Apparatus and computer-implemented method for operating mobile terminal, and apparatus and method for supporting mobile terminal in method

Abstract

An apparatus and a computer-implemented method for operating a mobile terminal (102-1), in which a quantity characterizing a radio connection between the mobile terminal (102-1) and an access point of a telecommunications network is determined in the mobile terminal (102-1), wherein a model (204) is used in the mobile terminal (102-1) which is designed to predict a property of the radio connection and / or a property of a radio connection between the mobile terminal (102-1) and another access point of the telecommunications network on the basis of the quantity and at least one model parameter, a prediction of the characteristic is determined on the basis of the quantity and the at least one model parameter, a reference quantity being determined in the mobile terminal (102-1), the reference quantity characterizing the radio connection for which the prediction has been determined; wherein a measure of the quality of the prediction is determined in the mobile terminal (102-1) as a function of the prediction and the reference quantity, and wherein the measure is transmitted by the mobile terminal (102-1) to a receiver outside the mobile terminal (102-1), wherein at least one model parameter for the model (204) is received by the mobile terminal (102-1) from a transmitter outside the mobile terminal (102-1), the model parameter being determined outside the mobile terminal (102-1) as a function of the measure of quality, at least one model parameter of the model (204) is replaced by at least one model parameter for the model (204). The invention relates to a device and a method for supporting a mobile terminal (102-1) in a method for operating the mobile terminal (102-1).

Description

Technical Field

[0001] The present invention relates to a device and a computer-implemented method for operating a mobile terminal device, as well as a device and a method for supporting the mobile terminal device in this method. Background Art

[0002] Traditionally, passive measurements of the characteristics of a radio connection are performed by the mobile terminal device itself and are used to predict the characteristics of the radio connection. Summary of the Invention

[0003] A computer-implemented method for operating a mobile terminal device provides for determining, in the mobile terminal device, a quantity characterizing a radio connection between the mobile terminal device and an access point of a telecommunication network, wherein in the mobile terminal device, a model configured to predict characteristics of the radio connection and / or characteristics of a radio connection between the mobile terminal device and another access point of the telecommunication network based on the determined quantity and at least one model parameter is used, and a prediction of the characteristics is determined based on the determined quantity and the at least one model parameter, wherein in the mobile terminal device, a reference quantity is determined, which reference quantity characterizes the radio connection for which the prediction is made, wherein in the mobile terminal device, a measure of the quality of the prediction is determined based on the prediction and the reference quantity, wherein the measure is sent from the mobile terminal device to a receiver outside the mobile terminal device, wherein at least one model parameter for the model is received by the mobile terminal device from a transmitter outside the mobile terminal device, the at least one model parameter for the model being determined outside the mobile terminal device based on the measure of the quality, and wherein in the mobile terminal device, at least one of the model parameters of the model is replaced by the at least one model parameter for the model. Thereby, the model is trained in a distributed manner, i.e., using a learning step performed outside the mobile terminal device. This means that in the learning step, the model parameters of the local model of the mobile terminal device are determined on a central learning unit.

[0004] For inference, it is provided that another quantity is determined in the mobile terminal device, and wherein in the mobile terminal device, another prediction is determined based on the another quantity and the model in which at least one of the model parameters of the model has been replaced by the at least one model parameter for the model, wherein the another quantity and the another prediction characterize a radio connection between the mobile terminal device and the access point and / or a radio connection between the mobile terminal device and the another access point and / or another radio connection between the mobile terminal device and other access points of the telecommunication network.

[0005] Preferably, the quantity is determined at a first time point and a first position, wherein the prediction is determined for a second time point and a second position, wherein the reference quantity is determined at the second time point and the second position, or wherein the other quantity is determined at a third time point and a third position, and wherein the other prediction is determined for a fourth time point and a fourth position.

[0006] Preferably, it is provided that in a mobile terminal device, a first estimate of a characteristic of a predefined position is determined using the model based on the quantity or the other quantity, wherein a second estimate of the characteristic of the predefined position is received in the mobile terminal device from a transmitter outside the mobile terminal device, wherein the second estimate is determined outside the mobile terminal device based on a quantity characterizing a radio connection between another mobile terminal device and an access point of a telecommunications network, and wherein the prediction is determined based on the first estimate and the second estimate. Thereby, the prediction is determined in cooperation with another mobile terminal device. Thereby, in particular, when the first mobile terminal device moves from a cell of a cell-based telecommunications network in which the first mobile terminal device is located to another cell of the telecommunications network in which the other mobile terminal device is located, a more robust prediction can be achieved.

