Root cause analysis for service degradation in computer networks

1. A method of determining the root cause of service degradation in a network, the method comprising:

determining a window of time;

determining one or more abnormal Key Quality Indicators (KQIs) in the window;

determining one or more abnormal Key Performance Indicators (KPIs) in the window;

calculating a conditional probability that each of one or more KPIs is abnormal when a Key Quality Indicator (KQI) is normal;

calculating a conditional probability that the each of one or more KPIs is abnormal when the KQI is abnormal;

calculating a score for each KPI based upon a divergence of

a Beta distribution for the conditional probability that each of one or more KPIs is abnormal when a KQI is normal, and

a Beta distribution for the conditional probability that the each of one or more KPIs is abnormal when the KQI is abnormal; and

generating a representative root-cause list based upon the score for each KPI.

2. The method of claim 1 , wherein the step of determining one or more abnormal KQIs in the window comprises determining anomalous behavior of the KQI.

3. The method of claim 1 , wherein the step of determining one or more abnormal KQIs in the window comprises determining network alarms of the KQI and determining network alarms of the KPI.

4. The method of claim 1 , further comprising generating two or more clusters of KQIs based on root cause scores of the KPIs of each KQI, wherein each cluster comprises at least one KQI.

5. The method of claim 4 , wherein the step of generating a representative root-cause list based upon the score for each KPI comprises calculating a weighted average score of each KPI type in each cluster.

6. The method of claim 5 , wherein the step of generating a representative root-cause list based upon the score for each KPI comprises ranking the scores for each of the one or more KPIs.

7. The method of claim 4 , further comprising:

determining the size of each cluster; and

prioritizing two or more root cause recovery actions based on the size of each cluster.

8. The method of claim 1 , wherein the step of generating a representative root-cause list based upon the score for each KPI comprises ranking the scores for each of the one or more KPIs.

9. The method of claim 8 , further comprising modifying the rank of the scores for each of the one or more KPIs based upon a cost to repair each of the one or more KPIs.

10. The method of claim 1 , further comprising determining a KPI with the highest priority.

11. The method of claim 10 , wherein determining a KPI with the highest priority comprises determining the KPI with the highest rank, impact and lowest repair costs.

12. The method of claim 10 , wherein determining a KPI with the highest priority further comprises:

determining the size of each cluster of KQIs; and

prioritizing two or more recovery actions based upon the number of KQIs

determined in the size of each cluster of KQIs.

13. An administrative device for determining the root cause of service degradation in a network, the device comprising:

a network interface configured to communicate with other devices in a network;

a memory; and

a processor in communication with the network interface and the memory, the processor configured to:

determine a window of time;

determine one or more abnormal Key Quality Indicators (KQIs) in the window;

determine one or more abnormal Key Performance Indicators (KPIs) in the window;

calculate a conditional probability that each of one or more KPIs is abnormal when a Key Quality Indicator (KQI) is normal;

calculate a conditional probability that the each of one or more KPIs is abnormal when the KQI is abnormal;

calculate a score for each KPI based upon a divergence of

a Beta distribution for the conditional probability that each of one or more KPIs is abnormal when a KQI is normal, and

a Beta distribution for the conditional probability that the each of one or more KPIs is abnormal when the KQI is abnormal; and

generate a representative root-cause list based upon the score for each KPI.

14. The administrative device of claim 13 , the processor further configured to, when determining one or more abnormal KQIs in the window, determine anomalous behavior of the KQI.

15. The administrative device of claim 13 , the processor further configured to, when determining one or more abnormal KQIs in the window,

determine network alarms of the KQI; and

determine network alarms of the KPI.

16. The administrative device of claim 13 , the processor further configured to generate two or more clusters of KQIs based on root cause scores of the KPIs of each KQI, wherein each cluster comprises at least one KQI.

17. The administrative device of claim 16 , the processor further configured to, when generating a representative root-cause list based upon the score for each KPI, calculate a weighted average score of each KPI type in each cluster.

18. The administrative device of claim 17 , the processor further configured to, when generating a representative root-cause list based upon the score for each KPI, rank the scores for each of the one or more KPIs.

19. The administrative device of claim 16 , the processor further configured to:

determine the size of each cluster; and

prioritize two or more root cause recovery actions based on the size of each cluster.

20. The administrative device of claim 13 , the processor further configured to, when generating a representative root-cause list based upon the score for each KPI, rank the scores for each of the one or more KPIs.