Position determination

Abstract

An apparatus, method, and computer program are disclosed. The apparatus can include means for receiving, at a first time instance, from a target device, a first set of measurement data associated with each of a plurality of base stations, and determining a first position of the target device based on the received first set of measurement data. The means can also receive, at each of one or more subsequent time instances, from the target device, a second set of measurement data associated with each of one base station or a smaller number of base stations, and determine a respective position of the target device at each of the one or more subsequent time instances based on the position determined at a previous time instance and the second set of measurement data.

Description

Technical Field

[0001] The example embodiments relate to location determination, such as apparatus and methods for determining the geographical location of a user device. Background Technology

[0002] Determining the location of a target device can be useful in a variety of applications. For example, the target device could be a mobile phone, and location determination could support locating the origin of an emergency phone call and / or performing optimizations at the telephone network, such as for Radio Resource Management (RRM).

[0003] In the context of mobile phone networks, Global Navigation Satellite Systems (GNSS) can be used to provide positioning in outdoor environments, assuming the target device has a GNSS receiver and / or sufficient satellite signals to determine location. This may not be the case if the target device is indoors or in a specially constructed environment. Using radio signals between the target device and one or more base stations can also be used as an alternative method for determining location, including in indoor environments. Summary of the Invention

[0004] The scope of protection sought by the various embodiments of the present invention is set forth in the independent claims. Embodiments and features described in the specification that do not fall within the scope of the independent claims (if any) are to be interpreted as examples useful for understanding the various embodiments of the invention.

[0005] According to one aspect, an apparatus is provided, comprising components for: receiving, at a first time, a first set of measurement data associated with each of a plurality of base stations from a target device; determining a first location of the target device based on the received first set of measurement data; receiving, at each of one or more subsequent time instances, a second set of measurement data associated with one of the base stations or each of a smaller number of the plurality of base stations; and determining a corresponding location of the target device at each of one or more subsequent time instances based on the location determined at a previous time instance and the second set of measurement data.

[0006] The apparatus may further include components for enabling the target device to transmit a first set of measurement data at a first time instance and a second set of measurement data at subsequent time instances. The apparatus may also include components for enabling the target device to transmit the first set of measurement data at repeating time periods and to transmit (or more) second set of measurement data at one or more sub-intervals within each time period. Only one second set of measurement data associated with a serving base station for the target device may be received. The first set of measurement data may be associated with the serving base station for the target device and one or more neighboring base stations. The first and second sets of measurement data may represent a reference signal received power (RSRP) measurement determined by the target device against a reference signal from a base station. The location determined at a previous time instance may be stored in the apparatus's storage device. Determining the corresponding location of the target device at each of one or more subsequent time instances may be based on a machine learning model that determines a functional relationship between the corresponding location and the location determined at the previous time instance and the second set of measurement data. The machine learning model may include convolutional neural networks and recurrent neural networks. The apparatus may be a positioning node in a radio communication network.

[0007] According to another aspect, a method may be provided, comprising: receiving, at a first time instance, a first set of measurement data associated with each of a plurality of base stations from a target device; determining a first location of the target device based on the received first set of measurement data; receiving, at each of one or more subsequent time instances, a second set of measurement data associated with one of the plurality of base stations or each of a smaller number of the plurality of base stations from the target device; and determining a corresponding location of the target device at each of one or more subsequent time instances based on the location determined at the previous time instance and the second set of measurement data.

[0008] The method may further include causing the target device to transmit a first set of measurement data at a first time instance and a second set of measurement data at subsequent time instances. The method may also include causing the target device to transmit the first set of measurement data at repeated time periods and to transmit (or more) second sets of measurement data at one or more sub-intervals within each time period. Only one second set of measurement data associated with a serving base station for the target device may be received. The first set of measurement data may be associated with the serving base station for the target device and one or more neighboring base stations. The first and second sets of measurement data may represent a reference signal received power (RSRP) measurement determined by the target device for a reference signal from a base station. The method may further include storing the location determined at the previous time instance in a memory device. Determining the corresponding location of the target device at each of the one or more subsequent time instances may be based on a machine learning model that determines a functional relationship between the corresponding location and the location determined at the previous time instance and the second set of measurement data. The machine learning model may include convolutional neural networks and recurrent neural networks. The method may be performed at a location node in a radio communication network.

