Methods, architectures, apparatuses and systems for data-driven user equipment (ue)-specific reference signal operation

Patent metadata

Field	Value
Publication number	US-2026074860-A1
Application number	US-202319101830-A
Country	US
Kind code	A1
Filing date	Aug 7, 2023
Priority date	Aug 8, 2022
Publication date	Mar 12, 2026
Grant date	—

Abstract

Procedures, methods, architectures, apparatuses, systems, devices, and computer program products for configuring, selecting, and/or feeding back data-driven UE-specific RS are provided. One method may include a WTRU determining and/or indicating, to a network element (e.g., gNB), data-driven UE-specific RS related information, such as bundling type, RS pattern, RS position, RS density, RS signaling feedback overhead, and/or performance requirements.

First claim

1 . A method, comprising:
receiving, by a wireless transmit/receive unit (WTRU), configuration information indicating (1) a set of parameters for determining a reference signal (RS) configuration and (2) a parameter to use within the set, wherein the set of parameters comprise at least two bundling types; receiving a transmission using a first RS pattern of a first RS configuration type; performing one or more measurements on the transmission using the first RS pattern, wherein the one or more measurements include any of: an effective channel matrix for one or more slots or resource blocks, channel estimation accuracy, and RS signaling overhead; determining a bundling type associated with the RS configuration, based on (i) the set of parameters and (ii) the one or more measurements; and transmitting a first indication of the determined bundling type.

2 . The method of claim 1 , wherein the bundling type indicates resource bundling properties across one or more of time and frequency.

3 . The method of claim 1 , wherein the determining of the bundling type is performed using an artificial intelligence/machine learning (AI/ML) model.

4 . The method of claim 1 , wherein the set of parameters further comprise one or more of: a RS position, RS pattern, maximum allowed RS density, minimum allowed RS density, multi-user multiple input multiple output (MU-MIMO) setting, RS configuration signaling feedback overhead, and performance requirements.

5 . The method of claim 1 , comprising:
determining one or more of a second RS pattern and a second RS configuration type, based on (i) the set of parameters and (ii) the one or more measurements; and transmitting a second indication of the determined one or more of the second RS pattern and the second RS configuration type.

6 . The method of claim 5 , wherein one or more of the second RS pattern and the second RS configuration type is determined using an artificial intelligence/machine learning (AI/ML) model.

7 . The method of claim 6 , wherein:
the AI/ML model is trained with labelled data comprising input and output, the input comprises one or more of: (i) the effective channel matrix for one or more slots or resource blocks, and (ii) the set of parameters for determining the RS configuration, and the output comprises one or more of: the bundling type, position, pattern, and density for the second RS pattern based on performance objectives.

8 . The method of claim 1 , wherein the first RS configuration type comprises one or more of:
new radio (NR) legacy demodulation reference signals (DM-RS); a uniform RS pattern; a deterministic non-uniform RS pattern; a random non-uniform RS pattern; and a non-orthogonal RS pattern.

9 . The method of claim 1 , wherein transmitting the first indication or the second indication comprises:
implicitly transmitting one or more of the first indication and the second indication; explicitly transmitting one or more of the first indication and the second indication; transmitting one or more of the first indication and the second indication using inter-node data exchange; periodically transmitting one or more of the first indication and the second indication; or semi-periodically transmitting one or more of the first indication and the second indication.

10 . The method of claim 1 , wherein the transmitting of the second indication is triggered based on one or more of: (i) channel estimation performance accuracy meeting at least one threshold, and (ii) channel statistics meeting at least one threshold.

11 . A wireless transmit/receive unit (WTRU) comprising:
circuitry including any of a processor, memory, transmitter and receiver, the circuitry configured to:
receive configuration information indicating (1) a set of parameters for determining a reference signal (RS) configuration and (2) a parameter to use within the set, wherein the set of parameters comprise at least two bundling types;
receive a transmission using a first RS pattern of a first RS configuration type;
perform one or more measurements on the transmission using the first RS pattern, wherein the one or more measurements include any of: an effective channel matrix for one or more slots or resource blocks, channel estimation accuracy, and RS signaling overhead;
determine a bundling type associated with the RS configuration, based on (i) the set of parameters and (ii) the one or more measurements; and
transmit a first indication of the determined bundling type.

12 . The WTRU of claim 11 , wherein the bundling type indicates resource bundling properties across one or more of time and frequency.

13 . The WTRU of claim 11 , wherein the bundling type is determined using an artificial intelligence/machine learning (AI/ML) model.

14 . The WTRU of claim 11 , wherein the set of parameters further comprise one or more of: a RS position, RS pattern, maximum allowed RS density, minimum allowed RS density, multi-user multiple input multiple output (MU-MIMO) setting, RS configuration signaling feedback overhead, and performance requirements.

15 . The WTRU of claim 11 , wherein the circuitry is configured to:
determine one or more of a second RS pattern and a second RS configuration type, based on (i) the set of parameters and (ii) the one or more measurements; and transmit a second indication of the determined one or more of the second RS pattern and the second RS configuration type.

16 . The WTRU of claim 15 , wherein the circuitry is configured to determine one or more of the second RS pattern and the second RS configuration type using an artificial intelligence/machine learning (AI/ML) model.

17 . The WTRU of claim 16 , wherein:
the AI/ML model is trained with labelled data comprising input and output, the input comprises one or more of: (i) the effective channel matrix for one or more slots or resource blocks, and (ii) the set of parameters for determining the RS configuration, and the output comprises one or more of: the bundling type, position, pattern, and density for the second RS pattern based on performance objectives.

18 . The WTRU of claim 11 , wherein the first RS configuration type comprises one or more of:
new radio (NR) legacy demodulation reference signals (DM-RS); a uniform RS pattern; a deterministic non-uniform RS pattern; a random non-uniform RS pattern; and a non-orthogonal RS pattern.

19 . The WTRU of claim 11 , wherein the circuitry is configured to:
implicitly transmit one or more of the first indication and the second indication; explicitly transmit one or more of the first indication and the second indication; transmit one or more of the first indication and the second indication using inter-node data exchange; periodically transmit one or more of the first indication and the second indication; or semi-periodically transmit one or more of the first indication and the second indication.

20 . The WTRU of claim 11 , wherein transmitting the second indication is triggered based on one or more of: (i) channel estimation performance accuracy meeting at least one threshold, and (ii) channel statistics meeting at least one threshold.

Assignees

• Interdigital Patent Holdings Inc

Inventors

• Shojaeifard Arman
• Beluri Mihaela
• Tooher Patrick
• Lee Moon-Il
• Narayanan Thangaraj Yugeswar Deenoo
• Ibrahim Mohamed Salah
• Hemadeh Ibrahim
• Malhotra Akshay

Classifications

• G06N3/08

  Learning methods · CPC title

• G06N3/0455

  Auto-encoder networks; Encoder-decoder networks · CPC title

• H04L25/0254

  using neural network algorithms · CPC title

• H04L25/0232

  by interpolation between sounding signals · CPC title

• H04L5/0051 Primary

  of dedicated pilots, i.e. pilots destined for a single user or terminal · CPC title