Determining measurement configurations in a wireless communication network

By determining measurement configurations based on prediction accuracy, the method optimizes signaling in wireless networks by adjusting measurement frequency, reducing overhead and enhancing UE performance, particularly in areas with poor coverage.

WO2026235577A1PCT designated stage Publication Date: 2026-11-12TELEFONAKTIEBOLAGET LM ERICSSON (PUBL)
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
PCT/SE2025/051015
Authority / Receiving Office
WO · WO
Patent Type
Applications
Current Assignee / Owner
Priority Date
2025-05-07
Filing Date
2025-11-11
Publication Date
2026-11-12

AI Technical Summary

Technical Problem

The challenge in modern wireless communication networks is the high signaling overhead due to frequent measurements for traffic steering and mobility, which is exacerbated by the uncertainty in prediction accuracy of signal quality, making it difficult to determine when explicit measurements are needed and to what degree.

Method used

A method and computing device that determine a measurement configuration for a UE based on prediction accuracy, adjusting measurement granularity inversely proportional to prediction accuracy, reducing unnecessary signaling by limiting measurements when predictions are reliable and increasing them when predictions are unreliable.

Benefits of technology

This approach optimizes signaling resources by efficiently allocating them to UEs that require frequent measurements, improving UE performance and supporting services in areas with poor coverage, such as remote driving, by minimizing unnecessary signaling and maximizing resource allocation.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure SE2025051015_12112026_PF_FP_ABST
    Figure SE2025051015_12112026_PF_FP_ABST
Patent Text Reader

Abstract

A method (100) performed by a computing device for determining a measurement configuration for a user equipment, UE, in a wireless communication network. The method (100) comprising obtaining (110) a request for a measurement configuration for a UE. The method (100) further comprises determining (120) a prediction accuracy of a predicted signal measurement for the UE based on baseband information of the UE, and determining (130) a measurement configuration for the UE based on the prediction accuracy. The measurement configuration indicates a measurement granularity and the measurement granularity is determined to be inversely proportional to the prediction accuracy. The method (100) further comprises providing (140) a response indicating the determined measurement configuration for the UE.
Need to check novelty before this filing date? Find Prior Art

Description

[0001] DETERMINING MEASUREMENT CONFIGURATIONS IN A WIRELESS COMMUNICATION NETWORK

[0002] Technical field

[0003] The present disclosure relates to methods for determining a measurement configuration for a user equipment (UE) in a wireless communication network. The methods may be performed by a computing device in the wireless communication network, and the present disclosure also relates to a computing device and a computer program product configured to carry out methods for determining a measurement configuration for a UE in a wireless communication network.

[0004] Background

[0005] Traffic steering and mobility are two important concepts for controlling a UE in a radio access network (RAN) of a wireless communication network. Traffic steering refers to directing user traffic to different bands or frequency layers in a coverage area to optimize network performance and improve user experience. The UEs are connected to the RAN via cells realized by frequency layers and each UE can be connected to one or more cells. Traffic steering comprises moving a UE between cells to maximize a target function, e.g., maximizing the user throughput or energy saving. In this way, quality-of-experience (QoE) expectations of the UE can be meet, the load to different bands or frequency layers can be balanced, and a high spectral and energy efficiency can be achieved. Mobility refers to directing UEs to different cells to keep the UEs connected to the RAN and securing coverage when the UEs moves in the wireless communication network. Thereby, ensuring a continuous and stable network connection for UEs moving between different geographical areas and hence different coverage areas in the wireless communication network.

[0006] An important input to traffic steering and mobility decisions is information about signal quality between the UE and different cells in a coverage area. The signal quality information is gathered from measurements on reference signals which consumes a lot of signaling resources.

[0007] With the evolution of wireless communication networks, from second generation (2G) networks to fifth generation (5G) new radio (NR) networks, the number, capacity, and diversity of available cells for serving traffic have changed dramatically. In coverage areas with co-located cells, the cells may differ in technology, frequency, coverage and bandwidths / capacities. The high number of heterogeneous cells increases the complexityof handling traffic steering and mobility. Even without mobility, the presence of other users in the cells and changing traffic load affects the situation and it becomes very challenging for the network to find an optimal set of cells to serve a UE. The network needs to run frequent serving cell optimizations for the UE leading to a high measurement load and a lot of signalling between the UE and the network.

[0008] Consequently, traffic steering and mobility management in the newer generations requires more measurements and consumes more signalling resources than the previous generations. The signalling overhead reduces the gain from optimal traffic steering and mobility. There is hence a need to find a trade-off between the performance improvements achieved with optimal traffic steering and mobility and the amount of signalling and measurements needed for the traffic steering and mobility procedures.

[0009] Summary

[0010] One option to reduce measurements and hence measurement related signalling is to use predictions of signal quality instead of actual measurements. Machine learning (ML) can be used for this type of predictions in wireless communication networks, where a ML model can be trained to predict signal quality between UEs and cells based on a set of input parameters. However, depending on the parameters used as input, the predictions of signal quality in different areas can be of different levels of accuracy. For areas where the accuracy is low and the predictions are unreliable, the signal quality needs to be obtained by explicit measurements which leads to a costly signalling overhead.

[0011] The uncertainty in prediction accuracy makes it difficult to rely on predicted signal quality alone and furthermore difficult to determine when predicted signal quality is reliably and when explicit measurements are needed, and to what degree.

[0012] Embodiments of the present disclosure address these and other issues by defining methods for determining a measurement configuration for a UE which takes prediction accuracy of a predicted signal measurement into account. In this way, a measurement granularity indicating how often to measure can be set based on how accurate a prediction is likely to be. This is beneficial for e.g., traffic steering and mobility procedure and can minimize the signaling overhead needed for these procedures.