[0009] According to another aspect, an apparatus may be provided, comprising at least one processor, at least one memory directly connected to the at least one processor, the at least one memory including computer program code, and the at least one processor, together with the at least one memory and the computer program code, being arranged to perform the following method: receiving, at a first time instance, a first set of measurement data associated with each of a plurality of base stations from a target device; determining a first location of the target device based on the received first set of measurement data; receiving, at each of one or more subsequent time instances, a second set of measurement data associated with one of the base stations or each of a smaller number of the plurality of base stations from the target device; and determining a corresponding location of the target device at each of one or more subsequent time instances based on the location determined at the previous time instance and the second set of measurement data.

[0010] According to another aspect, a computer program product including an instruction set, which, when executed on a device, is configured to cause the device to perform the following methods: receiving, at a first time instance, a first set of measurement data associated with each of a plurality of base stations from a target device; determining a first location of the target device based on the received first set of measurement data; receiving, at each of one or more subsequent time instances, a second set of measurement data associated with one of the base stations or each of a smaller number of the base stations from the plurality of base stations; and determining a corresponding location of the target device at each of one or more subsequent time instances based on the location determined at the previous time instance and the second set of measurement data.

[0011] According to another aspect, a non-transient computer-readable medium may be provided, including program instructions stored thereon for performing a method comprising: receiving, at a first time instance, a first set of measurement data associated with each of a plurality of base stations from a target device; determining a first location of the target device based on the received first set of measurement data; receiving, at each of one or more subsequent time instances, a second set of measurement data associated with one of the base stations or each of a smaller number of the base stations; and determining a corresponding location of the target device at each of one or more subsequent time instances based on the location determined at the previous time instance and the second set of measurement data. Attached Figure Description

[0012] Exemplary embodiments will be described in a non-limiting manner with reference to the accompanying drawings, wherein:

[0013] Figure 1 This is a partial schematic diagram of a radio network, including three base stations, one user equipment unit, and a portion of the core network;

[0014] Figure 2 This is a schematic diagram illustrating how the location of a user equipment is determined as a function of time via a received signal power measurement report.

[0015] Figure 3 This is a timing diagram illustrating how the location of a user equipment is triggered to send a measurement report according to an example embodiment;

[0016] Figure 4 This is a flowchart illustrating operations that can be performed at a portion of the core network to determine the location of a user equipment, according to an example embodiment;

[0017] Figure 5This is a schematic diagram illustrating how the location of a user equipment is determined via a received signal power measurement report as a function of time, according to an example embodiment.

[0018] Figure 6 This is a schematic diagram of a neural network that can be trained and used for inference according to an example embodiment;

[0019] Figure 7 This is a bar graph showing the average position error for different time instances of three positioning method types, including the method according to the example embodiment;

[0020] Figure 8A This is a bar chart showing the percentage increase in overhead for the positioning method and the reference positioning method at different time instances according to the example embodiment;

[0021] Figure 8B This is a bar chart showing the percentage accuracy gain of the positioning method and the reference positioning method at different time instances according to the example embodiment;

[0022] Figure 9 This is a flowchart illustrating the processing operations according to an example embodiment;

[0023] Figure 10 It is a device that can be configured or used to perform the example embodiments; and

[0024] Figure 11 It is a non-transitory medium for storing computer-readable code that can execute example embodiments. Detailed Implementation

[0025] The example embodiments relate to positioning systems and methods.

[0026] The following abbreviations may be used in this article:

[0027] 3G / 4G / 5G refers to third-generation, fourth-generation, and fifth-generation cellular network technologies.

[0028] AMF access and mobility features;

[0029] BS base station;

[0030] BSC (Base Station Controller)

[0031] eNodeB is an enhanced NodeB;

[0032] ESM Enhanced Measurement Set;

[0033] FCN Fully Connected Network;

[0034] gNB next-generation base station;

[0035] LCS location services;

[0036] LMF location management function;

[0037] LUT (Lookup Table);

[0038] RAN (Radio Access Network);

[0039] RNC (Radio Network Controller);

[0040] RNN (Recurrent Neural Network)

[0041] RSRP reference signal received power; and

[0042] UE (User Equipment)