[0007] Preferably, the first estimate is determined using a first part of the model, and wherein the prediction is determined using a second part of the model based on the first estimate and the second estimate.

[0008] Preferably, the second part of the model is configured to weight the first estimate according to a weight, wherein the weight is determined based on the timeliness of the first estimate, and the prediction is determined based on the first estimate weighted with the weight, and / or the second part of the model is configured to weight the second estimate according to a weight, wherein the weight is determined based on the timeliness of the second estimate, and the prediction is determined based on the second estimate weighted with the weight. This additionally improves the quality of the prediction.

[0009] Preferably, a third estimate of the characteristic of the predefined position is received by the mobile terminal device from a transmitter outside the mobile terminal device or another transmitter, wherein the timeliness of the third estimate is determined, and wherein if the timeliness of the third estimate is more recent than the timeliness of the second estimate, the prediction is determined based on the third estimate instead of the second estimate. This additionally improves the prediction.

[0010] The quantity preferably characterizes a radio connection in a first cell of a cell-based telecommunications network, while the other quantity characterizes a radio connection in the first cell or a second cell of the cell-based telecommunications network. Thereby, the prediction is determined for the same cell or another cell.

[0011] A computer-implemented method for supporting a mobile terminal device in a method for operating the mobile terminal device provides for receiving a measure of the predicted quality of a radio connection between the mobile terminal device and an access point of a telecommunications network, wherein the measure is determined based on a prediction of the characteristic and a reference quantity, wherein the prediction is determined in the mobile terminal device using a model having at least one model parameter, wherein at least one model parameter for the model is determined outside the mobile terminal device based on the measure and based on at least one other measure of the predicted quality of a radio connection between another mobile terminal device and an access point of the telecommunications network, and wherein at least one model parameter for the model is transmitted from a transmitter outside the mobile terminal device to the mobile terminal device. In training, the mobile terminal device is supported in distributed learning by centrally determining the model parameters.

[0012] Preferably, a quantity characterizing a radio connection between another mobile terminal device and an access point of the telecommunications network is received and transmitted to the mobile terminal device. In training, the mobile terminal device is supported in determining a reference quantity for determining a measure of quality by forwarding the quantity when determining the prediction for the location at which the quantity for the other mobile terminal device is determined.

[0013] Preferably, an estimate of the characteristics of a radio connection between another mobile terminal device and an access point of the telecommunications network at a pre-given location is received and transmitted to the mobile terminal device.

[0014] The mobile terminal device is configured to execute the method for operating the mobile terminal device.

[0015] A device for supporting a mobile terminal device in a method for operating the mobile terminal device is configured to execute the method for support.

[0016] A computer program, which includes instructions executable by a computer that, when executed by the computer, run a corresponding method; and a corresponding computer-readable medium on which the computer program is stored, the computer program and the computer-readable medium having advantages corresponding to the advantages of these methods. Description of the Drawings

[0017] Other advantageous embodiments can be learned from the following description and the drawings. In the drawings: Figure 1 A schematic diagram of a telecommunications network is shown, Figure 2 A schematic diagram of a first embodiment of a system for operating a mobile terminal device is shown, Figure 3 Steps in a method according to the first embodiment are shown, Figure 4Schematic diagram showing a second embodiment of a system for operating a mobile terminal device. Figure 5 Schematic diagram showing a third embodiment of a system for operating a mobile terminal device. Detailed implementation

[0018] In Figure 1 a telecommunications network 100 is schematically shown.

[0019] The telecommunications network 100 includes a plurality of access points, among which Figure 3 a first access point 100-1, a second access point 100-2, and a third access point 100-3 are shown.

[0020] The telecommunications network 100 is a cell-based telecommunications network 100, where in this example, a cell is defined for each access point. The first access point 100-1 provides a first cell 101-1. The second access point 100-2 provides a second cell 101-2. The third access point 100-3 provides a third cell 101-3.

[0021] In this example, these cells partially overlap each other. The first cell provides a radio connection to the services provided in the telecommunications network 100 for the mobile terminal device located in the first cell via the first access point 100-1. The second cell provides a radio connection to the service for the mobile terminal device located in the second cell via the second access point 100-2. The third cell provides a radio connection to the service for the mobile terminal device located in the third cell via the third access point 100-1. In this example, the mobile terminal device is located in one or more cells according to its respective position. In this example, the mobile terminal device is connected to one of these access points.

