Using bandwidth allocation data to determine a breakdown of VoLTE calls by AMR coding mode

1. A computer-implemented method comprising:

accessing a first database of call detail records (CDRs) containing information associated with a plurality of communication sessions established over a mobile telephony network;

selecting, from the CDRs, a first set of CDRs that are associated with communication sessions that used a long term evolution (LTE) radio access technology (RAT);

selecting, from the first set of CDRs, a second set of CDRs that are associated with communication sessions having a Quality of Service Class Indicator (QCI) value that corresponds to voice over long term evolution (VoLTE);

selecting, from the second set of CDRs, a third set of CDRs that are associated with communication sessions having durations greater than a threshold period of time;

selecting, from the third set of CDRs, a fourth set of CDRs associated with communication sessions that were released from a bearer of the communication sessions corresponding to the fourth set of CDRs;

selecting, from the fourth set of CDRs, a fifth set of CDRs associated with communication sessions that were allocated an amount of bandwidth greater than a threshold bandwidth;

referencing a second database containing correlations between adaptive multi-rate (AMR) speech coding modes and bandwidth values;

classifying the CDRs of the fifth set of CDRs into groups of different AMR speech coding modes based at least in part on the referencing; and

recommending a capacity planning metric or a resource allocation metric based at least in part on the classifying.

2. The computer-implemented method of claim 1 , further comprising, prior to referencing the second database, creating the correlations in the second database by:

tracing a plurality of VoLTE calls, each of the plurality of VoLTE calls using a known adaptive multi-rate (AMR) speech coding mode;

for each of the plurality of VoLTE calls:

extracting, from a trace of the VoLTE call, an amount of bandwidth allocated for the VoLTE call; and

correlating the known AMR speech coding mode with the amount of the bandwidth allocated for the VoLTE call.

3. The computer-implemented method of claim 1 , further comprising determining a sixth set of CDRs associated with one or more groups of the groups of different AMR speech coding modes that are based on an AMR Wideband (WB) codec, wherein the capacity planning metric or the resource allocation metric is based at least in part on a number of the CDRs in the sixth set of CDRs.

4. The computer-implemented method of claim 1 , wherein the resource allocation metric comprises a recommended amount of bandwidth for serving customers of the mobile telephony network over a future period of time and a recommended number of available resources for providing the recommended amount of bandwidth.

5. The computer-implemented method of claim 1 , wherein the selecting the first set of CDRs comprises filtering the CDRs by a RAT type value that corresponds to the LTE RAT.

6. The computer-implemented method of claim 1 , wherein the threshold bandwidth is about 30,000 bits per second (bps).

7. A computer-implemented method comprising:

accessing a plurality of call detail records (CDRs) containing information associated with a plurality of corresponding communication sessions established over a mobile telephony network;

filtering the plurality of CDRs to obtain a subset of CDRs that are associated with communication sessions that used voice over long term evolution (VoLTE) services;

referencing a data structure containing correlations between adaptive multi-rate (AMR) speech coding modes and bandwidth values;

classifying the CDRs of the subset of CDRs into groups of different AMR speech coding modes based at least in part on the referencing; and

recommending a capacity planning metric or a resource allocation metric based at least in part on numbers of the CDRs in the subset of CDRs within each of the groups of the different AMR speech coding modes.

8. The computer-implemented method of claim 7 , wherein filtering the plurality of CDRs comprises:

filtering the plurality of CDRs by a radio access technology (RAT) type value that corresponds to a long term evolution (LTE) RAT.

9. The computer-implemented method of claim 7 , wherein filtering the plurality of CDRs comprises:

filtering the plurality of CDRs by a Quality of Service Class Indicator (QCI) value that corresponds to voice over long term evolution (VoLTE).

10. The computer-implemented method of claim 7 , wherein filtering the plurality of CDRs comprises:

filtering the plurality of CDRs by session duration such that the subset of CDRs is associated with communication sessions having durations greater than a threshold period of time.

