Positioning method, LMF network element, NWDAF network element and positioning system

Abstract

The invention provides a positioning method, an LMF network element, an NWDAF network element and a positioning system, and relates to the technical field of position positioning, and the method executed by the LMF network element comprises the following steps: receiving a positioning request for user equipment, the positioning request carrying a positioning demand; according to the positioning requirement, judging whether the user equipment needs to be positioned by utilizing a machine learning positioning mode or not; and under the condition that the user equipment needs to be positioned in the machine learning positioning mode, positioning parameters are input into the target machine learning model to position the user equipment, and the positioning parameters comprise parameters required when the user equipment is positioned in multiple positioning modes.

Description

Technical Field

[0001] The present disclosure relates to the field of position positioning technology, and in particular to a positioning method, a LMF network element, a NWDAF network element, and a positioning system. Background Art

[0002] With the development of communication technology, the demand for precise positioning of user equipment (UE) is increasing. Summary of the invention

[0003] In the related art, the positioning result of the UE is inaccurate.

[0004] After analysis, it is found that in the related technology, when the UE needs to be positioned, generally one of multiple traditional positioning methods is selected to position the UE. Here, the traditional positioning method can position the UE by using parameters related to signal transmission time generated when the UE interacts with the base station.

[0005] Further analysis found that traditional positioning methods cannot accurately locate UE in some cases.

[0006] For example, when there are obstructions or obstacles between the base station and the UE (i.e., in a non-line-of-sight environment), signal transmission is interfered by the obstructions or obstacles, resulting in the inability to obtain accurate signal transmission time, making it impossible for traditional positioning methods to accurately locate the UE.

[0007] In order to solve the above problems, the embodiments of the present disclosure propose the following solutions.

[0008] According to one aspect of an embodiment of the present disclosure, a positioning method is provided, which is executed by a location management function LMF network element, including: receiving a positioning request for a user equipment, the positioning request carrying a positioning requirement; judging whether it is necessary to locate the user equipment using a machine learning positioning method according to the positioning requirement; and in the case where it is necessary to locate the user equipment using a machine learning positioning method, inputting positioning parameters into a target machine learning model to locate the user equipment, the positioning parameters including the parameters required for each of the multiple positioning methods when locating the user equipment respectively.

[0009] In some embodiments, the target machine learning model is trained by a network data analysis function NWDAF network element, and the first information of the LMF network element matches the second information of the NWDAF network element, wherein: the first information includes the identifier of the LMF network element, the format of the machine learning model supported by the LMF network element, and at least one of the operating environments of the machine learning model supported by the LMF network element; the second information includes at least one of the identifier of the LMF network element supported by the NWDAF network element, the format of the machine learning model trained by the NWDAF network element, and the operating environment of the machine learning model trained by the NWDAF network element.

[0010] In some embodiments, when there is no need to use a machine learning positioning method to locate the user equipment, any one of the multiple positioning methods is used to locate the user equipment.

[0011] In some embodiments, the positioning requirement includes the positioning accuracy required for the user device; judging whether it is necessary to use a machine learning positioning method to locate the user device based on the positioning requirement includes: when the positioning accuracy is greater than a preset accuracy, judging that it is necessary to use a machine learning positioning method to locate the user device; and when the positioning accuracy is less than or equal to the preset accuracy, judging that it is not necessary to use a machine learning positioning method to locate the user device.

[0012] In some embodiments, whether the LMF network element has a machine learning model that matches the user device is determined based on third information of the user device, the third information including at least one of the type of area in which the user device is located and an identifier of a network slice to which the user device accesses; if the LMF network element does not have a machine learning model that matches the user device, a subscription request is sent to a network data analysis function NWDAF network element, the subscription request including the third information; and the target machine learning model is received from the NWDAF network element, wherein the target machine learning model is trained by the NWDAF network element based on the third information.

[0013] In some embodiments, the actual location of the user device is obtained; an evaluation index of the target machine learning model is determined based on the actual location of the user device and the location output by the target machine learning model, the evaluation index including at least one of positioning accuracy and average positioning error; and the evaluation index is sent to a positioning monitoring demander, the positioning monitoring demander including a network data analysis function NWDAF network element that provides the target machine learning model and at least one of the initiators of the positioning request.

[0014] In some embodiments, when the evaluation index does not meet the first preset requirement, the use of the target machine learning model to locate the user device is stopped.

[0015] According to another aspect of an embodiment of the present disclosure, a positioning method is provided, which is executed by a network data analysis function NWDAF network element, including: obtaining a sample position of a sample user device; obtaining positioning parameters of the sample user device, the positioning parameters including parameters required for respectively positioning the sample user device at the sample position using a plurality of positioning methods; using the positioning parameters as input and the sample position as output, training a machine learning model to obtain a target machine learning model; and sending the target machine learning model to a location management function LMF network element, so that the LMF network element uses the target machine learning model to locate the user device that needs to be located.

[0016] In some embodiments, the first information of the LMF network element matches the second information of the NWDAF network element, wherein: the first information includes at least one of the identification of the LMF network element, the format of the machine learning model supported by the LMF network element, and the operating environment of the machine learning model supported by the LMF network element; and the second information includes at least one of the identification of the LMF network element supported by the NWDAF network element, the format of the machine learning model trained by the NWDAF network element, and the operating environment of the machine learning model trained by the NWDAF network element.

[0017] In some embodiments, a subscription request sent by the LMF network element is received, the subscription request including third information of the user equipment, the third information including at least one of the area type in which the user equipment is located and an identifier of a network slice accessed by the user equipment; and the sample user equipment is determined based on the third information, wherein the third information of the sample user equipment is the same as the third information of the user equipment.

[0018] In some embodiments, an evaluation index of the target machine learning model is received from the LMF network element, and the evaluation index includes at least one of positioning accuracy and average positioning error; and when the evaluation index does not meet the second preset requirement, the target machine learning model is retrained, and the retrained target machine learning model is sent to the LMF network element.