[0022] In this example, a first vehicle 103-1 is shown, which includes a first mobile terminal device 101-1. In this example, the first mobile terminal device 101-1 is integrated into the first vehicle 103-1. In this example, a second vehicle 103-2 is shown, which includes a second mobile terminal device 101-2. In this example, the second mobile terminal device 101-2 is integrated into the second vehicle 103-2. In this example, a third vehicle 103-3 is shown, which includes a third mobile terminal device 101-3. In this example, the third mobile terminal device 101-3 is integrated into the third vehicle 103-3.

[0023] The quality of the service available at the location of the mobile terminal device in the cell, i.e., the quality of service (QoS), is characterized, for example, by a quantity that characterizes the radio connection between the mobile terminal device and the access point of the telecommunications network 100.

[0024] For example, the quality is characterized by the bandwidth of the transmission via the radio connection at that location.

[0025] For example, the quality is characterized by an indicator of the received field strength at that location. For example, the radio connection is provided using a reference signal (RSRP) of the received field strength at the mobile terminal device, and the received field strength characterizes the quality.

[0026] For example, the quality is characterized by an indicator of the received power in the frequency channel of the radio connection. For example, the radio connection is provided using the broadband power including thermal noise (RSSI) received in the frequency channel, and the broadband power characterizes the quality.

[0027] For example, the quality is characterized by a calculated ratio value derived from the values of RSRP and RSSI. For example, the radio connection is provided using the ratio value RSRQ calculated from these values, and the ratio value characterizes the quality.

[0028] For example, RSRP, RSSI, and RSRQ are defined according to the 3rd Generation Partnership Project (3GPP) Release 15. RSRP and RSSI can be measured in the mobile terminal device. Other quantities that can be measured in the mobile terminal device and characterize QoS can also be used. Other quantities that can be measured outside the mobile terminal device and characterize QoS can also be used, and these quantities are provided to the mobile terminal device via services outside the mobile terminal device and applications (such as MobileInsight) in the mobile terminal device.

[0029] For example, MobileInsight is described in "Mobileinsight: extracting and analyzing cellular network information on smartphones (Yuanjie Li, Chunyi Peng, Zengwen Yuan, Jiayao Li, Haotian Deng, Tao Wang, MobiCom '16: Proceedings of the 22nd Annual International Conference on Mobile Computing and Networking, October 2016, pp. 202 - 215, https: / / dl.acm.org / doi / 10.1145 / 2973750.2973751)".

[0030] In this example, the first mobile terminal device 102-1 is located in the first cell 101-1. In this example, the second mobile terminal device 102-2 is located in the second cell 101-2. In this example, the third mobile terminal device 102-3 is located in the third cell 101-3.

[0031] The first mobile terminal device 102-1 is configured to execute a method for operating the first mobile terminal device 102-1.

[0032] In this example, the telecommunications network 100 includes a device 104, which is configured to support the first mobile terminal device 102-1 in this method. For example, the device 104 is connected to the first access point 100-1, the second access point 100-2, and the third access point 100-3 via an optical or wired communication connection 105.

[0033] It can be specified that the second mobile terminal device 102-2 is configured to execute a method for operating the second mobile terminal device 102-2. It can be specified that the second mobile terminal device 102-2 is configured to support the first mobile terminal device 102-1 when executing this method, while the second mobile terminal device 102-2 itself does not execute this method. It can be specified that the third mobile terminal device 102-3 is configured to execute a method for operating the third mobile terminal device 102-3. It can be specified that the third mobile terminal device 102-3 is configured to support the first mobile terminal device 102-1 when executing this method, while the third mobile terminal device 102-3 itself does not execute this method.

[0034] In the method of operating the first mobile terminal device 102-1, the first mobile terminal device 102-1 detects at least one quantity, and the at least one quantity characterizes the quality of the radio connection to the access points reachable by the first mobile terminal device 102-1. In this method, the first mobile terminal device 102-1 determines a prediction of the quality of the radio connection between the first mobile terminal device 102-1 and at least one access point of the telecommunications network 100. This access point is, for example, the access point currently used by the first mobile terminal device 102-1 for its radio connection to the telecommunications network 100, or another access point that is currently or may be used in the future.

[0035] Figure 2 Schematically shows a system 200 for operating a mobile terminal device according to a first embodiment.

