IP Library Granted Patent US 12,726,830
Granted Patent B2
US 12,726,830 · App. 17/929,213 · Granted Sep 1, 2026

Optimizing a cellular network using a machine learning model

Inventors: Yaniv Vaknin (Tel Aviv, IL); Daniel Yellin (Raanana, IL); David Ronen (Kfar Saba, IL); Ehud Mashraky (Bruchin, IL)
Assignee: QUALCOMM Incorproated
H04W16/18H04W84/042
View Patent ↗
Loading inventors, assignments & file history…
Monitor This Case
Get email alerts when status or documents change.
Order Certified Copies
Most orders are placed with the USPTO same day — all within 24 business hours.
Order via The Patent Place →
Pre-filled with this patent's details
Quick Facts
Patent No.
US 12,726,830
App. No.
17/929,213
Granted
Sep 1, 2026
Kind
B2
Abstract

Various aspects of the present disclosure generally relate to wireless communication. In some aspects, a network node may obtain a machine learning (ML) model trained to provide one or more predicted channel interference (CI) indicators informative of channel interference between cells of a cellular network. The network node may calculate, using the ML model, the one or more predicted CI indicators using data characterizing a given cell and one or more neighbor cells of the given cell. The network node may provide the one or more predicted CI indicators. Numerous other aspects are described.

Claims (57)

1 . A method performed by a network node, comprising:

obtaining a machine learning (ML) model trained to provide one or more predicted channel interference (CI) indicators informative of channel interference between cells of a cellular network;

selecting, based on an operating bandwidth of a given cell, a first sub-model or a second sub-model as the ML model, wherein the ML model comprises a first sub-model trained for serving cells operating in accordance with a 4G protocol, and a second sub-model trained for serving cells operating in accordance with a 5G protocol;

calculating, using the ML model, the one or more predicted CI indicators using data characterizing the given cell and one or more neighbor cells of the given cell, the data used as input data to the ML model and comprising at least one of a latitude, a longitude, an azimuth of an antenna, a height of the antenna, a tilt of the antenna, or a beam configuration of the antenna of one or both of a given cell or one or more neighbor cells; and

providing the one or more predicted CI indicators.

2 . The method of claim 1 , wherein the one or more predicted CI indicators indicate an impact of the given cell on the one or more neighbor cells, or an impact of the one or more neighbor cells on the given cell.

3 . The method of claim 2 , further comprising:

configuring a cellular network based at least in part on the one or more predicted CI indicators.

4 . The method of claim 3 , wherein configuring the cellular network further comprises at least one of:

adding or removing a cell as a neighbor cell of the given cell,

adding or removing a cell as a 5G-4G anchor,

performing physical cell identifier planning,

performing root sequence index planning,

performing coverage and capacity optimization,

performing mobility load balancing,

performing self-healing, or

performing non-terrestrial network planning.

5 . The method of claim 3 , wherein configuring the cellular network further comprises identifying a placement of a cell based at least in part on the one or more predicted CI indicators.

6 . The method of claim 1 , further comprising selecting the given cell and the one or more neighbor cells as a set of cells of interest, wherein calculating, using the ML model, the one or more predicted CI indicators further comprises:

calculating the one or more predicted CI indicators for the given cell and the one or more neighbor cells based at least in part on the given cell and the one or more neighbor cells being the set of cells of interest.

7 . The method of claim 6 , wherein selecting the given cell and the one or more neighbor cells as the set of cells of interest is based at last in part on a use case associated with configuring a cellular network including the given cell and the one or more neighbor cells.

8 . The method of claim 1 , wherein calculating, using the ML model, the one or more predicted CI indicators further comprises calculating the one or more predicted CI indicators further based at least in part on:

a carrier frequency,

a bandwidth, or

a transmission power of the given cell or the one or more neighbor cells.

9 . The method of claim 1 , wherein the ML model is trained using a training set generated based at least in part on a model of a network of deployed cells, wherein the deployed cells are characterized by a location, a height, an antenna tilt, a carrier frequency, an operating bandwidth, or a transmission power.

10 . An apparatus for wireless communication at a network node, comprising:

a memory; and

one or more processors coupled to the memory, the memory and the one or more processors configured to:

obtain a machine learning (ML) model trained to provide one or more predicted channel interference (CI) indicators informative of channel interference between cells of a cellular network;

select, based on an operating bandwidth of a given cell, a first sub-model or a second sub-model as the ML model, wherein the ML model comprises a first sub-model trained for serving cells operating in accordance with a 4G protocol, and a second sub-model trained for serving cells operating in accordance with a 5G protocol;

calculate, using the ML model, the one or more predicted CI indicators using data characterizing the given cell and one or more neighbor cells of the given cell, the data used as input data to the ML model and comprising at least one of a latitude, a longitude, an azimuth of an antenna, a height of the antenna, a tilt of the antenna, or a beam configuration of the antenna of one or both of a given cell or one or more neighbor cells; and

provide the one or more predicted CI indicators.

11 . The apparatus of claim 10 , wherein the one or more predicted CI indicators indicate an impact of the given cell on the one or more neighbor cells, or an impact of the one or more neighbor cells on the given cell.

12 . The apparatus of claim 11 , wherein the memory and the one or more processors are further configured to configure a cellular network based at least in part on the one or more predicted CI indicators.

13 . The apparatus of claim 12 , wherein the memory and the one or more processors configured to configure the cellular network further comprises the memory and the one or more processors further configured to at least one of:

add or remove a cell as a neighbor cell of the given cell,

add or remove a cell as a 5G-4G anchor,

perform physical cell identifier planning,

perform root sequence index planning,

perform coverage and capacity optimization,

perform mobility load balancing,

perform self-healing, or

perform non-terrestrial network planning.

