CN-122029860-A - Graph learning subband allocation

Abstract

Methods, apparatus, devices, and computer readable storage media for graph learning subband allocation. The method comprises receiving (710) a report of identifiers of a predetermined number of neighboring subnets of the subnet from a second device, the second device being located in the subnet, the predetermined number of neighboring subnets being associated with a predefined interference metric, generating (720) a subnet deployment conflict graph based on the reported identifiers, obtaining (730) an inference model associated with the subband allocation, and transmitting (740) to the second device at least one of at least one target subband, the at least one target subband being determined based at least on the inference model and being allocated to the subnet, or the inference model together with the subnet deployment conflict graph.

Inventors

• D. Abbott
• R. B. abru
• L Salon
• G Berardi Lee
• R. Ojikule Adgon
• T. H. Jacobson

Assignees

• 诺基亚技术有限公司

Dates

Publication Date

20260512

Application Date

20231011

Claims (20)

1. 1. A first apparatus, comprising: at least one processor, and At least one memory storing instructions that, when executed by the at least one processor, cause the first device to at least: Receiving a report of identifiers of a predetermined number of neighboring subnets of a subnet from a second device, the second device being located in the subnet, the predetermined number of neighboring subnets being associated with a predefined interference metric; generating a subnet deployment conflict graph based on the reported identifiers; acquiring an inference model associated with the subband assignment, and Transmitting to the second device at least one of: at least one target subband determined based at least on the inference model and assigned to the subnet, or The inference model is configured with the subnet deployment conflict graph.
2. 2. The first device of claim 1, wherein the first device is caused to: determining a configuration associated with a measurement of a reference signal from a neighboring subnet, wherein the configuration indicates at least one interference metric and a rule by which the second apparatus determines the identifiers of the predetermined number of neighboring subnets based on the at least one interference metric, and The configuration is sent to the second device.
3. 3. The first apparatus of claim 2, wherein the rule indicates at least one of: The identifiers of the predetermined number of neighboring subnets associated with the strongest interference are to be determined, or The identifiers of the predetermined number of neighboring subnets associated with the lowest signal-to-interference ratio are to be determined.
4. 4. A first apparatus according to any of claims 1-3, wherein the inference model for the subband allocation has been trained based at least on at least one historical subnet deployment conflict graph, a reported historical identifier, and a configured or predetermined loss function.
5. 5. The first device of claim 4, wherein the first device is caused to: Determining the at least one target subband assigned to the subnet based on the inference model and the identifier, Wherein the identifier is an input of the inference model and the at least one target subband is an output of the inference model.
6. 6. The first device of any of claims 1-5, wherein the first device is caused to: Obtaining information from the at least one second device related to respective historical identifiers of the predetermined number of subnets, the predetermined number of subnets being adjacent to the respective subnet in which the at least one second device is located, the predetermined number of subnets having a predefined interference metric; generating the at least one historical subnet deployment conflict graph based on the respective historical identifiers, and The inference model for the subband assignment is trained based on the at least one historical subnet deployment and a predetermined or configured loss function.
7. 7. The first apparatus of any of claims 1-6, wherein the inference model comprises a machine learning model associated with a graph neural network.
8. 8. The first apparatus of any of claims 1-7, wherein the inference model is trained using a Potts model-based loss function.
9. 9. The first device of any of claims 1-8, wherein the predetermined number is associated with a number of available subbands.
10. 10. The first apparatus of any of claims 1-9, wherein the first apparatus comprises a centralized access point or base station and the second apparatus comprises a distributed access point or user equipment located in the subnet.
11. 11. A second device in a sub-network, comprising: at least one processor, and At least one memory storing instructions that, when executed by the at least one processor, cause the second device to at least: Transmitting a report of identifiers of a predetermined number of neighboring subnets of the subnet to a first device, the predetermined number of neighboring subnets being associated with a predefined interference metric; Receiving from the first device at least one of: at least one target subband determined based on the identifier and allocated to the subnet, or An inference model for subband assignment along with a subnet deployment conflict graph associated with the identifier.
12. 12. The second device of claim 11, wherein the second device is caused to: Receiving a configuration associated with a measurement of a reference signal from a neighboring subnet from the first apparatus, wherein the configuration indicates at least one interference metric and a rule by which the second apparatus determines the identifiers of the predetermined number of neighboring subnets based on the at least one interference metric, and Measuring respective strengths of the reference signals from the neighboring subnets based on the configuration, and The report is generated based on the measured respective intensities and the rules.
13. 13. The second apparatus of claim 11 or 12, wherein the rule indicates at least one of: The identifiers of the predetermined number of neighboring subnets associated with the strongest interference are to be determined, or The identifiers of the predetermined number of neighboring subnets associated with the lowest signal-to-interference ratio are to be determined.
14. 14. A second device according to any of claims 11-13, wherein the second device is caused to: At least one sub-band available to the sub-network in which the second device is located is determined based on the inference model for the sub-band allocation and a sub-network deployment conflict graph.
15. 15. The second device of claim 14, wherein the second device is caused to: transmitting node embeddings of at least one other second device located in the neighboring subnetwork to the second device; Receiving, from the at least one further second device, a respective further node embedding of the at least one further second device; iteratively updating the node embedment of the second device based on the respective other node embedment by using the inference model and the subnet deployment conflict graph, and The at least one subband available to the subnet is determined based on the updated node embedding of the second device.
16. 16. The second apparatus of any of claims 11-15, wherein the inference model comprises a machine learning model associated with a graph neural network.
17. 17. The second apparatus of any of claims 11-16, wherein the inference model is trained using a Potts model-based loss function.
18. 18. The first device of any of claims 11-17, wherein the predetermined number is associated with a number of the available subbands.
19. 19. The first apparatus of any of claims 11-18, wherein the first apparatus comprises a centralized access point or base station and the second apparatus comprises a distributed access point or user equipment located in the subnet.
20. 20. A method, comprising: Receiving a report of identifiers of a predetermined number of neighboring subnets of a subnet from a second device, the second device being located in the subnet, the predetermined number of neighboring subnets being associated with a predefined interference metric; generating a subnet deployment conflict graph based on the reported identifiers; acquiring an inference model associated with the subband assignment, and Transmitting to the second device at least one of: at least one target subband determined based at least on the inference model and assigned to the subnet, or The inference model is configured with the subnet deployment conflict graph.

