KPI anomaly detection for radio access networks

1. A computing device operating as a network node in a core network of a cellular network, the computing device being configured to monitor a radio access network (RAN) of the cellular network, the RAN comprising a plurality of clusters, each cluster comprising a plurality of sites and a plurality of cells, each site communicates over one or more frequency or time divisions with cellular devices that are located in one or more cells, the computing device comprising:

one or more processors; and

a computer-readable medium storing a plurality of computer-executable components that are executable by the one or more processors to perform a plurality of actions, the plurality of actions comprising:

receiving a plurality of key performance indicator (KPI) measurements from the plurality of clusters, each of the plurality of KPI measurements generated for one of a plurality of KPI types;

receiving information identifying anomalous KPI measurements in the received KPI measurements;

for a cluster of the RAN, identifying one or more common anomalous KPI types that satisfy a ubiquity criterion, wherein a particular KPI type satisfies the ubiquity criterion when more than a threshold percentage of the sites in the cluster have anomalous KPI measurements generated for the particular KPI type;

determining, for each of the identified common anomalous KPI types for the cluster, a classification from a group of classifications, wherein at least one classification corresponds to two or more of the identified common anomalous KPI types;

ranking the identified common anomalous KPI types for the cluster based on an anomaly metric that is derived from the anomalous KPI measurements generated for each identified common anomalous KPI type;

outputting a list of sites of the cluster that are ranked based on anomaly metrics that are derived from the anomalous KPI measurements for the sites of the cluster;

receiving a selection of a site from the list of sites;

outputting a list of KPI types for the selected site that is ranked based on anomaly metrics that are derived from the anomalous KPI measurements of the plurality of KPI types for the site; and

outputting a list of common anomalous KPI types and corresponding classifications for the cluster based on the ranking.

2. The computing device of claim 1 , wherein the plurality of actions further comprises:

receiving a selection of a KPI type from the identified common anomalous KPI types of the cluster; and

outputting one or more sites having anomalous KPI measurements of the selected KPI type.

3. The computing device of claim 1 , wherein the plurality of actions further comprises:

outputting a list of cells of the cluster that is ranked based on anomaly metrics that are derived from the anomalous KPI measurements for the cells of the cluster.

4. The computing device of claim 1 , wherein the plurality of actions further comprises:

receiving a predefined set of KPI types; and

for each cluster of a set of clusters in the RAN:

determining an anomaly metric that is derived from the anomalous KPI measurements for each KPI type of the predefined set of KPI types; and

indicating a KPI status for each KPI type of the predefined set of KPI types based on the determined anomaly metric.

5. The computing device of claim 4 , wherein the plurality of actions further comprises:

receiving a selection of a KPI type in the predefined set of KPI types; and

outputting an additional list of sites of the cluster that is ranked based on an anomaly metric that is derived from the anomalous KPI measurements of the selected KPI type from each site.

6. A computer-implemented method for monitoring a radio access network (RAN) comprising a plurality of clusters, each cluster comprising a plurality of sites and a plurality of cells, each site communicating over one or more frequency or time divisions with cellular devices that are located in one or more cells, the method comprising:

receiving, at a network node of a core network associated with the RAN, a plurality of key performance indicator (KPI) measurements from the plurality of clusters, each of the plurality of KPI measurements generated for one of a plurality of KPI types;

receiving information identifying anomalous KPI measurements in the received KPI measurements;

for a cluster of the RAN, identifying one or more common anomalous KPI types that satisfy a ubiquity criterion, wherein a particular KPI type satisfies the ubiquity criterion when more than a threshold percentage of the sites in the cluster have anomalous KPI measurements generated for the particular KPI type;

determining, for each of the identified common anomalous KPI types for the cluster, a classification from a group of classifications, wherein at least one classification corresponds to two or more of the identified common anomalous KPI types;

ranking the identified common anomalous KPI types for the cluster based on an anomaly metric that is derived from the anomalous KPI measurements generated for each identified common anomalous KPI type;

outputting a list of sites of the cluster that are ranked based on anomaly metrics that are derived from the anomalous KPI measurements for the sites of the cluster;

receiving a selection of a site from the list of sites;

outputting a list of KPI types for the selected site that is ranked based on anomaly metrics that are derived from the anomalous KPI measurements of the plurality of KPI types for the site; and

outputting a list of common anomalous KPI types and corresponding classifications for the cluster based on the ranking.

