System and method for identifying deviating cell-sites

1. A method of identifying idiosyncratic performance-outliers amongst a plurality of cell-sites in a wireless network, the method comprising:

selecting cell-sites;

selecting, for each cell-site, at least one sector-carrier;

selecting, for each sector-carrier, one or more key performance indicators (KPIs);

determining, for each KPI, a historical value and a recent value based on samples of the KPI taken during a historical period and a recent period, respectively, the recent period being a subset of the historical period;

determining, for each KPI, a delta representing a difference between the historical value and the recent value thereby determining KPI-specific sets of deltas, respectively;

normalizing each set of deltas;

selecting, for each set of deltas, a KPI-specific threshold;

comparing, for each set of deltas, members thereof against the corresponding threshold; and

identifying, for each set of deltas, which combinations of cell-site and sector-carrier are KPI-specific performance outliers based on the comparison, respectively.

2. The method of claim 1 , further comprising:

sorting each normalized set of deltas into numerical order.

3. The method of claim 1 , wherein at least one of the following is true:

the historical period is 60 days; or

the recent period is 24 hours.

4. The method of claim 3 , wherein:

the historical value is representative on a daily basis.

5. The method of claim 4 , wherein:

the historical value is one of:

average;

correlation; or

median.

6. The method of claim 1 , wherein the at least one KPI includes at least one of:

a count of instances of a dropped connection between the cell-site and a wireless device;

a count of instances of a failed attempt to establish a connection between the cell-site and a wireless device;

a count of minutes of connection time for all instances of a connection between the cell-site and a wireless device;

a count of data consumed for all instances of a connection between the cell-site and a wireless device;

a value representative of throughput on a per-user basis for all instances of a connection between the cell-site and a wireless device; or

a count of simultaneously active instances of a connection between the cell-site and a wireless device.

7. The method of claim 1 , wherein the determining a historical value for each KPI includes:

filtering the samples taken during the historical period.

8. The method of claim 7 , wherein the filtering, for each KPI, includes:

determining a standard deviation (STD) for the historical period; and

excluding samples having a magnitude (m), K 1 *STD≦m, where K 1 is a real number and K 1 is about 5.

9. The method of claim 7 , wherein the filtering, for each KPI, includes:

determining an average (AVG) for the historical period; and

excluding samples having a magnitude (m), K 2 *AVG≦m, where K 2 is a real number and K 2 is about 1.9.

10. The method of claim 1 , wherein the normalizing includes:

using a non-linear transfer function to normalize a given set of deltas.

11. The method of claim 10 , wherein:

the non-linear transfer function approximates an exponential function, y=ab x ;

wherein:

a is non-zero; and

b is a positive real number, b>0.

12. The method of claim 10 , wherein

the normalizing produces, for members of each set of deltas, corresponding scaled values; and

the normalizing, for a given set of deltas, further includes:

using a range, R, for the scaled values, where R is an integer and 0≦R≦max(range), and max(range)=100; and

sub-normalizing, within the scaled range, into sub-ranges including:

sub-normalizing instances of delta having values, X, onto a sub-range, r 1 , of the range R, where X ⊂ given_set_of_deltas, 0≦r 1 <BOUNDARY 1 <max(range), 0≦X<REF 1 , and BOUNDARY 1 and REF 1 are real numbers, respectively;

sub-normalizing instances of delta having values, Y, onto a sub-range, r 2 , of the range R, where Y ⊂ given_set_of_deltas, max(X)<min(Y), BOUNDARY1≦r 2 <BOUNDARY 2 , BOUNDARY 2 <max(range), REF 1 ≦Y<REF 2 , and BOUNDARY 2 and REF 2 are real numbers, respectively; and

sub-normalizing instances of delta having values, Z, onto a sub-range, r 3 , of the range R, where Z ⊂ given_set_of_deltas, max(Y)<min(Z), BOUNDARY2≦r 3 ≦max(range), REF 2 ≦Z≦max(delta), max(delta) is a maximum value of delta, and max(range) and max(delta) are real numbers.

13. The method of claim 1 , further comprising:

assigning weights to at least one of:

the selected one or more KPIs, respectively; or

the selected at least one sector-carrier, respectively.

14. The method of claim 1 , further comprising:

recommending, for each performance-outlying combination of cell-site and sector-carrier, a change in one or more operational settings intended to decrease the corresponding degree to which the combination is a performance-outlier.

15. The method of claim 1 , further comprising:

selecting, for each cell-site, at least one cell-sector; and

wherein the selecting at least one sector-carrier includes:

selecting, for each cell-sector, at least one sector-carrier.

16. The method of claim 1 , further comprising:

selecting at least one cluster, each cluster including cell-sites, respectively; and

wherein the selecting cell-sites includes:

selecting, for each of the at least one cluster, amongst the cell-sites thereof.