[0036] The following describes the first embodiment by taking the first mobile terminal device 102-1 integrated in the first vehicle 103-1 as an example. The system 200 includes the first mobile terminal device 102-1 and the device 104.

[0037] The first mobile terminal device 102-1 includes: an interface 202 for radio connection, a model 204 for predicting the characteristics of the radio connection, and a device 206 for parameterizing the model 204. The model 204 includes at least one model parameter. The model 204 is configured to determine a prediction based on at least one model parameter and based on the quantity and the at least one model parameter. The device 206 for parameterizing the model 204 is configured to determine at least one model parameter in the model 204.

[0038] The device 104 includes an interface 208 for communicating with the first mobile terminal device 102-1 via a communication connection 210 to the interface 202 in the first mobile terminal device 102-1. The device 104 includes a device 212 for supporting the parameterization of the model 204.

[0039] In the first embodiment, the model 204 includes an artificial neural network, the weights of which are defined by the model parameters. The device 206 for parameterizing the model 204 is configured, for example, to determine at least one of the model parameters in the artificial neural network.

[0040] The device 212 for supporting the parameterization of the model 204 is configured to determine at least one model parameter in an optimization method in which the loss function is minimized using the gradient descent method. In this example, the loss function is defined based on the metric.

[0041] The device 212 for supporting the parameterization of the model 204 is configured to receive metrics from a plurality of mobile terminal devices, the plurality of mobile terminal devices being configured to determine respective metrics of the quality of their radio connections to the telecommunications network 100. The device 212 for supporting the parameterization of the model 204 is configured to minimize the loss function based on these metrics. The loss function is, for example, the average of these metrics. The device 212 for supporting the parameterization of the model 204 is configured to determine the gradient for the gradient descent method based on the loss function and to determine at least one model parameter for which the gradient is less than a threshold.

[0042] It may be provided that the device 212 for supporting the parameterization of the model 204 is configured to send at least one model parameter to the plurality of mobile terminal devices. For example, the plurality of mobile terminal devices includes a second mobile terminal device 102-2 and / or a third mobile terminal device 102-3.

[0043] The device 104 includes an output 214 for the prediction.

[0044] In Figure 3 the steps of a first embodiment of a computer-implemented method for operating the first mobile terminal device 102-1 are shown.

[0045] In step 302, a quantity characterizing the radio connection between the first mobile terminal device 102-1 and the access point of the telecommunication network 100 is determined in the first mobile terminal device 102-1.

[0046] For example, RSRP or RSSI is measured as the quantity, or RSRQ is calculated from these as the quantity. It can also be stipulated that the quantity is another quantity characterizing QoS that can be measured in the first mobile terminal device 102-1. It can also be stipulated that the quantity is another quantity characterizing QoS that is measured outside the first mobile terminal device 102-1, and the other quantity is provided to the first mobile terminal device 102-1 via services outside the mobile terminal device and an application (such as MobileInsight) in the first mobile terminal device 102-1.

[0047] In this example, the quantity characterizes the radio connection between the first mobile terminal device 102-1 and the first access point 100-1.

[0048] In this example, the quantity is determined at a first time point and at a first location. The first location is the location of the first mobile terminal device 102-1 at the first time point, and corresponds to the location of the first vehicle 103-1 at the time point when the quantity is determined.

[0049] In step 304, a prediction is determined in the first mobile terminal device 102-1 using the model 204. In this example, the prediction is determined for a second time point and a second location. The second location is the location of the first mobile terminal device 102-1 at the second time point, and corresponds to the location of the first vehicle at the second time point.

[0050] In one example, a prediction of the characteristics of the radio connection via which the first mobile terminal device 102-1 is connected to the first access point 100-1 is determined.

[0051] Alternatively or additionally, it can be stipulated that a prediction of the characteristics of the radio connection between the first mobile terminal device 102-1 and another access point of the telecommunication network 100 (such as the second access point 100-2 or the third access point 100-3) is determined.

[0052] In step 306, a reference quantity is determined in the first mobile terminal device 102-1, and the reference quantity characterizes the radio connection for which the prediction is determined. In this example, the reference quantity characterizes, for example, the same characteristic as the quantity, that is, the quantity and the reference quantity are, for example, both bandwidth, RSRP or RSSI or RSRQ, or one of the other quantities measured or provided.

[0053] In this example, a reference quantity is determined at a second time point and at a second location. In this example, it may be stipulated that the second time point or the second location when determining the reference quantity coincides precisely or at least substantially with the second time point or the second location for determining the prediction. It may be stipulated that the second time point or the second location when determining the reference quantity deviates from the second time point or the second location for determining the prediction by a given tolerance.

