A device and a methods for managing a set of positioning models, in particular artificial intelligence or machine learning positioning models

Abstract

A device (102) and a method for managing a set of positioning models, in particular artificial intelligence or machine learning positioning models, wherein the method comprises sending a deactivation signal (206) to a first positioning model (202) of the set of positioning models, in particular sending a first message comprising the deactivation signal (206), and sending an activation signal (208) to a second positioning model (204) of the set of positioning models, in particular sending a second message comprising the activation signal (208). A method for operating the first positioning model (202).

Description

[0001] R.414604

[0002] - 1 -

[0003] Description

[0004] Title

[0005] A device and a methods for managing a set of positioning models, in particular artificial intelligence or machine learning positioning models

[0006] Background

[0007] The invention relates to a device and a methods for managing a set of positioning models, in particular artificial intelligence or machine learning positioning models.

[0008] Artificial intelligence (Al) or machine learning (ML) based positioning systems in wireless communication networks reguire deployment and management of AI / ML positioning models that reguire efficient lifecycle management (LCM). Effective LCM is crucial for maintaining the accuracy and reliability of the AI / ML positioning models, particularly in dynamic environments where conditions can change rapidly and unpredictably.

[0009] Disclosure of the invention

[0010] The device and the method according to the independent claims provide a framework for managing a set of positioning models, in particular artificial intelligence or machine learning positioning models, in a wireless communication network, in particular according to their performance.

[0011] A method for managing a set of positioning models, in particular artificial intelligence (Al) or machine learning (ML) positioning models, in a wireless communication network, wherein the method comprises sending a deactivation signal to a first positioning model of the set of positioning models, in particular sending a first message comprising the deactivation signal, and sending an activation signal to a second positioning model of the set of positioning models, in particular sending a second message comprising the activation signal. This R.414604

[0012] - 2 - framework enables to switch between the positioning models, i.e., to deactivate the first positioning model and to activate the second positioning model, ensuring continuous positioning service. The method may be part of a location management function (LMF).

[0013] The method may comprise determining a performance of the first positioning model, and sending the deactivation signal to the first positioning model and the activation signal to the second positioning model when the performance is below a performance threshold. This means, switching from the first positioning model to the second positioning model is triggered by switching signals if the performance of the first positioning model has dropped below a predefined threshold.

[0014] The method may comprise determining the performance depending on an accuracy and / or a latency and / or a resource usage of the first positioning model. This means, switching from the first positioning model to the second positioning model is triggered based on the accuracy, latency or resource usage.

[0015] The method may comprise receiving the accuracy and / or the latency and / or the resource usage of the first positioning model from the first positioning model, in particular receiving a third message comprising the accuracy, the latency or the resource usage of the first positioning model. This means, the first positioning model reports its status with respect to the accuracy, latency or resource usage.

[0016] The method may comprise sending an instruction for adjusting at least one model parameter of the second positioning model to the second positioning model, in particular sending a fourth message comprising the instruction. The instruction may be sent based on the determined performance of the first positioning model, preferably when the performance is below the performance threshold. The at least one model parameter comprises for example a configuration parameter for initializing the second positioning model. This enables to activate the second positioning model with the initialized second positioning model.

[0017] A method for operating the first positioning model, in particular artificial intelligence or machine learning positioning model, in the wireless communication network, comprises receiving the deactivation signal in particular receiving the R.414604

[0018] - 3 - first message comprising the deactivation signal, and deactivating the first positioning model upon receipt of the deactivation signal, or receiving the activation signal, in particular receiving a second message comprising the activation signal, and activating the first positioning model upon receipt of the activation signal. This activates or deactivates the first positioning model seamlessly.

[0019] The method for operating the first positioning model may comprise determining and sending an accuracy and / or a latency and / or a resource usage of the first positioning model, in particular sending a third message comprising the accuracy, the latency or the resource usage of the first positioning model. This reports the status of the first positioning model.