[0013] In a first aspect, a computer implemented method is performed by a computing device in a wireless communication network. The method comprises obtaining a request for ameasurement configuration for a UE. The method further comprises determining a prediction accuracy of a predicted signal measurement for the UE based on baseband information of the UE and determining a measurement configuration for the UE based on the prediction accuracy. The measurement configuration indicates a measurement granularity. The measurement granularity is inversely proportional to the prediction accuracy. The method further comprises providing a response indicating the determined measurement configuration for the UE.

[0014] In a second aspect, there is provided a computing device for a wireless communication network. The computing device comprising processing circuitry and a memory, the memory containing instructions executable by the processing circuitry whereby the computing device is operable to obtain a request for a measurement configuration for a UE. The computing device is further operable to determine a prediction accuracy of a predicted signal measurement for the UE based on baseband information of the UE and determine a measurement configuration for the UE based on the prediction accuracy. The measurement configuration indicates a measurement granularity. The measurement granularity is inversely proportional to the prediction accuracy. The computing device is further operable to provide a response indicating the determined measurement configuration for the UE.

[0015] In a third aspect, there is provided a computer program, comprising instructions which, when executed on at least one processor, cause the at least one processor to carry out a method according to an embodiment of the first aspect.

[0016] In a fourth aspect, there is provided a computer-readable medium comprising instructions that, when executed on at least one processor, cause the at least one processor to perform a method according to an embodiment of the first aspect.

[0017] Aspects of the present disclosure provide methods and a computing device that can improve measurement functionality by adapting a measurement configuration and specifically a measurement granularity for a UE based on a prediction accuracy. When the prediction accuracy is high, the measurement granularity is low indicating less frequent measurements and vice versa. In this way, measurements and reporting of the measurements can be limited when predicted measurements are providing the desired accuracy. Unnecessary signalling can thereby be reduced, providing more resources for signalling necessary for more accurate and timely measurements.One advantage of the improved measurement functionality according to the embodiments is that it enables identification of UEs for which predictions are accurate and no or limited measurements are needed and UEs for which predictions are unreliable and frequent measurements are needed to be efficiently served. Based on this, signalling resources can be allocated more efficiently between the UEs. In use cases such as mobility, carrier aggregation and load balancing, more signalling resources can be allocated to cell borders and to high-speed UEs, as signalling resources can be saving from UEs which can use predicted measurements, e.g., UEs in the centre of the cell or stationery UEs.

[0018] One advantage of the improved measurement functionality according to the embodiments is an improved UE performance, due to the ability to adapt signalling distribution e.g., to give UEs with channel aging problems and areas with poor coverage more resources.

[0019] One advantage of the improved measurement functionality according to the embodiments is that it can better support services such as remote drive steering, autonomous driving, etc. in areas with poor coverage.

[0020] BRIEF DESCRIPTION OF THE DRAWINGS

[0021] For a better understanding of examples of the present disclosure, and to show more clearly how the examples may be carried into effect, reference will now be made, by way of example only, to the following drawings in which:

[0022] Figure 1 is a flow chart illustrating a method according to embodiments of the present disclosure;

[0023] Figure 2 is a schematic diagram illustrating a conventional serving cell selection function;

[0024] Figure 3 is a schematic diagram illustrating a serving cell selection function according to embodiments of the present disclosure;

[0025] Figure 4 is a schematic signaling diagram illustrating the method of Figure 1 performed by a network node according to embodiments of the present disclosure; Figure 5 illustrates an implementation of embodiments of the present disclosure in a 5G 3GPP network;

[0026] Figure 6 illustrates an implementation of computing devices according to embodiments of the present disclosure in a 5G 3GPP network; and

[0027] Figure 7 illustrates a computing device according to embodiments of the present disclosure.DETAILED DESCRIPTION

[0028] Particular embodiments are described more fully with reference to the accompanying drawings. Other embodiments, however, are contained within the scope of the subject matter disclosed herein. The disclosed subject matter should not be construed as limited to only the embodiments set forth herein; rather, these embodiments are provided by way of example to convey the scope of the subject matter to those skilled in the art.

[0029] Embodiments of the present disclosure propose methods and a computing device for enhanced measurement functionality to avoid signalling overhead. The proposed methods and computing device address the above discussed problems of not knowing when predictions of signal quality can be relied on and when measurements are needed. Embodiments of the present disclosure enable a computing device to determine a measurement configuration for a UE based on a prediction accuracy for a predicted signal measurement for the UE.

[0030] Figure 1 is a flow chart illustrating process steps in a computer implemented method 100 for determining a measurement configuration for a UE. The method 100 is performed by a computing device in a wireless communication network. The computing device may be comprised in a network node such as e.g., a radio network node or a core network node. The computing device may further comprise a physical or virtual node, and may be implemented in a server apparatus and / or in a virtualized environment, for example in a cloud, edge cloud or fog deployment. The UE for which the measurement configuration is to be determined may further be referred to as a wireless computing device, a mobile device, a mobile station, or similar.

[0031] In step 110, the method 100 comprises obtaining a request for a measurement configuration for a UE. A measurement configuration can herein be understood to mean a configuration indicating which measurements the UE should perform. The measurement configuration may be associated with one or more cells. The measurement configuration may comprise parameters and conditions indicating measurement to be performed or supported by the UE, e.g., which reference signals to measure and which reference signals to send. The measurement configuration may be used by a network node serving the UE, e.g., a radio access network node, to configure the UE to perform measurements to support e.g., cell selection.The request for the measurement configuration for the UE may be received from another computing device. The request may e.g., be comprised in a control message from the other computing device. However, the request may also be obtained internally in the computing device, e.g., trigger by a cell selection function in the computing device. The request for the measurement configuration for the UE may e.g., be received from a baseband unit.

[0032] In step 120, the method 100 comprises determining a prediction accuracy of a predicted signal measurement for the UE based on baseband information of the UE. The predicted signal measurement indicates a quality of a signal between the UE and a cell. The signal measurement may be predicted using a rule-based function or a ML model with the baseband information as input. For example, the predicted signal measurement for the UE may be obtained by inputting the baseband information to a rule-based function which maps the baseband information to a predicted signal measurement. Alternatively, the baseband information may be input to a ML model trained to predict signal measurements based on the baseband information. The baseband information may comprise one or more of Timing Advance (TA), Channel Quality Indicator (CQI), Uplink_Pathloss, Downlink_pathloss, Power_Ratio, Beamjndex, Precoding Matrix Indicator (PMI). The prediction accuracy may then be determined from the predicted signal measurement and / or an accuracy associated with a method used to perform the prediction, as will be further described below.