[0019] In some embodiments, the positioning parameters include multiple parameters of uplink arrival time difference, downlink arrival time difference, uplink arrival angle, downlink departure angle, and multipath travel time.

[0020] In some embodiments, the NWDAF network element is discovered by the LMF network element through a network storage function NRF network element storing the first information and the second information.

[0021] According to another aspect of an embodiment of the present disclosure, a location management function network element is provided, comprising: a module configured to execute the method performed by the LMF network element as described in any one of the above embodiments.

[0022] According to another aspect of an embodiment of the present disclosure, a location management function network element is provided, comprising: a memory; and a processor coupled to the memory, configured to execute the method performed by the LMF network element as described in any one of the above embodiments based on instructions stored in the memory.

[0023] According to another aspect of an embodiment of the present disclosure, a network data analysis function network element is provided, comprising: a module configured to execute the method performed by the NWDAF network element as described in any one of the above embodiments.

[0024] According to another aspect of an embodiment of the present disclosure, a network data analysis function network element is provided, comprising: a memory; and a processor coupled to the memory, configured to execute the method performed by the NWDAF network element as described in any one of the above embodiments based on instructions stored in the memory.

[0025] According to another aspect of the embodiments of the present disclosure, a positioning system is provided, comprising: a location management function network element as described in any one of the above embodiments; and a network data analysis function network element as described in any one of the above embodiments.

[0026] According to another aspect of an embodiment of the present disclosure, a computer-readable storage medium is provided, comprising computer program instructions, wherein when the computer program instructions are executed by a processor, the steps of the method described in any one of the above embodiments are implemented.

[0027] According to another aspect of the embodiments of the present disclosure, a computer program product is provided, including a computer program, wherein when the computer program is executed by a processor, the steps of the method described in any one of the above embodiments are implemented.

[0028] In the disclosed embodiment, when the location management function (LMF) network element determines that a machine learning positioning method is needed for positioning based on positioning requirements, the LMF network element will use multiple positioning methods to respectively position the UE and input the parameters required by each method into the target machine learning model to position the UE. In this way, when positioning the UE, the machine learning model can comprehensively consider the parameters required by each of the multiple positioning methods, thereby improving the accuracy of UE positioning.

[0029] The technical solution of the present disclosure is further described in detail below through the accompanying drawings and embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0030] In order to more clearly illustrate the embodiments of the present disclosure or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present disclosure. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0031] Figure 1 It is a flowchart of a positioning method according to some embodiments of the present disclosure.

[0032] Figure 2 It is a flowchart of a positioning method according to some other embodiments of the present disclosure.

[0033] Figure 3 It is a flowchart of a positioning method according to some embodiments of the present disclosure.

[0034] Figure 4 It is a schematic diagram of the structure of an LMF network element according to some embodiments of the present disclosure.

[0035] Figure 5 It is a schematic diagram of the structure of a NWDAF network element according to some embodiments of the present disclosure.

[0036] Figure 6 is a schematic diagram of the structure of an electronic device according to some embodiments of the present disclosure. DETAILED DESCRIPTION

[0037] The following will be combined with the drawings in the embodiments of the present disclosure to clearly and completely describe the technical solutions in the embodiments of the present disclosure. Obviously, the described embodiments are only part of the embodiments of the present disclosure, rather than all the embodiments. Based on the embodiments in the present disclosure, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present disclosure.

[0038] Unless specifically stated otherwise, the relative arrangement of components and steps, the numerical expressions and numerical values ​​set forth in these embodiments do not limit the scope of the present disclosure.

[0039] At the same time, it should be understood that for the convenience of description, the sizes of the various parts shown in the drawings are not necessarily drawn according to the actual proportional relationship.

[0040] Technologies, methods, and equipment known to ordinary technicians in the relevant art may not be discussed in detail, but where appropriate, the technologies, methods, and equipment should be considered as part of the specification.

[0041] In all examples shown and discussed herein, any specific values ​​should be interpreted as merely exemplary and not as limiting. Therefore, other examples of the exemplary embodiments may have different values.

[0042] It should be noted that like reference numerals and letters refer to similar items in the following figures, and therefore, once an item is defined in one figure, it need not be further discussed in subsequent figures.

[0043] In addition, in the description of the present disclosure, the terms "first", "second", "third", etc. are used for descriptive purposes only and should not be understood as indicating or implying relative importance and order. Similarly, although operations are depicted in a particular order in the accompanying drawings, this should not be understood as requiring that such operations be performed in the particular order shown or in a sequential order, or requiring that all illustrated operations be performed to achieve the desired results. In some cases, multitasking and parallel processing can be advantageous.

[0044] Figure 1 1 is a flow chart of a positioning method according to some embodiments of the present disclosure. Here, the positioning method is executed by an LMF network element.

[0045] In step 102, a positioning request for the UE is received. Here, the positioning request carries a positioning requirement. For example, the positioning requirement includes a requirement for positioning accuracy.

[0046] As some embodiments, the LMF network element receives a positioning request for the UE from an access and mobility management function (AMF) network element.

[0047] For example, when the UE needs to locate itself, the AMF network element can send a positioning request to the LMF network element to request the positioning of the UE in response to the positioning request sent by the UE via an uplink non-access layer transport message (UL NAS TRANSPORT message).

[0048] For another example, when a service provider that needs the UE location to provide services needs to locate the UE, the AMF network element can respond to the positioning request sent by the Gateway Mobile Location Center (GMLC) through the Namf Location Providing Positioning Information (Namf_Location_ProvidePositioningInfo) service, and send a positioning request to the LMF network element to request the positioning of the UE.

[0049] Here, the GMLC may send a positioning request to the AMF network element in response to a request for positioning the UE sent by a service provider. The service provider may include, for example, a navigation service provider and / or an emergency service provider. The service provider may interact with the GMLC, for example, through an application function (AF) network element or a location service (LCS) client.

[0050] As some implementation methods, the AMF network element may send a positioning request for the UE to the LMF network element via an Nlmf location determination request (Nlmf_Location_DetermineLocation_request).

