System and method for real-time mobile networks monitoring
Real-time cell performance information for each of cells in a mobile communication network, referred to herein as network topology information, is produced using IP addresses of data sessions in the cells, measured performance of the data sessions, and IP address to cell identifier mappings. The network topology information may be used to determine, in real time, a level of monitoring for each of the cells. The network topology information may be used to determine, in real time, performance of video streams in each cell. The network topology information may be used to determine, in real time, that a cell is experiencing a performance anomaly. Determinations made using the network topology information may be used to configure, maintain, and optimize the performance of the mobile communication network.
1. A method for monitoring a network, the method comprising:
producing respective User Equipment (UE) location information for a plurality of UEs by tracking each of the plurality of UEs on each of a plurality of cells of the network;
producing a plurality of data session analytics including respective UE data session performance information for respective data sessions of the plurality of UEs, the UE data session performance information including a throughput, a size, a latency, or a combination thereof for the corresponding data session;
producing, using the UE location information and the plurality of data session analytics, a plurality of tagged data session analytics by associating each of the data session analytics with an identifier of a network location corresponding to a UE associated with that data session analytics; and
producing, using the plurality of tagged data session analytics, real-time network metrics, the real-time network metrics including respective real-time performance information for the plurality of cells.
2. The method of claim 1 , wherein the real-time network metrics include network topology information including per-cell data session performance information for the plurality of cells.
3. The method claim 1 , comprising:
identifying, in real-time using the real-time network metrics, a cell of the plurality of cells having anomalous performance.
4. The method claim 1 , comprising:
determining, using the real-time network metrics, whether a cell in the plurality of cells is experiencing a performance anomaly;
in response to determining that the cell is experiencing the performance anomaly, reporting performance information about the cell at an enhanced interval; and
in response to determining that the cell is not experiencing the performance anomaly, reporting performance information about the cell at a normal interval, the normal interval being longer than the enhanced interval.
5. The method of claim 4 , comprising:
periodically, according to an anomaly detection interval, determining whether the cell in the plurality of cells is experiencing the performance anomaly.
6. The method claim 1 , comprising:
determining, using the real-time network metrics, that a data stream in a cell of the plurality of cells is carrying video data;
determining, using the real-time network metrics, second by second (S×S) rate data for the data stream;
determining, in real-time using the S×S rate data, video statistics, the video statistics including a bitrate estimate for the video data, a video resolution of the video data, an average video throughput, a buffer fill level corresponding to the video data, an average start time corresponding to the video data, stall information corresponding to the video data, or combinations thereof; and
reporting the video statistics in real-time.
7. The method claim 6 , comprising:
determining that the data stream is carrying video data according to an address of the data stream and information about addresses of sources of video data.
8. The method claim 6 , comprising:
determining, using machine learning, that the data stream is carrying video data when a pattern in the real-time network metrics corresponding to the data stream corresponds to a pattern for video-carrying data streams.
9. The method of claim 1 , wherein the real-time performance information for the plurality of cells includes average user throughput, average active users, average connected users, video performance metrics, traffic composition, cell congestion level, or combinations thereof.
10. A non-transitory computer readable medium (CRM) having computer-executable instructions embodied thereon that when executed by at least one processor cause the processor to perform a method for monitoring a network, the method comprising:
producing respective User Equipment (UE) location information for a plurality of UEs by tracking each of the plurality of UEs on each of a plurality of cells of the network;
producing a plurality of data session analytics including respective UE data session performance information for respective data sessions of the plurality of UEs, the UE data session performance information including a throughput, a size, a latency, or a combination thereof for the corresponding data session;
producing, using the UE location information and the plurality of data session analytics, a plurality of tagged data session analytics by associating each of the data session analytics with an identifier of a network location corresponding to a UE associated with that data session analytics; and
producing, using the plurality of tagged data session analytics, real-time network metrics, the real-time network metrics including respective real-time performance information for the plurality of cells.
11. The CRM of claim 10 , wherein the real-time network metrics include network topology information including per-cell data session performance information for the plurality of cells.
12. The CRM of claim 10 , wherein the method comprises:
identifying, in real-time using the real-time network metrics, a cell of the plurality of cells having anomalous performance.
13. The CRM of claim 10 , wherein the method comprises:
determining, using the real-time network metrics, whether a cell in the plurality of cells is experiencing a performance anomaly;
in response to determining that the cell is experiencing the performance anomaly, reporting performance information about the cell at an enhanced interval; and
in response to determining that the cell is not experiencing the performance anomaly, reporting performance information about the cell at a normal interval, the normal interval being longer than the enhanced interval.
14. The CRM of claim 13 , wherein the method comprises:
periodically, according to an anomaly detection interval, determining whether the cell in the plurality of cells is experiencing the performance anomaly.
15. The CRM of claim 10 , wherein the real-time performance information for the plurality of cells includes average user throughput, average active users, average connected users, video performance metrics, traffic composition, cell congestion level, or combinations thereof.
16. A non-transitory computer readable medium (CRM) having computer-executable instructions embodied thereon that when executed by at least one processor cause the processor to perform a method for monitoring a network, the method comprising:
producing User Equipment (UE) location information by tracking UEs on each of a plurality of cells of the network; and
producing, using the UE location information, real-time network metrics, the real-time network metrics including respective real-time performance information for the plurality of cells;
determining, using the real-time network metrics, that a data stream in a cell of the plurality of cells is carrying video data;
determining, using the real-time network metrics, second by second (S×S) rate data for the data stream;
determining, in real-time using the S×S rate data, video statistics, the video statistics including a bitrate estimate for the video data, a video resolution of the video data, an average video throughput, a buffer fill level corresponding to the video data, an average start time corresponding to the video data, stall information corresponding to the video data, or combinations thereof; and
reporting the video statistics in real-time.
17. The CRM of claim 16 , wherein the method comprises:
determining that the data stream is carrying video data according to an address of the data stream and information about addresses of sources of video data.
18. The CRM of claim 16 , wherein the method comprises:
determining, using machine learning, that the data stream is carrying video data when a pattern in the real-time network metrics corresponding to the data stream corresponds to a pattern for video-carrying data streams.