[0033] In step 130, the method 100 comprises determining a measurement configuration for the UE based on the prediction accuracy determined in step 120. The measurement configuration indicates a measurement granularity. The measurement granularity in turn indicates how often measurements are to be performed. When the measurement granularity is low, the UE performs measurements less frequently, i.e., a time interval between measurements is long. When the measurement granularity is high, the UE performs measurements more frequently, i.e., a time interval between measurements is short. The measurement granularity may further be referred to as a measurement frequency or measurement pattern and indicate a granularity, frequency, or pattern of measurements, respectively. The performing of measurements may include one or more of sending of reference signals, measuring reference signals, processing measurements, and reporting measurements.

[0034] In the determined measurement configuration, the measurement granularity is determined to be inversely proportional to the prediction accuracy. In other words, if the prediction accuracy increases, the measurement granularity decreases. In embodiments, a higherprediction accuracy leads to a lower measurement granularity associated with less frequent measurements and a lower prediction accuracy leads to a higher measurement granularity associated with more frequent measurements. A higher prediction accuracy can herein be understood to mean a more accurate prediction, i.e., that the predicted value is more reliable and hence more likely to correspond to the actual value. And a lower prediction accuracy can herein be understood to mean a less accurate prediction, i.e., that the predicted value is less reliable and hence less likely to correspond to the actual value.

[0035] In embodiments, if the prediction accuracy is above a first accuracy value, the measurement granularity is determined to be a minimum value. A prediction accuracy above the first accuracy value may indicate that the predicted signal measurement is reliable and that no or a minimum of measurements are required. The measurement granularity can hence be set to the minimum value. The minimum value may correspond to no measurements or as few measurements as possible (some measurements and / or reference signals may be mandatory and not possible to turn off).

[0036] In step 140, the method 100 comprises providing a response indicating the determined measurement configuration for the UE. When the request is received from another computing device, the response is provided to that other computing device. The determined measurement configuration may further be directly or indirectly provided to the UE for configuring the UE to perform measurements according to the determined measurement configuration.

[0037] In some embodiments, the prediction accuracy can be determined based on the predicted signal measurement, i.e., from the predicted value of the signal measurement. In this case, the prediction accuracy may be determined to be higher when the predicted signal measurement indicates a higher signal quality, and the prediction accuracy may be determined to be lower when the predicted signal measurement indicates a lower signal quality. That the prediction accuracy is determined to be higher may be understood to mean that the prediction accuracy is given or set to a higher value. Thus, the prediction accuracy may be given a higher value for a predicted signal measurement indicating a higher signal quality than a predicted signal measurement indicating a lower signal quality.

[0038] In embodiments, the prediction accuracy may be determined by comparing the value of the predicted signal measurement for the UE with one or more threshold values. In this case, the prediction accuracy may be given a value based on whether it is above or below theone or more threshold values. For example, if the value of the predicted signal measurement is equal to or above a first threshold value, the prediction accuracy may be determined to be a first value. And, if the value of the predicted signal measurement is below the first threshold value, the prediction accuracy may be determined to be a second value lower than the first value. The first value may be associated with no or limited measurements, as the predicted signal measurement is reliable and indicates that the UE is in coverage. The second value may be associated with frequent measurements, as the predicted signal measurement is unreliable, and it is ambiguous whether the UE is in coverage or out of coverage.

[0039] However, this ambiguity related to whether the UE is in coverage or out of coverage disappears when the predicted signal measurement becomes very poor, as this indicates that the UE is clearly out of coverage. Thus, although the prediction accuracy may be determined to be lower when the predicted signal measurement indicates a lower signal quality, this may in embodiments only be true as long as the signal quality is above a lower threshold. When the predicted signal measurement indicates a signal quality below the lower threshold, the prediction accuracy may again be set to a higher value. This time indicating that the UE is clearly out of coverage of the cell and no or limited measurements are needed for that cell. Thus, in embodiments, the first threshold value and a second threshold value may be used. The prediction accuracy may be set to the first value, if the predicted signal measurement is above the first threshold value or below the second threshold value. The prediction accuracy may be set to the second value lower than the first value, if the predicted signal measurement is between the first threshold value and the second threshold value. As before, the first value may be associated with no or limited measurements. This time as it can be assumed that the UE is clearly in coverage, alternatively clearly out of coverage. The second value may be associated with frequent measurements, as it is ambiguous whether the UE is in coverage or out of coverage from the predicted signal measurement and explicit measurements are needed.

[0040] As previously stated, the predicted signal measurement may be determined using a rulebased function or a ML model with the baseband information of the UE as input. The signal measurement may be any type of signal strength, signal quality or signal level measurement. Some non-limiting examples of signal measurement are Received Signal Strength Indicator (RSSI), Reference Signals Received Power (RSRP), and Reference Signal Received Quality (RSRQ).As an alternative or complement to using the predicted signal measurement to determine the prediction accuracy, the prediction accuracy may be determined based on an accuracy associated with the method used to perform the prediction. When a ML model is used to determine the predicted signal measurement for the UE, the accuracy of the ML model may be determined based on a quality metric associated with the ML model. In embodiments, the prediction accuracy may hence be determined based on a quality metric associated with the ML model. The prediction accuracy may be determined to be higher when the quality metric indicates a more accurate prediction, and the prediction accuracy may be determined to be lower when the quality metric indicates a less accurate prediction. In other words, the prediction accuracy is given or set to a higher value for a quality metric indicating a more accurate prediction than a quality metric indicating a less accurate prediction.