[0051] In step 104, based on the positioning requirements carried in the positioning request, it is determined whether it is necessary to use a machine learning positioning method to locate the UE.

[0052] As some implementation methods, different positioning requirements correspond to different positioning methods for positioning the UE. The following will introduce how to position the UE according to different positioning requirements.

[0053] It should be understood that the machine learning positioning method refers to using a machine learning model to locate the UE.

[0054] In step 106, when the UE needs to be positioned using a machine learning positioning method, the positioning parameters are input into the target machine learning model to position the UE. Here, the positioning parameters include parameters required for each of the multiple positioning methods to position the UE.

[0055] As some implementation methods, the LMF network element may return the positioning result of the UE to the AMF network element through an Nlmf location positioning response (Nlmf_Location_DetermineLocation response), so that the AMF network element returns the positioning result to the UE or the service provider.

[0056] As some implementations, the LMF network element may have an analytics logical function (AnLF) so that the LMF network element utilizes the target machine learning model using AnLF.

[0057] As some embodiments, the positioning parameters include multiple parameters of uplink time difference of arrival (UL-TDoA), downlink time difference of arrival (DL-TDoA), uplink angle of arrival (UL-AoA), downlink angle of departure (DL-AoD), and multi-path travel time (Multi-RTT).

[0058] For example, the positioning parameters include any two, any three, any four, or all of UL-TDoA, DL-TDoA, UL-AoA, DL-AoD, and Multi-RTT.

[0059] It should be understood that the multiple positioning methods include different positioning methods for positioning the UE. For example, each positioning method in the multiple positioning methods only uses one parameter among the above multiple parameters to position the UE, and different positioning methods each use different parameters to position the UE.

[0060] In the above embodiment, when the LMF network element determines that it needs to use a machine learning positioning method for positioning based on positioning requirements, the LMF network element will use multiple positioning methods to respectively position the UE and input the parameters required by each method into the target machine learning model to position the UE. In this way, when positioning the UE, the machine learning model can comprehensively consider the parameters required by each of the multiple positioning methods, thereby improving the accuracy of UE positioning.

[0061] Next, in combination with some embodiments of the present disclosure, it is introduced how to select an LMF network element for positioning the UE using a machine learning positioning method.

[0062] In some embodiments, the LMF network element that uses machine learning positioning to locate the UE is selected by the AMF network element based on the current load of the LMF network element and / or the distance between the AMF network element and the LMF network element.

[0063] For example, the LMF network element used to locate the UE is the LMF network element that is closest to the AMF network element; for another example, the LMF network element used to locate the UE is the LMF network element that currently has the lowest processing load; for another example, the LMF network element used to locate the UE is the LMF network element that is closest to the AMF network element and currently has the smallest processing load.

[0064] In the above embodiments, on the one hand, the LMF network element used to locate the UE is selected according to the current load of the LMF network element, and the LMF network element with a smaller current load can be selected to locate the UE, thereby helping to improve the efficiency of UE positioning; on the other hand, the LMF network element used to locate the UE is selected according to the distance between the AMF network element and the LMF network element, which helps to reduce the delay when the AMF network element interacts with the LMF network element, thereby helping to improve the efficiency of UE positioning.

[0065] In some embodiments, the LMF network elements in the first part of the multiple LMF network elements do not support positioning using machine learning positioning methods, and the LMF network elements in the second part of the multiple LMF network elements have an identification that supports machine learning positioning methods (that is, this part of the LMF network elements supports positioning using machine learning positioning methods), and the LMF network elements used to locate the UE are selected from the LMF network elements in the second part of the multiple LMF network elements. In this way, the optional range when the AMF network element selects the LMF network element can be limited, so that the AMF network element can quickly select the LMF network element used to locate the UE, which helps to improve the efficiency of UE positioning.

[0066] Next, in combination with some embodiments of the present disclosure, it is introduced how the LMF network element determines whether it is necessary to use machine learning positioning to locate the UE.

[0067] In some embodiments, the positioning requirement carried in the positioning request includes the positioning accuracy required for the UE.

[0068] When the positioning accuracy is greater than the preset accuracy, it is determined that the UE needs to be positioned using the machine learning positioning method; and when the positioning accuracy is less than or equal to the preset accuracy, it is determined that the UE does not need to be positioned using the machine learning positioning method. In this way, when the positioning accuracy requirement is high, the machine learning positioning method is used to position the UE, which can take into account both the high positioning accuracy requirement and the positioning efficiency.

[0069] In some embodiments, when it is not necessary to use a machine learning positioning method to locate the UE, any one of a plurality of positioning methods is used to locate the UE. For example, a certain traditional positioning method is used to locate the UE. In this way, positioning services with different positioning accuracies can be provided according to positioning requirements, so as to more effectively balance positioning accuracy requirements and positioning efficiency.

[0070] Next, a target machine learning model for positioning the UE is introduced in conjunction with some embodiments of the present disclosure.

[0071] In some embodiments, the target machine learning model is trained by a network data analysis function (NWDAF) network element, and the first information of the LMF network element used to locate the UE matches the second information of the NWDAF network element. The target machine learning model is, for example, trained by a model training logical function (MTLF) of the NWDAF network element.

[0072] Here, the first information of the LMF network element includes at least one of the identification of the LMF network element, the format of the machine learning model supported by the LMF network element, and the operating environment of the machine learning model supported by the LMF network element. The identification of the LMF network element is, for example, the identification number of the manufacturer of the LMF.

[0073] The second information of the NWDAF network element includes at least one of the identifier of the LMF network element supported by the NWDAF network element, the format of the machine learning model trained by the NWDAF network element, and the operating environment of the machine learning model trained by the NWDAF network element. The first information and the second information may be referred to as machine learning model interoperability information, for example.

[0074] As some implementation methods, the first information includes any one, two or all of the identification of the LMF network element, the format of the machine learning model supported by the LMF network element, and the operating environment of the machine learning model supported by the LMF network element.