[0054] For example, the time point and the location for determining the reference quantity are determined according to the planned route of the first vehicle provided with the first mobile terminal device 102-1. For example, this time point is an exact absolute time point or a comparable time, such as the same moment on another day or the same day of another year. The reference quantity may also be determined at other times within a pre-given time period including this time point.

[0055] It may be stipulated that a quantity characterizing the radio connection between another mobile terminal device and its access point to the telecommunication network 100 is received, and the reference quantity is determined based on this quantity.

[0056] It may be stipulated that the telecommunication network 100, in particular the device 104, receives this quantity from another mobile terminal device and sends it to the mobile terminal device 102-1.

[0057] In step 308, a measure of the quality of the prediction is determined in the first mobile terminal device 102-1 based on the prediction and the reference quantity.

[0058] For example, a deviation is determined, in particular the difference between the prediction and the reference quantity.

[0059] In step 310, this measure is sent from the first mobile terminal device 102-1 to a receiver outside the first mobile terminal device 102-1.

[0060] In this example, the receiver is the device 104. In this example, this measure is transmitted from the interface 202 in the first mobile terminal device 102-1 to the interface 208 in the device 104.

[0061] It may be stipulated that this measure is received from a plurality of mobile terminal devices configured to determine respective measures of the quality of their radio connections to the telecommunication network 100.

[0062] In step 312, at least one model parameter among the model parameters is determined outside the first mobile terminal device 102-1.

[0063] In this example, at least one model parameter is determined by the device 104.

[0064] At least one model parameter is determined based on the measure received from the first mobile terminal device 102-1.

[0065] In this example, at least one model parameter is determined in an optimization method in which the gradient descent method is used to minimize a loss function. In this example, the loss function is defined based on metrics received from the first mobile terminal device 102-1.

[0066] It may be specified that the loss function is defined based on multiple metrics. For example, the loss function is the average of these metrics. For example, the gradient for the gradient descent method is determined based on the loss function. In this example, at least one of the following model parameters is determined for which the gradient is less than a threshold.

[0067] In step 314, the first mobile terminal device 102-1 receives at least one model parameter for the model 204. This means that the first mobile terminal device 102-1 receives the model parameter from a transmitter outside the first mobile terminal device 102-1.

[0068] It may be specified that the device 212 for supporting the parameterization of the model 204 is configured to determine at least one model parameter and send it to a plurality of mobile terminal devices including the first mobile terminal device 102-1.

[0069] In this example, the transmitter is the device 104. In this example, at least one model parameter is transmitted from the interface 208 in the device 104 to the interface 202 in the first mobile terminal device 102-1.

[0070] In step 316, in the first mobile terminal device 102-1, at least one of the model parameters of the model 204 is replaced with at least one of the model parameters of the model 204.

[0071] Through steps 302 to 316, the model 204 is trained. These steps can be repeatedly executed to further train the model 204.

[0072] It may be specified that the method then ends.

[0073] In this example, it is specified to use the model 204.

[0074] In step 318, another quantity is determined in the first mobile terminal device 102-1. In one example, this other quantity characterizes the radio connection between the first mobile terminal device 102-1 and the first access point 100-1. It may be specified that this other quantity alternatively characterizes the radio connection between the first mobile terminal device 102-1 and another access point (such as the second access point 100-2 or the third access point 100-3).

[0075] In step 320, another prediction is determined in the first mobile terminal device 102-2 based on this other quantity and the model 204.

[0076] In model 204, at least one model parameter for model 204 replaces at least one model parameter of model 204.

[0077] In this example, the other prediction characterizes the radio connection between the first mobile terminal device 102-1 and the first access point 100-1. It may be provided that the other quantity characterizes the radio connection between the first mobile terminal device 102-1 and another access point (such as the second access point 100-2 or the third access point 100-3).

[0078] In this example, the other quantity is determined at a third time point and at a third location. In this example, the other prediction is determined for a fourth time point and a fourth location.

[0079] In this example, the quantity characterizes the radio connection in the first cell 101-1. In this example, the other quantity characterizes the radio connection in the first cell 101-1. It may also be provided that the other quantity characterizes the radio connection in the second cell 101-2 or the third cell 101-3.

[0080] For example, device 104 is a central server. It may be provided that device 104 is configured to provide a part of the server for each cell.