[0041] In embodiments, the prediction accuracy may be determined by comparing the quality metric with one or more thresholds. In this case, the prediction accuracy may be given a value based on whether the quality metrics is above or below the one or more thresholds. For example, if a value of the quality metric is equal to or above a third threshold value, the prediction accuracy may be determined to be a third value. And, if the value of quality metric is below the third threshold value, the prediction accuracy may be determined to be a fourth value lower than the third value.

[0042] The quality metric may be an error metric indicating prediction accuracy or a portion of correct predictions depending on the type of ML model used. When the ML model is a regression model, an error metric may be used. In this case, an error metric indicating less errors indicates a more accurate prediction and hence leads to a higher prediction accuracy. Common error metrics include e.g., Mean Absolute Error (MAE), Mean Squared Error (MSE), Root Mean Squared Error (RMSE) and R-squared. MAE measures the average absolute difference between actual and predicted values, MSE penalizes larger errors more heavily than MAE, RMSE provides a more interpretable scale compared to MSE, and R-squared measures how well the ML model explains the variance in the data given the mean value.

[0043] When the ML model is a classification model, a portion of correct predictions may be used as a quality metric. A higher portion of correct prediction indicates a more accurate prediction and hence leads to a higher prediction accuracy. The portion of correct predictions may be given by the number of correct predictions divided by the total number of predictions and may e.g., be indicated as a percentage value.In embodiments, the prediction accuracy is determined further based on one or more of characteristics of the UE and a location of the UE. In addition to using the baseband information, information related to characteristics and / or a location of the UE may hence be used to determine the prediction accuracy. Characteristics of the UE may comprise one or more of a device type, a device capability, a priority associated with the UE and a type of service used by the UE. The location of the UE may be a geographical location and / or a radio location of the UE.

[0044] The information used to determine the prediction accuracy may be obtained in a number of different ways. In embodiments, the request for the measurement configuration indicates one or more of: the baseband information of the UE, characteristics of the UE, and a location of the UE. Thus, the information needed to determine the prediction accuracy may be obtained in the request for the measurement configuration received in step 110. However, the baseband information of the UE, the characteristics of the UE, and / or the location of the UE may further be obtained from information available to the computing device, e.g., information stored in the computing device, or which can be fetched from another node.

[0045] The baseband information of the UE may e.g., be obtained from a baseband unit communicating with the UE. The baseband unit may be comprised in the computing device or in communication with the computing device.

[0046] The characteristics of the UE may e.g., be obtained from UE grouping class, which indicates traffic priority of a UE or type of the ongoing session (streaming, calling, gaming, etc.), or UE chipset class, which indicates the measurement accuracy of a UE and potentially the type of the device, wearable, mobile phone, etc. The UE grouping class and UE chipset class may be reported by the UE or obtained based on other UE information from e.g., a subscriber data base or vendor.

[0047] The location of the UE may be obtained from information and measurements exchanged between the UE and the wireless communication network either directly or indirectly through various known positioning techniques. Examples of such information and measurements are:

[0048] Timing measurements, such as Time of Arrival (ToA), Time Difference of Arrival (TDoA), and Round-Trip Time (RTT).

[0049] Signal strength measurements, such as e.g., RSSI, RSRQ etc.Angle measurements, such as e.g., Angle of Arrival (AoA).

[0050] Cell identifier information, such as Cell Global Identity (CGI), and Enhanced Cell-ID (E-CID).

[0051] Global Navigation Satellite System (GNSS) Information e.g., GPS, Galileo, etc. Location Services (LCS) Assistance Data

[0052] These measurements and information types can be used individually or in combination, depending on the specific positioning method employed by the network. For instance, in long term evolution (LTE) and NR networks, a combination of RSRP, RTT, and TDoA is often used to determine the location of a UE more accurately. In more advanced scenarios, the wireless communication network may also use ML algorithms to improve location accuracy by combining these measurements in sophisticated ways.

[0053] In embodiments, the measurement configuration is requested as part of a cell selection procedure for the UE. The cell selection procedure may be performed to select one or more serving cells for the UE for traffic steering or mobility purposes. An implementation of such an embodiment will now be described with reference to figures 2 and 3.

[0054] Figure 2 illustrates a conventional serving cell selection function 200. The serving cell selection function 200 is used in variety of layer 3 functions for traffic steering and mobility. The objective of the serving cell selection function 200 is to find a configuration for a UE. The configuration comprises a serving cell set and indicates which cells the UE should measure and how often to measure each cell. This configuration corresponds to the measurement configuration according to embodiments of the present disclosure.

[0055] With reference to figure 2, network information related to network configuration and network resources is input to a configuration finder 201. Based on the network information, the configuration finder 201 identifies one or more possible configurations for the UE and suggests a configuration to a capability checker 202. The capability checker 202 checks whether the suggested configuration is supported or not by the UE based on UE information provided to the capability checker 202, and provides a response to the configuration finder 201, as indicated in figure 2. This process is performed iteratively and continues until all possible configurations are checked or until a timeout. If the serving cell selection function 200 is unable to find a configuration that is supported by the UE, a UE capability failure is declared. If the serving cell selection function 200 finds a supported configuration, theserving cell selection function 200 outputs the selected configuration. The selected configuration is then provided to the UE.

[0056] Once configured, the UE starts measuring on the serving cell set according to the selected configuration. The measurements are monitored, and events are generated when certain conditions are fulfilled. The conditions are e.g., related to a serving cell or a neighbouring cell becoming better or worse than a threshold. Based on the generated events, the serving cell selection function 200 decides on actions related to traffic steering or mobility.

[0057] The measurements performed by the UE comes with a cost as they lead to signalling overhead and might also need measurement gaps, which cause throughput degradation. The gain from optimal traffic steering and mobility may hence be reduced due to the measurements needed to achieve the optimal traffic steering and mobility. For efficient traffic steering and mobility, a trade-off between the performance of the traffic steering or mobility procedures and the amount of measurements required needs to be found.

