine the beamforming vector further based on the at least one user specific direction.

9 . The apparatus according to claim 1 , wherein the instructions, when executed by the at least one processor, further cause the apparatus at least to:

perform at least a portion of the determination of the at least one estimated strongest direction of the common covariance matrix by utilizing a first deep neural network, DNN.

10 . The apparatus according to claim 1 , wherein the instructions, when executed by the at least one processor, further cause the apparatus at least to:

perform a combination phase of each subarray in the at least one SABFF-panel by utilizing a second DNN.

11 . The apparatus according to claim 1 , wherein at least a portion of the at least one estimated strongest direction of the common covariance matrix is based on a substantial sparsity of the at least one DBF-channel and the at least one SABBF-channel.

12 . A receiver comprising:

the apparatus according to claim 1 ;

the at least one DBF-panel; and

the at least one SABBF-panel.

13 . A network node device comprising the receiver of claim 12 .

14 . A method, comprising:

receiving at least one uplink pilot associated with a receiver, wherein the at least one up-link pilot is transmitted over at least one time-frequency unit, and wherein the receiver comprises at least one digital beamforming, DBF, -panel and at least one subarray-based beamforming, SABBF, -panel, the at least one DBF-panel having an associated DBF-channel and the at least one SABBF-panel having an associated SABBF-channel;

based on channel distributions of the DBF-channel and the SABFF-channel and further based on the at least one uplink pilot, determining at least one estimated strongest direction of a common covariance matrix associated with the at least one DBF-panel and the at least one SABBF-panel; and

determining a beamforming vector for at least one transmission phase or at least one reception phase based on the at least one estimated strongest direction of the common covariance matrix.

15 . A non-transitory computer-readable medium comprising program instructions which when executed by an apparatus cause the apparatus to perform the method of claim 14 .

Assignments (6)

Priority Claims (1)

Continuity (1)

References Cited (53)