Downlink interference detection and identification of aggressor cells
A system can include a network analysis platform that applies models to identify downlink interference at a network cell, such as at a base station. For a session at a cell, an expected performance with normalized downlink interference can be compared to an actual performance to determine whether the session is impacted. This can include normalizing channel quality index (“CQI”) and negative-acknowledgement (“NACK”) rate. Overshooting aggressor cells can be identified as the source of the downlink interference based on a useless overlap fraction exceeding a threshold. The impacted sessions and root causes can be displayed on a graphical user interface (“GUI”).
1. A method for detecting downlink interference in a telco network, comprising:
receiving telemetry data;
determining an actual performance value for a first session at a first cell of a plurality of cells in the telco network;
using a trained performance model, predicting an expected performance value for the first session based on normalized downlink interference, wherein the normalized downlink interference is based on a path loss of the first session and at least one factor relative to the plurality of cells;
determining an impact based on the expected performance value differing from the actual performance value by at least a threshold amount; and
indicating that downlink interference exists with respect to the first cell based on the impact.
2. The method of claim 1 , wherein the normalized downlink interference is determined based on at least one of: an average channel quality indicator (CQI) over the network, a negative acknowledgement rate across the plurality of cells, and a ratio of Rank 2 transmissions for the network, and wherein the actual and expected performance values are for throughput.
3. The method of claim 1 , further comprising identifying a second cell as responsible for the downlink interference based on over-transmission.
4. The method of claim 3 , wherein identifying the second cell as responsible for the downlink interference further comprises determining that the second cell has a useless overlap fraction that exceeds an overlap threshold, wherein the useless overlap fraction is a measure of sessions in first cell receiving strong signals from both the first and second cells, and wherein the strong signals are defined by signals having a Reference Signal Receive Power (RSRP) above a power threshold.
5. The method of claim 4 , identifying the second cell includes matching the second cell against an identifier in a resource control measurement report from the first cell.
6. The method of claim 4 , wherein the overlap fraction is defined as the 75 th percentile of useless overlap fractions among the plurality of cells.
7. The method of claim 3 , further comprising indicating that at least one of a power configuration and an electronic tilt configuration should be adjusted at the second cell.
8. A non-transitory, computer-readable medium containing instructions that, when executed by a hardware-based processor, performs stages for detecting downlink interference in a telco network, the stages comprising:
receiving telemetry data;
determining an actual performance value for a first session at a first cell of a plurality of cells in the telco network;
using a trained performance model, predicting an expected performance value for the first session based on normalized downlink interference, wherein the normalized downlink interference is based on a path loss of the first session and at least one factor relative to the plurality of cells;
determining an impact based on the expected performance value differing from the actual performance value by at least a threshold amount; and
indicating that downlink interference exists with respect to the first cell based on the impact.
9. The non-transitory, computer-readable medium of claim 8 , wherein the normalized downlink interference is determined based on at least one of: an average channel quality indicator (CQI) over the network, a negative acknowledgement rate across the plurality of cells, and a ratio of Rank 2 transmissions for the network, and wherein the actual and expected performance values are for throughput.
10. The non-transitory, computer-readable medium of claim 8 , the stages further comprising identifying a second cell as responsible for the downlink interference based on over-transmission.
11. The non-transitory, computer-readable medium of claim 10 , wherein identifying the second cell as responsible for the downlink interference further comprises determining that the second cell has a useless overlap fraction that exceeds an overlap threshold, wherein the useless overlap fraction is a measure of sessions in first cell receiving strong signals from both the first and second cells, and wherein the strong signals are defined by signals having a Reference Signal Receive Power (RSRP) above a power threshold.
12. The non-transitory, computer-readable medium of claim 11 , identifying the second cell includes matching the second cell against an identifier in a resource control measurement report from the first cell.
13. The non-transitory, computer-readable medium of claim 11 , wherein the overlap fraction is defined as the 75 th percentile of useless overlap fractions among the plurality of cells.
14. The non-transitory, computer-readable medium of claim 10 , the stages further comprising indicating that at least one of a power configuration and an electronic tilt configuration should be adjusted at the second cell.
15. A system for detecting downlink interference in a telco network, comprising:
a memory storage including a non-transitory, computer-readable medium comprising instructions; and
a computing device including a hardware-based processor that executes the instructions to carry out stages comprising:
receiving telemetry data;
determining an actual performance value for a first session at a first cell of a plurality of cells in the telco network;
using a trained performance model, predicting an expected performance value for the first session based on normalized downlink interference, wherein the normalized downlink interference is based on a path loss of the first session and at least one factor relative to the plurality of cells;
determining an impact based on the expected performance value differing from the actual performance value by at least a threshold amount; and
indicating that downlink interference exists with respect to the first cell based on the impact.
16. The system of claim 15 , wherein the normalized downlink interference is determined based on at least one of: an average channel quality indicator (CQI) over the network, a negative acknowledgement rate across the plurality of cells, and a ratio of Rank 2 transmissions for the network, and wherein the actual and expected performance values are for throughput.
17. The system of claim 15 , the stages further comprising identifying a second cell as responsible for the downlink interference based on over-transmission.
18. The system of claim 17 , wherein identifying the second cell as responsible for the downlink interference further comprises determining that the second cell has a useless overlap fraction that exceeds an overlap threshold, wherein the useless overlap fraction is a measure of sessions in first cell receiving strong signals from both the first and second cells, and wherein the strong signals are defined by signals having a Reference Signal Receive Power (RSRP) above a power threshold.
19. The system of claim 17 , identifying the second cell includes matching the second cell against an identifier in a resource control measurement report from the first cell.
20. The system of claim 18 , wherein the overlap fraction is defined as the 75 th percentile of useless overlap fractions among the plurality of cells.