[0043] The example embodiments involve determining the location of a target device (e.g., a UE) based on signals associated with a base station. A base station can include any node of a RAN that has a known location and provides radio transmission (and typically reception) functionality to an associated coverage area via one or more antennas. A base station may be referred to herein as a NodeB, but the term is interchangeable with any such node described above. In the context of 4G, a base station is sometimes referred to as an eNodeB, while in the context of 5G, a base station is sometimes referred to as a gNB. The example embodiments are applicable to any next-generation cellular network technology. The example embodiments may involve the use of multiple spatially separated base stations, each having a coverage area, typically referred to as a cell, within which it can reliably transmit and receive signals carrying data. Some coverage overlap may exist between adjacent cells. Each base station may transmit a reference signal, which may be used at least partially to estimate the location of the UE. It is common for a UE to receive signals from two or more base stations. Typically, based on some measurement parameters such as signal strength and / or quality of service, one of the base stations will be designated as the serving base station, and its corresponding cell will be designated as the serving cell. The serving base station / cell may sometimes be referred to as the primary base station / cell and may change over time, such as during handover.

[0044] Any form of UE capable of receiving data from spatially separated base stations via a network can be used, and the UE may include smartphones, tablets, laptops, personal computers, etc.

[0045] Each cell can be identified by an identifier within its local radio area, which is broadcast throughout the cell. Base stations communicate with UEs within their range via an air interface operating on radio frequencies. In some versions of the RAN, several base stations are typically connected (e.g., via terrestrial or microwave) to a controller node (such as an RNC or BSC), which oversees and coordinates the various activities of the multiple base stations connected to it. The RNC is typically connected to one or more core networks.

[0046] Many methods exist for determining UE location, some of which are based on Observed Time Difference of Arrival (OTDOA) and Location Enhancement Cell ID (E-CID). The example embodiments described herein involve the use of Reference Signal Received Power (RSRP) measurements. The concept and measurement of RSRP will be known. Embodiments may be applicable to other location techniques. In summary, a database of expected RSRPs for all relevant base stations within a coverage area can be created. Message Transmission Reports (MRs) from the UE (where MRs may include RSRP measurements for at least the serving cell and possibly neighboring cells) can be correlated with the database to identify the best match point as the UE's current location. This location method can be referred to as fingerprinting, thus each RSRP measurement stored in the database is a form of fingerprint regarding location. Other measurement types may be used.

[0047] In the context of 5G, base stations (gNBs) will utilize antenna arrays to transmit through one or more corresponding beams. (Reference) Figure 1 The diagram illustrates a scenario involving three gNBs 10, 11, and 12. The first gNB is the serving base station for the serving cell of the target UE 13. The other two, 11 and 12, are neighboring gNBs for adjacent cells. In practice, UE 13 can measure the quality of the transmit beam 14 from both the serving cell and neighboring cells. However, generally, only a subset of these measurements can be reported to the serving cell. This subset may correspond only to the optimal measurement beam for the serving cell. Under predefined triggering conditions, UE 13 can also report measurements from neighboring cells to its serving cell (e.g., in the case of a handover-triggered event). This limitation in measurement reporting is for signaling overhead optimization.

[0048] In the example embodiment, beam RSRP measurements available at service gNB10 are used as input for positioning.

[0049] The following portions of this disclosure will only refer to gNB 10, 11, 12 as examples of base stations. However, other embodiments may use other forms of base stations.

[0050] Each of gNBs 10, 11, and 12 can connect to the management portion of the network. This can be referred to as the core network. The core network may include (in a 5G environment) AMF 15 and LMF 6. AMF 36 is part of the 3GPP 5G architecture and has the primary tasks of registration management, connection management, reachability management, mobility management, and various functions related to security, access management, and authorization. Other technologies may have equivalent functionality, and the specifications of AMF 15 are not required for the example implementation.

[0051] LMF 38 is part of the network that determines the location of UE 13 and can actually trigger location measurements.