14 . The apparatus of claim 12 , wherein the memory and the one or more processors configured to configure the cellular network further comprises the memory and the one or more processors further configured to identify a placement of a cell based at least in part on the one or more predicted CI indicators.

15 . The apparatus of claim 10 , wherein the memory and the one or more processors are further configured to select the given cell and the one or more neighbor cells as a set of cells of interest, and calculate the one or more predicted CI indicators for the given cell and the one or more neighbor cells based at least in part on the given cell and the one or more neighbor cells being the set of cells of interest.

16 . The apparatus of claim 15 , wherein the memory and the one or more processors are configured to select the given cell and the one or more neighbor cells as the set of cells of interest based at last in part on a use case associated with configuring a cellular network including the given cell and the one or more neighbor cells.

17 . The apparatus of claim 10 , wherein the memory and the one or more processors are further configured to calculate, using the ML model, the one or more predicted CI indicators further based at least in part on

a transmission power of the given cell or the one or more neighbor cells.

18 . A non-transitory computer-readable medium storing a set of instructions for wireless communication, the set of instructions comprising:

one or more instructions that, when executed by one or more processors of an apparatus, cause the apparatus to:

obtain a machine learning (ML) model trained to provide one or more predicted channel interference (CI) indicators informative of channel interference between cells of a cellular network;

select, based on an operating bandwidth of a given cell, a first sub-model or a second sub-model as the ML model, wherein the ML model comprises a first sub-model trained for serving cells operating in accordance with a 4G protocol, and a second sub-model trained for serving cells operating in accordance with a 5G protocol;

calculate, using the ML model, the one or more predicted CI indicators using data characterizing the given cell and one or more neighbor cells of the given cell, the data used as input data to the ML model and comprising at least one of a latitude, a longitude, an azimuth of an antenna, a height of the antenna, a tilt of the antenna, or a beam configuration of the antenna of one or both of a given cell or one or more neighbor cells; and

provide the one or more predicted CI indicators.

19 . The non-transitory computer-readable medium of claim 18 , wherein the one or more predicted CI indicators indicate an impact of the given cell on the one or more neighbor cells, or an impact of the one or more neighbor cells on the given cell.

20 . The non-transitory computer-readable medium of claim 19 , wherein the one or more instructions further cause the apparatus to configure a cellular network based at least in part on the one or more predicted CI indicators.

Assignments (4)
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Jan 3, 2024
From: VAKNIN, YANIV; YELLIN, DANIEL; RONEN, DAVID; MASHRAKY, EHUD
To: CELLWIZE WIRELESS TECHNOLOGIES LTD.
Reel/Frame 066007/0470 →
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Jul 7, 2023
From: CELLWIZE WIRELESS TECHNOLOGIES LTD.
To: QGT INTERNATIONAL, INC.
Reel/Frame 064187/0172 →
CORRECTIVE ASSIGNMENT TO CORRECT THE FOURTH INVENTOR'S FIRST NAME PREVIOUSLY RECORDED AT REEL: 061710 FRAME: 0704. ASSIGNOR(S) HEREBY CONFIRMS THE ASSIGNMENT . Recorded Nov 23, 2022
From: VAKNIN, YANIV; YELLIN, DANIEL; RONEN, DAVID; MASHRAKY, EHUD
To: CELLWIZE WIRELESS TECHNOLOGIES LTD.
Reel/Frame 061995/0375 →
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Nov 9, 2022
From: VAKNIN, YANIV; YELLIN, DANIEL; RONEN, DAVID; MASHRAKY, UDI
To: QUALCOMM INCORPORATED
Reel/Frame 061710/0704 →
Continuity (2)
Provisional Application 63240009 · Sep 2, 2021
Related Publication 20230063522A1 · Mar 2, 2023
References Cited (22)
US 6799044B1 · Wesby · 2004 [cited by examiner]
US 9392471B1 · Thomas · 2016 [cited by examiner]
US 9980187B2 · Dribinski et al. · 2018 [cited by applicant]
US 12207114B2 · Breuer · 2025 [cited by examiner]
US 20140126403A1 · Siomina · 2014 [cited by examiner]
US 20160262065A1 · Axmon · 2016 [cited by examiner]
US 20170222746A1 · Kang · 2017 [cited by examiner]
US 20190215729A1 · Oyman · 2019 [cited by examiner]
US 20190357057A1 · Cirkic · 2019 [cited by examiner]
US 20200076520A1 · Jana · 2020 [cited by examiner]
US 20200106536A1 · Bedekar · 2020 [cited by applicant]
US 20200169895A1 · Chen et al. · 2020 [cited by applicant]
US 20200314799A1 · Henry · 2020 [cited by examiner]
US 20200322960A1 · Monajemi · 2020 [cited by examiner]
US 20200366385A1 · Ge et al. · 2020 [cited by applicant]
US 20200413316A1 · Isaksson et al. · 2020 [cited by applicant]
US 20210037540A1 · Tsuchiya · 2021 [cited by examiner]
CN 112818596A · 2021 [cited by examiner]
WO WO2016095826A1 · 2016 [cited by applicant]
WO WO2020214007A1 · 2020 [cited by examiner]
WO 2021064713A1 · 2021 [cited by applicant]
International Search Report and Written Opinion—PCT/US2022/075926—ISA/EPO—Dec. 8, 2022. [cited by applicant]