7. The computer-implemented method of claim 6 , further comprising:

receiving a selection of a KPI type from the identified common anomalous KPI types of the cluster; and

outputting one or more sites having anomalous KPI measurements of the selected KPI type.

8. The computer-implemented method of claim 6 , further comprising:

outputting a list of cells of the cluster that is ranked based on anomaly metrics that are derived from the anomalous KPI measurements for the cells of the cluster.

9. The computer-implemented method of claim 6 , further comprising:

receiving a predefined set of KPI types; and

for each cluster of a set of clusters in the RAN:

determining an anomaly metric that is derived from the anomalous KPI measurements for each KPI type of the predefined set of KPI types; and

indicating a KPI status for each KPI type of the predefined set of KPI types based on the determined anomaly metric.

10. The computer-implemented method of claim 9 , further comprising:

receiving a selection of a KPI type in the predefined set of KPI types;

outputting an additional list of sites of the cluster that is ranked based on an anomaly metric that is derived from the anomalous KPI measurements of the selected KPI type from each site.

11. One or more non-transitory computer storage media of a computing device storing computer-executable instructions for monitoring a radio access network (RAN) comprising a plurality of clusters, each cluster comprising a plurality of sites and a plurality of cells, each site communicating over one or more frequency or time divisions with cellular devices that are located in one or more cells, wherein upon execution the computer-executable instructions causing one or more processors to perform acts comprising:

receiving, at a network node of a core network associated with the RAN, a plurality of key performance indicator (KPI) measurements from the plurality of clusters, each of the plurality of KPI measurements generated for one of a plurality of KPI types;

receiving information identifying anomalous KPI measurements in the received KPI measurements;

for a cluster of the RAN, identifying one or more common anomalous KPI types that satisfy a ubiquity criterion, wherein a particular KPI type satisfies the ubiquity criterion when more than a threshold percentage of the sites in the cluster have anomalous KPI measurements generated for the particular KPI type;

determining, for each of the identified common anomalous KPI types for the cluster, a classification from a group of classifications, wherein at least one classification corresponds to two or more of the identified common anomalous KPI types;

ranking the identified common anomalous KPI types for the cluster based on an anomaly metric that is derived from the anomalous KPI measurements generated for each identified common anomalous KPI type;

outputting a list of sites of the cluster that are ranked based on anomaly metrics that are derived from the anomalous KPI measurements for the sites of the cluster;

receiving a selection of a site from the list of sites;

outputting a list of KPI types for the selected site that is ranked based on anomaly metrics that are derived from the anomalous KPI measurements of the plurality of KPI types for the site; and

outputting a list of common anomalous KPI types and corresponding classifications for the cluster based on the ranking.

12. The one or more non-transitory computer-readable media of claim 11 , wherein the acts further comprise:

receiving a selection of a KPI type from the identified common anomalous KPI types of the cluster; and

outputting one or more sites having anomalous KPI measurements of the selected KPI type.

13. The one or more non-transitory computer-readable media of claim 11 , wherein the acts further comprise:

outputting a list of cells of the cluster that is ranked based on anomaly metrics that are derived from the anomalous KPI measurements for the cells of the cluster.

14. The one or more non-transitory computer-readable media of claim 11 , wherein the acts further comprise:

receiving a predefined set of KPI types; and

for each cluster of a set of clusters in the RAN:

determining an anomaly metric that is derived from the anomalous KPI measurements for each KPI type of the predefined set of KPI types; and

indicating a KPI status for each KPI type of the predefined set of KPI types based on the determined anomaly metric.

15. The one or more non-transitory computer-readable media of claim 11 , wherein the acts further comprise:

receiving a selection of a KPI type in the predefined set of KPI types;

outputting an additional list of sites of the cluster that is ranked based on an anomaly metric that is derived from the anomalous KPI measurements of the selected KPI type from each site.