# US 2026/0230148 A1 · Channel Quality Indicator (CQI) Determination for Channel State Information (CSI) Compression Using Artificial Intelligence(AI)/Machine Learning (ML) Models in Wireless Communication Networks

> Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt

- Number: US 2026/0230148 A1
- Document ID: US20260230148A1
- Status: Published application
- Kind: A1
- Application number: 19/226,141
- Filed: 2025-06-02
- Published: 2026-08-06
- Earliest claimed: 2025-01-31
- Assignees: Apple Inc.
- Applicants: Apple Inc.
- Inventors: Konstantinos Sarrigeorgidis; Yang Tang; Xiang Chen; Jie Cui
- Web page: https://patentgrep.com/patent/US20260230148A1

## CPC

- H04B 7/0632: {Channel quality parameters, e.g. channel quality indicator [CQI]}
- H04B 7/0626: {Channel coefficients, e.g. channel state information [CSI]}
- H04B 7/0639: {Using selective indices, e.g. of a codebook, e.g. pre-distortion matrix index [PMI] or for beam selection}
- H04B 17/346: {Noise values (signal-to-interference ratio [SIR] or carrier-to-interference ratio [CIR] H04B17/336)}
- H04L 41/16: using machine learning or artificial intelligence
- H04B: TRANSMISSION
- H04L: TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION (arrangements common to telegraphic and telephonic communication H04M)

## Abstract

Apparatuses, systems, and methods for determining Channel Quality Indicator (CQI) for Channel State Information (CSI) compression using AI/ML-based models are described including systems, methods, and mechanisms for a user equipment device (UE) to report an adjusted Channel Quality Indicator (CQI) to a base station (BS). In one example, adjusted Channel Quality Indicator (CQI) may be based on differences in precoding matrices.

## Claims

### Claim 1

A method for determining Channel Quality Indicator (CQI) for Channel State Information (CSI) compression using artificial intelligence (AI)/machine learning (ML)-based models at a User Equipment (UE) in a wireless communications network, the method comprising:
encoding an optimal precoding matrix according to an AI/ML-based model;
sending the encoded optimal precoding matrix to a base station (BS);
receiving, at the UE, a precoding matrix reconstructed from the encoded optimal precoding matrix from the base station (BS); and
calculating a Channel Quality Indicator (CQI) adjustment value based on differences between the optimal precoding matrix and the precoding matrix reconstructed from the encoded optimal precoding matrix.

### Claim 2 (depends on 1)

The method of claim 1, wherein calculating a Channel Quality Indicator (CQI) adjustment value based on differences between the optimal precoding matrix and the precoding matrix reconstructed from the encoded optimal precoding matrix includes calculating a capacity loss of a downlink (DL) channel based on differences in channel conditions provided by the optimal precoding matrix and the reconstructed precoding matrix.

### Claim 3 (depends on 2)

The method of claim 2, wherein calculating a capacity loss includes approximating a capacity loss based on values provided by the optimal precoding matrix and the reconstructed precoding matrix.

### Claim 4 (depends on 2)

The method of claim 2, wherein calculating a capacity loss includes calculating a capacity loss based on a Signal-to-Interference-plus-Noise Ratio (SINR) calculation provided for a receiver model.

### Claim 5 (depends on 4)

The method of claim 4, wherein a receiver model includes one of a Zero Forcing (ZF) receiver model, a Minimum Mean Square Error (MMSE) receiver model, and a Singular Value Decomposition (SVD) receiver model.

### Claim 6 (depends on 2)

The method of claim 2, wherein calculating a Channel Quality Indicator (CQI) adjustment value further includes mapping a calculated capacity loss to Channel Quality Indicator (CQI) difference values.

### Claim 7 (depends on 6)

The method of claim 6, wherein calculating a Channel Quality Indicator (CQI) adjustment value further includes calculating the Channel Quality Indicator (CQI) adjustment value by averaging a number of Channel Quality Indicator (CQI) difference values, wherein the number of Channel Quality Indicator (CQI) difference values is based on a number of iterations performed during a calibration phase.

### Claim 8 (depends on 7)

The method of claim 7, wherein each iteration corresponds to a time interval specified by the base station (BS).

