Computing system to create radio frequency plans for wireless network deployments

• US 10,939,302 B2
• Filed: 10/31/2018
• Issued: 03/02/2021
• Est. Priority Date: 10/31/2018
• Status: Active Grant

• Current Assignee: Hewlett Packard Enterprise Development LP (Hewlett-Packard Enterprise Company)
• Original Assignee: Hewlett Packard Enterprise Development LP (Hewlett-Packard Enterprise Company)
• Inventors: Agarwal, Gopal, Gupta, Gopal
• Primary Examiner(s): Arevalo, Joseph
• Application Number:
  US16/176,636
• Publication Number:
  US 20200137585A1
• Time in Patent Office: 853 Days
• Field of Search: 455416, 455509, 4554141, 4554561, 455446, 4554566, 455423, 455450, 707769
• CPC Class Codes: G06N 20/00 Machine learning
• CPC Class Codes: H04W 16/18 Network planning tools

Abstract

The present disclosure provides an improved computing system to determine a suitability of a RF plan and a dynamic wireless network system. The method includes receiving uploads of RF plans associated with wireless network deployments which provided wireless network capability throughout each respective deployment area. Further, storing the uploaded RF plans with in a network storage cloud. In addition, training a predictive model with the stored uploaded RF plans. The dynamic wireless network system disclosed herein includes a plurality of internet-of-things devices within a deployment area. The system further includes at least one network device communicatively coupled to each of the plurality of internet-of-things devices, wherein the at least one network device has access to a trained predictive model that is to determine a number of access points and a layout of the access points within the deployment area based on a set of characteristics associated with the deployment area.

Claims

1. A computing device comprising a display screen, at least one processor, and memory having machine readable instructions stored thereon when executed causes the computing device to be configured to determine a suitability of a radio frequency (RF) plan, comprising:

receiving uploads of RF plans associated with wireless network deployments which provided wireless network capability throughout each respective deployment area;

storing the uploaded RF plans within a network storage cloud in an anonymized form such that a source of the RF plans is unknown to subsequent users of a predictive model, wherein the source comprises names, addresses, other indicia of ownership, or a combination thereof;

categorizing each of the uploaded RF plans according one of a plurality of deployment area types; and

training the predictive model with RF plans of the stored uploaded RF plans which are categorized as a particular deployment type of the plurality of deployment area types and associated with successful deployments of that particular deployment type.

2. The computing device of claim 1, employing the trained predictive model to determine the suitability of the RF plan based on the received input.

3. The computing device of claim 1, further comprising re-training the trained predictive model upon storing additional RF plans received after an initial training of the trained predictive model and employing the re-trained predictive model to determine the suitability of the RF plan based on the received input.

4. A computing device comprising a display screen, at least one processor, and memory having machine readable instructions stored thereon when executed causes the computing device to be configured to determine a suitability of a radio frequency (RF) plan, comprising:

receiving input in a trained predictive model, wherein the received input comprises a set of characteristics associated with a deployment area which is to host a deployment of a wireless network;

employing the trained predictive model to determine the suitability of the RF plan based on the received input, wherein the employing comprises performing multi-variate regression;

outputting a prediction of the employment of the trained predictive model;

monitoring the deployment area to detect a change in the set of characteristics associated with the deployment area; and

based on a determination that the change in the set of characteristics associated with the deployment area renders a present number of access points (APs) and their layout superfluous for the wireless network, issuing a notification to an administrator to remove one or more of the APs from the wireless network.

5. The computing device of claim 4, wherein the received input in the trained predictive model includes a plurality of RF plans associated with successful deployments of wireless networks.

6. The computing device of claim 5, wherein the plurality of RF plans are uploaded to a cloud network management system.

7. The computing device of claim 6, further comprising instructions to re-train the trained predictive model upon receipt of a pre-determined addition of RF plans associated with wireless network deployments.

8. The computing device of claim 4, wherein the received input comprises at least one of a floor plan of the deployment area, a description of partitions within the deployment area, a user density inside of the deployment area, and a target degree of RF interference within the deployment area.

9. The computing device of claim 4, wherein the received input comprises a proposed RF plan.

10. The computing device of claim 4, further comprising instructions to employ the trained predictive model in an event that there is a change in the set of characteristics associated with the deployment area and outputting a new prediction.

11. The computing device of claim 4, wherein the set of characteristics associated with the deployment area comprises a target floor plan of the deployment area, a description of partitions of the deployment area, a target user density associated with the deployment area, wall attributes of the deployment area, an amount of RF interference within the deployment area, or a combination thereof.

12. The computing device of claim 4, further comprising instructions to:

based on a determination that the change in the set of characteristics associated with the deployment area renders the present number of APs and their layout unsuitable for the wireless network,

generate a prediction of a suitable number, type, and layout of APs for the

deployment area with respect to the change in the set of characteristics; and

issue a notification to the administrator based on the generated prediction.

13. A computing device comprising a display screen, at least one processor, and memory having machine readable instructions stored thereon when executed by the at least one processor causes the computing device to:

receive input in a trained predictive model, wherein the received input comprises a set of characteristics associated with a deployment area which is to host a deployment of a wireless network;

employ the trained predictive model to determine a suitability of an RF plan based on the received input;

output a prediction of the employment of the trained predictive model;

monitor the deployment area to detect a change in the set of characteristics associated with the deployment area;

based on a determination that the change in the set of characteristics associated with the deployment area renders a present number of access points (APs) and their layout superfluous for the wireless network, issue a notification to an administrator to remove one or more of the APs from the wireless network; and

based on a determination that the change in the set of characteristics associated with the deployment area renders the present number of APs and their layout unsuitable for the wireless network,

generate a prediction of a suitable number, type, and layout of APs for the deployment area with respect to the change; and

issue a notification to the administrator based on the generated prediction.

14. The computing device of claim 13, wherein the received input in the trained predictive model includes a plurality of RF plans associated with successful deployments of wireless networks.

15. The computing device of claim 13, wherein the plurality of RF plans are uploaded to a cloud network management system.