17. The method of claim 16 , further comprising:

selecting at least one market, each market including clusters, respectively; and

wherein the selecting at least one cluster includes:

selecting, for each of the at least one market, amongst the clusters thereof.

18. The method of claim 17 , further comprising:

selecting at least one region, each region including markets, respectively; and

wherein the selecting at least one market includes:

selecting, for each of the at least one region, amongst the markets thereof.

19. An apparatus for identifying idiosyncratic performance-outliers amongst a plurality of cell-sites in a wireless network, the apparatus comprising:

a user-interface (UI) configured to:

facilitate a selection of cell-sites;

facilitate a selection, for each cell-site, of at least one sector-carrier; and

facilitate a selection, for each sector-carrier, of one or more KPIs;

an idiosyncrasy generator configured to determine, for each KPI, a historical value and a recent value based on samples of the KPI taken during a historical period and a recent period, respectively, the recent period being a subset of the historical period;

a difference unit configured to determine, for each KPI, a delta representing a difference between the historical value and the recent value thereby determining KPI-specific sets of deltas, respectively;

a normalizer configured to normalize each set of deltas;

wherein the UI is further configured to:

facilitate a selection, for each set of deltas, of a KPI-specific threshold;

a comparator configured to compare, for each set of deltas, members thereof against the corresponding threshold; and

an identifier unit configured to identify, for each set of deltas, which combinations of cell-site and sector-carrier are KPI-specific performance outliers based on the comparison, respectively.

20. The apparatus of claim 19 , further comprising:

a sorter configured to sort each normalized set of deltas into numerical order.

21. The apparatus of claim 19 , wherein at least one of the following is true:

the historical period is 60 days; or

the recent period is 24 hours.

22. The apparatus of claim 19 , wherein:

the historical value is one of:

average;

correlation; or

median.

23. The apparatus of claim 19 , wherein the at least one KPI includes at least one of:

a count of instances of a dropped connection between the cell-site and a wireless device;

a count of instances of a failed attempt to establish a connection between the cell-site and a wireless device;

a count of minutes of connection time for all instances of a connection between the cell-site and a wireless device;

a count of data consumed for all instances of a connection between the cell-site and a wireless device;

a value representative of throughput on a per-user basis for all instances of a connection between the cell-site and a wireless device; or

a count of simultaneously active instances of a connection between the cell-site and a wireless device.

24. The apparatus of claim 19 , wherein the idiosyncrasy generator is further configured to:

filter the samples taken during the historical period.

25. The apparatus of claim 24 , wherein the idiosyncrasy generator is further configured to:

determine a standard deviation (STD) for the historical period; and

exclude samples having a magnitude (m), K 1 *STD≦m, where K 1 is a real number and K 1 is about 5.

26. The apparatus of claim 24 , wherein the idiosyncrasy generator is further configured to:

determine an average (AVG) for the historical period; and

exclude samples having a magnitude (m), K 2 *AVG≦m, where K 2 is a real number and K 2 is about 1.9.

27. The apparatus of claim 19 , wherein the normalizer is further configured to:

use a non-linear transfer function to normalize a given set of deltas.

28. The apparatus of claim 27 , wherein:

the non-linear transfer function approximates an exponential function, y=abx;

wherein:

a is non-zero; and

b is a positive real number, b>0.

29. The apparatus of claim 27 , wherein

the normalizing produces, for members of each set of deltas, corresponding scaled values; and

the normalizer, for a given set of deltas, is further configured to:

use a range, R, for the scaled values, where R is an integer and 0≦R≦max(range), and max(range)=100; and

sub-normalize, within the scaled range, into sub-ranges including:

sub-normalizing instances of delta having values, X, onto a sub-range, r 1 , of the range R, where X ⊂ given_set_of_deltas, 0≦r 1 <BOUNDARY 1 <max(range), 0≦X<REF 1 , and BOUNDARY 1 and REF 1 are real numbers, respectively;

sub-normalizing instances of delta having values, Y, onto a sub-range, r 2 , of the range R, where Y ⊂ given_set_of_deltas, max(X)<min(Y), BOUNDARY1≦r 2 <BOUNDARY 2 , BOUNDARY 2 <max(range), REF 1 ≦Y<REF 2 , and BOUNDARY 2 and REF 2 are real numbers, respectively; and

sub-normalizing instances of delta having values, Z, onto a sub-range, r 3 , of the range R, where Z ⊂ given_set_of_deltas, max(Y)<min(Z), BOUNDARY2≦r 3 ≦max(range), REF 2 ≦Z≦max(delta), max(delta) is a maximum value of delta, and max(range) and max(delta) are real numbers.

30. The apparatus of claim 19 , wherein the UI is further configured to:

facilitate an assignment of weights to at least one of:

the selected one or more KPIs, respectively; or

the selected at least one sector-carrier.