Change roll out in wireless networks
An approach for change roll out in wireless networks that utilizes a diverse set of features, such as software/hardware configuration, radio parameters, user population, mobility patterns, network topology and automatically identifies the test locations that would improve the predictability between the performance impacts during testing and network-wide deployment. Through automated and effective analysis of a wide variety of features, the approach for change roll out in wireless networks reflects the impacts observed during testing and predicts the performance of the post-test wide-scale deployment.
1. A server comprising:
a processor; and
a memory coupled with the processor, the memory comprising executable instructions that when executed by the processor cause the processor to effectuate operations comprising:
generating a first list of a plurality of features based on a hamming distance, the first list of the plurality of features associated with a communications network;
determining a first plurality of test locations based on the first list of the plurality of features, the test locations comprising a plurality of upgradable devices;
providing instructions to upgrade the first plurality of test locations;
responsive to the upgrade, assessing performance of the first plurality of test locations; and
based on the performance of the first plurality of test locations, determining a second list of the plurality of features, the second list a subset of the first list.
2. The server of claim 1 , wherein the generating the first list of the plurality of features further comprises: determining a second subset list of the first list based on a first maximum minimum hamming distance over the first subset list.
3. The server of claim 2 , wherein the first list comprises a third subset list of the first list, wherein the third subset list is based on a second maximum minimum hamming distance from the first subset list and the second subset list.
4. The server of claim 1 , wherein the hamming distance is weighed based on a threshold number of test locations with a feature in the first list.
5. The server of claim 1 , wherein the first list comprises a cell search associated feature.
6. The server of claim 1 , wherein the first list comprises an uplink noise associated feature or physical resource block utilization associated feature.
7. The server of claim 1 , wherein the generating the first list of the plurality of features further comprises: discretizing to numerical values a first subset list of the first list.
8. The server of claim 1 , wherein the generating the first list of the plurality of features further comprises: reducing a number of features based clustering features into equivalence classes.
9. The server of claim 1 , wherein the upgrade comprises a change of software or hardware in the plurality of test locations.
10. The server of claim 1 , further operations comprising determining a second plurality of test locations based on the second list of the plurality of features.
11. A computer readable storage medium comprising computer executable instructions that when executed by a computing device cause said computing device to effectuate operations comprising:
generating a first list of a plurality of features based on a hamming distance, the first list of the plurality of features associated with a communications network;
determining a first plurality of test locations based on the first list of the plurality of features, the test locations comprising a plurality of upgradable devices;
providing instructions to upgrade the first plurality of test locations;
responsive to the upgrade, assessing performance of the first plurality of test locations; and
based on the performance of the first plurality of test locations, determining a second list of the plurality of features, the second list a subset of the first list.
12. The computer readable storage medium of claim 11 , wherein the generating the first list of the plurality of features further comprises: determining a second subset list of the first list based on a first maximum minimum hamming distance over the first subset list.
13. The computer readable storage medium of claim 12 , wherein the first list comprises a third subset list of the first list, wherein the third subset list is based on a second maximum minimum hamming distance from the first subset list and the second subset list.
14. The computer readable storage medium of claim 11 , wherein the hamming distance is weighed based on a threshold number of test locations with a feature in the first list.
15. The computer readable storage medium of claim 11 , wherein the first list comprises a power control associated feature.
16. A method comprising:
generating, by a server, a first list of a plurality of features based on a hamming distance, the first list of the plurality of features associated with a communications network;
determining a first plurality of test locations based on the first list of the plurality of features, the test locations comprising a plurality of upgradable devices;
providing instructions to upgrade the first plurality of test locations;
responsive to the upgrade, assessing performance of the first plurality of test locations;
based on the performance of the first plurality of test locations, determining a second list of the plurality of features, the second list a subset of the first list;
determining a second plurality of test locations based on the second list of the plurality of features;
providing instructions to upgrade the second plurality of test locations;
responsive to the upgrade, assessing performance of the second plurality of test locations; and
based on assessing performance of the second plurality of test locations, providing a feature of the second list that is the root cause of a degradation of service of the communications network.
17. The method of claim 16 , wherein the generating the first list of the plurality of features further comprises: determining a second subset list of the first list based on a first maximum minimum hamming distance over the first subset list.
18. The method of claim 17 , wherein the first list comprises a third subset list of the first list, wherein the third subset list is based on a second maximum minimum hamming distance from the first subset list and the second subset list.
19. The method of claim 16 , wherein the hamming distance is weighed based on a threshold number of test locations with a feature in the first list.
20. The method of claim 16 , wherein the first list comprises a medium access layer associated feature.