[0058] Figure 3 illustrates an embodiment where the method 100 is implemented in a cell selection procedure performed by a serving cell selection function 300. In addition to a configuration finder 301 and a capability checker 302 performing the same or similar functions to the convention configuration finder 201 and capability checker 202, respectively, a new module named Prediction Accuracy to Signaling Overhead Mapper (PASOM) 303 is introduced in the serving cell selection function 300. The PASOM 303 performs one or more of the steps of the method 100. For example, the determination of the prediction accuracy and the measurement configuration for the UE, as described with reference to step 120 and 130 in figure 1. With reference to figure 3, the PASOM 303 receives the supported configuration from the capability checker 301 and the UE information. Based on this input, the PASOM 303 determines for which components of the serving cell set in the supported configuration the UE should perform measurements and how often, i.e., a measurement granularity per cell in the serving cell set. This determination is based on the prediction accuracy, as previously described. The supported configuration with measurement requirements, which corresponds to the measurement configuration determined in step 130 of the method 100, is provided to the configuration finder 301 and then output from the serving cell selection function 300 as the selected configuration. The UE information provided to the PASOM 303 includes the baseband information and may further include one or more of characteristics of the UE, such as e.g., UE grouping class, UE chipset class, and location of the UE.The determining of the predicted signal measurement according to the method 100, e.g., by the PASOM module, may be implemented using a rule-based function or a ML model. In an example of a rule-based function, the predicted signal measurement may be obtained from a set of tables mapping input information such as baseband information to a predicted signal measurement. The tables may be stored at the PASOM or accessible to the PASOM, e.g., from a Network Data Analytics Function (NWDAF). The tables could be derived in computing devices, such as e.g., edge devices, based on coverage maps of service areas, locations of base stations, statistical mobility and / or carrier aggregation bitrates, and measurement results from UEs. The tables may be built and used continuously based on received events (measurement-based) and measurements. They may also be built once a radio network node starts operation, and then be exploited until the need for an update of the tables is triggered.

[0059] Assume the signal measurement to be predicted is RSRP. A configuration may be sent from the baseband unit to the PASOM to check the need for measurements for the cells associated with the configuration. Using baseband information of the UE as input to the set of tables, a prediction of RSRP at the radio location of the UE can be made. The radio location being obtained from the baseband information of the UE.

[0060] By comparing the predicted RSRP with thresholds values, the prediction accuracy of the predicted RSRP can be determined. For example, a first threshold may be associated with a good signal and a second threshold may be associated with a poor signal. If the predicted RSRP is above the first threshold or below the second threshold value, the prediction accuracy is determined to be high. The predicted RSRP can be trusted, meaning that the UE can reliably be assumed to be in coverage or out of coverage, respectively, and a minimum of measurements are needed. On the other hand, if the predicted RSRP is between the first threshold and the second threshold, the prediction accuracy is determined to be low, meaning that it is ambiguous if the UE is in or out of coverage. The predicted RSRP cannot be trusted, and measurements are needed.

[0061] From the RSRP predicted by the tables, the PASOM may hence identify whether measurements are needed or not and if measurements are not needed, whether the UE is in coverage or not.

[0062] In an example of using a ML model for predicting signal measurements, the ML model may be a deep neural network. The deep neural network may be trained on similar events andmeasurements as described for the rule-based function. Information corresponding to the rule-based tables may be captured and learned by weights of the neurons of the neural network. Non-limiting examples of input for training the neural network are serving frequency, Physical Cell ID (PCI), beamwidth, radial indicator measurements (e.g. timing advance or pathloss), angle of arrival indicator measurements, e.g. PMI, RSRP to the serving cell, PCIs and RSRP measurement associated with neighbouring cells.

[0063] The choice of neural network layers and number of neurons per layer can vary based on the type of network node the ML model is expected to be deployed in. Factors defining the network architecture may e.g., include type of cells (macro, micro, etc.), type of area (urban, rural, etc.) and number of antennas in the nodes. Depending on the complexity and size of the service area, and dynamic and / or static blockages in the service area, learning of the coverage in the service area could be different from one network to another. Based on a proof of concept, a 4 layer fully connected neural network including (2,10,10, 2) neurons with pathloss and PMI as input, and giving a Q-value of in-coverage or out-of-coverage as output can work for an ordinary urban cell (sector) of size 2000m2.

[0064] ML model(s) for coverage prediction could be built and used continuously based on the received events (measurement-based) and measurements. They can be also built once a radio network node starts operation and be exploited until the need for retraining is triggered based on performance degradation.

[0065] For the purposes of the present disclosure, the term “ML model” encompasses within its scope the following concepts: ML algorithms, comprising processes or instructions through which data may be used in a training process to generate a model artefact for performing a given task, or for representing a real world process or system; the model artefact that is created by such a training process, and which comprises the computational architecture that performs the task; and the process performed by the model artefact in orderto complete the task.

[0066] As previously described, the measurement configuration may be requested as part of a cell selection procedure for the UE. The cell selection procedure may be performed to select one or more serving cells for the UE. Figure 4 illustrates the process of determining a measurement configuration for a UE according to an embodiment in a cell selection scenario in a 3GPP wireless communication network. The cell selection scenario may be for selecting a secondary cell (SCell) for the UE in a dual connectivity or carrier aggregationscenario. By using the method 100, the signalling overhead can be adapted to get a good trade-off between signalling and suitable cells serving the UE. In the illustrated embodiment, the method 100 is performed by a first network node 410 upon a request from a second network node 420. The second network node 420 serves the UE 430 and may e.g., control a cell being a primary cell (PCell) for the UE 430. The first network node 410 may comprise one or more cells which are potential SCells for the UE 430.

[0067] In step I in figure 4, the second network node 420 receives UE and radio information from the UE 430. The UE information may indicate the capabilities of the UE 430. The radio information may e.g., include pathloss, TA, Beam Index (Bl), and other radio-related information. In step II in figure 4, the second network node 420 sends a UE measurement configuration request to the first network node 410. The UE measurement configuration request corresponds to the request for measurement for the UE according to embodiments of the present disclosure and may comprise the UE and radio information which the second network 420 node received from the UE 430.

