Automatically using configuration management analytics in cellular networks
A method includes partitioning a set of configuration management (CM) data for one or more cellular network devices into multiple distinct time intervals, each time interval associated with a distinct set of CM settings at the one or more cellular network devices, the CM data comprising multiple CM parameters. The method also includes determining a regression model based on the set of CM data. The method also includes applying the regression model to compute a distinct set of scores and compare the set of scores to estimate whether a performance of the one or more cellular network devices has changed during a second time interval relative to a first time interval.
1. A method comprising:
partitioning a set of configuration management (CM) data for one or more cellular network devices into multiple distinct time intervals, each time interval associated with a distinct set of CM settings at the one or more cellular network devices, the CM data comprising multiple CM parameters;
determining a regression model based on the set of CM data; and
applying the regression model to compute a distinct set of scores and compare the set of scores to estimate whether a performance of the one or more cellular network devices has changed during a second time interval relative to a first time interval.
2. The method of claim 1 , further comprising:
in response to estimating that the performance of the one or more cellular network devices has degraded during the second time interval, performing root cause analysis (RCA) including statistically analyzing behavior of a network cell before and after a CM change to determine whether the degradation in the performance of the one or more cellular network devices is attributable to the CM change.
3. The method of claim 1 , further comprising:
determining at least one recommended parameter configuration to optimize a target key performance indicator (KPI) or key quality indicator (KQI).
4. The method of claim 3 , further comprising:
performing parameter sensitivity analysis comprising computing a respective sensitivity of each CM parameter of the set of CM data based on an average impact of that CM parameter on the target KPI or KQI while maintaining at least some other factors.
5. The method of claim 1 , wherein determining the regression model comprises:
selecting one or more regressor key performance indicators (KPIs) for the regression model to avoid masking one or more CM parameter effects.
6. The method of claim 1 , further comprising:
performing at least one of multiple data preprocessing operations on the set of CM data, the multiple data preprocessing operations comprising (i) removing invalid data samples, (ii) normalizing or scaling the CM data, (iii) removing trends or seasonality in the CM data, (iv) generating additional synthetic features from existing KPIs in the CM data, and (v) selecting a subset of the CM data associated with a specific timeframe or a specific group of the cellular network devices.
7. The method of claim 1 , wherein the regression model comprises a one-group regression or a two-group regression.
8. A device comprising:
a transceiver configured to transmit and receive information; and
a processor operably connected to the transceiver, the processor configured to:
partition a set of configuration management (CM) data for one or more cellular network devices into multiple distinct time intervals, each time interval associated with a distinct set of CM settings at the one or more cellular network devices, the CM data comprising multiple CM parameters;
determine a regression model based on the set of CM data; and
apply the regression model to compute a distinct set of scores and compare the set of scores to estimate whether a performance of the one or more cellular network devices has changed during a second time interval relative to a first time interval.
9. The device of claim 8 , wherein the processor is further configured to:
in response to estimating that the performance of the one or more cellular network devices has degraded during the second time interval, perform root cause analysis (RCA) including statistically analyzing behavior of a network cell before and after a CM change to determine whether the degradation in the performance of the one or more cellular network devices is attributable to the CM change.
10. The device of claim 8 , wherein the processor is further configured to:
determine at least one recommended parameter configuration to optimize a target key performance indicator (KPI) or key quality indicator (KQI).
11. The device of claim 10 , wherein the processor is further configured to:
perform parameter sensitivity analysis in which the processor computes a respective sensitivity of each CM parameter of the set of CM data based on an average impact of that CM parameter on the target KPI or KQI while maintaining at least some other factors.
12. The device of claim 8 , wherein to determine the regression model, the processor is further configured to:
select one or more regressor key performance indicators (KPIs) for the regression model to avoid masking one or more CM parameter effects.
13. The device of claim 8 , wherein the processor is further configured to:
perform at least one of multiple data preprocessing operations on the set of CM data, the multiple data preprocessing operations comprising (i) removing invalid data samples, (ii) normalizing or scaling the CM data, (iii) removing trends or seasonality in the CM data, (iv) generating additional synthetic features from existing KPIs in the CM data, and (v) selecting a subset of the CM data associated with a specific timeframe or a specific group of the cellular network devices.
14. The device of claim 8 , wherein the regression model comprises a one-group regression or a two-group regression.
15. A non-transitory computer readable medium comprising program code that, when executed by a processor of a device, causes the device to:
partition a set of configuration management (CM) data for one or more cellular network devices into multiple distinct time intervals, each time interval associated with a distinct set of CM settings at the one or more cellular network devices, the CM data comprising multiple CM parameters;
determine a regression model based on the set of CM data; and
apply the regression model to compute a distinct set of scores and compare the set of scores to estimate whether a performance of the one or more cellular network devices has changed during a second time interval relative to a first time interval.
16. The non-transitory computer readable medium of claim 15 , further comprising program code that, when executed by the processor of the device, causes the device to:
in response to estimating that the performance of the one or more cellular network devices has degraded during the second time interval, perform root cause analysis (RCA) including statistically analyzing behavior of a network cell before and after a CM change to determine whether the degradation in the performance of the one or more cellular network devices is attributable to the CM change.
17. The non-transitory computer readable medium of claim 15 , further comprising program code that, when executed by the processor of the device, causes the device to:
determine at least one recommended parameter configuration to optimize a target key performance indicator (KPI) or key quality indicator (KQI).
18. The non-transitory computer readable medium of claim 17 , further comprising program code that, when executed by the processor of the device, causes the device to:
perform parameter sensitivity analysis in which the processor computes a respective sensitivity of each CM parameter of the set of CM data based on an average impact of that CM parameter on the target KPI or KQI while maintaining at least some other factors.
19. The non-transitory computer readable medium of claim 15 , wherein the program code that causes the device to determine the regression model comprises program code that causes the device to:
select one or more regressor key performance indicators (KPIs) for the regression model to avoid masking one or more CM parameter effects.
20. The non-transitory computer readable medium of claim 15 , further comprising program code that, when executed by the processor of the device, causes the device to:
perform at least one of multiple data preprocessing operations on the set of CM data, the multiple data preprocessing operations comprising (i) removing invalid data samples, (ii) normalizing or scaling the CM data, (iii) removing trends or seasonality in the CM data, (iv) generating additional synthetic features from existing KPIs in the CM data, and (v) selecting a subset of the CM data associated with a specific timeframe or a specific group of the cellular network devices.