[0075] As some implementation methods, the second information includes any one, two or all of the identifiers of the LMF network elements supported by the NWDAF network element, the format of the machine learning model trained by the NWDAF network element, and the operating environment of the machine learning model trained by the NWDAF network element.

[0076] It should be understood that the type of content included in the first information is the same as the type of content included in the second information. For example, the type of content included in the first information is the identifier of the LMF network element, and the type of content included in the second information is the identifier of the LMF network element supported by the NWDAF network element; for another example, the type of content included in the first information is the format of the machine learning model supported by the LMF network element, and the type of content included in the second information is the format of the machine learning model trained by the NWDAF network element, and so on.

[0077] When each item of the first information and the second information is the same, the first information and the second information match. For example, when the first information includes two items of content, the second information includes two items of content, and each of the two items of the first information and each of the two items of the second information are the same, the first information and the second information match. For another example, when the first information includes three items of content, the second information includes three items of content, and each of the three items of the first information and each of the three items of the second information are the same, the first information and the second information match.

[0078] In the above embodiment, the LMF network element and the NWDAF network element are matched using the machine learning model interoperability information, which can ensure that the target machine learning model trained by the NWDAF network element can run on the LMF network element.

[0079] Next, how to determine the target machine learning model is introduced in conjunction with some embodiments of the present disclosure.

[0080] In some embodiments, the LMF network element for positioning the UE determines whether the LMF network element has a machine learning model that matches the UE based on the third information of the UE; if the LMF network element does not have a machine learning model that matches the UE, sends a subscription request to the NWDAF network element; and receives a target machine learning model from the NWDAF network element. For example, the LMF network element for positioning the UE can send a subscription request through the Nnwdaf machine learning model provision subscription (Nnwdaf_MLModelProvision_Subscribe) service.

[0081] Here, the third information includes at least one of the area type where the UE is located and the identifier of the network slice accessed by the UE. As some embodiments, the area type includes urban and suburban areas.

[0082] The subscription request sent by the LMF network element for positioning the UE includes the third information, and the target machine learning model is trained by the NWDAF network element based on the third information. The third information can be called machine learning model filtering information, for example.

[0083] As some implementation methods, when the LMF network element used to locate the UE has a machine learning model that matches the UE, the machine learning model that matches the UE is used as the target machine learning model.

[0084] As some implementation methods, the NWDAF network element determines a sample UE based on the third information, and trains a target machine learning model based on the positioning parameters of the sample UE.

[0085] As some implementation methods, when the sample UE used in training the machine learning model has the same third information as the UE to be located, the UE to be located matches the trained machine learning model; as other implementation methods, when the sample UE used in training the machine learning model does not have the same third information as the UE to be located, the UE to be located does not match the trained machine learning model.

[0086] In the above embodiment, the LMF network element determines whether it has a machine learning model that matches the UE based on the third information of the UE to be located, and subscribes to the NWDAF network element for a target machine learning model trained based on the third information if it does not have a matching machine learning model. In this way, the training samples based on which the target machine learning model is trained are closer to the UE to be located, which can improve the accuracy of the target machine learning model in locating the UE.

[0087] Next, how to determine the NWDAF network element for training the target machine learning model is introduced in combination with some embodiments of the present disclosure.

[0088] In some embodiments, the NWDAF network element is discovered by the LMF network element through a network storage function (NRF) network element storing the first information and the second information.

[0089] That is to say, when the LMF network element used to locate the UE does not have a machine learning model that matches the UE, the LMF network element can discover the NWDAF network element that supports training the machine learning model for UE positioning through the NRF network element.

[0090] As some implementation methods, in response to a registration request sent by the LMF network element and carrying the first information, the NRF network element stores the first information. For example, the registration request of the LMF network element may also carry an identifier that supports the machine learning positioning method, and the NRF network element may also store the identifier.

[0091] As some implementations, in response to a registration request sent by the NWDAF network element and carrying the second information, the NRF network element stores the second information. For example, the registration request of the NRF network element may also carry an identifier that supports training a machine learning model for UE positioning, and the NRF network element may also store the identifier.

[0092] That is to say, the NRF network element can perform the above-mentioned machine learning model interoperability information matching process.

[0093] In the above embodiment, the NRF network element that stores relevant information of the LMF network element and the NWDAF network element is used to discover the NWDAF network element that supports the training of the machine learning model for UE positioning. The LMF network element is not required to perform the process of matching the interoperability information of the machine learning model, which can reduce the processing pressure of the LMF network element and help improve the efficiency of positioning the UE.

[0094] Next, how to monitor the positioning performance of a target machine learning model is introduced in conjunction with some embodiments of the present disclosure.

[0095] In some embodiments, the LMF network element obtains the actual location of the UE, determines the evaluation index of the target machine learning model based on the actual location of the UE and the location output by the target machine learning model, and sends the evaluation index of the target machine learning model to the positioning monitoring demander.

[0096] Here, the evaluation index of the target machine learning model includes at least one of the positioning accuracy and the average positioning error, and the positioning monitoring demander includes at least one of the NWDAF network element that provides the target machine learning model and the initiator of the positioning request. The initiator of the positioning request is, for example, the UE itself or a service provider that needs the UE location to provide a service.

[0097] For example, the evaluation index includes positioning accuracy or average positioning error; for another example, the evaluation index includes positioning accuracy and average positioning error.

[0098] As some implementations, the LMF network element may receive the actual location of the UE provided by the user of the UE or the initiator of the positioning request. For example, the actual location of the UE may be reported when the user of the UE or the initiator of the positioning request finds that the location output by the target machine learning model is inaccurate.

[0099] It should be understood that the LMF network element can receive multiple positioning requests for multiple UEs and use the target machine learning model to locate the multiple UEs. Accordingly, the LMF network element can receive the actual locations of the multiple UEs provided by the users of the multiple UEs or the initiators of the multiple positioning requests.