[0068] Upon receiving the request for a UE measurement configuration from the second network node 420, the first network node 410 determines a measurement configuration for the UE 430 in step III in figure 4. Step III comprises the first network node 410 obtaining information about prediction accuracy to determine whether the UE 430 should perform measurements and when this is the case how often these measurements should be performed and / or reported.

[0069] With reference to Figure 4, obtaining the prediction accuracy may comprise the first network node 410 checking with a core node 400 in the network what the measurement granularity for the UE should be. The core node 400 may store data mapping the UE baseband information and optionally UE information to a prediction accuracy. Thus, the first network node 410 may send a prediction accuracy request to the core node 400 in step Illa. The prediction accuracy request may comprise UE baseband information and optionally UE information. The UE baseband information and / or UE information may be obtained from the request received in step II and / or from other information available to the first network node 410. The core network 400 maps the UE baseband information to a prediction accuracy and sends a prediction accuracy response comprising the prediction accuracy to the first network node 410 in step lllb. If the prediction accuracy is high, measurements may be switched off for a certain period. If the prediction accuracy is low, the measurements maybe intensified. In general, the level of prediction accuracy may map to a certain level of measurement granularity for the UE.

[0070] The first network node 410 may receive additional information about interference (RSRQ), load, signal strength, etc., per carrier-radio location. The additional information may be received from one or more baseband units. This additional information may be used by the first network node 410 to determine the amount of measurement signalling that effectively will meet the quality-of-service (QoS) of the UE. Thus, in step III in figure 4, the first network node 410 evaluates the prediction accuracy from the core node 400 and optionally available additional information to determine a measurement configuration for the UE 430 including a measurement granularity.

[0071] In step IV in figure 4, the first network node 410 sends the determined UE measurement configuration to the second network node 420. The second network node 420 instructs the UE 430 on the configuration measurement setup it should apply by sending a UE measurement configuration setup request to the UE 430 in step V in figure 4. Once the UE 430 actuates the measurement configuration successfully it sends an acknowledgement to the second network node 420 in step VI in figure 4. The UE 430 then performs measurements according to the measurement granularity indicated by the measurement configuration.

[0072] In embodiments, the UE 430 may further send location information, such as GPS information, along with the UE and radio information allowing mapping of geolocation and measurements. This would allow for effectively distributing the signalling overhead for measurements among the UEs. In such a case, non-line of sight (NLOS) users or highly mobile users can be configured for more frequent signalling for measurements, while line of sight (LOS) and stationary users could receive no or less frequent signalling for measurements.

[0073] The first network node 410 may further receive information from other network nodes about the UE 430, e.g. TA and AoA, and other radio-related measurements. The other network nodes may belong to the same first network node 410, e.g., be cells controlled by the first network node 410, or be neighbouring network nodes.

[0074] In embodiments, information may be exchanged between first network nodes 410 and second network nodes 420 realizing PCells or SCells for the UE 430. The above UEbaseband information can be requested by multiple second network nodes 420 for a UE 430. The first network node(s) 310 to which the second network nodes 420 belong may determine collaboratively a measurement configuration taking into account the multiple connections of the UE 430 to the network, i.e., the UE baseband information gathered by the second network nodes 420. Through the collaboration between PCell(s) and SCell(s) the signalling overhead can be reduced.

[0075] In another embodiment, the derived measurement configuration may impact the target of cell set selection, and hence, the cell selection and measurement configuration derivation may be done iteratively. For example, to maximize UE throughput, a first cell set has been selected, and subsequently a measurement configuration has been derived. Now, the required measurements may include some measurements that reduce the throughput by x% for a period of time. Thus, the expected benefit from the first cell set may be less than the benefit from another cell set that has a lower measurement requirement, e.g., a second cell set which may be the second best in throughput. The second cell set may hence become the best option and be selected. Thereby, enabling the selection of a cell set with a higher achievable throughput when taking also the signalling overhead into account.

[0076] Figure 5 illustrates an implementation of an embodiment of the invention in a 5G 3GPP network 500. In 5G, one or more 5G Node Bs (gNBs) 510 may be connected to the 5G Core 520 via the NG interface. The gNB 510 may further be split into a gNB-Central Unit (gNB-CU) 511 and one or more gNB-Distributed Unit(s) (gNB-DU) 512 connected via the F1 interface. Two gNBs 510 may be connected to each other via the Xn interface.

[0077] In figure 5, the computing device, e.g., the PASOM, may be implemented in the gNB-CUs 511 and / or the gNB-DUs 512. As indicated in figure 5, the measurement configuration request and the measurement configuration response may be exchanged between the gNB-CU 511 and any of the gNB-DUs 512 over the F1 interface. Furthermore, the prediction accuracy request and the prediction accuracy response may be exchanged over the NG interface between the gNBs 510 and the 5G Core 520.

[0078] This implementation enables the prediction accuracy to be determined over multiple cells, e.g. mutliple Scell candidates. A DU 512 of a gNB is associated with a cell (Pcell or Scell) and the PASOM at the DU 512 may indicate the prediction accuracy of that particular cell. By connecting with the PASOM at the CU 511 and / or a NWDAF 522 in the 5G core 520, the prediction accuracy over multiple cells can be determined, see also figure 6. In addition,for certain UEs the level of confidence in terms of prediction accuracy may be rather low. In this case, combining prediction accuracy over multiple cells can provide safety margins to serve that UE. Combining prediction accuracy over multiple cells may further be useful for traffic load varying use cases where cells are turned on and off. For example, when switching off a Pcell or Scell where the prediction accuracy is high and the UE is moved to a Pcell or Scell where the prediction accuracy may be lower.