[0052] Figure 2 A schematic example of UE 13 transmitting RSRP measurement reports 20 as a function of time is depicted, which also corresponds to the motion trajectory. At each instant t... i UE position (X) t It can be estimated at AMF 15 as a function of the reported beam measurement (BR), using {BR}. t The achieved positioning accuracy is highly dependent on the beam RSRP measurements considered, typically depending on the number of reported beams and the number of cells (serving cell plus best neighboring cells). For example, it can be observed that using the best four beams, a positioning accuracy of approximately 2.6m can be achieved through beam measurements from the serving cell. It can also be observed that using beam measurements from the serving cell and two best neighboring cells, where each cell reports four beams, a positioning accuracy of approximately 1.5m can be achieved.

[0053] These results can be achieved using neural network-based methods (e.g., two hidden layers, each with 500 nodes, and an activation function of tanh).

[0054] It can then be concluded that higher positioning accuracy requires a larger set of beam measurements from the serving cell and neighboring cells. For example, due to neighboring cell measurement reports, exemplary embodiments provide methods and apparatus for achieving this higher positioning accuracy while reducing or minimizing associated signaling overhead.

[0055] Still referencing Figure 2 Each measurement report {BR} t This can refer to UE 13 at the corresponding time instance t. i RSRP reported to its current service gNB 10. Can be reported every δ t Reports are generated in seconds. In practice, reports can be generated for each measurement. t Only a subset of beams serving gNB 10 is reported, for example, four. Under some predefined conditions, more beams, such as twenty, can be reported from neighboring cells to provide higher accuracy.

[0056] To reduce signaling overhead, the example implementation uses a predefined duration T or time period, such that every T seconds a measurement report including an Enhanced Measurement Set (ESM) is used to estimate the location of UE 13 with higher accuracy. For example, the measurement report includes RSRP measurements from the serving cell and one or more neighboring cells. For instance, RSRP measurements from two neighboring cells could be used.

[0057] Between consecutive ESM reports, the reporting of measurements is limited to a small number of measurements, such as a minimum requirement. For example, if an ESM report is made at an instant t0, then the sub-intervals within the time period T (e.g., t0+δ) are also considered. t , t0+2δ t ...t0+T-δ t At t+δ, only the minimum required measurement report can be executed. UE 13 at t+δ t The location at a point can be estimated as its estimated location at a previous time instance t, using the minimum measurement report {BR}. t+δt The function f, i.e.

[0058] X t+δt =f(X) t ,BR t+δt )

[0059] In some example implementations, machine learning models can be used to approximate the function f.

[0060] Position X of UE 13 can be executed at LMF 16. t The determination of the measurement. At a specific instant, LMF 16 or some other node in the network can trigger UE 13 to report a measurement. In particular, triggering can be performed in response to the need to enhance positioning using RSRP measurements from neighboring cells. Other reports following ESM reports are those that have already been performed, or those that have traditionally been performed using beams from only the serving cell or a small number of cells.

[0061] Figure 3 A network timing diagram according to an example embodiment is provided. The timing diagram represents signals sent and / or received by one or more LCS clients 30, which can be any service requesting the current location of UE 13, as well as LMF 16 and AMF 15. Upon receiving a location request from LCS client 30, LMF 16 can determine which positioning method (e.g., ESM or non-ESM) is required. LMF 16 can forward a reporting request indicating whether an ESM report is required to AMF 15. If an ESM report is required, AMF 15 can then send a trigger message to UE 13 for ESM reporting, i.e., sending RSRP measurements of a predetermined number of beams associated with serving gNB 10 and one or more adjacent gNBs 11, 12. This means that the ESM method can be triggered only when needed, while ensuring location accuracy in the interval between triggers. This results in a significant gain in signaling overhead. In any case, non-ESM reporting can be performed as part of scheduling, and this method can be used in the interval between ESM reports.

[0062] Figure 4This is a flowchart indicating the operations that can be performed at LMF 16 or some other node, in hardware, software, or a combination thereof. The operation numbers do not necessarily indicate the order of processing. In some embodiments, more or fewer operations may be provided.

[0063] Reference numeral 4.1 indicates the start condition, which may correspond to receiving a location request from LCS client 30. At this stage, a time period T can be determined; this may be a predetermined number, or based on the type of LCS client 30 or the number of specific requests from LCS client 30. In some embodiments, the value of the time period parameter T may be estimated offline and stored in the network in the form of a LUT (e.g., at LMF 16), where the value of T is associated with a pairing of desired location accuracy and permissible signaling overhead.

[0064] In step 4.2, determine whether an ESM report is required. This can correspond to time instances t=0, T, 2T, etc.