[0081] Figure 4 Schematically shown is a system 200 for operating a mobile terminal device according to a second embodiment. The elements included in both embodiments are in Figure 4 denoted by the same reference numerals as in Figure 2 the same.

[0082] The second embodiment is particularly suitable for using information about the radio connection to an access point of the telecommunications network 100 that cannot be directly reached by the first mobile terminal device 102-1 at the time point or location at which the prediction is determined. The corresponding method is also configured accordingly.

[0083] For example, the first mobile terminal device 102-1 is located in the first cell 101-1 outside the second cell 101-2. For example, the second mobile terminal device 102-2 is located in the second cell 101-2 and is able to evaluate the radio connection of its instantaneous position to the second access point 100-2.

[0084] Different from the first embodiment, the model 204 according to the second embodiment includes a first part 216 for determining a first estimate 218 of the characteristics of a pre-given position based on a quantity characterizing the radio connection between the first mobile terminal device 102-1 and an access point of the telecommunications network 100.

[0085] Different from the first embodiment, the model 204 according to the second embodiment includes a second part 220 for determining a prediction based on the first estimate 218 and a second estimate 222 of the characteristics of a pre-given location received from a transmitter outside the first mobile terminal device 102-1 (e.g., via the interface 202).

[0086] For example, the second mobile terminal device 102-2 is configured to determine the second estimate 222. For example, the second mobile terminal device 102-2 is configured to provide the second estimate 222 to other mobile terminal devices in the telecommunications network 100. For example, the telecommunications network 100 is configured to receive the second estimate 222 from the second mobile terminal device 102-2 and forward it to the first mobile terminal device 102-1. For example, the device 104 is configured to receive and forward the second estimate 222 of the pre-given location. For example, the telecommunications network 100 is configured to send the second estimate 222 required for the prediction of the location. For example, the interface 202 is configured to send a query for the second estimate of the pre-given location to the telecommunications network 100.

[0087] For example, the second part 220 of the model 204 is configured to weight the first estimate 218 according to a weight. In one embodiment, the model 204 is configured to determine the weight according to the timeliness of the first estimate. For example, the second part 220 is configured to determine the prediction based on the first estimate 218 weighted by the weight.

[0088] For example, the second part 220 of the model 204 is configured to weight the second estimate 222 according to a weight. In one embodiment, the model 204 is configured to determine the weight according to the timeliness of the second estimate 222. For example, the second part 220 is configured to determine the prediction based on the second estimate 222 weighted by the weight.

[0089] For example, the first part 216 includes an artificial neural network that is configured to determine the first estimate 218.

[0090] For example, the second part 220 includes an artificial neural network that is configured to map the first estimate 218 and the second estimate 220 to a prediction.

[0091] For example, the device 206 for parameterizing the model 204 is configured to determine at least one of the model parameters in the first and / or second artificial neural networks.

[0092] The method for operating the first mobile terminal device 102-1 according to the second embodiment differs from the method according to the first embodiment in that, in step 304, a first estimate 218 is determined in the first mobile terminal device 102-1 for a predefined location, a second estimate 222 of the predefined location is received from outside the first mobile terminal device 102-1, and a prediction is determined in the first mobile terminal device 102-1 based on the two estimates of the predefined location.

[0093] Step 306 according to the second embodiment differs from step 306 according to the first embodiment in that a reference quantity is determined at or substantially at the predefined location.

[0094] Step 320 according to the second embodiment differs from step 320 according to the first embodiment in that a first estimate 218 is determined in the first mobile terminal device 102-1 for a predefined location, a second estimate 222 of the predefined location is received from outside the first mobile terminal device 102-1, and a prediction is determined in the first mobile terminal device 102-1 based on the two estimates of the predefined location.

[0095] For example, the second part 220 of the model 204 weights the first estimate 218 according to the weight of the second estimate 218. For example, the second part 220 determines this weight according to the timeliness of the first estimate 218. This means that the prediction is determined based on the first estimate 218 weighted by this weight.

[0096] For example, the second part 220 of the model 204 weights the second estimate 222 according to the weight of the second estimate 218. For example, the second part 220 determines this weight according to the timeliness of the second estimate 222. This means that the prediction is determined based on the second estimate 222 weighted by this weight.

[0097] It can be provided that the second part 220 of the model 204, in particular the artificial neural network representing this part, is configured to map a plurality of estimates received from outside the first mobile terminal device 102-1 (e.g., via the interface 202), weighted or unweighted as described for the second estimate 222, together with the first estimate 218 and the second estimate 222 to the prediction.

