Granted Patent B2

US 12,507,088 · App. 18/364,930 · Granted Dec 23, 2025

Inventors: Denzel Daniel Arthur (Stewartsville, NJ); Andrew James McNeil (Morristown, NJ); Tsung-Ying Lu (Parsippany, NJ)

Assignee: T-Mobile USA, Inc.

H04W24/02H04W28/16

Abstract

Claims (104)

1 . A non-transitory, computer-readable storage medium storing instructions recorded thereon that, when executed by at least one data processor of a system of a telecommunications network, cause the system to:

detect network performance in a region of the telecommunications network,

wherein the region includes a set of cell areas that support subscribers of the telecommunications network;

determine one or more low-performing cell areas and one or more high-performing cell areas of the set of cell areas by using a network performance model,

wherein the network performance model assigns performance scores to cell areas based on user experience resources,

wherein the one or more low-performing cell areas comprise performance scores that are less than a lower bound of an acceptable range, and

wherein the one or more high-performing cell areas comprise performance scores that are higher than an upper bound of the acceptable range;

determine common subscribers between the one or more low-performing cell areas and the one or more high-performing cell areas,

wherein the common subscribers subscribe to both a low-performing cell area of the one or more low-performing cell areas and a high-performing cell area of the one or more high-performing cell areas;

identify a recipient cell area of the one or more low-performing cell areas and a donor cell area of the one or more high-performing cell areas based on a number of common subscribers between the recipient cell area and the donor cell area,

wherein the number of common subscribers between the recipient cell area and the donor cell area is greater than numbers of common subscribers of other cell area pairings between the one or more low-performing cell areas and the one or more high-performing cell areas;

determine a reallocation of donor user experience resources of the donor cell area to the recipient cell area by using the network performance model to assign an increased performance score to the recipient cell area and a decreased performance score to the donor cell area based on changes to one or more cell site parameters of the donor cell area and to one or more cell site parameters of the recipient cell area,

wherein the increased performance score and the decreased performance score are both above the lower bound of the acceptable range; and

generate an indication that the reallocation of the donor user experience resources will improve a first user experience for a first group of subscribers of the recipient cell area without impairing a second user experience for a second group of subscribers of the donor cell area.

2 . The non-transitory, computer-readable storage medium of claim 1 , wherein the user experience resources comprise at least one of:

a total number of user connections in a cell area;

an average speed among user connections in the cell area; or

an average payload supported by the cell area for the user connections.

3 . The non-transitory, computer-readable storage medium of claim 1 , wherein the one or more cell site parameters comprise at least one of:

power to an antenna;

parameter used to move subscribers between the one or more low-performing cell areas and the one or more high-performing cell areas;

parameter to define user preference toward specific cell area; or

a tilt of the antenna.

4 . The non-transitory, computer-readable storage medium of claim 1 , further causing the system to:

present the indication as a recommendation at a user device,

wherein the recommendation is associated with a control configured to accept an input from a user of the user device to accept, refute, or revise the reallocation.

5 . The non-transitory, computer-readable storage medium of claim 1 , wherein the increased performance score is a predicted improvement, wherein the decreased performance score is a predicted decrease, and wherein the system is further caused to:

cause the reallocation of the user experience resources of the donor cell area to the recipient cell area; and

determine an actual improved performance score assigned to the recipient cell area and an actual decreased performance score assigned to the donor cell area using the network performance model.

6 . The non-transitory, computer-readable storage medium of claim 5 , further causing the system to:

determine an evaluation of the network performance model based on a first comparison of the predicted improvement and the actual improved performance score, and based on a second comparison of the predicted decrease and the actual decreased performance score; and

based on the evaluation, revise the network performance model to account for the actual improved performance score and for the actual decreased performance score.

7 . The non-transitory, computer-readable storage medium of claim 1 , wherein determining the common subscribers further causes the system to:

rank subscribers for the one or more low-performing cell areas and for the one or more high-performing cell areas based on a number of attempted connections made by the subscribers over a time period;

identify top subscribers who occupy top positions in both the one or more high-performing cell areas and the one or more low-performing cell areas,

wherein the top positions are positions above a baseline number of attempted connections, or above a baseline position; and

determine the common subscribers based on the top subscribers.

8 . The non-transitory, computer-readable storage medium of claim 1 , wherein the network performance model comprises a set of weights based on a desired performance metric, wherein the desired performance metric biases the user experience resources in favor of one or more preferred resources over one or more unpreferred resources, and wherein prior to using the network performance model to determine low-performing cell areas and high-performing cell areas, the system is caused to determine the desired performance metric.

9 . The non-transitory, computer-readable storage medium of claim 8 , wherein determining the desired performance metric is based on an input to a user device received from a user, or on the user experience resources of a cell area over a time period.

10 . The non-transitory, computer-readable storage medium of claim 1 , wherein the system detects the network performance in the region by receiving an alert, and wherein the alert is triggered by the user experience resources of the set of cell areas of the region crossing a performance threshold.