### Claim 9

A method for determining Channel Quality Indicator (CQI) for Channel State Information (CSI) compression using artificial intelligence (AI)/machine learning (ML)-based models at a User Equipment (UE) in a wireless communications network, the method comprising:
encoding an optimal precoding matrix according to an AI/ML-based model;
sending the encoded optimal precoding matrix to a base station (BS);
calculating a Channel Quality Indicator (CQI) value, wherein the Channel Quality Indicator (CQI) value is based on the optimal precoding matrix;
applying a Channel Quality Indicator (CQI) adjustment value to the calculated Channel Quality Indicator (CQI) value; and
reporting an adjusted Channel Quality Indicator (CQI) to the base station (BS).

### Claim 10 (depends on 9)

The method of claim 9, wherein applying a Channel Quality Indicator (CQI) adjustment value includes reducing the calculated Channel Quality Indicator (CQI) value.

### Claim 11

An apparatus of a user equipment (UE) configured to determine Channel Quality Indicator (CQI) for Channel State Information (CSI) compression using artificial intelligence (AI)/machine learning (ML)-based models at a User Equipment (UE) in a wireless communications network, the apparatus comprising:
one or more processors configured to:
encode an optimal precoding matrix according to an AI/ML-based model;
send the encoded optimal precoding matrix to a base station (BS);
calculate a Channel Quality Indicator (CQI) value, wherein the Channel Quality Indicator (CQI) value is based on the optimal precoding matrix;
apply a Channel Quality Indicator (CQI) adjustment value to the calculated Channel Quality Indicator (CQI) value; and
report an adjusted Channel Quality Indicator (CQI) to the base station (BS).

### Claim 12 (depends on 11)

The apparatus of claim 11, wherein applying a Channel Quality Indicator (CQI) adjustment value includes reducing the calculated Channel Quality Indicator (CQI) value.

### Claim 13 (depends on 11)

The apparatus of claim 11, wherein the one or more processors are further configured to:
receive a precoding matrix reconstructed from the encoded optimal precoding matrix from the base station (BS); and
calculate the Channel Quality Indicator (CQI) adjustment value based on differences between the optimal precoding matrix and the precoding matrix reconstructed from the encoded optimal precoding matrix.

### Claim 14 (depends on 13)

The apparatus of claim 13, wherein calculating a Channel Quality Indicator (CQI) adjustment value based on differences between the optimal precoding matrix and the precoding matrix reconstructed from the encoded optimal precoding matrix includes calculating a capacity loss of a downlink (DL) channel based on differences in channel conditions provided by the optimal precoding matrix and the reconstructed precoding matrix.

### Claim 15 (depends on 14)

The apparatus of claim 14, wherein calculating a capacity loss includes approximating a capacity loss based on values provided by the optimal precoding matrix and the reconstructed precoding matrix.

### Claim 16 (depends on 14)

The apparatus of claim 14, wherein calculating a capacity loss includes calculating a capacity loss based on a Signal-to-Interference-plus-Noise Ratio (SINR) calculation provided for a receiver model.

### Claim 17 (depends on 16)

The apparatus of claim 16, wherein a receiver model includes one of a Zero Forcing (ZF) receiver model, a Minimum Mean Square Error (MMSE) receiver model, and a Singular Value Decomposition (SVD) receiver model.

### Claim 18 (depends on 13)

The apparatus of claim 13, wherein calculating a Channel Quality Indicator (CQI) adjustment value further includes mapping a calculated capacity loss to a Channel Quality Indicator (CQI) difference value.

### Claim 19 (depends on 18)

The apparatus of claim 18, wherein calculating a Channel Quality Indicator (CQI) adjustment value further includes calculating the Channel Quality Indicator (CQI) adjustment value by averaging a number of Channel Quality Indicator (CQI) difference values, wherein the number of Channel Quality Indicator (CQI) difference values is based on a number of iterations performed during a calibration phase.

### Claim 20 (depends on 19)

The apparatus of claim 19, wherein each iteration corresponds to a time interval specified by the base station (BS).

### Claim at position 21

-27. (canceled)