[0020] The method for operating the first positioning model may comprise receiving an instruction for adjusting at least one model parameter of the first positioning model, in particular receiving a fourth message comprising the instruction, and adjusting the at least one model parameter according to the instruction. This adjusts the first positioning model accordingly.

[0021] The method for operating the first positioning model may comprise determining a performance of the first positioning model, and sending a switching signal, in particular sending a fifth message comprising the switching signal, to a second positioning model, when the performance is below a performance threshold. This switching signal enables a seamless model transition from the first positioning model to another positioning model.

[0022] The method for operating the first positioning model may comprise determining an accuracy of the first positioning model and determining the performance depending on the accuracy. Additionally, or alternatively, the method may comprise determining a latency of the first positioning model and determining the performance depending on the latency. Additionally, or alternatively, the method may comprise determining a resource usage of the first positioning model and determining the performance of the positioning model depending on the resource usage. R.414604

[0023] - 4 -

[0024] The method for operating the first positioning model may comprise receiving a switching signal, in particular receiving the fifth message comprising the switching signal, and activating the positioning model upon receipt of the switching signal. This switching signal enables a seamless model transition from another positioning model to the first positioning model.

[0025] A device for managing a set of positioning models, in particular artificial intelligence or machine learning positioning models, or for operating a first positioning model, in particular an artificial intelligence or machine learning positioning model, in a wireless communication network, is configured to execute the method(s). The device is for example a network node that provides the LMF.

[0026] A computer program may be provided, wherein the computer program comprises computer readable instructions that, when executed by a computer and / or a device according to an embodiment, cause the computer and / or the device to execute the method.

[0027] Some examples relate to a computer-readable storage medium comprising instructions which, when executed by a computer and / or the device, cause the computer and / or the device to carry out the method according to the disclosure.

[0028] Some examples relate to a data carrier signal carrying and / or characterizing the computer program according to the disclosure.

[0029] Further exemplary embodiments are derived from the following description and the drawing. In the drawing:

[0030] Fig. 1 schematically depicts a wireless communications network, Fig. 2 depicts messages for managing a set of positioning models.

[0031] Figure 1 depicts a wireless communications network 100 schematically. The wireless communication network 100 is for example a cellular communications network 100.

[0032] The wireless communications network 100 may be based on and / or adheres at least partially to at least one third generation partnership project, 3GPP, radio R.414604

[0033] - 5 - standard such as 4G (fourth generation), 5G (fifth generation), or 6G (sixth generation), or to another other radio access technology.

[0034] The wireless communications network 100 comprises a device 102, e.g., a network node, that is configured for providing and for managing a set of positioning models. The device 102 may provide the LMF.

[0035] Figure 2 depicts messages for managing the set of positioning models. The positioning models are for example artificial intelligence (Al) or machine learning (ML) positioning models.

[0036] The wireless communications network 100 may comprise a positioning model repository 200 that is configured for provisioning the positioning models to the communications network 100, in particular via the device 102.

[0037] A method for managing the set of positioning models is described for a first positioning model 202 and a second positioning model 204 of the set of positioning models. The method may collocate with or be part of the location management function (LMF) the communications network 100. The method may run as its own network function in the communications network 100. An instance of the LMF or the own network function may be associated with a network function (NF) instance identification (ID).

[0038] The method is not limited to managing two positioning models. The method is applied to more than two positioning models of the set of positioning models as described for the exemplary two positioning models.

[0039] The method may comprise an automated model deployment that comprises for example the following automated procedures:

[0040] Deployment automation:

[0041] Deployment automation involves using, e.g., 3GPP-defined Configuration and Provisioning of Network Functions (CPNF), messages to set up the AI / ML positioning models. This ensures that the AI / ML positioning models are R.414604

[0042] - 6 - configured, instantiated, and validated automatically, e.g., according to 3GPP SA5 specifications.

[0043] According to an example, the method uses a deployment script. The deployment script for example uses the 3GPP NFV-MANO architectural frameworks to orchestrate the instantiation of the AI / ML positioning models via Virtual Network Functions (VNF) and Cloud-native Network Functions (CNF) and the lifecycle management. The scripts for example use standard 3GPP interfaces for model deployment checks and validations.