[0098] In one embodiment, the telecommunication network 100, in particular the device 104, is configured to detect the prediction of the mobile terminal device and jointly determine the location of the prediction, and provide it as the second estimate 222 to another mobile terminal device. In one embodiment, it is provided that the prediction is stored together with the identification of the cell for which the prediction is determined in a central database in particular, and the prediction containing this identification is provided as the second estimate 222 when queried by the mobile terminal device.

[0099] The prediction, i.e., the second estimate 222, is associated with the respective pre-given position or the respective cell. In a second embodiment, the prediction, i.e., the second estimate 222, can be time-independent.

[0100] In Figure 5 FIG. schematically shows a system 200 according to a third embodiment.

[0101] In the third embodiment, it is provided that the prediction, i.e., the second estimate 222, is associated with the time point at which the prediction is determined. The elements included in the second and third embodiments are denoted in Figure 5 by the same reference numerals as in Figure 4 FIG.

[0102] The third embodiment is particularly suitable for using the most up-to-date information possible regarding the radio connection to the access point of the telecommunications network 100. The corresponding method is also constituted accordingly.

[0103] According to the third embodiment, the device 104 includes a device 224 for selecting an estimate for determining the prediction. The device 224 for selecting an estimate is configured to determine the time point associated with the second estimate 222. In this example, the interface 202 is configured to receive the second estimate 222 and this time point. For example, the telecommunications network 100 is configured to send the second estimate 222 and its time point. For example, the interface 202 is configured to send a query for the second estimate and its time point to the telecommunications network 100.

[0104] The device 224 for selecting an estimate is configured to compare the current time point or the time point at which the first estimate 216 is valid with the received time point, and if the second estimate 222 is the same as or more up-to-date than the first estimate 216 in terms of timeliness, then select the second estimate 222 for determining the prediction. Otherwise, the device 224 for selecting an estimate is configured not to select the second estimate 222 for determining the prediction.

[0105] In an example where multiple estimates are received from outside the first mobile terminal device 102-1, it can be provided that the device 224 is configured to select one or more estimates for determining the prediction. For example, select the most up-to-date estimate, or select an estimate that is more up-to-date than the first estimate 216.

[0106] For example, if at time point t0, a bandwidth of 10 Mbit is determined as an estimate by the second mobile terminal device 102-2 at location X, and at time point t0 + x ms, a bandwidth of 17 Mbit is determined as an estimate by the third mobile terminal device 102-3 at location X, and then the first mobile terminal device 102-1 arrives at location X, then the first mobile terminal device 102-1 uses the estimate from the third mobile terminal device 102-3 as the most up-to-date estimate.

[0107] It may also be stipulated that if the time interval between these two estimates is less than a threshold value, these two estimates are used.

Claims

1. A computer-implemented method for operating a mobile terminal device (102-1), characterized in that, In the mobile terminal device (102-1), a quantity characterizing a radio connection between the mobile terminal device (102-1) and an access point (100-1) of a telecommunication network (100) is determined (302), wherein in the mobile terminal device (102-1), a model (204) configured to predict characteristics of the radio connection and / or characteristics of a radio connection between the mobile terminal device (102-1) and another access point (100-2, 100-3) of the telecommunication network based on the determined quantity and at least one model parameter is utilized, and a prediction of the characteristics is determined (304) based on the determined quantity and the at least one model parameter, wherein in the mobile terminal device (102-1), a reference quantity is determined (306), the reference quantity characterizing the radio connection for which the prediction is determined, wherein in the mobile terminal device (102-1), a measure of the quality of the prediction is determined (308) based on the prediction and the reference quantity, wherein the measure is sent (310) by the mobile terminal device (102-1) to a receiver outside the mobile terminal device (102-1), wherein the mobile terminal device (102-1) receives (314) from a transmitter outside the mobile terminal device (102-1) at least one model parameter for the model (204), the at least one model parameter for the model being determined outside the mobile terminal device (102-1) based on the measure of the quality, and wherein in the mobile terminal device (102-1), at least one model parameter of the model (204) is replaced (316) with the at least one model parameter for the model (204).