[0100] As some implementation methods, the LMF network element may determine the evaluation index of the target machine learning model in response to a subscription request for monitoring the positioning performance of the target machine learning model issued by the positioning monitoring demander, and send the evaluation index to the positioning monitoring demander. The positioning monitoring demander may, for example, subscribe to the evaluation index of the target machine learning model through the Nnwdaf Analytics Subscription (Nnwdaf_AnalyticsSubscription_Subscribe) service.

[0101] For example, when a UE user or the initiator of a positioning request finds that the location output by the target machine learning model is inaccurate, in order to determine whether to still use the target machine learning model to locate the UE, the UE user or the initiator of the positioning request can send a subscription request to the LMF network element to monitor the positioning performance of the target machine learning model.

[0102] In the above embodiment, the LMF network element can monitor the positioning performance of the target machine learning model and send evaluation indicators related to the positioning performance to the positioning monitoring demander, which helps the positioning monitoring demander to judge and analyze the positioning performance of the target machine learning model.

[0103] In some embodiments, the subscription request for positioning performance carries an identifier of the target machine learning model, a predetermined time period for monitoring the positioning performance of the target machine learning model, and a reporting object for the evaluation indicators of the target machine learning model.

[0104] As some embodiments, the predetermined time period may be a period of time in the past, i.e., the subscription request for positioning performance indicates monitoring the positioning performance of the target machine learning model within a period of time in the past; as other embodiments, the predetermined time period may be a period of time in the future, i.e., the subscription request for positioning performance indicates monitoring the positioning performance of the target machine learning model within a period of time in the future.

[0105] As some implementation methods, the reporting object of the evaluation indicators of the target machine learning model can be the positioning monitoring demand party.

[0106] In the above embodiment, the information carried in the subscription request for positioning performance can accurately monitor the positioning performance of the target machine learning model that the positioning monitoring demander wants to monitor, which helps the positioning monitoring demander to accurately judge and analyze the positioning performance of the target machine learning model.

[0107] In some embodiments, when the evaluation index of the target machine learning model does not meet the first preset requirement, the LMF network element stops using the target machine learning model to locate the UE. As some embodiments, when the evaluation index of the target machine learning model meets the first preset requirement, the LMF network element continues to use the target machine learning model to locate the user equipment.

[0108] As some implementation methods, the first preset requirement may be a preset positioning accuracy or a preset average positioning error. It should be understood that the first preset requirement may be set according to actual conditions and is not limited here.

[0109] As some implementation methods, when the LMF network element stops using the target machine learning model to locate the UE, the LMF network element may use other positioning methods (for example, one of the multiple positioning methods mentioned above) to locate the UE.

[0110] In the above embodiment, when the evaluation index of the target machine learning model is too low, the LMF network element itself stops using the target machine learning model to locate the UE, which can reduce the output of inaccurate positioning results and prevent the inaccurate positioning results from being further used.

[0111] The present disclosure also proposes a positioning method performed by an NWDAF network element. Next, the positioning method is introduced in conjunction with other embodiments of the present disclosure.

[0112] Figure 2 FIG. 1 is a flow chart of a positioning method according to some other embodiments of the present disclosure. Here, the positioning method is executed by a NWDAF network element.

[0113] In step 202, sample locations of sample UEs are obtained. In some implementation modes, the sample locations may be obtained through a DCCF network element.

[0114] In step 204, the positioning parameters of the sample UE are obtained. Here, the positioning parameters include parameters required for positioning the sample UEs at the sample positions respectively using multiple positioning methods.

[0115] As some implementation methods, the positioning parameters may be acquired through a data collection and coordination (DCCF) network element. For example, the positioning parameters include any two, any three, any four, or all of UL-TDoA, DL-TDoA, UL-AoA, DL-AoD, and Multi-RTT.

[0116] In step 206, the positioning parameters are used as input and the sample positions are used as output to train the machine learning model to obtain a target machine learning model.

[0117] In step 208, the target machine learning model is sent to the LMF network element so that the LMF network element uses the target machine learning model to locate the UE that needs to be located.

[0118] As some implementations, the NWDAF network element may send the target machine learning model to the LMF network element via the Nnwdaf machine learning model provision notification (Nnwdaf_MLModelProvision_Notify) service.

[0119] In the above embodiment, by taking the positioning parameters of the sample user equipment as input and the sample positions of the sample user equipment as output, the machine learning model is trained to obtain the target machine learning model, and the target machine learning model is sent to the LMF network element. This allows the LMF network element to comprehensively consider multiple traditional positioning methods when positioning the UE, thereby improving the accuracy of UE positioning.

[0120] In some embodiments, Figure 1 The corresponding embodiment is similar, the first information of the LMF network element matches the second information of the NWDAF network element. The first information and the second information can refer to the above introduction, and will not be repeated here.

[0121] In this way, by matching LMF network elements and NWDAF network elements using machine learning model interoperability information, it can be ensured that the target machine learning model trained by the NWDAF network element can run on the LMF network element.

[0122] In some embodiments, the NWDAF network element receives a subscription request sent by the LMF network element, and determines a sample UE based on the third information.

[0123] Here, the subscription request includes the third information of the UE that needs to be positioned, and the third information includes the type of area where the UE is located and at least one of the identifiers of the network slice accessed by the UE. The third information of the sample UE is the same as the third information of the UE that needs to be positioned.

[0124] In the above embodiment, the NWDAF network element receives the subscription request sent by the LMF network element, and determines the sample UE according to the third information in the subscription request. In this way, the training sample based on which the target machine learning model is trained is closer to the UE to be located, which can improve the accuracy of the target machine learning model in locating the UE.

[0125] In some embodiments, the NWDAF network element receives an evaluation index of a target machine learning model from the LMF network element; if the evaluation index does not meet the second preset requirement, the target machine learning model is retrained, and the retrained target machine learning model is sent to the LMF network element. Here, the evaluation index includes at least one of positioning accuracy and average positioning error.

[0126] As some implementation methods, the NWDAF network element may send a subscription request to the LMF network element to monitor the positioning performance of the target machine learning model to receive evaluation indicators of the target machine learning model from the LMF network element.

