Granted Patent B2

US 12,574,841 · App. 17/920,178 · Granted Mar 10, 2026

Inventors: Konstantinos Vandikas (Solna, SE); Lackis Eleftheriadis (Valbo, SE); Alexandros Nikou (Stockholm, SE); Yifei Jin (Solna, SE); Ioannis Fikouras (Stockholm, SE)

Assignee: Telefonaktiebolaget LM Ericsson (publ)

H04W52/0206H04W16/22H04W24/02H04W28/0221Y02D30/70

Abstract

Claims (88)

1 . A method of operating a radio network node to adjust power consumption of a telecommunications network, the method comprising:

determining a traffic prediction representing how each node of a set of nodes will interact with the radio network node over a period of time using a combined traffic model based on a traffic model of each node in the set of nodes, wherein the set of nodes comprises a portion of communication devices within a coverage area of the radio network node, and wherein determining the traffic prediction comprises formulating patterns that include one or more periods of inactivity of each node in the set of nodes and expected traffic at the radio network node; and

determining to enable or disable at least one power related feature of the radio network node based on the traffic prediction.

2 . The method of claim 1 , further comprising:

determining that a portion of the traffic prediction associated with a communication device is more than a threshold value different than an output of a traffic model generated by the communication device of the set of nodes; and

responsive to determining that the traffic prediction is more than a threshold value different than the output of the traffic model associated with the communication device, transmitting a retrain message to the communication device requesting the communication device retrain the traffic model associated with the communication device.

3 . The method of claim 1 , wherein the at least one power related feature comprises at least one of:

a discontinuous reception;

a reduction in transmission power; and

a reduction in reception power.

4 . The method of claim 1 , wherein the traffic model comprises a machine learning model,

wherein determining the traffic prediction comprises determining the traffic prediction from the machine learning model based on an input to the machine learning model, and

wherein the input comprises one or more of:

a location of each node in the set of nodes;

a distance of each node in the set of nodes from the radio network node;

a measured reference signal received power (RSRP);

a measured reference signal received quality (RSRQ);

a measured amount of bits per a time unit sent or received by each node in the set of nodes; or

a signal to noise ratio (SNR).

5 . The method of claim 1 , wherein determining the traffic prediction comprises generating the combined traffic model by averaging the traffic models of each node.

6 . The method of claim 1 , further comprising:

determining the set of nodes based on the communication devices that most often connect to the radio network node;

transmitting a request message to each node of the set of nodes, each request message requesting the traffic model be generated and provided to the radio network node;

responsive to transmitting the request message to each node of the set of nodes, receiving a response message from each node of the set of nodes, each response message including the traffic model; and

responsive to receiving the response message from each node of the set of nodes, generating the combined traffic model based on the traffic model from each node of the set of nodes,

wherein the request message comprises an indication of a type of the traffic model, an indication of at least one communication feature to be measured and modeled, and/or an amount of resources to be allocated by each node of the set of nodes to train the traffic model.

7 . The method of claim 6 , wherein the request message comprises an indication that each communication device in the set of nodes determines its power class and generates the traffic model based on the power class.

8 . The method of claim 6 , wherein the request message comprises the amount of resources to be allocated by each communication device in the set of nodes to train the traffic model, which includes at least one of: an amount of time to generate the traffic model; an amount of resources to use to generate the traffic model; and a threshold battery level at which to stop generating the traffic model.

9 . The method of claim 1 , wherein determining to enable or disable comprises determining to change a state of the radio network node based on comparing an output of the combined traffic model to a predetermined threshold value, the method further comprising:

responsive to determining to enable or disable, causing the state of the radio network node to transition between an active state and a sleep state.

10 . The method of claim 1 , wherein the radio network node is a next generation base station (gNB) and the telecommunications network is a new radio (NR) network.

11 . The method of claim 1 , wherein the set of nodes further comprises a portion of neighboring radio network nodes of the radio network node, the method further comprising:

determining the set of nodes based on the neighboring radio network nodes that most often perform handovers with the radio network node,

wherein a request message transmitted to each neighboring radio network node of the set of nodes comprises an indication of a type of the traffic model and/or an indication that the traffic model predicts uplink and/or downlink traffic at the radio network node originating from the neighboring radio network nodes.