[0065] If an ESM report is required, LMF 16 can be used in operation 4.3 to report the ESM report {BR}. t The network requests a set of measurements.

[0066] In operation 4.4, LMF 16 can receive the ESM report {BR} for that time instance. t Receive ESM report {BR} from UE13 via AMF 15 t .

[0067] In Operation 4.5, LMF 16 can determine the location (X) of UE 13 from which it received the ESM report based on the measurements in the ESM report. t For example, the ESM report {BR} t Measurements may be included, including RSRP associated with each of the four beams serving gNB 10 and RSRP associated with each of the four beams from each of the two adjacent gNBs 11, 12. The two adjacent gNBs 11, 12 may be selected as “best neighbors” based on the values ​​of their RSRP.

[0068] In operation 4.6, the position (X) can be... t Report back to LCS client 30.

[0069] In operation 4.7, position (X) t The location (X) can be stored in memory along with the corresponding time instance. This memory can be the local memory of the LMF 16 or a different memory. For example, in operation 4.8, the location (X) t () can be stored in a location database.

[0070] Sub-intervals within time period T (e.g., t0+δ) t , t0+2δ t ...t0+T-δ t At this location, non-ESM reporting can be performed. As explained, this method can use regular RSRP measurements from, for example, those serving only gNB10, which are already available because they are part of regular signaling used for resource allocation. For example, these could include RSRP measurements for the four beams serving gNB10. Therefore, specific measurement reports are not required during these sub-segments. Instead, the corresponding position at the sub-segment is based on the stored previous position (X). t-δt The function f and the available measurements associated with service gNB 10.

[0071] Therefore, in Operation 4.2, if an ESM report is not required, then in Operation 4.9, via Operations 4.4 and 4.5, a request is made or simply a measurement report is made based on the standard number of RSRP measurements {BR}. t Sum (X) t-δt To estimate the current position of the UE, as provided in Operation 4.10.

[0072] Operation 4.11 represents an evaluation of the current time instance, which may involve monitoring the clock. For subsequent time instances, the process can return to operation 4.2.

[0073] Figure 5 and Figure 2 The similarity lies in that it schematically depicts an example of UE 13 sending an RSRP measurement report 50 as a function of time, which also corresponds to the motion trajectory. In this case, according to the example embodiment, at each time instance t, the RSRP measurement report 50 depends on whether ESM is requested or needed, for example, only at a predetermined time instance with a time period T. The value of T can be considered as an adjustment parameter, as it adjusts the positioning accuracy. The UE 13's sub-intervals within the time period T (e.g., t0 + δ) t , t0+2δ t …t0+T-δ t The location at point ) is determined as a function f(60) of its estimated location at a previous time instance and the corresponding measurement at the reported current time. Alternatively:

[0074] X t+δt =f(X) t ,BR t+δt ).

[0075] The function f60 can be determined using a machine learning model. The machine learning model can be a trained model. For example, it can include a deep neural network architecture based on convolutional neural networks (CNNs) and recurrent neural networks (RNNs). This architecture takes into account the current measurement report and previous estimates of the location from UE 13.

[0076] In the simulation, a training program was developed that has... Figure 6 The deep learning model with the architecture shown is illustrated. Data points on grid 61 represent training data, which is passed through six CNN layers 62, one flattening layer 63, then through three dense fully connected network layers 63, and finally to an RNN layer 65.

[0077] Example parameters for deep learning models are specified in Table 1 below.

[0078] layer parameter Convolution 1 <![CDATA[N f =128,k=(3,3),s=(1,1)]]> Convolution 2 <![CDATA[N f =64,k=(3,3),s=(2,2)]]> Convolution 3 <![CDATA[N f =32,k=(2,3),s=(1,1)]]> Convolution 4 <![CDATA[N f =16,k=(3,3),s=(2,2)]]> Convolution 5 <![CDATA[N f =8,k=(3,2),s=(1,1)]]> Convolution 6 <![CDATA[N f =2,k=(3,3),s=(1,1)]]> Dense 1 <![CDATA[N L =64]]> Dense 2 <![CDATA[N L =16]]> Dense 3 <![CDATA[N L =2]]> RNN <![CDATA[N o =2]]>

[0079] Table 1

[0080] Key:

[0081] N f = The number of filters;

[0082] K = kernel size;

[0083] S = stride length;

[0084] N L = The number of neurons; and

[0085] N o = Output Dimension

[0086] For training Figure 6 The model's purpose is to use a dataset comprising 16,800 UE data values, consisting of beam RSRP values ​​from the serving cell and four optimal beams from two best neighboring cells. The test data consists of 4,200 samples.