[0127] As some implementations, when the evaluation index meets the second preset requirement, the target machine learning model is not retrained, that is, no other operations are performed. The second preset requirement is, for example, the same as the first preset requirement.

[0128] In the above embodiment, the NWDAF network element can receive evaluation indicators related to positioning performance from the LMF network element, and when the evaluation indicators do not meet the second preset requirements, retrain the target machine learning model, and send the retrained target machine learning model to the LMF network element, so that the LMF network element uses the retrained target machine learning model to locate the UE, thereby improving the accuracy of positioning the UE.

[0129] In some embodiments, the LMF network element also sends the retrained target machine learning model to another LMF network element. For example, when multiple LMF network elements use the same target machine learning model for positioning, and the evaluation index of the target machine learning model does not meet the second preset requirement, in addition to sending the retrained machine learning model to the LMF network element that sent the evaluation index of the target machine learning model, the NWDAF network element can also send the retrained target machine learning model to other LMF network elements in the multiple LMF network elements.

[0130] In the above embodiment, when multiple LMF network elements use the same target machine learning model for positioning and the output of the target machine learning model is inaccurate, each LMF network element using the target machine learning model can receive and use the retrained target machine learning model to locate the UE that needs to be located, thereby improving the accuracy of positioning the UE.

[0131] Next, combine Figure 3 This section describes the overall process of UE positioning.

[0132] Figure 3 It is a flowchart of a positioning method according to some embodiments of the present disclosure.

[0133] like Figure 3 As shown, in step S1a, the service provider (AF network element or LCS client) sends a request for positioning the UE to the GMLC, and the GMLC then sends a request for positioning to the AMF network element through the Namf_Location_ProvidePositioningInfo service. For example, the AF network element or the LCS client can send a request for positioning the UE to the GMLC via the Network Exposure Function (NEF) network element.

[0134] In step S1b, the UE sends a request to locate the UE to the AMF network element (e.g., via a UL NAS TRANSPORT message).

[0135] In step S2, the AMF network element selects the LMF network element used to locate the UE.

[0136] In step S3, the AMF network element sends a positioning request for the UE to the selected LMF network element (for example, through Nlmf_Location_DetermineLocation_request).

[0137] In step S4, after receiving the positioning request, the LMF network element determines whether it is necessary to use machine learning positioning to locate the UE.

[0138] In the case where the UE needs to be positioned using a machine learning positioning method, it is checked whether a machine learning model matching the UE is stored. If there is no matching machine learning model, an NWDAF network element (i.e., an NWDAF network element with MTLF) that supports training a machine learning model for UE positioning is discovered through an NRF network element (not shown).

[0139] The LMF network element sends a subscription request to the discovered NWDAF network element through the Nnwdaf_MLModelProvision_Subscribe service to subscribe to the target machine learning model. Here, the subscription request includes machine learning model filtering information and the machine learning model interoperability information of the LMF network element.

[0140] When there is no need to use machine learning positioning methods to locate the UE, the traditional positioning methods are used to locate the UE.

[0141] In step S5, after receiving the subscription request sent by the LMF network element, the MLTF in the NWDAF network element collects training data from the radio access network (RAN), sample UEs and LMF network elements and trains the machine learning model to obtain a target machine learning model. Here, the training data includes positioning parameters of the sample UE and sample locations of the sample UE.

[0142] In step S6, after the MLTF in the NWDAF network element completes the machine learning model training, the target machine learning model is sent to the LMF network element through the Nnwdaf_MLModelProvision_Notify service.

[0143] In step S7, after receiving the target machine learning model, the LMF network element inputs the data used for positioning into the target machine learning model to output the positioning result of the UE. Here, the data used for positioning includes the positioning parameters of the UE.

[0144] In step S8, the LMF network element returns the UE positioning result to the AMF network element through Nlmf_Location_DetermineLocationresponse.

[0145] In step S9a, the AMF network element returns the positioning result of the UE to the UE. Here, step S9a corresponds to step S1b.

[0146] In step S9b, the AMF network element returns the positioning result of the UE to the AF network element or the LCS client, for example, the positioning result of the UE is returned to the AF network element or the LCS client via the GMLC and the NEF network element in sequence. Here, step S9b corresponds to step S1a.

[0147] In step S10, the positioning monitoring demander sends a subscription request for monitoring the positioning performance of the target machine learning model to the LMF network element through the Nnwdaf_AnalyticsSubscription_Subscribe service to subscribe to the evaluation indicators of the target machine learning model.

[0148] In steps S11 to S12, when the LMF network element receives a subscription request for monitoring the positioning performance of the target machine learning model, the LMF network element verifies the identification of the target machine learning model, the scheduled time period for monitoring the positioning performance of the target machine learning model, and the reporting object of the evaluation indicators of the target machine learning model, so as to accurately monitor the positioning performance of the target machine learning model that the positioning monitoring demander wants to monitor.

[0149] As some implementation methods, after verification, the LMF network element will collect data for generating evaluation indicators.

[0150] In step S13, the LMF network element returns the evaluation indicators of the target machine learning model to the positioning monitoring demander through Nnwdaf_AnalyticsSubscription_Subscribe. The NWDAF network element can determine whether to retrain the target machine learning model based on the evaluation indicators. The initiator of the positioning request (UE, AF network element or LCS client) can decide whether to continue using the target machine learning model for positioning.

[0151] In step S14, when the evaluation index of the target machine learning model does not meet the first preset requirement, the LMF network element can stop using the target machine learning model to locate the UE.

[0152] Each embodiment in this specification is described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the embodiments can be referred to each other. For the device embodiment, since it basically corresponds to the method embodiment, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiment.

[0153] In some embodiments, the LMF network element includes a module for executing the method performed by the LMF network element in any of the above embodiments.

[0154] Figure 4 It is a schematic diagram of the structure of an LMF network element according to some embodiments of the present disclosure.

[0155] like Figure 4 As shown, the LMF network element includes a receiving module 401, a judging module 402 and a positioning module 403.

