Method And Apparatus For Linear Combination Codebook Design And CSI Feedback In Mobile Communications

• US 20190326965A1
• Filed: 07/04/2019
• Published: 10/24/2019
• Est. Priority Date: 06/16/2017
• Status: Abandoned Application

• Current Assignee: MediaTek, Inc.
• Original Assignee: MediaTek, Inc.
• Inventors: Yang, Weidong, Tsai, Lung-Sheng
• Application Number:
  US16/503,583
• Publication Number:
  US 20190326965A1
• Time in Patent Office: Days
• CPC Class Codes: H04B 7/0417 Feedback systems
• CPC Class Codes: H04B 7/0452 Multi-user MIMO systems
• CPC Class Codes: H04B 7/0478 Special codebook structures directed to feedback optimisation
• CPC Class Codes: H04B 7/0479 for multi-dimensional arrays, e.g. horizontal or vertical pre-distortion matrix index [PMI]

Abstract

Techniques and examples pertaining to linear combination codebook design and channel state information (CSI) feedback in mobile communications are described. Frequency-dependent parameterization is supported to reduce signaling overhead in reporting beam selection and to enhance CSI resolution. Additionally, the number of candidates for beam selection depends on the strength of each beam to be selected such that more candidates are considered for a stronger beam than for a weaker beam. Moreover, amplitude quantization and/or phase quantization for beams of different strengths are different to reduce quantization error.

Claims

1. A method, comprising:

measuring, by a processor of a user equipment (UE), one or more reference signals from a network node of a wireless network;

selecting, by the processor, a set of one or more selected beams from a plurality of beams by;

determining, based on the measuring, at least one high-power beam among the plurality of beams as a chosen beam; and

searching in a spatial region around the chosen beam to identify one or more other beams each having a power no greater than that of the chosen beam, the chosen beam and the one or more other beams being the one or more selected beams;

generating, by the processor, a report indicating the one or more selected beams; and

transmitting, by the processor, the report to the network node.

2. The method of claim 1, wherein the report comprises bit-fields that separately indicate the chosen beam with high power and other beams as the one or more selected beams.

3. The method of claim 2, wherein the report further indicates quantized linear combination codebook coefficients associated with the one or more selected beams.

4. The method of claim 2, wherein the spatial region comprises an array of multiple geometric shapes arranged in a M×

N dimension of M rows and N columns centered around the chosen beam with the chosen beam in one of the geometric shapes in a center of the array, and wherein each of M and N is a positive integer.

5. The method of claim 4, wherein a shape of each of the geometric shapes is a circle, an ellipse or a polygon.

6. The method of claim 1, wherein the one or more selected beams comprise beams of different strengths, and wherein the generating of the report indicating the quantized linear combination codebook coefficients associated with the one or more selected beams comprises either or both of:

quantizing an amplitude of each linear combination codebook coefficient associated with each of the one or more selected beams using different amplitude quantization schemes based on either the different strengths of the one or more selected beams or a strength order of the one or more selected beams; and

quantizing a phase of each linear combination codebook coefficient associated with each of the one or more selected beams using different phase quantization schemes based on either the different strengths of the one or more selected beams or the strength order of the one or more selected beams.

7. The method of claim 1, wherein the one or more selected beams comprise beams of different strengths, and wherein the generating of the report indicating the quantized linear-combination codebook coefficients associated with the one or more selected beams comprises:

generating a plurality of channel realizations;

performing computation for linear combination codebook amplitude coefficients and linear combination codebook phase coefficients associated with the one or more selected beams without quantizing the amplitude coefficients and the phase coefficients;

collecting statistics regarding the amplitude coefficients and the phase coefficients for the one or more selected beams;

fitting the collected statistics to one or more distribution curves; and

identifying an optimal distribution and a corresponding optimal quantizer based on a result of the fitting.

8. The method of claim 7, wherein the fitting of the collected statistics to one or more distribution curves comprises applying a Lloyd-Max iterative algorithm on the collected statistics.

9. The method of claim 1, further comprising:

utilizing a first quantizer for the coefficients for a strong beam; and

utilizing a second quantizer for the coefficients for a weak beam,

wherein the first quantizer and the second quantizer have different settings of quantization ranges, different numbers of quantization levels, or different quantizing step sizes.