Network path selection based on power consumption

1 . A system comprising:

user equipment configured to connect with a service application via a wireless network;

a plurality of network devices of the wireless network configured to form a plurality of network paths between the user equipment and the service application; and

a base station of the wireless network configured to:

receive a routing request from the user equipment indicating a request to connect with the service application;

determine a power consumption associated with each network path of the plurality of network paths;

identify a constraint of the service application, the constraint being a latency constraint, a bandwidth constraint, a packet loss constraint, or a regulatory constraint;

determine a subset of network paths of the plurality of network paths that satisfy the constraint;

generate a ranking of the subset of network paths based on the power consumption associated with each network path in the subset;

select a network path of the subset of network paths based on the ranking; and

enable the network path between the user equipment and the service application.

2 . The system of claim 1 , wherein the base station is further configured to:

determine that the network path is associated with a lowest power consumption; and

select the network path in response to determining that the network path is associated with the lowest power consumption.

3 . The system of claim 1 , wherein the base station is further configured to:

receive a second routing request from second user equipment;

determine a second plurality of network paths of the wireless network, each network path of the second plurality of network paths comprising a route between the second user equipment and a second service application associated with the second routing request;

determine a second power consumption associated with each network path of the second plurality of network paths; and

select the network path of the plurality of network paths and a second network path of the second plurality of network paths for the power consumption associated with the network path and the second power consumption associated with the second network path, wherein the network path and the second network path distribute power consumption across the wireless network.

4 . The system of claim 3 , wherein the base station is further configured to select the network path and the second network path by:

inputting the plurality of network paths, the second plurality of network paths, the power consumption associated with each network path of the plurality of network paths, and the second power consumption associated with each network path of the second plurality of network paths into a machine-learning model trained to select network paths for user equipment-to-service application connections; and

receiving an output of the machine-learning model indicating the network path and the second network path.

5 . The system of claim 1 , wherein the wireless network comprises a fourth-generation long-term evolution network, a fifth-generation network, or a WiFi-6 network.

6 . The system of claim 1 , wherein the base station is further configured to:

periodically re-evaluate the power consumption associated with each network path of the plurality of network paths; and

in response to detecting a change in power consumption or network conditions:

determine an updated subset of network paths that satisfy the constraint;

generating a new ranking of the updated subset; and

select a new network path of the updated subset based on the new ranking.

7 . A method comprising:

receiving, by a base station of a wireless network, a routing request from user equipment;

determining, by the base station, a plurality of network paths of the wireless network, each network path of the plurality of network paths comprising a route between the user equipment and a service application associated with the routing request;

determining, by the base station, a power consumption associated with each network path of the plurality of network paths;

identifying, by the base station, a constraint of the service application, the constraint being a latency constraint, a bandwidth constraint, a packet loss constraint, or a regulatory constraint;

determining, by the base station, a subset of network paths of the plurality of network paths that satisfy the constraint;

generating, by the base station, a ranking of the subset of network paths based on the power consumption for each network path;

selecting, by the base station, a network path of the subset of network paths based on the ranking; and

enabling the network path between the user equipment and the service application for the routing request.

8 . The method of claim 7 , further comprising:

determining that the network path is associated with a lowest power consumption; and

selecting the network path in response to determining that the network path is associated with the lowest power consumption.

9 . The method of claim 7 , further comprising:

receiving a second routing request from second user equipment;

determining a second plurality of network paths of the wireless network, each network path of the second plurality of network paths comprising a second route between the second user equipment and a second service application associated with the second routing request;

determining a second power consumption associated with each network path of the second plurality of network paths; and

selecting the network path of the plurality of network paths and a second network path of the second plurality of network paths for the power consumption associated with the network path and the second power consumption associated with the second network path, wherein the network path and the second network path distribute power consumption across the wireless network.

10 . The method of claim 9 , wherein selecting the network path and the second network path comprises:

inputting the plurality of network paths, the second plurality of network paths, the power consumption associated with each network path of the plurality of network paths, and the second power consumption associated with each network path of the second plurality of network paths into a machine-learning model trained to select network paths for user equipment-to-service application connections; and

receiving an output of the machine-learning model indicating the network path and the second network path.

11 . The method of claim 7 , wherein the wireless network comprises a fourth-generation long-term evolution network, a fifth-generation network, or a WiFi-6 network.

12 . A non-transitory computer-readable medium comprising program code executable by a processing device for causing the processing device to:

receive, by a base station of a wireless network, a routing request from user equipment;

determine, by the base station, a plurality of network paths of the wireless network, each network path of the plurality of network paths comprising a route between the user equipment and a service application associated with the routing request;

determine, by the base station, a power consumption associated with each network path of the plurality of network paths;

identify, by the base station, a constraint of the service application, the constraint being a latency constraint, a bandwidth constraint, a packet loss constraint, or a regulatory constraint;

determine, by the base station, a subset of network paths of the plurality of network paths that satisfy the constraint;

generate, by the base station, a ranking of the subset of network paths based on the power consumption for each network path;

select, by the base station, a network path of the subset of network paths based on the ranking; and

enable the network path between the user equipment and the service application for the routing request.

13 . The non-transitory computer-readable medium of claim 12 , further comprising program code executable by the processing device for causing the processing device to:

determine that the network path is associated with a lowest power consumption; and

select the network path in response to determining that the network path is associated with the lowest power consumption.

14 . The non-transitory computer-readable medium of claim 12 , further comprising program code executable by the processing device for causing the processing device to:

receive a second routing request from second user equipment;

determine a second plurality of network paths of the wireless network, each network path of the second plurality of network paths comprising a second route between the second user equipment and a second service application associated with the second routing request;

determine a second power consumption associated with each network path of the second plurality of network paths; and

select the network path of the plurality of network paths and a second network path of the second plurality of network paths for the power consumption associated with the network path and the second power consumption associated with the second network path, wherein the network path and the second network path distribute power consumption across the wireless network.

15 . The non-transitory computer-readable medium of claim 14 , further comprising program code executable by the processing device for causing the processing device to select the network path and the second network path by:

inputting the plurality of network paths, the second plurality of network paths, the power consumption associated with each network path of the plurality of network paths, and the second power consumption associated with each network path of the second plurality of network paths into a machine-learning model trained to select network paths for user equipment-to-service application connections; and

receiving an output of the machine-learning model indicating the network path and the second network path.