Detecting stable wireless conditions to rebalance AP loads in large (conference) rooms

1. A method, comprising:

receiving, at a supervisory service for a wireless network, telemetry data indicative of client movement of wireless clients of the network within a location;

determining, by the supervisory service, a measure of client movement stability for the location, by using a time series of the received telemetry data as input to a regression model;

selecting, by the supervisory service, one or more of the clients for wireless access point reassignment, when the determined measure of client movement stability, using the regression model, is predicted to be above or below a predefined threshold for a specified amount of time; and

causing, by the supervisory service, the selected one or more clients to switch from a first wireless access point at the location to a second wireless access point at the location.

2. The method as in claim 1 , wherein the supervisory service selects the one or more clients for wireless access point reassignment, when the regression model predicts that the measure of client movement stability for the location will remain above the predefined threshold for the specified amount of time.

3. The method as in claim 1 , further comprising:

calculating measures of client movement stability for each of the one or more clients, wherein the one or more clients are selected for wireless access point reassignment based on their measures of client movement stability and their locations relative to the second wireless access point.

4. The method as in claim 1 , wherein determining the measure of client movement stability comprises:

using the telemetry data as input to a machine learning-based anomaly detection model.

5. The method as in claim 1 , wherein causing the selected one or more clients to switch from the first wireless access point at the location to the second wireless access point at the location comprises:

disconnecting the selected one or more clients from the first wireless access point.

6. The method as in claim 1 , wherein causing the selected one or more clients to switch from the first wireless access point at the location to the second wireless access point at the location comprises:

instructing the first wireless access point to lower its transmit power.

7. The method as in claim 1 , wherein the first wireless access point is located closer to an entryway of the location than the second wireless access point is to the entryway.

8. The method as in claim 1 , wherein the one or more clients are selected for wireless access point reassignment based on traffic profiles of the one or more clients.

9. An apparatus, comprising:

one or more network interfaces to communicate with a wireless network;

a processor coupled to the network interfaces and configured to execute one or more processes; and

a memory configured to store a process executable by the processor, the process when executed configured to:

receive telemetry data indicative of client movement of wireless clients of the network within a location;

determine a measure of client movement stability for the location, by using the received telemetry data by using a time series of the received telemetry data as input to a regression model;

select one or more of the clients for wireless access point reassignment, when the determined measure of client movement stability, using the regression model, is predicted to be above or below a predefined threshold for a specified amount of time; and

cause the selected one or more clients to switch from a first wireless access point at the location to a second wireless access point at the location.

10. The apparatus as in claim 9 , wherein the apparatus selects the one or more clients for wireless access point reassignment, when the regression model predicts that the measure of client movement stability for the location will remain above the predefined threshold for the specified amount of time.

11. The apparatus as in claim 9 , wherein the process when executed is further configured to:

calculate measures of client movement stability for each of the one or more clients, wherein the one or more clients are selected for wireless access point reassignment based on their measures of client movement stability and their locations relative to the second wireless access point.

12. The apparatus as in claim 9 , wherein the apparatus determines the measure of client movement stability by:

using the telemetry data as input to a machine learning-based anomaly detection model.

13. The apparatus as in claim 9 , wherein the apparatus causes the selected one or more clients to switch from the first wireless access point at the location to the second wireless access point at the location by:

disconnecting the selected one or more clients from the first wireless access point.

14. The apparatus as in claim 9 , wherein the apparatus causes the selected one or more clients to switch from the first wireless access point at the location to the second wireless access point at the location by:

instructing the first wireless access point to lower its transmit power.

15. The apparatus as in claim 9 , wherein the first wireless access point is located closer to an entryway of the location than the second wireless access point is to the entryway.

16. The apparatus as in claim 9 , wherein the one or more clients are selected for wireless access point reassignment based on traffic profiles of the one or more clients.

17. A tangible, non-transitory, computer-readable medium storing program instructions that cause a supervisory service for a wireless network to execute a process comprising:

receiving, at the supervisory service for the wireless network, telemetry data indicative of client movement of wireless clients of the network within a location;

determining, by the supervisory service, a measure of client movement stability for the location, by using a time series of the received telemetry data as input to a regression model;

selecting, by the supervisory service, one or more of the clients for wireless access point reassignment, when the determined measure of client movement stability, using the regression model, is predicted to be above or below a predefined threshold for a specified amount of time; and

causing, by the supervisory service, the selected one or more clients to switch from a first wireless access point at the location to a second wireless access point at the location.

18. The computer-readable medium as in claim 17 , wherein determining the measure of client movement stability comprises:

using the telemetry data as input to a machine learning-based anomaly detection model.