12 . A radio network node in a telecommunications network, the radio network node comprising:

processing circuitry; and

memory coupled with the processing circuitry, wherein the memory includes instructions that when executed by the processing circuitry causes the radio network node to perform operations to adjust power consumption of the telecommunications network, the operations comprising:

determining a traffic prediction representing how each node of a set of nodes will interact with the radio network node over a period of time using a combined traffic model based on a traffic model of each node in the set of nodes, wherein the set of nodes comprises a portion of communication devices within a coverage area of the radio network node, and wherein determining the traffic prediction comprises formulating patterns that include one or more periods of inactivity of each node in the set of nodes and expected traffic at the radio network node; and

determining to enable or disable at least one power related feature of the radio network node based on the traffic prediction.

13 . The radio network node of claim 12 ,

wherein the at least one power related feature comprises at least one of:

a discontinuous reception;

a reduction in transmission power; and

a reduction in reception power,

wherein the traffic model comprises a machine learning model,

wherein determining the traffic prediction comprises determining the traffic prediction from the machine learning model based on an input to the machine learning model, and

wherein the input comprises one or more of:

a location of each node in the set of nodes;

a distance of each node in the set of nodes from the radio network node;

a measured reference signal received power (RSRP);

a measured reference signal received quality (RSRQ);

a measured amount of bits per a time unit sent or received by each node in the set of nodes; or

a signal to noise ratio (SNR).

14 . A method of operating a communication device in a telecommunication network to adjust power consumption of the telecommunications network, the method comprising:

receiving, a request message from a radio network node operating in the telecommunication network, the request message requesting the communication device generate and provide a traffic model to the radio network node, wherein the request message further comprises an indication of a type of the traffic model and an indication of at least one communication feature to be measured and modeled, wherein the type of the traffic model comprises a machine learning model;

responsive to receiving the request message, generating the traffic model, wherein generating the traffic model comprises measuring the at least one communication feature and training the machine learning model based on the at least one communication feature; and

responsive to generating the traffic model, transmitting a response message to the radio network node, the response message including the traffic model.

15 . The method of claim 14 , wherein the request message further comprises an amount of resources to be allocated by the communication device to generate the traffic model,

wherein the at least one communication feature comprises one or more of:

a power class of the communication device;

a location of the communication device;

a distance of the communication device from the radio network node;

a measured reference signal received power (RSRP);

a measured reference signal received quality (RSRQ);

a measured amount of bits per a time unit transmitted or received by the communication device; or

a signal to noise ratio (SNR), and

wherein the request message comprises the amount of resources to be allocated by the communication device to generate the traffic model, which includes at least one of: an amount of time to generate the traffic model, an amount of resources to use to generate the traffic model, and a threshold battery level at which to stop generating the traffic model.

16 . The method of claim 14 , further comprising:

receiving a third message from the radio network node, the third message requesting that the communication device retrain the traffic model;

responsive to receiving the third message, generating a new traffic model; and

responsive to retraining the traffic model, transmitting a fourth message to the radio network node including the new traffic model.

17 . The method of claim 14 , wherein the radio network node is a next generation base station (gNB) and the telecommunications network is a new radio, NR, network.

18 . A communication device in a telecommunications network, the communication device comprising:

processing circuitry; and

memory coupled with the processing circuitry, wherein the memory includes instructions that when executed by the processing circuitry causes the communication device to perform operations to adjust power consumption of the telecommunications network, the operations comprising:

receiving, a request message from a radio network node operating in the telecommunication network, the request message requesting the communication device generate and provide a traffic model to the radio network node, wherein the request message further comprises an indication of a type of the traffic model and an indication of at least one communication feature to be measured and modeled, wherein the type of the traffic model comprises a machine learning model;

responsive to receiving the request message, generating the traffic model, wherein generating the traffic model comprises measuring the at least one communication feature and training the machine learning model based on the at least one communication feature; and

responsive to generating the traffic model, transmitting a response message to the radio network node, the response message including the traffic model.

19 . The communication device of claim 18 , wherein the request message further comprises an amount of resources to be allocated by the communication device to generate the traffic model,

wherein the at least one communication feature comprises one or more of:

a power class of the communication device;

a location of the communication device;

a distance of the communication device from the radio network node;

a measured reference signal received power (RSRP);

a measured reference signal received quality (RSRQ);

a measured amount of bits per a time unit transmitted or received by the communication device; or

a signal to noise ratio (SNR), and

wherein the request message comprises the amount of resources to be allocated by the communication device to generate the traffic model, which includes at least one of: an amount of time to generate the traffic model, an amount of resources to use to generate the traffic model, and a threshold battery level at which to stop generating the traffic model.

Assignments (1)

Continuity (1)

References Cited (13)