[0087] For the following three cases, record the average position error value of the test data for different instantaneous time t values ​​varying from 1 to 10:

[0088] • ESM method at all time instances (3 cells * 4 beam RSRP for all t values);

[0089] • No ESM method at any time instance (1 cell * 4 beam RSRP for all t values); and

[0090] • The ESM method is used only at the first time instance of time period T (3 cells * 4 beam RSRP at t = 0, and 1 cell * 4 beam RSRP at t > 0).

[0091] Figure 7 This is a graphical comparison of the average position error achieved under three different scenarios. The x-axis range corresponds to a single time interval T; therefore, in the third scenario, the ESM method is used at t=0, and not from t=1 until the start of the next time interval T. In the third scenario, the larger the value of T, the larger the position error. However, the advantages of using the third scenario are evident from... Figure 8A and Figure 8B It is evident that they respectively show the percentage of signaling overhead and the percentage of positioning accuracy as a function of time.

[0092] exist Figure 8A In this context, signaling overhead is shown as the percentage increase in overhead compared to the second scenario where ESM is not used at any time. Figure 8B In this context, positioning accuracy is shown as a percentage gain compared to the second case where ESM is not used at any time.

[0093] Conclusions that can be drawn from the results include that the third case provides less accuracy than the first case, but this difference depends on the time period T considered. For low time periods (T<4), the difference is less than 10%, and reaches 40% for T=9. However, the benefits of using the third case become clearer when examining signaling overhead, and increase over time.

[0094] The simulation results clearly demonstrate the advantages of using ESM only at periodic intervals, such as the advantage in signaling overhead (only a 20% increase in overhead for T=9), while ensuring positioning accuracy superior to the reference scheme based on the traditional report measurement set.

[0095] Figure 9 This is a flowchart indicating operations that can be performed at LMF 16 or some other node, in hardware, software, or a combination thereof. The operation numbers do not necessarily indicate the order of processing. In some embodiments, more or fewer operations may be provided.

[0096] The first operation 9.1 may include receiving, at a first time instance, a first set of measurement data associated with each of the plurality of base stations from the target device.

[0097] The second operation 9.2 may include determining the first position of the target device based on the received first set of measurement data.

[0098] The third operation 9.3 may include receiving a second set of measurement data from the target device at each of one or more subsequent time instances, the second set of measurement data being associated with one of the multiple base stations or each of a smaller number of the multiple base stations.

[0099] The fourth operation 9.4 may include determining the corresponding location of the target device at each of one or more subsequent time instances based on the location determined at the previous time instance and a second set of measurement data.

[0100] The operation may also include the target device transmitting a first set of measurement data at a first time instance and a second set of measurement data at subsequent time instances. The first set of measurement data may be transmitted at repeated time periods, and multiple sets of second measurement data may be transmitted at one or more sub-intervals within each time period. A second set of measurement data associated with a serving base station of the target device may be received. The first set of measurement data may be associated with the serving base station of the target device and one or more neighboring base stations. These sets may include measurement data associated with multiple beams transmitted by the base stations. The first and second sets of measurement data may represent RSRP measurements determined by the target device against reference signals from the base stations.

[0101] Figure 10 An apparatus according to an example embodiment is shown. This apparatus can be configured to perform the operations described herein, for example, with reference to… Figure 3 , Figure 4 and / or Figure 9 The described operation. The device may include at least one processor 100 and at least one memory 120 directly or closely connected to the processor. The memory 120 includes at least one random access memory (RAM) 120b and at least one read-only memory (ROM) 120a. Computer program code (software) 125 is stored in the ROM 120b. The device may be connected to a receiver path of a base station to obtain LPP layer signals, etc. The device may be connected to a user interface (UI) for instructing the device and / or for outputting results. However, instead of a UI, instructions may be input from, for example, a batch file, and the output may be stored in non-volatile memory. At least one processor 100, together with at least one memory 120 and computer program code 125, is arranged to cause the device to at least perform operations according to... Figure 5 The method or any of its variations as disclosed herein.