11 . A system comprising:

at least one hardware processor; and

at least one non-transitory memory storing instructions that, when executed by the at least one hardware processor, cause the system to:

detect network performance in a set of cells of a telecommunications network;

determine low-performing cells and high-performing cells of the set of cells by using a performance model to assign performance scores to the set of cells based on user resources and cell parameters of the set of cells,

wherein low performance scores of the low-performing cells are below a first bound of a range,

and wherein high-performance scores of the high-performing cells are above a second bound of the range;

determine common subscribers between the low-performing cells and the high-performing cells,

wherein the common subscribers subscribe to both the low-performing cells and the high-performing cells;

identify a recipient cell of the low-performing cells and a donor cell of the high-performing cells based on a number of common subscribers shared by the recipient cell and the donor cell,

wherein the number of common subscribers is greater than numbers of common subscribers shared by other cells of the low-performing cells and the high-performing cells;

determine a reallocation of donor user resources of the donor cell to the recipient cell to cause an improvement to the recipient cell without causing an impairment to the donor cell,

wherein the improvement is measurable by a first performance score above the first bound of the range assigned to the recipient cell, and

wherein the impairment is measurable by a second performance score above the first bound of the range assigned to the donor cell; and

generate an indication that the reallocation of the donor user resources will improve first user experiences of the recipient cell without diminishing second user experiences of the donor cell.

12 . The system of claim 11 , wherein the user resources of the set of cells comprise at least one of:

a total number of user connections in a cell area;

an average speed among user connections in the cell area; or

an average payload supported by the cell area for the user connections,

wherein the cell parameters comprise power to an antenna and a tilt of the antenna.

13 . The system of claim 11 , further caused to:

present the indication as a recommendation at a user device,

wherein the recommendation is associated with a control that is configured to accept an input from a user of the user device to accept, refute, or revise the reallocation.

14 . The system of claim 11 , wherein the improvement is a predicted improvement, wherein the impairment is a predicted decrease, and wherein the system is further caused to:

cause the reallocation of donor user resources to the recipient cell;

determine an actual improved performance score assigned to the recipient cell and an actual decreased performance score assigned to the donor cell using the performance model;

determine an evaluation of the performance model based on a first comparison of the predicted improvement and the actual improved performance score, and based on a second comparison of the predicted decrease and the actual decreased performance score; and

based on the evaluation, revise the performance model to account for the actual improved performance score and for the actual decreased performance score.

15 . The system of claim 11 , wherein determining the common subscribers further causes the system to:

rank subscribers for the low-performing cells and for the high-performing cells based on a number of attempted connections made by the subscribers over a time period;

identify top subscribers from subscribers who occupy top positions in both the high-performing cells and the low-performing cells,

wherein the top positions are positions above a baseline number of attempted connections, or above a baseline position; and

determine the common subscribers based on the top subscribers.

16 . The system of claim 11 :

wherein the performance model comprises a set of weights based on a desired performance metric,

wherein the desired performance metric biases the user resources in favor of one or more preferred resources over one or more unpreferred resources, and

wherein prior to using the performance model to determine low-performing cells and high-performing cells, the system is caused to:

determine the desired performance metric,

wherein determining the desired performance metric is based on an input received from an entity or based on a measure of the user resources of a cell area over a time period.

17 . The system of claim 11 , wherein the system detects the network performance by receiving an alert triggered by user resources of the set of cells crossing a performance threshold.

18 . A method comprising:

detecting network performance in a set of cell areas of a telecommunications network,

determining low-performing areas and high-performing areas of the set of cell areas by using a performance model to assign performance scores to the set of cell areas based on user experience resources and parameters of the set of cell areas,

wherein the low-performing areas comprise low scores below a first bound of a range,

wherein the high-performing areas comprise high scores above a second bound of a range;

determining common subscribers between the low-performing areas and the high-performing areas,

wherein the common subscribers subscribe to both a low-performing area of the low-performing areas and a high-performing area of the high-performing cell areas;

identifying a recipient area of the low-performing area and a donor area of the high-performing area based on a number of common subscribers between the recipient area of the donor area,

wherein the number of common subscribers between the recipient area and the donor area is greater than numbers of common subscribers of other area pairings between the low-performing areas and the high-performing areas;

determining a reallocation of resources from the donor area to recipient area by using the network performance model to assign an increased performance score to the recipient area and a decreased performance score to the donor area based on changes to one or more cell site parameters of the donor area and to one or more cell site parameters of the recipient area to cause an improvement to the low-performing area without impairing the high-performing area;

wherein the increased performance score and the decreased performance score are both above the first bound of the range; and

generating an indication that the reallocation will improve user experiences of the recipient area without diminishing user experiences of the donor area.

19 . The method of claim 18 , wherein identifying a low-performing area and a high-performing area based on common subscribers comprises:

ranking subscribers for the low-performing areas and for the high-performing areas based on a number of attempted connections made by the subscribers over a time period;

identifying top subscribers from subscribers who occupy top positions in both the high-performing areas and the low-performing areas,

wherein top positions are positions above a baseline number of attempted connections, or above a baseline position; and

determining the common subscribers based on the top subscribers.

20 . The method of claim 18 :

wherein the performance model comprises a set of weights based on a desired performance metric,

wherein the desired performance metric biases the resources in favor of one or more preferred resources over one or more unpreferred resources, and

wherein prior to using the performance model to assign scores to the set of areas,

the method further comprises:

determining the desired performance metric, and

wherein determining the desired performance metric is based on an input received from an entity or based on a measure of the resources of a cell area over a time period.

Assignments (1)

Continuity (1)

References Cited (80)