[0156] The receiving module 401 is configured to receive a positioning request for a UE. Here, the positioning request carries a positioning requirement.

[0157] The determination module 402 is configured to determine whether it is necessary to locate the UE using a machine learning positioning method according to positioning requirements.

[0158] The positioning module 403 is configured to input the positioning parameters into the target machine learning model to locate the UE when the machine learning positioning method is needed to locate the UE. Here, the positioning parameters include the parameters required for each of the multiple positioning methods to locate the UE.

[0159] In some embodiments, the LMF network element may further include other modules to execute the positioning method performed by the LMF network element in any of the above embodiments.

[0160] In some embodiments, the NWDAF network element includes a module for executing the method performed by the NWDAF network element in any one of the above embodiments.

[0161] Figure 5 It is a schematic diagram of the structure of a NWDAF network element according to some embodiments of the present disclosure.

[0162] like Figure 5 As shown, the NWDAF network element includes a first acquisition module 501, a second acquisition module 502, a training module 503 and a sending module 504.

[0163] The first acquisition module 501 is configured to acquire a sample position of a sample UE.

[0164] The second acquisition module 502 is configured to acquire positioning parameters of the sample UEs. Here, the positioning parameters include parameters required for positioning the sample UEs at the sample locations using multiple positioning methods.

[0165] The training module 503 is configured to take the positioning parameters as input and the sample positions as output, and train the machine learning model to obtain a target machine learning model.

[0166] The sending module 504 is configured to send the target machine learning model to the LMF network element so that the LMF network element uses the target machine learning model to locate the UE that needs to be located.

[0167] In some embodiments, the NWDAF network element may further include other modules to execute the positioning method performed by the NWDAF network element in any of the above embodiments.

[0168] Figure 6 1 is a schematic diagram of the structure of an electronic device according to some embodiments of the present disclosure. In some embodiments, the electronic device is a LMF network element or a NWDAF network element.

[0169] like Figure 6 As shown, the electronic device 600 includes a memory 601 and a processor 602 coupled to the memory 601 .

[0170] In some embodiments, when the electronic device 600 is a LMF network element, the processor 602 is configured to execute the method performed by the LMF network element in any of the aforementioned embodiments based on instructions stored in the memory 601.

[0171] In some embodiments, when the electronic device 600 is a NWDAF network element, the processor 602 is configured to execute the method performed by the NWDAF network element in any of the aforementioned embodiments based on instructions stored in the memory 601 .

[0172] The memory 601 may include, for example, a system memory, a fixed non-volatile storage medium, etc. The system memory may store, for example, an operating system, an application program, a boot loader, and other programs.

[0173] In some embodiments, the electronic device 600 may further include an input / output interface 603, a network interface 604, a storage interface 605, etc. The input / output interface 603, the network interface 604, the storage interface 605, and the memory 601 and the processor 602 may be connected, for example, via a bus 606. The input / output interface 603 provides a connection interface for input / output devices such as a display, a mouse, a keyboard, and a touch screen. The network interface 604 provides a connection interface for various networked devices. The storage interface 605 provides a connection interface for external storage devices such as an SD card and a USB flash drive.

[0174] The embodiments of the present disclosure also provide a positioning system, including the LMF network element of any one of the above embodiments, and the NWDAF network element of any one of the above embodiments.

[0175] In some embodiments, the positioning system further includes an NRF network element. The NRF network element is configured to store the first information of the LMF network element and the second information of the NWDAF network element, and discover the LMF network element used to locate the UE based on the first information and the second information. That is, the NRF network element can perform the above-mentioned machine learning model interoperability information matching process based on the first information and the second information.

[0176] In some embodiments, the positioning system further comprises an AMF network element. The AMF network element is configured to select the LMF network element according to the current load of the LMF network element and / or the distance between the AMF network element and the LMF network element.

[0177] For example, the LMF network element used to locate the UE is the LMF network element that is closest to the AMF network element; for another example, the LMF network element used to locate the UE is the LMF network element that currently has the lowest processing load; for another example, the LMF network element used to locate the UE is the LMF network element that is closest to the AMF network element and currently has the smallest processing load.

[0178] In some embodiments, a first part of LMF network elements among multiple LMF network elements do not support positioning using a machine learning positioning method, a second part of LMF network elements among the multiple LMF network elements have an identification that supports a machine learning positioning method (i.e., this part of LMF network elements supports positioning using a machine learning positioning method), and the AMF network element selects an LMF network element for locating the UE from the second part of LMF network elements among the multiple LMF network elements.

[0179] In some embodiments, in response to a positioning request sent by the UE or a positioning request sent by the GMLC, the AMF network element sends a positioning request for the UE to the LMF network element. Here, the GMLC may send a positioning request to the AMF network element in response to a request sent by the service provider to locate the UE.

[0180] In some embodiments, the AMF network element may receive the positioning result of the UE sent by the LMF network element, and return the positioning result to the UE or the service provider.

[0181] The embodiment of the present disclosure further provides a computer-readable storage medium, including computer program instructions, which implement the steps of the method of any one of the above embodiments when executed by a processor.

[0182] The embodiments of the present disclosure further provide a computer program product, including a computer program, which implements the steps of the method of any one of the above embodiments when executed by a processor.

[0183] So far, various embodiments of the present disclosure have been described in detail. In order to avoid obscuring the concept of the present disclosure, some details known in the art are not described. Based on the above description, those skilled in the art can fully understand how to implement the technical solution disclosed here.

[0184] Those skilled in the art will appreciate that the embodiments of the present disclosure may be provided as methods, systems, or computer program products. Therefore, the present disclosure may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Furthermore, the present disclosure may take the form of a computer program product implemented on one or more computer-usable non-transient storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0185] The present disclosure is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present disclosure. It should be understood that the functions specified in one or more processes in the flowchart and / or one or more blocks in the block diagram can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the functions specified in the flowchart. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0186] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture including an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0187] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0188] Although some specific embodiments of the present disclosure have been described in detail by way of examples, it should be understood by those skilled in the art that the above examples are for illustration only and are not intended to limit the scope of the present disclosure. It should be understood by those skilled in the art that the above embodiments may be modified or some technical features may be replaced by equivalents without departing from the scope and spirit of the present disclosure. The scope of the present disclosure is defined by the appended claims.