2. The method according to claim 1, characterized in that, In the mobile terminal device (102-1), another quantity is determined (318), and wherein in the mobile terminal device (102-1), another prediction is determined (320) based on the another quantity and the model (204) with at least one model parameter of the model (204) replaced by the at least one model parameter for the model (204), wherein the another quantity and the another prediction characterize a radio connection between the mobile terminal device (102-1) and the access point (100-1) and / or a radio connection between the mobile terminal device (102-1) and the another access point (100-2) and / or another radio connection between the mobile terminal device (102-1) and other access points (100-3) of the telecommunication network (100).

3. The method according to claim 1 or 2, characterized in that The quantity is determined (302) at a first time point and at a first location, wherein the prediction is determined (304) for a second time point and a second location, wherein the reference quantity is determined (306) at the second time point and at the second location, or wherein the another quantity is determined (318) at a third time point and at a third location, and wherein the another prediction is determined (320) for a fourth time point and a fourth location.

4. The method according to any one of claims 1 to 3, characterized in that, In the mobile terminal device (102-1), a first estimate (218) of a characteristic of a predefined location is determined using the model (204) based on the quantity or the other quantity, wherein a second estimate (222) of the characteristic of the predefined location is received in the mobile terminal device (102-1) from a transmitter outside the mobile terminal device (102-1), wherein the second estimate (222) is determined outside the mobile terminal device (102-1) based on a quantity characterizing a radio connection between another mobile terminal device (102-2) and an access point (100-1) of the telecommunication network (100), and wherein the prediction is determined based on the first estimate (216) and the second estimate (222).

5. The method according to claim 4, characterized in that The first estimate (218) is determined using a first part (216) of the model (204), and the prediction is determined using a second part (220) of the model (204) based on the first estimate and the second estimate.

6. The method according to claim 5, wherein The second part (220) of the model (204) is configured to weight the first estimate (218) according to a weight, wherein the weight is determined based on the timeliness of the first estimate (218), and the prediction is determined based on the first estimate (218) weighted by the weight, and / or the second part (220) of the model (204) is configured to weight the second estimate (222) according to a weight, wherein the weight is determined based on the timeliness of the second estimate (222), and the prediction is determined based on the second estimate (222) weighted by the weight.

7. The method according to claim 6, wherein A third estimate of the characteristic of the predefined location is received by the mobile terminal device (102-1) from a transmitter outside the mobile terminal device (102-1) or another transmitter, wherein the timeliness of the third estimate is determined, and if the timeliness of the third estimate is newer than the timeliness of the second estimate (222), the prediction is determined based on the third estimate instead of the second estimate (222).

8. The method according to any one of the above claims, characterized in that, The quantity characterizes a radio connection in a first cell of the cell-based telecommunication network (100), and the other quantity characterizes a radio connection in the first cell (101-1) or the second cell (101-2) of the cell-based telecommunication network (100).

9. A computer-implemented method for supporting a mobile terminal device (102-1) in a method of operating the mobile terminal device (102-1), characterized in that, Receive a measure of the quality of a prediction of characteristics of a radio connection between a mobile terminal device (102-1) and an access point of a telecommunications network (100), wherein the measure is determined based on the prediction of the characteristics and a reference quantity of the characteristics, wherein the prediction is determined in the mobile terminal device (102-1) using a model (204) having at least one model parameter, and wherein at least one model parameter for the model (204) is determined outside the mobile terminal device (102-1) based on the measure and at least one other measure of the quality of a prediction of characteristics of a radio connection between another mobile terminal device (102-2) and its access point to the telecommunications network, and wherein the at least one model parameter for the model (204) is sent from a transmitter outside the mobile terminal device (102-1) to the mobile terminal device (102-1).

10. The method according to claim 9, characterized in that, Receive a quantity characterizing a radio connection between another mobile terminal device (102-2) and its access point to the telecommunications network (100), and send it to the mobile terminal device (102-1).

11. The method according to claim 8 or 9, characterized in that, Receive an estimate of characteristics of a radio connection between another mobile terminal device and its access point to the telecommunications network (100) at a pre-given location, and send it to the mobile terminal device (102-1).

12. A mobile terminal device (102-1), characterized in that, The mobile terminal device (102-1) is configured to perform the method according to any one of claims 1 to 8.

13. An apparatus (104) for supporting a mobile terminal device (102-1) in a method of operating the mobile terminal device (102-1), characterized in that, The apparatus is configured to perform the method according to any one of claims 9 to 11.

14. A computer program, characterized in that, The computer program includes computer-executable instructions that, when executed by a computer, will run the method according to any one of claims 1 to 8 or 9 to 11.

15. A computer-readable medium having stored thereon the computer program according to claim 14.

Images