Systems and methods for orchestration and optimization of wireless networks
A system described herein may provide for the use of artificial intelligence/machine learning (“AI/ML”) techniques to generate models for various locations or regions (e.g., sectors) associated with one or more radio access networks (“RANs”) of a wireless network. The system may determine Key Performance Indicators (“KPIs”) or other attributes that are of particular relevance or importance for a given sector model, and may determine actions to perform with respect to particular sectors in order to enhance performance according to the KPIs that are of particular relevance to a sector model determined with respect to the particular sectors.
1. A device, comprising:
one or more processors configured to:
maintain a plurality of sector models, wherein each sector model is associated with a respective set of radio access network (“RAN”) attributes;
maintain a plurality of optimization goals, wherein each sector model is associated with one or more optimization goal of the plurality of optimization goals;
select, based on a particular set of RAN attributes of a particular RAN, a particular sector model of the plurality of sector models;
select a particular optimization goal, of the plurality of optimization goals, with which the selected particular sector model is associated;
identify a set of actions to perform based on the particular optimization goal; and
implement the set of actions, associated with the particular optimization goal, at the particular RAN.
2. The device of claim 1 ,
wherein the particular set of RAN attributes includes at least one of:
performance metrics, or
energy consumption metrics,
wherein selecting the particular sector model includes determining that measure of similarity, between performance metrics or energy consumption metrics of the particular sector model and of the particular RAN, exceeds a threshold measure of similarity.
3. The device of claim 1 ,
Wherein the particular set of RAN attributes includes a set of locale features associated with the particular RAN,
Wherein selecting the particular sector model includes determining that measure of similarity, between locale features of the particular sector model and of the particular RAN, exceeds a threshold measure of similarity.
4. The device of claim 3 , wherein the locale features associated with the particular RAN include at least one of:
building density in a geographical area with which the particular RAN is associated,
topographical features of the geographical area with which the particular RAN is associated, or
air quality metrics of the geographical area with which the particular RAN is associated.
5. The device of claim 1 , Wherein the one or wore processors are further configured to:
determine a plurality of sets of affinity scores between the plurality of optimization goals and a plurality of sets of actions, wherein a particular set of affinity scores is associated with:
the particular optimization goal, and
the plurality of sets of actions,
wherein each respective affinity score of the particular set of affinity scores is associated with the particular optimization goal and a respective action of the plurality of sets of actions.
6. The device of claim 5 , wherein identifying the set of actions is lased on the particular optimization goal or goals:
determining that the respective affinity score associated with the identified set of actions and the particular optimization goal is a highest affinity score of the particular set of affinity scores.
7. The device of claim 1 , wherein each optimization goal, of the plurality of optimization goals includes, a plurality of metrics that are each associated with a particular weight.
8. A non-transitory computer-readable medium, storing a plurality of processor-executable instructions to:
maintain a plurality of sector models, wherein each sector model is associated with a respective set of radio access network (“RAN”) attributes;
maintain a plurality of optimization goals, wherein each sector model is associated with one or more optimization goal of the plurality of optimization goals;
select, based on a particular set of RAN attributes of a particular RAN, a particular sector model of the plurality of sector models;
select a particular optimization goal, of the plurality of optimization goals, with which the selected particular sector model is associated;
identify a set of actions to perform based on the particular optimization goal; and
implement the set of actions, associated with the particular optimization goal, at the particular RAN.
9. The non-transitory computer-readable medium of claim 8 ,
Wherein the particular set of RAN attributes includes at least one of:
performance metrics, or
energy consumption metrics,
wherein selecting the particular sector model includes determining that measure of similarity, between performance metrics or energy consumption metrics of the particular sector model and of the particular RAN, exceeds a threshold measure of similarity.
10. The non-transitory computer-readable medium of claim 8 ,
wherein the particular set of RAN attributes includes a set of locale features associated with the particular RAN,
wherein selecting the particular sector model includes determining that measure of similarity, between locale features of the particular sector model and of the particular RAN, exceeds a threshold measure of similarity.
11. The non-transitory computer-readable medium of claim 10 , wherein the locale features associated with the particular RAN include at least one of:
building density in a geographical area with which the particular RAN is associated,
topographical features of the geographical area with which the particular RAN is associated, or
air quality metrics of the geographical area with which the particular RAN is associated.
12. The non-transitory computer-readable medium of claim 8 , wherein the plurality of processor-executable instructions further include processor-executable instructions to:
determine a plurality of sets of affinity scores between the plurality of optimization goals and a plurality of sets of actions, wherein a particular set of affinity scores is associated with:
the particular optimization goal, and
the plurality of sets of actions,
wherein each respective affinity score of the particular set of affinity scores is associated with the particular optimization goal and a respective action of the plurality of sets of actions.
13. The non-transitory computer-readable medium of claim 12 , wherein identifying the set of actions is based on the particular optimization goal or goals:
determining that the respective affinity score associated with the identified set of actions and the particular optimization goal is a highest affinity score of the particular set of affinity scores.
14. The non transitory computer-readable medium of claim 8 , wherein each optimization goal, of the plurality of optimization goals includes, a plurality of metrics that are each associated with a particular weight.
15. A method, comprising:
maintaining a plurality of sector models, wherein each sector model is associated with a respective set of radio access network (“RANs”) attributes;
maintaining a plurality of optimization goals, wherein each sector model is associated with one or more optimization goal of the plurality of optimization goals;
selecting, based on a particular set of RAN attributes of a particular RAN, a particular sector model of the plurality of sector models;
selecting a particular optimization goal, of the plurality of optimization goals, with which the selected particular sector model is associated;
identifying a set of actions to perform based on the particular optimization goal; and
implementing the set of actions, associated with the particular optimization goal, at the particular RAN.
16. The method of claim 15 ,
Wherein the particular set of RAN attributes includes at least one of:
performance metrics, or
energy consumption metrics,
wherein selecting the particular sector model includes determining that measure of similarity, between performance metrics of energy consumption metrics of the particular sector model and of the particular RAN, exceeds a threshold measure of similarity.
17. The method of claim 15 ,
wherein the particular set of RAN attributes includes a set of locale features associated with, the particular RAN,
wherein selecting the particular sector model includes determining that measure of similarity, between locale features of the particular sector model and of the particular RAN, exceeds a threshold measure of similarity.
18. The method of claim 15 , the method further comprising:
determining a plurality of sets of affinity scores between the plurality of optimization goals and a plurality of sets of actions, wherein a particular set of affinity scores is associated with:
the particular optimization goal, and
the plurality of sets of actions,
wherein each respective affinity score of the particular set of affinity scores is associated with the particular optimization goal and a respective action of the plurality of sets of actions.
19. The method of claim 18 wherein identifying the set of actions is based on the particular optimization goal or goals:
determining that the respective affinity score associated with the identified set of actions and the particular optimization goal is a highest affinity score of the particular set of affinity scores.
20. The method of claim 15 , wherein each optimization goal, of the plurality of optimization goals includes, a plurality of metrics that are each associated with a particular weight.