Precoding-Matched CSI Feedback via Rank Hypothesis Segmentation

Solution Overview

Problem

In mobile communications, there is a mismatch between user equipment (UE) channel quality indication (CQI) feedback and the base station's scheduled rank under antenna switching, leading to performance uncertainty and the inability to convert CQI values under lower-rank precoding to higher-rank multiple-input-multiple-output (MIMO) precoders.

Innovation Solution

The proposed solution involves measuring and generating CSI feedback with multiple CQI values based on hypotheses of rank and antenna port allocation, allowing UE to report multiple pairs of rank and CQI values, and using existing antenna-port tables for DMRS/CSI-RS port indication, enabling precoding-matched CSI feedback without requiring new measurement mechanisms or algorithms.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Device complexity

If UE reports CQI under lower-rank PMI for eigen-beamforming, then the feedback computation is simplified, but the scheduled rank by the network may be higher causing precoding mismatch

Engineering Contradiction:

ImproveCSI feedback computation complexityVSAvoidprecoding match accuracy

Core Design Contradiction:

Device complexityVSReliability

Solution Approach 1:

The patent segments the CQI feedback into multiple hypotheses, each corresponding to a different rank assumption. Instead of reporting a single CQI value, the UE generates multiple CQI reports (e.g., CQI for rank-2, CQI for rank-3, CQI for rank-4) based on different PMI assumptions. This segmentation allows the network to select the appropriate CQI that matches its scheduling decision, resolving the precoding mismatch problem while keeping individual CQI computations relatively simple.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The patent applies preliminary action by having the UE pre-compute multiple CQI values for different rank hypotheses before the actual scheduling decision. The UE generates a set of CQI reports covering various possible ranks (e.g., rank-2, rank-3, rank-4) that the network might schedule. This preliminary computation of multiple scenarios enables the network to immediately select the matching CQI without requiring complex real-time conversion algorithms.

Inventive Principle:

Principle #10Preliminary action

2Productivity

If the network schedules higher rank than UE reports, then system throughput is improved, but CQI accuracy for the scheduled rank becomes uncertain

Engineering Contradiction:

Improvesystem throughputVSAvoidCQI accuracy

Core Design Contradiction:

ProductivityVSMeasurement precision

Solution Approach 1:

The patent segments the CQI feedback into multiple hypotheses, each corresponding to a different rank assumption. Instead of reporting a single CQI value, the UE generates multiple CQI reports (e.g., CQI for rank-2, CQI for rank-3, CQI for rank-4) based on different PMI assumptions. This segmentation allows the network to select the appropriate CQI that matches its scheduling decision, resolving the precoding mismatch problem while keeping individual CQI computations relatively simple.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The patent changes the parameter of rank assumption in the CQI computation. The UE computes CQI values under different rank hypotheses (e.g., assuming rank-2, rank-3, or rank-4 precoding) by varying the PMI parameter accordingly. When the network schedules a specific rank, the corresponding pre-computed CQI with matching rank parameter provides accurate throughput prediction, enabling the network to schedule higher ranks with confidence in CQI accuracy.

Inventive Principle:

Principle #35Parameter changes

3Measurement precision

If proprietary conversion algorithms are developed to map CQI from lower-rank to higher-rank precoding, then CQI accuracy is improved, but device complexity and processing overhead increase

Engineering Contradiction:

ImproveCQI conversion accuracyVSAvoidprocessing algorithm complexity

Core Design Contradiction:

Measurement precisionVSDevice complexity

Solution Approach 1:

The patent applies preliminary action by having the UE pre-compute multiple CQI values for different rank hypotheses before the actual scheduling decision. The UE generates a set of CQI reports covering various possible ranks (e.g., rank-2, rank-3, rank-4) that the network might schedule. This preliminary computation of multiple scenarios enables the network to immediately select the matching CQI without requiring complex real-time conversion algorithms.

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The patent uses copying by having the UE generate multiple copies of CQI reports, each corresponding to a different rank hypothesis. Instead of developing complex conversion algorithms to transform a single CQI value across different ranks, the UE creates separate CQI copy versions (CQI for rank-2, CQI for rank-3, CQI for rank-4) based on different PMI assumptions. The network simply selects the appropriate copy that matches its scheduling decision, avoiding the need for proprietary conversion algorithms.

Inventive Principle:

Principle #26Copying

Data Source

PatentUS11563467B2Precoding-matched CSI feedback in mobile communications

Publication Date: 2023.01.24 MEDIATEK INC

AI summary

Examples pertaining to precoding-matched channel state information (CSI) feedback in mobile communications are described. An apparatus (e.g., UE) beamforms one or more of a plurality of ports used in reception of a channel state information reference signal (CSI-RS) with one or more precoders that are applied on one or more of the plurality of ports used in reception of a demodulation reference signal (DMRS) or a physical downlink shared channel (PDSCH). The apparatus generates a CSI feedback comprising at least a plurality of channel quality indication (CQIs) with respect to a plurality of hypothesized ranks. The apparatus then transmits the CSI feedback to a network.