[0102] Figure 11A non-transitory medium 130 according to some embodiments is shown. The non-transitory medium 130 is a computer-readable storage medium. It may be, for example, a CD, DVD, USB flash drive, Blu-ray disc, etc. The non-transitory medium 130 stores computer program code when... Figure 10 When a processor such as processor 100 executes, it causes the device to perform... Figure 3 , Figure 4 and / or Figure 9 The method or any of its variations as disclosed herein.

[0103] The names of network elements, protocols, and methods are based on current standards. In other versions or other technologies, the names of these network elements and / or protocols and / or methods may differ, as long as they provide the corresponding functionality. For example, embodiments may be deployed in 2G / 3G / 4G / 5G networks and further generations of 3GPP, but may also be deployed in non-3GPP radio networks such as WiFi. Therefore, base stations may be BTS, NodeB, eNodeB, gNB, WiFi access points, etc.

[0104] Memory can be volatile or non-volatile. It may be, for example, RAM, SRAM, flash memory, FPGA blockram, DCD, CD, USB flash drive, and Blu-ray disc. Unless otherwise stated or explicitly stated from the context, different statements about two entities mean that they perform different functions. This does not necessarily mean that they are based on different hardware. That is, each entity described in this specification may be based on different hardware, or some or all entities may be based on the same hardware. This does not necessarily mean that they are based on different software. That is, each entity described in this specification may be based on different software, or some or all entities may be based on the same software. Each entity described in this specification can be embodied in the cloud.

[0105] As a non-limiting example, implementations of any of the foregoing blocks, devices, systems, techniques, or methods include implementations as hardware, software, firmware, dedicated circuitry or logic, general-purpose hardware or controllers, or other computing devices, or combinations thereof. Some embodiments may be implemented in the cloud.

[0106] It should be understood that the above description represents what is currently considered a preferred embodiment. However, it should be noted that the description of the preferred embodiment is given by way of example only, and various modifications can be made without departing from the scope defined by the appended claims.

Claims

1. An apparatus comprising means for: receiving, from a target device at a first time instance, a first set of measurement data associated with each of a plurality of base stations, wherein the first set of measurement data is associated with a serving base station and one or more neighboring base stations for the target device; determining a first position of the target device based on the received first set of measurement data; receiving, from the target device at each of one or more subsequent time instances, a second set of measurement data associated with one or more of a subset of base stations of the plurality of base stations, wherein the subset of base stations comprises fewer base stations than the plurality of base stations, wherein the second set of measurement data is associated with the serving base station; determining, at each of the one or more subsequent time instances, a respective position of the target device based on the position determined at a previous time instance and the second set of measurement data.

2. The apparatus of claim 1, further comprising means for causing the target device to transmit the first set of measurement data at the first time instance and the second set of measurement data at the subsequent time instances.

3. The apparatus of claim 2, further comprising means for causing the target device to transmit the first set of measurement data at a repeating period and one or more second sets of measurement data at one or more sub-intervals within each period.

4. The apparatus of claim 1 of the preceding claims, wherein only one second set of measurement data associated with a serving base station for the target device is received.

5. The apparatus of claim 1 of the preceding claims, wherein the first set of measurement data and the second set of measurement data represent reference signal received power (RSRP) measurements determined by the target device for reference signals from the base stations.

6. The apparatus of claim 1 of the preceding claims, wherein the position determined at a previous time instance is stored in a memory device of the apparatus.

7. The apparatus of claim 1 of the preceding claims, wherein determining the respective position of the target device at each of the one or more subsequent time instances is based on a machine learning model that determines a functional relationship of the respective position with the position determined at a previous time instance and the second set of measurement data.

8. The apparatus of claim 7, wherein the machine learning model comprises a convolutional neural network and a recurrent neural network.