11. The computer-implemented method of claim 7 , wherein filtering the plurality of CDRs comprises:

filtering the plurality of CDRs by information indicating a successful bearer release such that the subset of CDRs is associated with communication sessions that were released from a bearer of the communication sessions.

12. The computer-implemented method of claim 7 , wherein filtering the plurality of CDRs comprises:

filtering the plurality of CDRs by bandwidth allocation such that the subset of CDRs is associated with communication sessions that were allocated an amount of bandwidth greater than a threshold bandwidth.

13. The computer-implemented method of claim 7 , further comprising, prior to referencing the data structure, creating the correlations in the data structure by:

tracing a plurality of VoLTE calls, each of the plurality of VoLTE calls using a known adaptive multi-rate (AMR) speech coding mode;

for each of the plurality of VoLTE calls:

extracting, from a trace of the VoLTE call, an amount of bandwidth allocated for the VoLTE call; and

correlating the known AMR speech coding mode with the amount of the bandwidth allocated for the VoLTE call.

14. One or more non-transitory computer-readable media storing computer-executable instructions that, when executed, cause one or more processors of a computing device to perform acts comprising:

accessing a plurality of call detail records (CDRs) containing information associated with a plurality of corresponding communication sessions established over a mobile telephony network;

filtering the plurality of CDRs to obtain a subset of CDRs that are associated with communication sessions that used voice over long term evolution (VoLTE) services;

referencing a data structure containing correlations between adaptive multi-rate (AMR) speech coding modes and bandwidth values;

classifying the CDRs of the subset of CDRs into groups of different AMR speech coding modes based at least in part on the referencing; and

recommending a capacity planning metric or a resource allocation metric based at least in part on numbers of the CDRs in the subset of CDRs within each of the groups of the different AMR speech coding modes.

15. The one or more non-transitory computer-readable media of claim 14 , wherein filtering the plurality of CDRs comprises:

filtering the plurality of CDRs by a radio access technology (RAT) type value that corresponds to a long term evolution (LTE) RAT to obtain a first set of CDRs;

filtering the first set of CDRs by a Quality of Service Class Indicator (QCI) value that corresponds to voice over long term evolution (VoLTE) to obtain a second set of CDRs;

filtering the second set of CDRs by session duration to obtain a third set of CDRs associated with communication sessions having durations greater than a threshold period of time;

filtering the third set of CDRs by information indicating a successful bearer release to obtain a fourth set of CDRs associated with communication sessions that were released from a bearer of the communication sessions; and

filtering the fourth set of CDRs by bandwidth allocation to obtain a fifth set of CDRs associated with communication sessions that were allocated an amount of bandwidth greater than a threshold bandwidth.

16. The one or more non-transitory computer-readable media of claim 15 , wherein the threshold bandwidth is about 30,000 bits per second (bps).

17. The one or more non-transitory computer-readable media of claim 14 , the acts further comprising determining a set of CDRs associated with one or more groups of the groups of different AMR speech coding modes that are based on an AMR Wideband (WB) codec, wherein the capacity planning metric or the resource allocation metric is based at least in part on a number of the CDRs in the set of CDRs.

18. The one or more non-transitory computer-readable media of claim 14 , wherein the resource allocation metric comprises a recommended amount of bandwidth for serving customers of the mobile telephony network over a future period of time and a recommended number of available resources for providing the recommended amount of bandwidths.

19. The one or more non-transitory computer-readable media of claim 14 , wherein the capacity planning metric comprises a recommended number of new resources for providing service to customers of the mobile telephony network.

20. The one or more non-transitory computer-readable media of claim 14 , the acts further comprising, prior to referencing the data structure, creating the correlations in the data structure by:

tracing a plurality of VoLTE calls, each of the plurality of VoLTE calls using a known adaptive multi-rate (AMR) speech coding mode;

for each of the plurality of VoLTE calls:

extracting, from a trace of the VoLTE call, an amount of bandwidth allocated for the VoLTE call; and

correlating the known AMR speech coding mode with the amount of the bandwidth allocated for the VoLTE call.