Claims

1. A positioning method, performed by a location management function LMF network element, comprising: receiving a positioning request for a user equipment, wherein the positioning request carries a positioning requirement; According to the positioning requirement, determining whether it is necessary to use a machine learning positioning method to locate the user equipment; as well as When it is necessary to locate the user device using a machine learning positioning method, the positioning parameters are input into a target machine learning model to locate the user device. The positioning parameters include the parameters required for each of the multiple positioning methods to locate the user device.

2. The method according to claim 1, wherein: The target machine learning model is trained by a network data analysis function NWDAF network element, and the first information of the LMF network element matches the second information of the NWDAF network element, wherein: The first information includes at least one of an identifier of the LMF network element, a format of a machine learning model supported by the LMF network element, and an operating environment of the machine learning model supported by the LMF network element, The second information includes at least one of an identifier of the LMF network element supported by the NWDAF network element, a format of the machine learning model trained by the NWDAF network element, and an operating environment of the machine learning model trained by the NWDAF network element.

3. The method according to claim 1, further comprising: When there is no need to use a machine learning positioning method to locate the user equipment, any one of the multiple positioning methods is used to locate the user equipment.

4. The method according to any one of claims 1 to 3, wherein: The positioning requirement includes the positioning accuracy required for the user equipment; The determining, according to the positioning requirement, whether it is necessary to use a machine learning positioning method to position the user equipment includes: When the positioning accuracy is greater than a preset accuracy, determining that it is necessary to use a machine learning positioning method to position the user equipment; as well as When the positioning accuracy is less than or equal to the preset accuracy, it is determined that there is no need to use the machine learning positioning method to locate the user equipment.

5. The method according to any one of claims 1 to 3, further comprising: Determining whether the LMF network element has a machine learning model matching the user equipment according to third information of the user equipment, the third information including at least one of an area type in which the user equipment is located and an identifier of a network slice accessed by the user equipment; In a case where the LMF network element does not have a machine learning model matching the user equipment, sending a subscription request to a network data analysis function NWDAF network element, wherein the subscription request includes the third information; as well as Receive the target machine learning model from the NWDAF network element, wherein the target machine learning model is trained by the NWDAF network element based on the third information.

6. The method according to any one of claims 1 to 3, further comprising: Obtaining the actual location of the user equipment; Determining an evaluation index of the target machine learning model according to the actual position of the user equipment and the position output by the target machine learning model, wherein the evaluation index includes at least one of a positioning accuracy rate and an average positioning error; and The evaluation index is sent to a positioning monitoring demander, wherein the positioning monitoring demander includes at least one of a network data analysis function NWDAF network element that provides the target machine learning model and an initiator of the positioning request.

7. The method according to claim 6, further comprising: When the evaluation index does not meet the first preset requirement, stop using the target machine learning model to locate the user equipment.

8. A positioning method, performed by a network data analysis function NWDAF network element, comprising: Obtaining a sample location of a sample user device; Acquire positioning parameters of the sample user equipment, where the positioning parameters include parameters required for positioning the sample user equipment at the sample location using multiple positioning methods respectively; Taking the positioning parameter as input and the sample position as output, training the machine learning model to obtain a target machine learning model; as well as The target machine learning model is sent to the location management function LMF network element so that the LMF network element can use the target machine learning model to locate the user equipment that needs to be located.

9. The method according to claim 8, wherein: The first information of the LMF network element matches the second information of the NWDAF network element, wherein: The first information includes at least one of an identifier of the LMF network element, a format of a machine learning model supported by the LMF network element, and an operating environment of the machine learning model supported by the LMF network element, The second information includes at least one of an identifier of the LMF network element supported by the NWDAF network element, a format of the machine learning model trained by the NWDAF network element, and an operating environment of the machine learning model trained by the NWDAF network element.

10. The method according to claim 8, further comprising: receiving a subscription request sent by the LMF network element, where the subscription request includes third information of the user equipment, where the third information includes at least one of an area type in which the user equipment is located and an identifier of a network slice accessed by the user equipment; as well as The sample user equipment is determined according to the third information, wherein the third information of the sample user equipment is the same as the third information of the user equipment.

11. The method according to claim 8, further comprising: Receiving an evaluation index of the target machine learning model from the LMF network element, the evaluation index comprising at least one of a positioning accuracy rate and a mean positioning error; as well as When the evaluation index does not meet the second preset requirement, the target machine learning model is retrained, and the retrained target machine learning model is sent to the LMF network element.

12. The method according to claim 1 or 8, wherein: The positioning parameters include multiple parameters of uplink arrival time difference, downlink arrival time difference, uplink arrival angle, downlink departure angle, and multipath travel time.

13. The method according to claim 2 or 8, wherein: The NWDAF network element is discovered by the LMF network element through a network storage function NRF network element storing the first information and the second information.

14. A location management function network element, comprising: A module configured to perform the method of any one of claims 1-7, 12 and 13.

15. A location management function network element, comprising: Memory; as well as A processor coupled to the memory is configured to execute the method of any one of claims 1-7, 12 and 13 based on instructions stored in the memory.

16. A network data analysis function network element, comprising: A module configured to execute the method according to any one of claims 8 to 13.

17. A network data analysis function network element, comprising: Memory; as well as A processor coupled to the memory is configured to execute the method of any one of claims 8 to 13 based on instructions stored in the memory.

18. A positioning system comprising: The location management function network element as described in claim 14 or 15; as well as The network data analysis functional network element as described in claim 16 or 17.

19. A computer-readable storage medium comprising a computer program, wherein: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 13 are implemented.

20. A computer program product comprising a computer program, wherein: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 13 are implemented.

Images