Transparent roaming in virtual access point (VAP) enabled networks

1. A method, comprising:

forming, by a supervisory device in a network, a virtual access point (VAP) for a node in the network, wherein a set of access points (APs) in the network are mapped to the VAP during a VAP mapping that allows the APs to be treated as a single AP, and wherein the node communicates with the network through the VAP;

receiving, at the supervisory device and from the APs in the VAP mapping, measurements gathered by the APs regarding communications associated with the node;

identifying, by the supervisory device, a movement of the node and a predicted future location of the node based on the received measurements from the APs in the VAP mapping, wherein a machine learning model determines the predicted future location of the node using the measurements received from the APs; and

adjusting, by the supervisory device, the set of APs in the VAP mapping based on the identified movement of the node and the predicted future location of the node.

2. The method as in claim 1 , wherein the measurements from the APs in the VAP mapping comprise at least one of: a signal strength or signal quality of the communications associated with the node.

3. The method as in claim 1 , wherein identifying the movement of the node based on the received measurements from the APs in the VAP mapping comprises:

computing, by the supervisory device, a time difference of arrival of communications from the node received by the APs in the VAP mapping.

4. The method as in claim 1 , wherein identifying the movement of the node based on the received measurements from the APs in the VAP mapping comprises:

matching, by the supervisory device, the node to a particular user; and

receiving, at the supervisory device, scheduling information regarding the user, wherein the identified movement of the node is based in part on the scheduling information regarding the user.

5. The method as in claim 1 , wherein adjusting the set of APs in the VAP mapping based on the identified movement of the node comprises:

using, by the supervisory device, a machine learning model to predict an optimal set of changes to the set of APs in the VAP mapping based on the predicted future location of the node, the optimal set of changes including an addition or a removal of an AP from the set of APs that are mapped to the VAP.

6. The method as in claim 5 , further comprising:

receiving, at the supervisory device, a notification that the predicted optimal set of changes to the set of APs in the VAP mapping caused the node to roam; and

retraining, by the supervisory device, the machine learning model used to predict the future location of the node or the machine learning model used to predict the optimal set of changes to the set of APs in the VAP mapping, based on the received notification that the changes to the set of APs in the VAP mapping caused the node to roam.

7. The method as in claim 5 , wherein the machine learning model to predict an optimal set of changes to the set of APs in the VAP mapping further bases the prediction on roaming criteria used by the node to determine when to roam.

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

determining, by the supervisory device, that the measurements from the APs in the VAP mapping indicate reduced signal performance of the communications associated with the node; and

varying, by the supervisory device, the set of APs in the VAP mapping to obtain test measurements regarding the communications associated with the node.

9. An apparatus, comprising:

one or more network interfaces to communicate with a network;

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

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

form a virtual access point (VAP) for a node in the network, wherein a set of access points (APs) in the network are mapped to the VAP during a VAP mapping that allows the APs to be treated as a single AP, and wherein the node communicates with the network through the VAP;

receive, from the APs in the VAP mapping, measurements gathered by the APs regarding communications associated with the node;

identify a movement of the node and a predicted future location of the node based on the received measurements from the APs in the VAP mapping, wherein a machine learning model determines the predicted future location of the node using the measurements received from the APs; and

adjust the set of APs in the VAP mapping based on the identified movement of the node and the predicted future location of the node.

10. The apparatus as in claim 9 , wherein the measurements from the APs in the VAP mapping comprise at least one of: a signal strength or signal quality of the communications associated with the node.

11. The apparatus as in claim 9 , wherein the apparatus identifies the movement of the node based on the received measurements from the APs in the VAP mapping by:

computing a time difference of arrival of communications from the node received by the APs in the VAP mapping.

12. The apparatus as in claim 9 , wherein the apparatus identifies the movement of the node based on the received measurements from the APs in the VAP mapping by:

matching the node to a particular user; and

receiving scheduling information regarding the user, wherein the identified movement of the node is based in part on the scheduling information regarding the user.

13. The apparatus as in claim 9 , wherein the apparatus adjusts the set of APs in the VAP mapping based on the identified movement of the node by:

using machine learning model to predict an optimal set of changes to the set of APs in the VAP mapping based on the predicted future location of the node, the optimal set of changes including an addition or a removal of an AP from the set of APs that are mapped to the VAP.

14. The apparatus as in claim 13 , wherein the process when executed is further configured to:

receive a notification that the predicted optimal set of changes to the set of APs in the VAP mapping caused the node to roam; and

retrain the machine learning model used to predict the future location of the node or the machine learning model used to predict the optimal set of changes to the set of APs in the VAP mapping, based on the received notification that the changes to the set of APs in the VAP mapping caused the node to roam.

15. The apparatus as in claim 13 , wherein the machine learning model to predict an optimal set of changes to the set of APs in the VAP mapping further bases the prediction on roaming criteria used by the node to determine when to roam.

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

determine that the measurements from the APs in the VAP mapping indicate reduced signal performance of the communications associated with the node; and

vary the set of APs in the VAP mapping to obtain test measurements regarding the communications associated with the node.

17. A tangible, non-transitory, computer-readable medium storing program instructions that, when executed by a supervisory device in a network, cause the supervisory device to perform a process comprising:

forming, by the supervisory device, a virtual access point (VAP) for a node in the network, wherein a set of access points (APs) in the network are mapped to the VAP during a VAP mapping that allows the APs to be treated as a single AP, and wherein the node communicates with the network through the VAP;

receiving, at the supervisory device and from the APs, measurements gathered by the APs regarding communications associated with the node;

identifying, by the supervisory device, a movement of the node and a predicted future location of the node based on the received measurements from the APs in the VAP mapping, wherein a machine learning model determines the predicted future location of the node using the measurements received from the APs; and

adjusting, by the supervisory device, the set of APs in the VAP mapping based on the identified movement of the node and the predicted future location of the node.