# US 12,603,820 B2 · System and method for cellular network prediction model analysis

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

- Number: US 12,603,820 B2
- Document ID: US12603820B2
- Status: Granted
- Kind: B2
- Application number: 18/669,100
- Filed: 2024-05-20
- Published: 2025-11-20
- Granted: 2026-04-14
- Earliest claimed: 2024-05-20
- Assignees: Boost SubscriberCo L.L.C.
- Applicants: Boost SubscriberCo L.L.C.
- Inventors: Ahmed Awwad Whdan; Arnold Agcaoili; Satish Kumar Sharma; Yifei Zhao; Carl Rickard Hardy Soederberg
- Web page: https://patentgrep.com/patent/US12603820B2

## CPC

- H04L 41/147: for predicting network behaviour
- H04L 41/16: using machine learning or artificial intelligence
- H04L: TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION (arrangements common to telegraphic and telephonic communication H04M)

## Abstract

Systems and methods are directed towards determining an optimum network prediction model to employ to predict traffic loads on target nodes. A plurality of possible network prediction models and a plurality of target requirements in which to assess the plurality of possible network prediction models are selected. The plurality of target requirements are then ranked for the target nodes. For each corresponding possible network prediction model of the plurality of possible network prediction models, a total score is calculated based on scores of each corresponding target requirement and the rankings of each corresponding target requirement. An optimum network prediction model is determined from the plurality of possible network prediction models for the target nodes based on the total score of each of the plurality of possible network prediction models. The optimum network prediction model is then employed to predict the traffic loads on the target nodes in the cellular network.

## Claims

### Claim 1

A method, comprising:
selecting a plurality of possible network prediction models to predict traffic loads on target nodes in a cellular network;
selecting a plurality of target requirements in which to assess the plurality of possible network prediction models for predicting the traffic loads on the target nodes in the cellular network;
for each corresponding possible network prediction model of the plurality of possible network prediction models, scoring each corresponding target requirement of the plurality of target requirements for the corresponding possible network prediction model;
selecting a first set of target nodes in the cellular network;
for the first set of target nodes:
ranking the plurality of target requirements in which to assess the plurality of possible network prediction models with respect to the first set of target nodes;
for each corresponding possible network prediction model of the plurality of possible network prediction models, calculating a first total score for the corresponding possible network prediction model based on the scores and the ranks of each corresponding target requirement for the corresponding possible network prediction model for the first set of target nodes;
determining a first optimum network prediction model from the plurality of possible network prediction models for the first set of target nodes based on the first total score of each of the plurality of possible network prediction models; and
employing the first optimum network prediction model to predict the traffic loads on the first set of target nodes in the cellular network;
selecting a second set of target nodes in the cellular network, wherein the second set of target nodes is different from the first set of target nodes; and
for the second set of target nodes:
re-ranking the plurality of target requirements in which to assess the plurality of possible network prediction models with respect to the second set of target nodes;
for each corresponding possible network prediction model of the plurality of possible network prediction models, calculating a second total score for the corresponding possible network prediction model based on the scores and re-ranks of each corresponding target requirement for the corresponding possible network prediction model for the second set of target nodes;
determining a second optimum network prediction model from the plurality of possible network prediction models for the second set of target nodes based on the second total score of each of the plurality of possible network prediction models; and
employing the second optimum network prediction model to predict the traffic loads on the second set of target nodes in the cellular network.

### Claim 2 (depends on 1)

The method of claim 1, wherein the plurality of target requirements includes at least two of accuracy of the corresponding possible network prediction model, seasonality incorporated into the corresponding possible network prediction model, manual input into the corresponding possible network prediction model, multi-variable input into the corresponding possible network prediction model, and user ease of use of the corresponding possible network prediction model.

### Claim 3 (depends on 1)

The method of claim 1, wherein selecting the first set of target nodes further comprising:
selecting the first set of target nodes to include at least one distributed unit from a plurality of distributed units utilized by the cellular network, wherein the plurality of distributed units provide real-time support for lower layers of a protocol stack for cellular communications across the cellular network.

### Claim 4 (depends on 1)

The method of claim 1, wherein selecting the first set of target nodes further comprising:
selecting the first set of target nodes to include at least one central unit from a plurality of central units utilized by the cellular network, wherein the plurality of central units provide real-time support for higher layers of a protocol stack for cellular communications across the cellular network.

### Claim 5 (depends on 1)

The method of claim 1, wherein selecting the first set of target nodes further comprising:
selecting the first set of target nodes to include at least one sector in which one or more cells provide access to the cellular network.

### Claim 6 (depends on 1)

The method of claim 1, wherein selecting the first set of target nodes further comprising:
selecting the first set of target nodes to include at least one cell of the cellular network.

### Claim 7 (depends on 1)

The method of claim 1, wherein selecting the first set of target nodes further comprising:
selecting the first set of target nodes to include at least one radio unit of the cellular network.

### Claim 8 (depends on 1)

The method of claim 1, wherein determining the first optimum network prediction model from the plurality of possible network prediction models comprises:
selecting the first optimum network prediction model as the corresponding possible network prediction model having a highest first total score of the plurality of possible network prediction models.

### Claim 9 (depends on 1)

The method of claim 1, wherein determining the first optimum network prediction model from the plurality of possible network prediction models comprises:
identifying the first optimum network prediction model based on a comparison of the first total score of each of the plurality of possible network prediction models, wherein the first optimum network prediction model has a highest first total score.