[0044] Dynamic model activation and deactivation:

[0045] The method comprises sending a deactivation signal 206 to the first positioning model 202.

[0046] The method for example comprises sending a first message comprising the deactivation signal 206. The method comprises sending an activation signal 208 to the second positioning model 204. The method for example comprises sending a second message comprising the activation signal 208.

[0047] The deactivation signal 206 may include a first model ID of the first positioning model 202. This means, the deactivation signal 206 comprises the ID of the model that shall be deactivated.

[0048] The first message may be a model function deactivation (MFD) message.

[0049] The first message may comprise a first field that comprises the NF instance ID.

[0050] The first message may comprise a second field that comprises a deactivation reason.

[0051] This first message is for example triggered by a network condition that is detected in particular through NF monitoring. The first message is for example triggered by a network condition that no longer justifies the use of a specific AI / ML positioning model for enhanced positioning accuracy. R.414604

[0052] - 7 -

[0053] The activation signal 208 may include a second model ID of the second positioning model 204. This means, the activation signal 208 comprises the ID of the model that shall be activated.

[0054] The second message may be a model function activation (MFA) message.

[0055] The second message may comprise a first field that comprises the NF instance ID.

[0056] The second positioning model 204 may comprise configuration parameters that are adjustable. The second message may comprise a second field that comprises the configuration parameters to be adjusted.

[0057] The second positioning model 204 may be configured for activating upon a trigger condition. The second message may comprise a third field comprising the trigger condition.

[0058] This second message is for example triggered by a network condition that is detected in particular through NF monitoring. The second message is for example triggered by a network condition that requires the activation of a specific AI / ML positioning model for enhanced positioning accuracy.

[0059] According to the example, the device 102 provides the first positioning model 202 and the second provisioning model 204. This means the deactivation signal 206 and the activation signal 208 are transmitted within the device 102.

[0060] The method is not limited to transmitting the deactivation signal 206 or the activation signal 208 within the device 102. The deactivation signal 206 may be sent to a network node of the wireless communications network 100 providing the second positioning model 204. The activation signal 208 may be sent to a network node of the wireless communications network 100 providing the first positioning model 202. The same network node may provide the first positioning model 202 and the second positioning model 204. The deactivation signal 206 and the activation signal 208 may be sent to the same network node. This framework enables the network nodes or the network node to deactivate the first R.414604

[0061] - 8 - positioning model 202 and to activate the second positioning model 204, ensuring continuous positioning service.

[0062] The method may comprise determining a performance of the first positioning model 202, and sending the deactivation signal 206 to the first positioning model and the activation signal 208 to the second positioning 204 model when the performance is below a performance threshold.

[0063] This means, switching from the first positioning model 202 to the second positioning model 204 is triggered by switching signals if the performance of the first positioning model 202 has dropped below a predefined threshold. The second positioning model 204 is for example activated as a fallback model for the first positioning model 202.

[0064] The method for example continuously monitors environmental conditions such as signal strength and interference levels in the communications . When conditions indicate that a specific positioning model should be activated (e.g., during high multipath interference), the method comprises sending the activation signal 208 with the positioning model ID and configuration parameters. Conversely, when conditions no longer require the positioning model, the deactivation signal 206 with the positioning model ID is sent to gracefully shut the positioning model down.

[0065] The method may comprise determining the performance depending on an accuracy, a latency or a resource usage of the first positioning model. This means, switching from the first positioning model to the second positioning model is triggered based on the accuracy, latency or resource usage.

[0066] The method may comprise receiving the accuracy, the latency or the resource usage of the first positioning model 202 from the first positioning model 202, in particular receiving a third message 210 comprising the accuracy, the latency or the resource usage of the first positioning model 202. This means, the first positioning model 202 reports its status with respect to the accuracy, latency or resource usage. R.414604

[0067] - 9 -

[0068] The method comprises determining and sending the accuracy, the latency or the resource usage of the first positioning model 202, in particular the third message.

