Managing access points of a cellular network
There are provided a method and system to control traffic in a cellular network comprising a plurality of access points (APs) serving a plurality of user equipment devices (UEs). The method comprises: using a machine learning (ML) model to classify at least part of the APs in accordance with a first part of AP metrics thereof, thereby giving rise to a plurality of classes, each comprising peering APs; for a given class, processing AP performance metrics of peering APs classified to the given class to identify, among them, one or more first APs with negative performance variations above a variation threshold and, thereby, requiring corrective actions; and enabling one or more corrective actions with regard to the identified one or more first APs.
1. A method of controlling traffic in a wireless network, the method comprising:
using, by a processing system comprising at least one processor and at least one memory, a machine learning (ML) model for classification of one or more access points (APs) in the wireless network in accordance with a first part of AP metrics to generate a plurality of classes, each comprising peering APs;
identifying, by the processing system, one or more first APs with negative performance variations above a first variation threshold, the first variation threshold defined as a predefined percentage of a mean value of one or more AP performance metrics of all peering APs classified to a given class, and identifying one or more second APs with positive performance variations above a second variation threshold; and
enabling, by the processing system, one or more corrective actions with regard to the one or more first APs, the one or more corrective actions including aligning a configuration corresponding to a subset of the peering APs with a configuration corresponding to the one or more second APs.
2. The method of claim 1 , wherein the AP performance metrics do not belong to the first part of AP metrics.
3. The method of claim 1 , further comprising identifying, by the processing system, class-based AP performance metrics of the given class in accordance with AP performance metrics of the peering APs classified, and identifying the one or more corrective actions in accordance with the class-based AP performance metrics.
4. The method of claim 1 , wherein the identifying and the enabling are provided continuously.
5. The method of claim 4 , wherein the using is provided continuously.
6. The method of claim 1 , further comprising identifying, by the processing system, a class-based configuration of the given class, the class-based configuration corresponding to the peering APs, the one or more corrective actions including aligning a configuration of the one or more first APs with the class-based configuration.
7. The method of claim 6 further comprising:
optimizing the class-based configuration in accordance with the configuration corresponding to the one or more second APs.
8. The method of claim 1 , wherein the ML model is a deep neural network trained to classify APs using at least part of the AP metrics thereof.
9. The method of claim 1 , wherein the one or more corrective actions include a configuration update.
10. The method of claim 1 , wherein the negative performance variations to be identified and AP performance metrics usable therefor are customized for each given class.
11. A method of controlling traffic in a wireless network, the method comprising:
using, by a processing system, comprising at least one processor and at least one memory, a machine learning (ML) model for classification of one or more access points (APs) in the wireless network belonging to a given class to generate a plurality of peering APs;
identifying, by the processing system, a class-based configuration in accordance with AP metrics of APs classified to the given class, the class-based configuration comprising a mean configuration of the plurality of peering APs, the mean configuration defining a first variation threshold corresponding to a predefined percentage of a mean value of one or more AP performance metrics of all peering APs classified to a given class, and identifying one or more APs with positive performance variations above a second variation threshold; and
enabling, by the processing system, applying the class-based configuration to the plurality of peering APs, the class-based configuration including aligning configurations of a subset of the peering APs with configurations corresponding to the one or more APs.
12. A processing system to control traffic in a wireless network, the processing system comprising at least one processor and at least one memory configured to:
use a machine learning (ML) model for classification of one or more access points (APs) in the wireless network in accordance with a first part of AP metrics to generate a plurality of classes, each comprising peering APs;
process one or more first APs with negative performance variations above a first variation threshold, the first variation threshold defined as a predefined percentage of a mean value of one or more AP performance metrics of all peering APs classified to a given class, and process one or more second APs with positive performance variations above a second variation threshold; and
enable one or more corrective actions with regard to the one or more first APs, the one or more corrective actions including aligning configurations of a subset of the peering APs with configurations corresponding to the one or more second APs.
13. The processing system of claim 12 , wherein the AP performance metrics do not belong to the first part of AP metrics.
14. The processing system of claim 12 , wherein the processing system is further configured to identify class-based AP performance metrics of the given class in accordance with AP performance metrics of the peering APs classified thereto, and identify the one or more corrective actions in accordance with the class-based AP performance metrics.
15. The processing system of claim 12 , wherein the processing system is further configured to identify a class-based configuration of the given class, the class-based configuration corresponding to the peering APs, the one or more corrective actions including aligning a configuration of the one or more first APs with the class-based configuration.
16. The processing system of claim 15 , wherein the processing system is further configured to:
optimize a class-based configuration in accordance with a configuration corresponding to the one or more second APs.
17. The processing system of claim 12 , wherein the negative performance variations to be processed and AP performance metrics usable therefor are customized for each given class.
18. A non-transitory computer readable medium usable by a processing system to control traffic in a wireless network, the non-transitory computer readable medium comprising instructions that, when executed by at least one processor and at least one memory of the processing system, cause the processing system to perform operations comprising:
using a machine learning (ML) model for classification of one or more access points (APs) in the wireless network in accordance with a first part of AP metrics to generate a plurality of classes, each comprising peering APs;
processing one or more first APs with negative performance variations above a first variation threshold, the first variation threshold defined as a predefined percentage of a mean value of one or more AP performance metrics of all peering APs classified to a given class, and processing one or more second APs with positive performance variations above a second variation threshold; and
enabling one or more corrective actions with regard to the one or more first APs, the one or more corrective actions including aligning configurations of a subset of the peering APs with configurations corresponding to the one or more second APs.
19. The non-transitory computer readable medium of claim 18 , wherein the ML model is a deep neural network trained to classify APs according to the AP metrics.
20. The non-transitory computer readable medium of claim 18 , further comprising instructions for continuously updating the classification of the APs using real-time AP performance metrics.