System and method for adaptive access and handover configuration based on prior history in a multi-RAT environment
View Patent ↗Systems and methods for adaptive access and handover configuration based on historical data are provided. Access and handover decisions are optimized in a multiple radio access technology environment using historical data associated with network performance. Future needs for access and handovers are predicted using historical data associated with the user and historical data associated with network performance. Performance metrics are received periodically or continuously from nodes in one or more networks at a centralized controller. The centralized multi RAT controller correlates these performance metrics and determines predicted handovers for a user device. Preparations for the predicted handovers can then be made prior to the handover event.
1. A method, in a network controller, for implementing handover decisions based on performance data, comprising:
predicting, by the network controller, an application that is to be run at a future time on a user device served by a first network access device based on an affinity group of user devices for the user device, wherein the affinity group is generated based on usage patterns for the user device that track use of the application that is to be run at the future time on the user device;
retrieving performance metrics for the application that is to be run at the future time on the user device, wherein the performance metrics for the application that is predicted, by the network controller, to be run at the future time on the user device are determined based on observations of a plurality of potential paths and each of the plurality of potential paths comprises a set of network nodes;
determining a handover path to a second network access device for the application that is to be run at the future time on the user device based on the performance metrics and historical travel patterns for the user device; and
communicating the handover path to the first network access device serving the user device.
2. The method of claim 1 , further comprising:
receiving the performance metrics for the application from a plurality of user devices.
3. The method of claim 2 , further comprising:
receiving the performance metrics for the application from a network node.
4. The method of claim 3 , further comprising:
compiling the performance metrics for the application; and
generating a pathway performance metrics profile for the application based on the compiled performance metrics for the application.
5. The method of claim 1 , wherein the performance metrics include observed network latencies for the plurality of potential paths, observed error rates for the plurality of potential paths, or observed quality of service (QoS) data for the plurality of potential paths.
6. The method of claim 1 , wherein the first network access device is a base station.
7. The method of claim 1 , wherein the second network access device is an access point.
8. The method of claim 4 , wherein the pathway performance metrics profile for the application includes pathway performance metrics for the plurality of potential paths.
9. The method of claim 8 , wherein the pathway performance metrics of a path of the plurality of potential paths includes performance metrics associated with each network node of the set of networks nodes of the path.
10. The method of claim 1 , wherein predicting the application that is to be run at the future time on the user device includes:
predicting the application that is to be run at the future time on the user device based on the usage patterns.
11. A network controller, comprising:
a processor; and
a storage coupled to the processor,
wherein the processor is configured to compile received performance metrics for an application and generate a pathway performance metrics profile for the application based on the compiled performance metrics for the application,
wherein the compiled performance metrics for the application are determined based on observations of a plurality of potential paths,
wherein the storage is configured to store the pathway performance metrics profile for the application,
wherein the processor is further configured to predict that the application is to be run at a future time on a user device based on an affinity group that identifies user devices based on the user device, retrieve the pathway performance metrics profile for the application from the storage, and determine a handover path for the user device based on the pathway performance metrics profile for the application and historical travel patterns for the user device, and
wherein the affinity group is generated based on usage patterns for the user device that track use of the application.
12. The network controller of claim 11 , wherein the pathway performance metrics profile for the application includes pathway performance metrics for the plurality of potential paths.
13. The network controller of claim 12 , wherein the pathway performance metrics of a path of the plurality of potential paths includes performance metrics associated with each network node of a set of networks nodes of the path.
14. The network controller of claim 11 , wherein the compiled performance metrics include observed network latency, observed error rates, or observed quality of service (QoS) data.
15. The network controller of claim 12 , wherein the processor is further configured to select a path from the plurality of potential paths as the handover path based on the pathway performance metrics for the plurality of potential paths.
16. The network controller of claim 11 , wherein the user device is served by a first network access device and the handover path includes a second network access device, and wherein the first network access device and the second access device utilize different radio access technologies.
17. The network controller of claim 11 , wherein the network controller is shared by a plurality of networks.
18. A method, in a network controller, for implementing handover decisions based on performance data, comprising:
retrieving performance metrics for an application that is to be run at a future time on a user device, wherein the performance metrics for the application are determined based on observations of a plurality of potential paths and each of the plurality of potential paths comprises a set of network nodes, wherein the application that is to be run at the future time on the user device is determined based on an affinity group that identifies user devices based on the user device, wherein the affinity group is generated based on usage patterns for the user device that track use of the application that is to be run at the future time on the user device;
determining a handover path for the application that is to be run at the future time on the user device based on the performance metrics for the application and historical travel patterns for the user device; and
communicating the handover path to a network access device serving the user device.
19. The method of claim 18 , further comprising:
receiving the performance metrics for the application from a plurality of user devices.
20. The method of claim 19 , further comprising:
receiving the performance metrics for the application from a network node.