# US 12,009,888 B2 · Network nodes and methods performed in a wireless communication network

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

- Number: US 12,009,888 B2
- Document ID: US12009888B2
- Status: Granted
- Kind: B2
- Application number: 17/779,201
- Filed: 2019-12-06
- Published: 2023-02-02
- Granted: 2024-06-11
- Earliest claimed: 2019-12-06
- Assignees: Telefonaktiebolaget LM Ericsson (publ)
- Applicants: Telefonaktiebolaget LM Ericsson (publ)
- Inventors: Maksym Girnyk
- Web page: https://patentgrep.com/patent/US12009888B2

## CPC

- G06N 3/08: Learning methods
- H04B 7/0456: Selection of precoding matrices or codebooks, e.g. using matrices antenna weighting
- G06N 3/0499: Feedforward networks
- G06N 3/09: Supervised learning
- H04L 25/0204: {of multiple channels}
- H04W 24/02: Arrangements for optimising operational condition
- G06N 20/10: using kernel methods, e.g. support vector machines [SVM]
- G06N: COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- H04B: TRANSMISSION
- H04L: TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION (arrangements common to telegraphic and telephonic communication H04M)
- H04W: WIRELESS COMMUNICATION NETWORKS (broadcast communication H04H; communication systems using wireless links for non-selective communication, e.g. wireless extensions H04M1/72)

## Abstract

Embodiments herein relate to a method performed by a network node for handling communication in a wireless communication network. The network node selects a precoder given a channel based on output of a trained computational model, trained with approximated precoders or channel matrices mapped to preferred precoders, and transmits data over the channel using the selected precoder.

## Claims

### Claim 1

A method performed by a network node for handling communication in a wireless communication network, the method comprising:
selecting a precoder given a channel based on output of a trained computational model, trained with one of approximated precoders and channel matrices mapped to preferred precoders;
transmitting data over the channel using the selected precoder; and
the approximated precoders comprises one of singular-value decomposition, SVD, based precoders and optimized precoders for lower-order discrete constellations.

### Claim 2 (depends on 1)

The method according to claim 1, further comprising:
training the computational model.

### Claim 3 (depends on 1)

The method according to claim 1, wherein selecting the precoder comprises selecting a precoder matrix.

### Claim 4 (depends on 1)

The method according to claim 1, wherein the selection is performed for a given signal constellation.

### Claim 5 (depends on 1)

The method according to claim 1, wherein the trained computational model is one of a neural network, a multiple output decision tree, and a random forest.

### Claim 6 (depends on 1)

The method according to claim 1, wherein one of:
the trained computational model is obtained; and
the output of the trained computational model is obtained from a model network node.

### Claim 7 (depends on 2)

The method according to claim 2, wherein selecting the precoder comprises selecting a precoder matrix.

### Claim 8 (depends on 2)

The method according to claim 2, wherein the selection is performed for a given signal constellation.

### Claim 9 (depends on 2)

The method according to claim 2, wherein the trained computational model is one of a neural network, a multiple output decision tree, and a random forest.

### Claim 10

A method performed by a model network node for handling communication in a wireless communication network, the method comprising:
training a computational model with one of approximated precoders and channel matrices mapped to preferred precoders for selecting a precoder given a channel based on an output of the trained computational model; and
transferring one of the trained computational model and the output of the computational model to a network node.

### Claim 11

A network node for handling communication in a wireless communication network, the network node being configured to:
select a precoder given a channel based on output of a trained computational model, trained with one of approximated precoders and channel matrices mapped to preferred precoders;
transmit data over the channel using the selected precoder; and
the approximated precoders comprising one of singular-value decomposition, SVD, based precoders, and optimized precoders for lower-order discrete constellations.

### Claim 12 (depends on 11)

The network node according to claim 11, further being configured to train the computational model.

### Claim 13 (depends on 11)

The network node according to claim 11, wherein the network node is configured to select the precoder by selecting a precoder matrix.

### Claim 14 (depends on 11)

The network node according to claim 11, wherein the network node is configured to select the precoder for a given signal constellation.

### Claim 15 (depends on 11)

The network node according to claim 11, wherein the trained computational model is one of a neural network, a multiple output decision tree, and a random forest.

### Claim 16 (depends on 11)

The network node according to claim 11, wherein the network node is configured to obtain the one of the trained computational model and the output of the trained computational model from a model network node.

### Claim 17

A model network node for handling communication in a wireless communication network, the model network node being configured to:
train a computational model with one of approximated precoders and channel matrices mapped to preferred precoders for selecting a precoder given a channel based on an output of the trained computational model, and
transfer one of the trained computational model and the output of the computational model to a network node.