9. The apparatus of any one of the preceding claims, comprising a positioning node of a wireless communication network.

10. A method comprising: receiving, from a target device at a first time instance, a first set of measurement data associated with each of a plurality of base stations, wherein the first set of measurement data is associated with a serving base station and one or more neighboring base stations for the target device; determining a first position of the target device based on the received first set of measurement data; receiving, from the target device at each of one or more subsequent time instances, a second set of measurement data associated with one or more of a subset of base stations of the plurality of base stations, wherein the subset of base stations comprises fewer base stations than the plurality of base stations, wherein the second set of measurement data is associated with the serving base station; determining, at each of the one or more subsequent time instances, a respective position of the target device based on the position determined at a previous time instance and the second set of measurement data. receiving, from the target device, a second set of measurement data at each of one or more subsequent time instances, the second set of measurement data being associated with one or more base stations of a subset of the plurality of base stations, wherein the subset of base stations comprises fewer base stations than the plurality of base stations, wherein the second set of measurement data is associated with the serving base station; determining, at each of the one or more subsequent time instances, a respective position of the target device based on the position determined at the previous time instance and the second set of measurement data.

11. The method of claim 10, further comprising causing the target device to transmit the first set of measurement data at the first time instance and the second set of measurement data at the subsequent time instance.

12. The method of claim 11, further comprising causing the target device to transmit a first set of measurement data at a repeating period and one or more second sets of measurement data at one or more sub-intervals within each period.

13. The method of claim 11, wherein only one second set of measurement data associated with a serving base station for the target device is received.

14. The method of claim 11, wherein the first set of measurement data and the second set of measurement data represent reference signal received power (RSRP) measurements determined by the target device for reference signals from the base stations.

15. The method of claim 11, further comprising storing the position determined at a previous time instance in a memory device.

16. The method of claim 11, wherein determining, at each of the one or more subsequent time instances, a respective position of the target device is based on a machine learning model that determines a functional relationship of the respective position with the position determined at a previous time instance and the second set of measurement data.

17. The method of claim 16, wherein the machine learning model comprises a convolutional neural network and a recurrent neural network.

18. The method of any one of claims 10 to 17, the method being performed at a positioning node of a radio communications network.

19. An apparatus comprising at least one processor, at least one memory directly connected to the at least one processor, the at least one memory comprising computer program code, and the at least one processor, with the at least one memory and the computer program code, being arranged to perform the following method: receiving, from a target device, a first set of measurement data associated with each of a plurality of base stations at a first time instance, wherein the first set of measurement data is associated with a serving base station and one or more neighboring base stations for the target device; determining a first position of the target device based on the received first set of measurement data; receiving, from the target device at each of one or more subsequent time instances, a second set of measurement data, the second set of measurement data being associated with one or more of a subset of base stations of the plurality of base stations, wherein the subset of base stations comprises fewer base stations than the plurality of base stations, wherein the second set of measurement data is associated with the serving base station; determining, at each of the one or more subsequent time instances, a respective position of the target device based on the position determined at the previous time instance and the second set of measurement data.

20. A computer program product comprising a set of instructions which, when executed on an apparatus, is configured to cause the apparatus to perform the following method: receiving, from a target device at a first time instance, a first set of measurement data associated with each of a plurality of base stations, wherein the first set of measurement data is associated with a serving base station and one or more neighboring base stations for the target device; determining a first position of the target device based on the received first set of measurement data; receiving, from the target device at each of one or more subsequent time instances, a second set of measurement data, the second set of measurement data being associated with one or more of a subset of base stations of the plurality of base stations, wherein the subset of base stations comprises fewer base stations than the plurality of base stations, wherein the second set of measurement data is associated with the serving base station; determining, at each of the one or more subsequent time instances, a respective position of the target device based on the position determined at the previous time instance and the second set of measurement data.

21. A non-transitory computer readable medium comprising program instructions stored thereon for performing a method, the method comprising: receiving, from a target device at a first time instance, a first set of measurement data associated with each of a plurality of base stations, wherein the first set of measurement data is associated with a serving base station and one or more neighboring base stations for the target device; determining a first position of the target device based on the received first set of measurement data; receiving, from the target device at each of one or more subsequent time instances, a second set of measurement data, the second set of measurement data being associated with one or more of a subset of base stations of the plurality of base stations, wherein the subset of base stations comprises fewer base stations than the plurality of base stations, wherein the second set of measurement data is associated with the serving base station; determining, at each of the one or more subsequent time instances, a respective position of the target device based on the position determined at the previous time instance and the second set of measurement data.

Images