System and method for access point selection and scoring based on machine learning

Patent metadata

Field	Value
Publication number	US-10820221-B2
Application number	US-201916410284-A
Country	US
Kind code	B2
Filing date	May 13, 2019
Priority date	Apr 26, 2017
Publication date	Oct 27, 2020
Grant date	Oct 27, 2020

Abstract

A device may receive information that identifies a first set of parameter values associated with a first set of access points. The first set of access points may be associated with a set of known access point quality scores. The device may generate a model based on the set of known access point quality scores and the first set of parameter values. The device may receive information that identifies a second set of parameter values associated with a second set of access points. The device may determine a set of access point quality scores, for the second set of access points, based on the second set of parameter values and the model. The device may provide information to permit an action to be performed in association with the second set of access points.

First claim

What is claimed is:

1. A device, comprising:
one or more memories; and
one or more processors, communicatively coupled to the one or more memories, to:
receive information that identifies a set of parameter values associated with a set of access points;
determine a set of access point quality scores based on a model and the information that identifies the set of parameters values,
the set of access point quality scores being indicative of network performance of a set of access networks that correspond to the set of access points, and
the model being trained based on a set of values that corresponds to a set of cumulative distribution function (CDF) curves associated with the set of parameter values; and

provide, based on the set of access point quality scores, information to permit an action to be performed.

2. The device of claim 1 , wherein the set of parameter values include one or more of:
a throughput value,
a latency value,
a bandwidth value,
a goodput value,
a jitter value,
an amount of packet loss,
a delay value,
a reference signal received power (RSRP) value,
a received signal code power (RSCP) value,
a received signal strength indicator (RSSI) value,
a reference signal received quality (RSRQ) value,
a block error rate (BLER) value,
a signal-to-interference-plus-noise ratio (SINR) value,
a path loss value,
a link rate value,
a duration of a connection between a user device and an access point, or
a number of downloaded bytes.

3. The device of claim 1 , wherein the model is generated based on a set of known access point quality scores and another set of parameter values,
the other set of parameter values being associated with another set of access points.

4. The device of claim 3 , wherein the one or more processors are further to:
determine a set of area under a curve values based on the other set of parameter values; and
determine a set of weight values based on the set of area under the curve values and the set of known access point quality scores; and
wherein the one or more processors, when generating the model, are to:
generate the model based on the set of weight values.

5. The device of claim 1 , wherein the one or more processors are further to:
generate a set of models based on a set of time frames,
the set of models including the model; and

wherein the one or more processors, when determining the set of access point quality scores, are to:
determine the set of access point quality scores based on the set of models.

6. The device of claim 1 , wherein the action corresponds to causing a location of an access point, of the set of access points, to be adjusted.

7. The device of claim 1 , where the set of access networks are wireless local area networks.

8. A non-transitory computer-readable medium storing instructions, the instructions comprising:
one or more instructions that, when executed by one or more processors, cause the one or more processors to:
train a model based on a set of known access point quality scores and a set of known parameter values,
the set of known access point quality scores and the set of known parameter values being associated with a first set of access points, and
the model being trained based on a set of values that corresponds to a set of cumulative distribution function (CDF) curves associated with the set of known parameter values;

determine a set of access point quality scores associated with a second set of access points,
the set of access point quality scores being determined based on the model and information that identifies a set of parameter values associated with the second set of access points; and

provide, based on the set of access point quality scores, information to permit an action to be performed in association with an access point in the second set of access points.

9. The non-transitory computer-readable medium of claim 8 , wherein the set of access point quality scores is indicative of network performance of a second set of access networks that correspond to the second set of access points.

10. The non-transitory computer-readable medium of claim 9 , where the second set of access networks are wireless local area networks.

11. The non-transitory computer-readable medium of claim 8 , wherein the one or more instructions, that cause the one or more processors to provide the information to cause the action to be performed, cause the one or more processors to:
provide, to a user device, an instruction to cause the user device to automatically connect to the access point based on an access point quality score, of the set of access point quality scores, that corresponds to the access point.

12. The non-transitory computer-readable medium of claim 8 , wherein the one or more instructions, when executed by the one or more processors, further cause the one or more processors to:
adjust a set of weight values based on the set of known access point quality scores and the set of known parameter values; and
wherein the one or more instructions, that cause the one or more processors to train the model, cause the one or more processors to:
train the model based on adjusting the set of weight values.

13. The non-transitory computer-readable medium of claim 8 , wherein the one or more instructions, when executed by the one or more processors, further cause the one or more processors to:
identify a set of time frames that correspond to the set of known parameter values; and
wherein the one or more instructions, that cause the one or more processors to train the model, cause the one or more processors to:
train the model based on the set of time frames.

14. A method, comprising:
generating, by a device, a model based on a set of known access point quality scores and a first set of parameter values,
the set of known access point quality scores and the first set of parameter values being associated with a first set of access points;

determining, by the device, a set of access point quality scores, for a second set of access points, based on a second set of parameter values and the model,
the set of access point quality scores being indicative of network performance of a second set of access networks that correspond to the second set of access points, and
the model being trained based on a set of values that corresponds to a set of cumulative distribution function (CDF) curves associated with the set of parameter values; and

providing, by the device, information to permit an action to be performed by a user device or an access point in the second set of access points.

15. The method of claim 14 , further comprising:
identifying a set of time frames associated with the first set of parameter values; and
wherein generating the model comprises:
generating the model, and at least one other model, based on the set of time frames.

16. The method of claim 14 , further comprising:
receiving, from the user device, information that identifies the second set of access networks; and
providing, to the user device, information that identifies the set of access point quality scores associated with the second set of access networks.

17. The method of claim 14 , wherein providing the information to permit the action to be performed comprises:
providing, to the user device, an instruction to cause the user device to connect to an access network in the second set of access networks.

18. The method of claim 14 , further comprising:
determining a set of area under a curve values based on the second set of parameter values;

Assignees

• Verizon Patent & Licensing Inc

Inventors

• Konstantinou Kyriaki
• Ouyang Ye

Classifications

• H04W24/08 Primary

  Testing, {supervising or monitoring} using real traffic · CPC title

• H04W24/02

  Arrangements for optimising operational condition · CPC title

• H04W24/04

  Arrangements for maintaining operational condition · CPC title

• H04W60/00

  Affiliation to network, e.g. registration; Terminating affiliation with the network, e.g. de-registration · CPC title