Method and system for using a digital twin of a mobile network to provide energy efficiency

Patent metadata

Field	Value
Publication number	US-12445869-B2
Application number	US-202318334753-A
Country	US
Kind code	B2
Filing date	Jun 14, 2023
Priority date	Jun 14, 2023
Publication date	Oct 14, 2025
Grant date	Oct 14, 2025

Abstract

A system for enhancing an energy efficiency of a mobile network includes a digital twin connected to the mobile network. The digital twin is programmed to: receive, from the mobile network, real-time network traffic data and base station data of base stations operating in the mobile network; generate a first prediction of network traffic at a first time-scale based on the real-time network traffic data; generate a second prediction of network traffic at a second time-scale based on the real-time network traffic data and the base station data; and generate a first power consumption prediction for the mobile network based on the first prediction, the second prediction, and a topology of the mobile network, and the digital twin emulates at least a portion of the mobile network.

First claim

What is claimed is:

1. A system for enhancing an energy efficiency of a mobile network, the system comprising:
a digital twin connected to the mobile network, wherein the digital twin is programmed to:
receive, from the mobile network, real-time network traffic data and base station data of a plurality of base stations operating in the mobile network;
generate a first prediction of network traffic at a first time-scale based on the real-time network traffic data;
generate a second prediction of network traffic at a second time-scale based on the real-time network traffic data and the base station data, wherein the second time-scale is shorter than the first time-scale; and
generate a first power consumption prediction for the mobile network based on the first prediction, the second prediction, and a topology of the mobile network, and
wherein the digital twin emulates at least a portion of the mobile network.

2. The system of claim 1 , wherein the digital twin is further programmed to:
compute a spatial correlation of traffic demand based on the real-time network traffic data and historical network traffic data; and
compute a time correlation of traffic demand based on the real-time network traffic data and historical network traffic data, and
wherein the first prediction is also based on the spatial correlation or the time correlation.

3. The system of claim 2 , wherein the first prediction is based on the spatial correlation and the time correlation.

4. The system of claim 1 , wherein the base station data comprises current power consumption data of each of the plurality of base stations.

5. The system of claim 1 , further comprising a sleep control agent programmed to:
generate an updated energy consumption mode for the plurality of base stations based on the first prediction and the second prediction.

6. The system of claim 5 , wherein the digital twin is further programmed to:
generate a second power consumption prediction for the mobile network based on the first prediction, the second prediction, the topology, and the updated energy consumption mode; and
generate a network power metric based on a comparison of the first power consumption prediction and the second power consumption prediction.

7. The system of claim 6 , further comprising a policy module programmed to:
generate an energy consumption mode penalty based on the network power metric and a quality of service metric;
make a determination that the energy consumption mode penalty is below a threshold; and
send, based on the determination, a signal to apply the updated energy consumption mode to the plurality of base stations.

8. The system of claim 6 , further comprising a policy module programmed to:
generate a first energy consumption mode penalty based on the network power metric and a quality of service metric;
make a first determination that the first energy consumption mode penalty is above a threshold;
wherein the sleep control agent is further programmed to:
generate a second updated energy consumption mode based on the first determination,

wherein the digital twin is further programmed to:
generate a third power consumption prediction for the mobile network based on the first prediction, the second prediction, the topology, and the second updated energy consumption mode; and
generate a second network power metric based on a comparison of the first power consumption prediction and the third power consumption prediction, and

wherein the policy module is further programmed to:
generate a second energy consumption mode penalty based on the second network power metric and the quality of service metric;
make a second determination that the second energy consumption mode penalty is below a threshold; and
send a signal to apply the second updated energy consumption mode to the plurality of base stations.

9. The system of claim 8 , wherein the sleep control agent is further programmed to:
update a learning model based on the first determination and the second determination, wherein the learning model is used to generate the updated energy consumption mode and the second updated energy consumption mode.

10. The system of claim 6 , wherein the updated energy consumption mode is also based on a proposed network change event, and wherein the digital twin is further programmed to:
generate an impact prediction based on the network power metric.

11. A method for enhancing an energy efficiency of a mobile network using a digital twin, the method comprising:
receiving, at the digital twin, network traffic data and base station data of a plurality of base stations of the mobile network;
computing, by the digital twin, a spatial correlation of traffic demand based on the network traffic data;
computing, by the digital twin, a time correlation of traffic demand based on the network traffic data;
generating, by the digital twin, a first prediction of network traffic at a first time-scale based on the spatial correlation and the time correlation;
generating, by the digital twin, a second prediction of network traffic at a second time-scale based on the network traffic data and the base station data, wherein the second time-scale is different than the first time-scale; and
performing, based on the first prediction and the second prediction, an energy efficiency enhancement action from an action set.

12. The method of claim 11 , wherein the action set comprises:
sending a signal to adjust an energy consumption mode of a base station of the plurality of base stations;
generating an energy consumption mode for the base station, wherein the energy consumption mode comprises setting the base station to: a first energy consumption mode for a first time period and a second energy consumption mode for a second time period; and
generating an impact prediction based on a mobile network change event.

13. The method of claim 11 , wherein network traffic data comprises real-time network traffic data and historical network traffic data.

14. The method of claim 11 , wherein the second time-scale is shorter than the first time-scale.

15. The method of claim 14 , wherein the second time-scale is a at a frame level of at least one of the base stations of the plurality of base stations.

16. The method of claim 11 , wherein the base station data comprises power consumption data of the plurality of base stations.

17. The method of claim 11 , wherein the time correlation is also based on the spatial correlation.

18. The method of claim 11 , further comprising:
generating a first power consumption prediction based on the first prediction and the second prediction and a topology of the mobile network;
generating a second power consumption prediction based on an updated energy consumption mode for the plurality of base stations, and the topology of the mobile network; and
generating a network power metric based on a comparison of the first power consumption prediction and the second power consumption prediction, and wherein the energy efficiency enhancement action is also based on the network power metric.

19. A method for enhancing an energy efficiency of a mobile network using a digital twin, the method comprising:
receiving, at the digital twin and from the mobile network, real-time network traffic data and base station data of a plurality of base stations of the mobile network, wherein the digital twin emulates at least a portion of the mobile network;
generating, by the digital twin, a plurality of predictions of network traffic, wherein

Assignees

• Dell Products Lp

Inventors

• Mitra Jeebak
• Poitau Gwenael
• Alhaol Ibrahim Abu
• Mirzaei Javad

Classifications

• H04W52/0203

  in the radio access network or backbone network of wireless communication networks · CPC title

• Y02D30/70

  in wireless communication networks · CPC title

• H04W24/02 Primary

  Arrangements for optimising operational condition · CPC title