### Claim 10

A computing device, comprising:
a memory configured to store computer instructions; and
a processor system configured to execute the computer instructions to:
select a plurality of possible network prediction models to predict traffic loads on target nodes in a cellular network;
select a plurality of target requirements in which to compare the plurality of possible network prediction models for predicting the traffic loads on the target nodes in the cellular network;
for each corresponding possible network prediction model of the plurality of possible network prediction models, score each corresponding target requirement of the plurality of target requirements for the corresponding possible network prediction model;
select a first set of target nodes in the cellular network;
for the first set of target nodes:
select a first rank for the plurality of target requirements in which to assess the plurality of possible network prediction models with respect to the first set of target nodes;
for each corresponding possible network prediction model of the plurality of possible network prediction models, calculate a first total score for the corresponding possible network prediction model based on the scores and the first rank of each corresponding target requirement for the corresponding possible network prediction model for the first set of target nodes;
determine a first optimum network prediction model from the plurality of possible network prediction models for the first set of target nodes based on the first total score of each of the plurality of possible network prediction models; and
employ the first optimum network prediction model to predict the traffic loads on the first set of target nodes in the cellular network;
select a second set of target nodes in the cellular network, wherein the second set of target nodes is different from the first set of target nodes; and
for the second set of target nodes:
select a second rank for the plurality of target requirements in which to assess the plurality of possible network prediction models with respect to the second set of target nodes;
for each corresponding possible network prediction model of the plurality of possible network prediction models, calculate a second total score for the corresponding possible network prediction model based on the scores and the second rank of each corresponding target requirement for the corresponding possible network prediction model for the second set of target nodes;
determine a second optimum network prediction model from the plurality of possible network prediction models for the second set of target nodes based on the second total score of each of the plurality of possible network prediction models; and
employ the second optimum network prediction model to predict the traffic loads on the second set of target nodes in the cellular network.

### Claim 11 (depends on 10)

The computing device of claim 10, wherein the plurality of target requirements includes at least two of accuracy of the corresponding possible network prediction model, seasonality incorporated into the corresponding possible network prediction model, manual input into the corresponding possible network prediction model, multi-variable input into the corresponding possible network prediction model, and user ease of use of the corresponding possible network prediction model.

### Claim 12 (depends on 10)

The computing device of claim 10, wherein the processor system selects the first set of target nodes by being configured to execute the computer instructions further to:
select the first set of target nodes to include at least one distributed unit from a plurality of distributed units utilized by the cellular network, wherein the plurality of distributed units provide real-time support for lower layers of a protocol stack for cellular communications across the cellular network.

### Claim 13 (depends on 10)

The computing device of claim 10, wherein the processor system selects the first set of target nodes by being configured to execute the computer instructions further to:
select the first set of target nodes to include at least one central unit from a plurality of central units utilized by the cellular network, wherein the plurality of central units provide real-time support for higher layers of a protocol stack for cellular communications across the cellular network.

### Claim 14 (depends on 10)

The computing device of claim 10, wherein the processor system selects the first set of target nodes by being configured to execute the computer instructions further to:
select the first set of target nodes to include at least one sector in which one or more cells provide access to the cellular network.

### Claim 15 (depends on 10)

The computing device of claim 10, wherein the processor system selects the first set of target nodes by being configured to execute the computer instructions further to:
select the first set of target nodes to include at least one cell of the cellular network.

### Claim 16 (depends on 10)

The computing device of claim 10, wherein the processor system selects the first set of target nodes by being configured to execute the computer instructions further to:
select the first set of target nodes to include at least one radio unit of the cellular network.

### Claim 17 (depends on 10)

The computing device of claim 10, wherein the processor system determines the first optimum network prediction model from the plurality of possible network prediction models by being configured to execute the computer instructions further to:
select the first optimum network prediction model as the corresponding possible network prediction model having a highest first total score of the plurality of possible network prediction models.

### Claim 18

A non-transitory computer-readable storage medium that stores instructions that, when executed by a processor in a computing system, cause the processor to perform actions, the actions comprising:
for each corresponding possible network prediction model of a plurality of possible network prediction models, scoring each corresponding target requirement of a plurality of target requirements for the corresponding possible network prediction model;
for a first set of target nodes in a cellular network:
ranking the plurality of target requirements in which to assess the plurality of possible network prediction models for predicting traffic loads on the first set of target nodes;
for each corresponding possible network prediction model of the plurality of possible network prediction models, calculating a first total score for the corresponding possible network prediction model based on the scores and the ranking of each corresponding target requirement for the corresponding possible network prediction model for the first set of target nodes; and
employing a first optimum network prediction model from the plurality of possible network prediction models to predict the traffic loads on the first set of target nodes in the cellular network, the first optimum network prediction model having a highest first total score; and
for a second set of target nodes in the cellular network, the second set of target nodes being different from the first set of target nodes:
re-ranking the plurality of target requirements in which to assess the plurality of possible network prediction models for predicting traffic loads on the second set of target nodes;
for each corresponding possible network prediction model of the plurality of possible network prediction models, calculating a second total score for the corresponding possible network prediction model based on the scores and the re-ranking of each corresponding target requirement for the corresponding possible network prediction model for the second set of target nodes; and
employing a second optimum network prediction model from the plurality of possible network prediction models to predict the traffic loads on the second set of target nodes in the cellular network, the second optimum network prediction model having a highest second total score.