Predicting an attribute of a wireless telecommunication network, such as a 5G network
The system obtains multiple KPIs and multiple configuration parameters directly from a wireless telecommunication network. The multiple KPIs indicate an observed performance associated with the wireless telecommunication network. The multiple configuration parameters indicate a configuration of the wireless telecommunication network. The system predicts a value of a difficult to predict attribute of the wireless telecommunication network, where the difficult to predict attribute depends on multiple other attributes associated with the wireless telecommunication network. To make the prediction, the system provides the multiple key performance indicators and the multiple configuration parameters to a machine learning model. The machine learning model predicts the value of the difficult to predict attribute associated with the wireless telecommunication network based on multiple key performance indicators and the multiple configuration parameters.
1. At least one computer-readable storage medium, excluding transitory signals and carrying instructions to predict a throughput of a wireless telecommunication network, which, when executed by at least one data processor of a system, causes the system to:
obtain multiple key performance indicators and multiple configuration parameters directly from the wireless telecommunication network,
wherein the multiple key performance indicators indicate an observed performance associated with the wireless telecommunication network,
wherein the multiple configuration parameters indicate a configuration of the wireless telecommunication network,
wherein the multiple key performance indicators include a number of users associated with the wireless telecommunication network, and
wherein the multiple configuration parameters include a generation of wireless technology associated with the wireless telecommunication network;
predict the throughput of the wireless telecommunication network by:
providing the multiple key performance indicators and the multiple configuration parameters to a machine learning model trained on previous observations of attributes during previous operation of the wireless telecommunication network, the attributes comprising at least six of: a quality of service class identifier, the number of users, modulation coding scheme, multiple-input multiple-output, carrier aggregation, band, bandwidth, and the generation of wireless technology associated with the wireless telecommunication network; and
predict, by the machine learning model, the throughput of the wireless telecommunication network based on the multiple key performance indicators and the multiple configuration parameters.
2. The computer-readable storage medium of claim 1 , wherein the instructions to obtain the multiple key performance indicators and the multiple configuration parameters comprise instructions to:
obtain observations of at least six of the attributes during operation of the wireless telecommunication network.
3. The computer-readable storage medium of claim 1 , wherein the multiple key performance indicators comprise observations of the quality of service class identifier, the modulation coding scheme, the multiple-input multiple-output, the carrier aggregation, and the number of users.
4. The computer-readable storage medium of claim 1 , wherein the multiple configuration parameters comprise the band, the bandwidth, and the generation of wireless technology associated with the wireless telecommunication network.
5. A method comprising:
obtaining multiple key performance indicators and multiple configuration parameters directly from a wireless telecommunication network,
wherein the multiple key performance indicators indicate an observed performance associated with the wireless telecommunication network,
wherein the multiple configuration parameters indicate a configuration of the wireless telecommunication network,
wherein the multiple key performance indicators and the multiple configuration parameters together comprise at least six of: a quality of service class identifier, a number of users, a modulation coding scheme, a multiple-input multiple-output, a carrier aggregation, a band, a bandwidth, and a generation of wireless technology associated with the wireless telecommunication network;
predicting a value of a difficult to predict attribute of the wireless telecommunication network,
wherein the difficult to predict attribute depends on multiple other attributes associated with the wireless telecommunication network,
the predicting including:
providing the multiple key performance indicators and the multiple configuration parameters to a machine learning model; and
predicting, by the machine learning model, the value of the difficult to predict attribute associated with the wireless telecommunication network based on multiple key performance indicators and the multiple configuration parameters.
6. The method of claim 5 , comprising:
obtaining previous observations of at least six attributes during previous operation of the wireless telecommunication network, the attributes comprising: the quality of service class identifier, the number of users, the modulation coding scheme, the multiple-input multiple-output, the carrier aggregation, the band, the bandwidth, and the generation of wireless technology associated with the wireless telecommunication network; and
training the machine learning model using the previous observations of at least six attributes.
7. The method of claim 5 , wherein obtaining the multiple key performance indicators and the multiple configuration parameters directly from the wireless telecommunication network produces a difference between the predicted value of the difficult to predict attribute and an observed value of the difficult to predict attribute having Mean Absolute Error less than 0.5, and R 2 greater than 0.5.
8. The method of claim 5 , wherein predicting the difficult to predict attribute of the wireless telecommunication network comprises predicting a throughput of the wireless telecommunication network.
9. The method of claim 5 , wherein predicting the difficult to predict attribute of the wireless telecommunication network comprises predicting a latency of the wireless telecommunication network.
10. The method of claim 5 , wherein the multiple key performance indicators comprise the quality of service class identifiers, the modulation coding scheme, the multiple-input multiple-output, the carrier aggregation, and the number of users.
11. The method of claim 5 , wherein the multiple configuration parameters comprise the band, the bandwidth, and the generation of wireless technology associated with the wireless telecommunication network.
12. A system comprising:
at least one hardware processor; and
at least one non-transitory memory storing instructions, which, when executed by the at least one hardware processor, cause the system to:
obtain multiple key performance indicators directly from a wireless telecommunication network,
wherein the multiple key performance indicators indicate an observed performance associated with the wireless telecommunication network,
wherein the multiple key performance indicators include a number of users associated with the wireless telecommunication network;
obtain multiple configuration parameters directly from the wireless telecommunication network,
wherein the multiple configuration parameters indicate a configuration of the wireless telecommunication network,
wherein the multiple configuration parameters include a generation of wireless technology associated with the wireless telecommunication network;
identify a difficult to predict attribute by identifying an attribute that depends on multiple other attributes associated with the wireless telecommunication network;
predict a value of the difficult to predict attribute associated with the wireless telecommunication network by:
providing the multiple key performance indicators and the multiple configuration parameters to a machine learning model trained on previous observations of attributes during previous operation of the wireless telecommunication network, the attributes comprising: a quality of service class identifier, the number of users, modulation coding scheme, multiple-input multiple-output, carrier aggregation, band, bandwidth, and the generation of wireless technology associated with the wireless telecommunication network; and
predicting, by the machine learning model, the value of the difficult to predict attribute associated with the wireless telecommunication network based on the multiple key performance indicators and the multiple configuration parameters.
13. The system of claim 12 , wherein the instructions to obtain the multiple key performance indicators and the multiple configuration parameters directly from the wireless telecommunication network produce a difference between the predicted value of the difficult to predict attribute and an observed value of the difficult to predict attribute having Mean Absolute Error less than 0.5, and R 2 greater than 0.5.
14. The system of claim 12 , the instructions to obtain the multiple key performance indicators and the multiple configuration parameters comprising the instructions to:
obtain observations of at least six of the attributes during operation of the wireless telecommunication network.
15. The system of claim 12 , wherein the instructions to predict the difficult to predict attribute of the wireless telecommunication network comprise instructions to predict a throughput or a latency of the wireless telecommunication network.
16. The system of claim 12 , wherein the multiple key performance indicators comprise observations of the quality of service class identifier, the modulation coding scheme, the multiple-input multiple-output, the carrier aggregation, and the number of users.
17. The system of claim 12 , wherein the multiple configuration parameters comprise the band, the bandwidth, and the generation of wireless technology associated with the wireless telecommunication network.