# US 12,604,278 B2 · 5G inter- and intra-network node coordination for power consumption optimization and reduction > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 12,604,278 B2 - Document ID: US12604278B2 - Status: Granted - Kind: B2 - Application number: 18/324,528 - Filed: 2023-05-26 - Published: 2024-11-28 - Granted: 2026-04-14 - Earliest claimed: 2023-05-26 - Assignees: Cisco Technology, Inc. - Applicants: Cisco Technology, Inc. - Inventors: Sukrit Dasgupta - Web page: https://patentgrep.com/patent/US12604278B2 ## CPC - H04W 52/285: {taking into account the mobility of the user} - H04W 8/02: Processing of mobility data, e.g. registration information at HLR [Home Location Register] or VLR [Visitor Location Register]; Transfer of mobility data, e.g. between HLR, VLR or external networks - H04W 52/0203: {in the radio access network or backbone network of wireless communication networks} - Y02D 30/70: in wireless communication networks - H04W: WIRELESS COMMUNICATION NETWORKS (broadcast communication H04H; communication systems using wireless links for non-selective communication, e.g. wireless extensions H04M1/72) - Y02D: CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE ## Abstract Association and mobility patterns corresponding to client devices within a 5G network are tracked in real-time. A machine learning model is trained to identify, based on these patterns, periods of time for powering down one or more 5G nodes within the 5G network. The machine learning model, based on these periods of time, generates a set of power saving profiles that are used to automatically define power saving modes for the one or more 5G nodes. The machine learning model is updated according to changes to the association and mobility patterns resulting from the power saving modes. ## Claims ### Claim 1 A computer-implemented method comprising: continuously monitoring association patterns corresponding to a set of client devices within a 5G network, wherein the 5G network includes a set of 5G nodes, and wherein the association patterns denote changes to associations between the set of client devices and the set of 5G nodes; tracking mobility patterns in the 5G network across the set of 5G nodes and the set of client devices, wherein the mobility patterns are tracked based on the association patterns; dynamically training a machine learning model to identify periods of time for powering down one or more 5G nodes of the set of 5G nodes, wherein the machine learning model is dynamically trained based on historical telemetry data associated with the 5G network; generating a set of power saving profiles for the set of 5G nodes, wherein the set of power saving profiles are generated using the association patterns and the mobility patterns as input to the machine learning model; automatically defining power saving modes for the set of 5G nodes, wherein the power saving modes are implemented according to the set of power saving profiles; and updating the machine learning model according to changes to the association patterns and the mobility patterns resulting from the power saving modes. ### Claim 2 (depends on 1) The computer-implemented method of claim 1, further comprising: generating a set of profiles corresponding to the set of client devices, wherein the set of profiles are generated based on the association patterns; and updating the historical telemetry data to include the set of profiles to dynamically train the machine learning model. ### Claim 3 (depends on 1) The computer-implemented method of claim 1, wherein the mobility patterns include time series patterns that track different association patterns corresponding to a client device with different 5G nodes within the 5G network. ### Claim 4 (depends on 1) The computer-implemented method of claim 1, further comprising: receiving a set of power saving schedules associated with one or more other 5G networks, wherein the one or more other 5G networks provide overlapping network coverage with the 5G network; and determining time periods during which a subset of the set of 5G nodes are powered down and serviced by the one or more other 5G networks. ### Claim 5 (depends on 1) The computer-implemented method of claim 1, further comprising: reducing power consumption for a set of higher transmit power frequencies as a result of a determination that the set of client devices can remain connected to the 5G network through a set of lower transmit power frequencies, wherein the determination is generated based on the association patterns and the mobility patterns. ### Claim 6 (depends on 1) The computer-implemented method of claim 1, wherein the set of power saving profiles define period of time when the set of 5G nodes are not servicing any client devices and can be powered down. ### Claim 7 (depends on 1) The computer-implemented method of claim 1, wherein automatically defining the power saving modes for the set of 5G nodes includes: generating schedules for the set of 5G nodes to enter the power saving modes; and incrementally communicating the schedules to the set of 5G nodes, wherein the schedules are incrementally communicated to limit network disruption amongst the set of client devices. ### Claim 8 A system, comprising: one or more processors; and memory storing thereon instructions that, as a result of being executed by the one or more processors, cause the system to: continuously monitor association patterns corresponding to a set of client devices within a 5G network, wherein the 5G network includes a set of 5G nodes, and wherein the association patterns denote changes to associations between the set of client devices and the set of 5G nodes; track mobility patterns in the 5G network across the set of 5G nodes and the set of client devices, wherein the mobility patterns are tracked based on the association patterns; dynamically train a machine learning model to identify periods of time for powering down one or more 5G nodes of the set of 5G nodes, wherein the machine learning model is dynamically trained based on historical telemetry data associated with the 5G network; generate a set of power saving profiles for the set of 5G nodes, wherein the set of power saving profiles