Description

Graph learning subband allocation Technical Field Various example embodiments of the present disclosure relate generally to the field of telecommunications and, in particular, relate to methods, apparatus, devices, and computer-readable storage media for graph learning subband allocation, particularly for generation 6 (6G) subnets. Background Intra-X (In-X) subnets (hereinafter also referred to as subnets) have been proposed as promising components to meet the extreme performance requirements In terms of delay, reliability and/or throughput envisaged for some short range scenarios In 6G radio access technologies. For example, the subnetworks may be installed in specific entities, e.g., in vehicles, in bodies, in houses, etc., to provide life critical data services with extreme performance over local capillary coverage. Disclosure of Invention In a first aspect of the present disclosure, a first apparatus is provided. The first apparatus includes at least one processor and at least one memory storing instructions that, when executed by the at least one processor, cause the first apparatus to at least receive a report of identifiers of a predetermined number of neighboring subnets of the subnet from a second apparatus, the second apparatus being located at the subnet, the predetermined number of neighboring subnets being associated with a predefined interference metric, generate a subnet deployment conflict graph based on the reported identifiers, obtain an inference model associated with a subband assignment, and send at least one of at least one target subband to the second apparatus, the at least one target subband being determined based at least on the inference model and assigned to the subnet, or the inference model along with the subnet deployment conflict graph. In a second aspect of the present disclosure, a second apparatus is provided. The second apparatus includes at least one processor and at least one memory storing instructions that, when executed by the at least one processor, cause the second apparatus to at least send a report of identifiers of a predetermined number of neighboring subnets of the subnet to the first apparatus, the predetermined number of neighboring subnets being associated with a predefined interference metric, receive at least one of at least one target subband from the first apparatus, the at least one target subband determined based on the identifier and assigned to the subnet, or an inference model for subband assignment along with a subnet deployment conflict graph associated with the identifier. In a third aspect of the present disclosure, a method is provided. The method includes receiving a report of identifiers of a predetermined number of neighboring subnets of the subnet from a second device, the second device being located in the subnet, the predetermined number of neighboring subnets being associated with a predefined interference metric, generating a subnet deployment conflict graph based on the reported identifiers, obtaining an inference model associated with a subband assignment, and transmitting to the second device at least one of at least one target subband determined based at least on the inference model and assigned to the subnet, or the inference model along with the subnet deployment conflict graph. In a fourth aspect of the present disclosure, a method is provided. The method includes transmitting a report of identifiers of a predetermined number of neighboring subnets of the subnet to the first device, the predetermined number of neighboring subnets being associated with a predefined interference metric, receiving at least one of at least one target subband from the first device, the at least one target subband being determined based on the identifiers and assigned to the subnet, or an inference model for subband assignment along with a subnet deployment conflict graph associated with the identifiers. In a fifth aspect of the present disclosure, a first apparatus is provided. The first apparatus includes means for receiving a report of identifiers of a predetermined number of neighboring subnets of the subnet from a second apparatus, the second apparatus being located in the subnet, the predetermined number of neighboring subnets being associated with a predefined interference metric, means for generating a subnet deployment conflict graph based on the reported identifiers, means for obtaining an inference model associated with a subband assignment, and means for transmitting to the second apparatus at least one target subband determined and assigned to the subnet based at least on the inference model, or the inference model along with the subnet deployment conflict graph. In a sixth aspect of the present disclosure, a second apparatus is provided. The second apparatus includes means for sending a report of identifiers of a predetermined number of neighboring subnets of the subnet to the first apparatus, the predetermined number of neighboring subn