[0079] Figure 6 illustrates an implementation of an embodiment where computing devices, here in the form of PASOMs 610, 620, is implemented both in the DUs 512 and in the CU 511 of a gNB. The PASOM in the CU 511 is here referred to as a backend PASOM 620. The DU PASOMs 610 may store the previously described tables or ML model for the cell realised by the respective DU 512. The tables or ML model stored in one of the PASOMs 610 may be enough to determine prediction accuracy for the cell realised by that DU 512. Thus, when the PASOM 610 receives a request for a measurement configuration for a UE 630 from a baseband unit 640, the DU PASOM 610 may determine the prediction accuracy and the measurement configuration for the UE 630 using information from the baseband unit 640 as input to the tables or ML model stored in the PASOM 610, and provide a response indicating the determined measurement configuration to the baseband unit, as indicated in figure 6.

[0080] When the prediction accuracy over multiple cells may be beneficial, the prediction accuracy may be determined at CU 511 level using tables or ML models stored in the backend PASOM 620 or stored in the 5G Core, e.g., in an edge Knowledge Base (KB) or an edge NWDAF (E-NDWAF) 650. These tables or ML models may be built using information from multiple baseband units 640, allowing information to be shared between multiple cells, e.g., multiple PCells and / or SCells. The use of the backend PASOM 620 is illustrated in figure 6 with the request-response exchange between the PASOMS 610 in the DUs 512 and the backend PASOM 620, as well as the request-response exchange between the backend PASOM 620 and the edge KB / E-NWDAF 650.

[0081] As discussed above, the method 100 is performed by a computing device, and the present disclosure provides a computing device that is adapted to perform any or all of the steps of the above discussed method. The computing device may comprise a physical or virtual node, and may be implemented in a radio access network node, radio base station, base transceiver station, node B, evolved node B, gNB, access point, access node, radio intelligence controller, computer system, computing device or server apparatus, etc., and / orin a virtualized environment, for example in a cloud, edge cloud, an Open Radio Access Network (O-RAN), or fog deployment. Examples of a virtual node may include a piece of software or computer program, a code fragment operable to implement a computer program, a virtualised function, or any other logical entity.

[0082] Figure 7 is a block diagram illustrating an example computing device 700. The computing device 700 may be configured to operate in the manner described herein in respect of a computing device, a first network node 410 and / or a PASOM 303, 610, 620. The computing device 700 comprises processing circuitry (or logic) 701. The processing circuitry 701 controls the operation of the computing device 700 and can implement examples of the methods described herein. The processing circuitry 701 can comprise one or more processors, processing units, multi-core processors or modules that are configured or programmed to control the computing device 700 in the manner described herein. In particular implementations, the processing circuitry 701 can comprise a plurality of software and / or hardware modules that are each configured to perform, or are for performing, individual or multiple steps of the method described herein in relation to the computing device 700. It will be appreciated that the computing device 700 may comprise one or more virtual machines running different software and / or processes. The computing device 700 may therefore comprise, or be implemented in or as one or more servers, switches and / or storage devices and / or may comprise cloud computing infrastructure that runs the software and / or processes.

[0083] The computing device 700 further comprises a memory 703. In some embodiments, the memory 703 of the computing device 700 can be configured to store instructions (e.g. program code) executable by the processing circuitry 701 of the computing device 700 whereby the apparatus is operable to perform one or more of the steps of the method 100 as described with reference to figure 1. Alternatively or in addition, the memory 703 of the computing device 700, can be configured to store any requests, resources, information, data, signals, or similar that are described herein. The processing circuitry 701 of the computing device 700 may be configured to control the memory 703 of the computing device 700 to store any requests, resources, information, data, signals, or similar that are described herein

[0084] In some embodiments, the computing device 700 may optionally comprise a communications interface 702. The communications interface 702 of the computing device 700 can be for use in communicating with other nodes, such as other virtual nodes. Forexample, the communications interface 702 of the computing device 700 can be configured to transmit to and / or receive from other nodes requests, resources, information, data, signals, or similar. The processing circuitry 701 of computing device 700 may be configured to control the communications interface 702 of the computing device 700 to transmit to and / or receive from other nodes requests, resources, information, data, signals, or similar. The communications interface 702 can use any suitable communication technology.

[0085] There is also provided a computer program comprising instructions which, when executed on a least one processor (such as the processing circuitry 701 of the computing device 700 described earlier), cause the processor to carry out at least part of the method(s) described herein. According to some embodiments there is provided a carrier containing the computer program. In some embodiments, the carrier can be any one of an electronic signal, an optical signal, an electromagnetic signal, an electrical signal, a signal, a microwave signal, or a computer-readable medium. There is also provided a (for example, tangible and / or non-transient) computer-readable medium comprising instructions which, when executed by at least one processor, cause the at least one processor to perform at least part of the method(s) described herein.

[0086] Functions implemented by some embodiments may be virtualized in a virtualization environment. In the present context, virtualizing means creating virtual versions of apparatuses or devices which may include virtualizing hardware platforms, storage devices and networking resources. As used herein, virtualization can be applied to any device described herein, or components thereof, and relates to an implementation in which at least a portion of the functionality is implemented as one or more virtual components. Some or all of the functions described herein may be implemented as virtual components executed by one or more virtual machines (VMs) implemented in one or more virtual environments hosted by one or more of hardware nodes, such as a hardware UE that operates as a network node, UE, core network node, or host. Further, in embodiments in which the virtual node does not require radio connectivity (e.g., a core network node or host), then the node may be entirely virtualized. In some embodiments, the virtualization environment includes components defined by the O-RAN Alliance, such as an O-Cloud environment orchestrated by a Service Management and Orchestration Framework via an 0-2 interface. Virtualization may facilitate distributed implementations of a network node, UE, core network node, or host.Applications (which may alternatively be called software instances, virtual appliances, network functions, virtual nodes, virtual network functions, etc.) can be run in the virtualization environment to implement some of the features, functions, and / or benefits of some of the embodiments disclosed herein.