are generated using the association patterns and the mobility patterns as input to the machine learning model; automatically define power saving modes for the set of 5G nodes, wherein the power saving modes are implemented according to the set of power saving profiles; and update the machine learning model according to changes to the association patterns and the mobility patterns resulting from the power saving modes. ### Claim 9 (depends on 8) The system of claim 8, wherein the instructions further cause the system to: generate a set of profiles corresponding to the set of client devices, wherein the set of profiles are generated based on the association patterns; and update the historical telemetry data to include the set of profiles to dynamically train the machine learning model. ### Claim 10 (depends on 8) The system of claim 8, wherein the mobility patterns include time series patterns that track different association patterns corresponding to a client device with different 5G nodes within the 5G network. ### Claim 11 (depends on 8) The system of claim 8, wherein the instructions further cause the system to: receive a set of power saving schedules associated with one or more other 5G networks, wherein the one or more other 5G networks provide overlapping network coverage with the 5G network; and determine time periods during which a subset of the set of 5G nodes are powered down and serviced by the one or more other 5G networks. ### Claim 12 (depends on 8) The system of claim 8, wherein the instructions further cause the system to: reduce power consumption for a set of higher transmit power frequencies as a result of a determination that the set of client devices can remain connected to the 5G network through a set of lower transmit power frequencies, wherein the determination is generated based on the association patterns and the mobility patterns. ### Claim 13 (depends on 8) The system of claim 8, wherein the set of power saving profiles define period of time when the set of 5G nodes are not servicing any client devices and can be powered down. ### Claim 14 (depends on 8) The system of claim 8, wherein the instructions that cause the system to automatically define the power saving modes for the set of 5G nodes further cause the system to: generate schedules for the set of 5G nodes to enter the power saving modes; and incrementally communicate the schedules to the set of 5G nodes, wherein the schedules are incrementally communicated to limit network disruption amongst the set of client devices. ### Claim 15 A non-transitory, computer-readable storage medium storing thereon executable instructions that, as a result of being executed by one or more processors of a computer system, cause the computer system to: continuously monitor association patterns corresponding to a set of client devices within a 5G network, wherein the 5G network includes a set of 5G nodes, and wherein the association patterns denote changes to associations between the set of client devices and the set of 5G nodes; track mobility patterns in the 5G network across the set of 5G nodes and the set of client devices, wherein the mobility patterns are tracked based on the association patterns; dynamically train a machine learning model to identify periods of time for powering down one or more 5G nodes of the set of 5G nodes, wherein the machine learning model is dynamically trained based on historical telemetry data associated with the 5G network; generate a set of power saving profiles for the set of 5G nodes, wherein the set of power saving profiles are generated using the association patterns and the mobility patterns as input to the machine learning model; automatically define power saving modes for the set of 5G nodes, wherein the power saving modes are implemented according to the set of power saving profiles; and update the machine learning model according to changes to the association patterns and the mobility patterns resulting from the power saving modes. ### Claim 16 (depends on 15) The non-transitory, computer-readable storage medium of claim 15, wherein the executable instructions further cause the computer system to: generate a set of profiles corresponding to the set of client devices, wherein the set of profiles are generated based on the association patterns; and update the historical telemetry data to include the set of profiles to dynamically train the machine learning model. ### Claim 17 (depends on 15) The non-transitory, computer-readable storage medium of claim 15, wherein the mobility patterns include time series patterns that track different association patterns corresponding to a client device with different 5G nodes within the 5G network. ### Claim 18 (depends on 15) The non-transitory, computer-readable storage medium of claim 15, wherein the executable instructions further cause the computer system to: receive a set of power saving schedules associated with one or more other 5G networks, wherein the one or more other 5G networks provide overlapping network coverage with the 5G network; and determine time periods during which a subset of the set of 5G nodes are powered down and serviced by the one or more other 5G networks. ### Claim 19 (depends on 15) The non-transitory, computer-readable storage medium of claim 15, wherein the executable instructions further cause the computer system to: reduce power consumption for a set of higher transmit power frequencies as a result of a determination that the set of client devices can remain connected to the 5G network through a set of lower transmit power frequencies, wherein the determination is generated based on the association patterns and the mobility patterns. ### Claim 20 (depends on 15) The non-transitory, computer-readable storage medium of claim 15, wherein the executable instructions that cause the computer system to automatically define the power saving modes for the set of 5G nodes further cause the computer system to: generate schedules for the set of 5G nodes to enter the power saving modes; and incrementally communicate the schedules to the set of 5G nodes, wherein the schedules are incrementally communicated to limit network disruption amongst the set of client devices.