[0069] The method may comprise receiving the accuracy, the latency or the resource usage of the second positioning model 204 from the second positioning model 204, in particular receiving the third message 210 comprising the accuracy, the latency or the resource usage of the second positioning model 202. This means, the second positioning model 204 reports its status with respect to the accuracy, latency or resource usage.

[0070] The method may comprise determining the accuracy of the first positioning model 202 and determining the performance depending on the accuracy.

[0071] The method may comprise determining the latency of the first positioning model 202 and determining the performance of the first positioning model 202 depending on the latency.

[0072] The method may comprise determining the resource usage of the first positioning model 202 and determining the performance of the first positioning model 202 depending on the resource usage.

[0073] Determining the performance may integrate 3GPP TS 28.552-defined management services that monitor performance of network functions and the AI / ML positioning models, providing real-time data for operational intelligence. Determining the performance may use telemetry data aligned with 3GPP-defined performance management (PM) services. Determining the performance may utilizes collected data to perform trend analysis and anomaly detection, initiating preventive or corrective actions as per 3GPP management and orchestration (MANO) standards.

[0074] For example, if an AI / ML positioning model's performance degrades below the defined Quality of Service (QoS) parameters of the communications network 100, a Session and Service Continuity (SSC) mode is triggered to switch to the fallback model. The second positioning model 204 may be selected as fallback model from the set of positioning models depending on its ability to perform with higher accuracy than the first positioning model 202. R.414604

[0075] - 10 -

[0076] The method may comprise sending an instruction 212 for adjusting at least one model parameter of the second positioning model 204 to the second positioning model 204.

[0077] The method may comprises sending a fourth message comprising the instruction 210.

[0078] The at least one model parameter comprises for example a configuration parameter for initializing the second positioning model 204. This enables the network node that provides the second positioning model 204 to activate the second positioning model 204 with the initialized second positioning model 204.

[0079] The fourth message may be a Model Adjustment Instruction (MAI). The fourth message may comprise a first field for identifying a target positioning model. The first field for example comprises the second model ID.

[0080] The fourth message may comprise a second field for the at least one model parameter adjustment. The second field for example comprises the adjustment.

[0081] The fourth message may comprise a third field for identifying an adjustment reason. The third field for example comprises the adjustment reason.

[0082] Sending the adjustments provides dynamic adjustments to the AI / ML positioning model receiving the adjustment. The adjustment may be based on real-time analysis, ensuring optimal performance and compliance with 3GPP QoS requirements.

[0083] The method comprises receiving the instruction, in particular the fourth message, and adjusting the at least one parameter in the second positioning model 204.

[0084] The method comprises receiving the deactivation signal 206, e.g., the first message comprising the deactivation signal 206, and deactivating the first positioning model 202 upon receipt of the deactivation signal 206. R.414604

[0085] - 11 -

[0086] The method comprises receiving the activation signal 208, e.g., the second message comprising the activation signal 206, and activating the second positioning model 204 upon receipt of the activation signal 208.

[0087] The method may comprise determining and sending the accuracy, the latency or the resource usage of the positioning models from the set of positioning models.

[0088] The method may comprise transmitting the third message from the positioning models from the set of positioning models.

[0089] The method for operating the first positioning model 202 may sending a switching signal 214, in particular sending a fifth message comprising the switching signal 214, to the second positioning model 204, when the performance of the first positioning model 202 is below a performance threshold.

[0090] The fifth message may comprise a first field for identifying the current positioning model. The first field for example comprises the first model ID. The fifth message may comprise a second field for identifying the target positioning model. The second field for example comprises the second model ID. The fifth message may comprise a third field for identifying a reason for switching. The reason may be lack of accuracy, too large latency, too large resource usage.

[0091] This switching signal 214 enables a seamless model transition from the first positioning model 202 to the second positioning model 204, ensuring uninterrupted positioning service and adherence to 3GPP SSC modes.