[0087] In some instances, detailed descriptions of well-known methods, nodes, interfaces, circuits, and devices are omitted so as not obscure the description with unnecessary detail. Those skilled in the art will appreciate that the functions described may be implemented in one or more nodes using hardware circuitry (e.g., analog and / or discrete logic gates interconnected to perform a specialized function, ASICs, PLAs, etc.) and / or using software programs and data in conjunction with one or more digital microprocessors or general-purpose computers. Nodes that communicate using the air interface may have suitable radio communications circuitry. Moreover, where appropriate the technology can additionally be considered to be embodied entirely within any form of computer-readable memory, such as (ROM, EEPROM, Flash memory, a memory disc, RAM etc.) solid-state memory, magnetic disk, or optical disk containing an appropriate set of computer instructions that would cause a processor to carry out the techniques described herein.

[0088] Hardware implementation may include or encompass, without limitation, digital signal processor (DSP) hardware, a reduced instruction set processor, hardware (e.g., digital or analogue) circuitry including but not limited to application specific integrated circuit(s) (ASIC) and / or field programmable gate array(s) (FPGA(s)), and (where appropriate) state machines capable of performing such functions.

[0089] Generally, all terms used herein are to be interpreted according to their ordinary meaning in the relevant technical field, unless a different meaning is clearly given and / or is implied from the context in which it is used. All references to a / an / the element, apparatus, component, means, step, etc. are to be interpreted openly as referring to at least one instance of the element, apparatus, component, means, step, etc., unless explicitly stated otherwise. The steps of any methods disclosed herein do not have to be performed in the exact order disclosed, unless a step is explicitly described as following or preceding another step and / or where it is implicit that a step must follow or precede another step. Any feature of any of the embodiments disclosed herein may be applied to any other embodiment, wherever appropriate. Likewise, any advantage of any of the embodiments may apply to any other embodiments, and vice versa. Other objectives, features and advantages of the enclosed embodimentswill be apparent from the following descriptionIt should be noted that the above-mentioned embodiments illustrate rather than limit the invention, and that those skilled in the art will be able to design many alternative embodiments without departing from the scope of the appended claims. The word “comprising” does not exclude the presence of elements or steps other than those listed in a claim, “a” or “an” does not exclude a plurality, and a single processor or other unit may fulfil the functions of several units recited in the claims. Any reference signs in the claims shall not be construed so as to limit their scope.

Claims

CLAIMS1. A computer implemented method (100) performed by a computing device in a wireless communication network, the method comprising:obtaining (110) a request for a measurement configuration for a user equipment, UE; determining (120) a prediction accuracy of a predicted signal measurement for the UE based on baseband information of the UE;determining (130) a measurement configuration for the UE based on the prediction accuracy, wherein the measurement configuration indicates a measurement granularity, the measurement granularity being inversely proportional to the prediction accuracy; and providing (140) a response indicating the determined measurement configuration for the UE.

2. The method (100) of claim 1, wherein a higher prediction accuracy leads to a lower measurement granularity associated with less frequent measurements and a lower prediction accuracy leads to a higher measurement granularity associated with more frequent measurements.

3. The method (100) of claim 1 or 2, wherein the prediction accuracy is determined based on the predicted signal measurement.

4. The method (100) of claim 3, wherein the prediction accuracy is determined to be higher when the predicted signal measurement indicates a higher signal quality, and the prediction accuracy is determined to be lower when the predicted signal measurement indicates a lower signal quality.

5. The method (100) of any of claim 1 to 4, wherein the predicted signal measurement is determined using a rule-based function with the baseband information of the UE as input.

6. The method (100) of any of claim 1 to 4, wherein the predicted signal measurement is determined using a machine learning, ML, model with the baseband information of the UE as input.

7. The method (100) of claim 6, wherein the prediction accuracy is determined based on a quality metric associated with the ML model, wherein the quality metric is an error metric indicating prediction accuracy or a portion of correct predictions.

8. The method (100) of claim 7, wherein the prediction accuracy is determined to be higher when the quality metric indicates a more accurate prediction, and the prediction accuracy is determined to be lower when the quality metric indicates a less accurate prediction.

9. The method (100) of any preceding claim, wherein if the prediction accuracy is above a threshold value, the measurement granularity is determined to be a minimum value.

10. The method (100) of any preceding claim, wherein the prediction accuracy is determined further based on one or more of characteristics of the UE and a location of the UE.

11. The method (100) of any preceding claim, wherein the request for the measurement configuration comprises one or more of: the baseband information of the UE, characteristics of the UE, and a location of the UE.

12. The method (100) of any preceding claim, wherein the baseband information comprises one or more of: Timing Advance (TA), Channel Quality Indicator (CQI), Uplink_Pathloss, Downlink_pathloss, Power_Ratio, Beamjndex, Precoding Matrix Indicator (PMI).

13. The method (100) of any preceding claim, wherein the measurement configuration is associated with a cell and the predicted signal measurement indicates a quality of a signal between the UE and the cell.

14. The method (100) of any preceding claim, wherein the measurement configuration is requested as part of a cell selection procedure for the UE.

15. A computing device (700) in a wireless communication network, the computing device (700) comprising processing circuitry (701) and a memory (703), the memory (703) containing instructions executable by the processing circuitry (701) whereby the computing device (700) is operable to:obtain a request for a measurement configuration for a user equipment, UE; determine a prediction accuracy of a predicted signal measurement for the UE based on baseband information of the UE;determine a measurement configuration for the UE based on the prediction accuracy, wherein the measurement configuration indicates a measurement granularity, the measurement granularity being inversely proportional to the prediction accuracy; andprovide a response indicating the determined measurement configuration for the UE.

16. The computing device (700) as claimed in claim 15, wherein the memory (703) further contains instructions executable by the processing circuitry (701) whereby the computing device (700) is operable to perform the method (100) as claimed in any one of claims 2 to 14.

17. A computer program, comprising instructions which, when executed on at least one processor, cause the at least one processor to carry out a method according to any of claims 1 to 14.

18. A computer program product comprising non transitory computer readable media having stored thereon a computer program according to claim 17.