US 12,476,869 B2
Device energy saving method and network device
Yongmei Zeng, Shanghai (CN); Xiangdong He, Chengdu (CN); Chunmei Qi, Chengdu (CN); Yibu Li, Shanghai (CN); and Jiang Zhu, Shanghai (CN)
Assigned to HUAWEI TECHNOLOGIES CO., LTD., Shenzhen (CN)
Filed by HUAWEI TECHNOLOGIES CO., LTD., Guangdong (CN)
Filed on May 28, 2024, as Appl. No. 18/676,214.
Application 18/676,214 is a continuation of application No. PCT/CN2022/119855, filed on Sep. 20, 2022.
Claims priority of application No. 202111473667.X (CN), filed on Nov. 29, 2021.
Prior Publication US 2024/0314033 A1, Sep. 19, 2024
Int. Cl. G06F 15/16 (2006.01); G06F 9/54 (2006.01); H04L 29/06 (2006.01); H04L 41/0833 (2022.01); H04L 41/5009 (2022.01)
CPC H04L 41/0833 (2013.01) [H04L 41/5009 (2013.01)] 20 Claims
OG exemplary drawing
 
1. A device energy saving method, comprising:
determining, by a first network device, a service data volume of a target time period based on a service data volume carried on N carriers in a target area in a historical time period;
obtaining, by the first network device, a target key performance indicator (KPI), a preset model group, and a preset condition based on the service data volume of the target time period;
generating, by the first network device, M resource information combinations based on carrier resource information of the N carriers, wherein each resource information combination comprises the carrier resource information of the N carriers, and both N and M are integers greater than 1;
sequentially selecting, by the first network device, a resource information combination from the M resource information combinations as a to-be-processed resource information combination;
selecting, by the first network device, input data of each model in the preset model group from a model data set, wherein the model data set comprises signal quality information of the historical time period, the service data volume of the historical time period, a resource information combination of the historical time period, the service data volume of the target time period, and the to-be-processed resource information combination;
inputting, by the first network device, the input data of each model into a corresponding model in the preset model group, wherein each model in the preset model group is for outputting a parameter value in a KPI of the to-be-processed resource information combination;
comparing, by the first network device, the KPI of each resource information combination with the target KPI;
selecting, by the first network device, at least one candidate resource information combination from the plurality of resource information combinations based on a comparison result that satisfies the preset condition;
inputting, by the first network device, the service data volume of the target time period and the candidate resource information combination into a preset resource utilization model, wherein the resource utilization model is for outputting a physical resource block utilization of the candidate resource information combination;
predicting, by the first network device, energy consumption of each candidate resource information combination based on the physical resource block utilization of each candidate resource information combination and a preset resource energy consumption model;
selecting, by the first network device, a target resource information combination from the at least one candidate resource information combination, wherein energy consumption of the target resource information combination is less than or equal to preset energy consumption.