[0092] The method for operating the second positioning model 204 may comprise receiving the switching signal 214, in particular receiving the fifth message comprising the switching signal 214, and activating the second positioning model 204 upon receipt of the switching signal 214.

[0093] The method my comprise operating the positioning models from the set of positioning models as described for the first positioning model 202 and / or the second positioning model 204.

Claims

R.414604- 12 -Claims1. A method for managing a set of positioning models, in particular artificial intelligence or machine learning positioning models, characterized in that the method comprises sending a deactivation signal (206) to a first positioning model (202) of the set of positioning models, in particular sending a first message comprising the deactivation signal (206), and sending an activation signal (208) to a second positioning model (204) of the set of positioning models, in particular sending a second message comprising the activation signal (208).

2. The method according to claim 1 , characterized in that the method comprises determining a performance of the first positioning model (202), and sending the deactivation signal (206) to the first positioning model (202) and the activation signal (208) to the second positioning model (204) when the performance is below a performance threshold.

3. The method according to claim 2, characterized in that the method comprises determining the performance depending on an accuracy and / or a latency and / or a resource usage of the first positioning model (202).

4. The method according to claim 3, characterized in that the method comprises receiving the accuracy and / or the latency and / or the resource usage of the first positioning model (202) from the first positioning model (202), in particular receiving a third message comprising the accuracy, the latency or the resource usage of the first positioning model (202).

5. The method according to one of the preceding claims, characterized in that the method comprisesR.414604- 13 - sending an instruction for adjusting at least one model parameter of the second positioning model (204) to the second positioning model (204), in particular sending a fourth message comprising the instruction.

6. A method for operating a first positioning model (202), in particular artificial intelligence or machine learning positioning model, in a wireless communication network, characterized in that the method comprises receiving a deactivation signal (206) in particular receiving a first message comprising the deactivation signal (206), and deactivating the first positioning model (202) upon receipt of the deactivation signal (206), or receiving an activation signal (208), in particular receiving a second message comprising the activation signal (208), and activating the first positioning model (202) upon receipt of the activation signal (208).

7. The method according to claim 6, characterized in that the method comprises determining and sending an accuracy and / or a latency and / or a resource usage of the first positioning model (202), in particular sending a third message comprising the accuracy, the latency or the resource usage of the first positioning model (202).

8. The method according to one of the claims 6 or 7, characterized in that the method comprises receiving an instruction for adjusting at least one model parameter of the first positioning model (202), in particular receiving a fourth message comprising the instruction, and adjusting the at least one model parameter according to the instruction.

9. The method according to one of the claims 6 to 8, characterized in that the method comprises determining a performance of the first positioning model (202), andR.414604- 14 - sending a switching signal, in particular sending a fifth message comprising the switching signal, to a second positioning model (204), when the performance is below a performance threshold.

10. The method according to claim 9, characterized in that the method comprises determining an accuracy of the first positioning model (202) and determining the performance depending on the accuracy, and / or determining a latency of the first positioning model (202) and determining the performance depending on the latency, and / or determining a resource usage of the first positioning model (202) and determining the performance of the positioning model depending on the resource usage.

11. The method according to one of the claims 6 to 10, characterized in that the method comprises receiving a switching signal, in particular receiving a fifth message comprising the switching signal, and activating the positioning model upon receipt of the switching signal.

12. A device (102) for managing a set of positioning models, in particular artificial intelligence or machine learning positioning models, or for operating a first positioning model (202), in particular an artificial intelligence or machine learning positioning model, in a wireless communication network, characterized in that the device is configured to execute the method(s) according to one of the preceding claims.

13. A computer program for, characterized in that the computer program comprises computer readable instructions that, when executed by a computer and / or the device (102) according to claim 12, cause the computer and / or the device (102) to execute the method(s) according to one of the claims 1 to 11.R.414604- 15 -14. A computer-readable storage medium comprising instructions which, when executed by a computer and / or the device (102) of claim 12, cause the computer and / or the device (102) to execute the method(s) according to one of the claims 1 to 11.

15. A data carrier signal carrying and / or characterizing the computer program of claim 13.

Images