Method and apparatus for remote estimation of physical antenna configuration

Abstract

Computer-implemented method, comprising receiving a plurality of measurement results each comprising an indication of a respective distance between a first antenna and a location where the respective measurement is performed and an indication of a respective value of a signal strength received from the first antenna at the location where the respective measurement is performed; estimating, for each of the measurement results, based on the respective distance and the respective value of the received signal strength, a respective preliminarily estimated physical configuration of the first antenna; averaging the preliminarily estimated physical configurations for the plurality of measurement results to obtain an estimated physical configuration of the first antenna, wherein the physical configuration comprises at least one of a direction of the first antenna, a downtilt of the first antenna, and a height of the first antenna.

Description

[0001] METHOD AND APPARATUS FOR REMOTE ESTIMATION OF PHYSICAL ANTENNA CONFIGURATION

[0002] TECHNICAL FIELD

[0003] The present disclosure relates to determining a physical configuration of an antenna from remote.

[0004] BACKGROUND

[0005] At several occasions, it may be necessary to determine the actual physical antenna configuration, such as a tilt of the antenna, a direction (azimuth) of the antenna, and a height of the antenna. For example, the physical antenna configuration may be determined when the antenna is just built or after some work on the antenna had been performed. Another occasion is after very heavy weather conditions, such as a strong storm, which might have changed the physical antenna configuration. Conventionally, in such cases the site of the antenna has to be visited. To get the most accurate results, a technician may climb the mast or get to the rooftop. There, the technician may determine the physical antenna configuration. Slightly less accurate, one may determine the physical configuration of the antenna from ground level.

[0006] SUMMARY

[0007] It is an object to improve the prior art.

[0008] According to a first aspect, there is provided a computer-implemented method, comprising receiving a plurality of measurement results each comprising an indication of a respective distance between a first antenna and a location where the respective measurement is performed and an indication of a respective value of a signal strength received from the first antenna at the location where the respective measurement is performed; estimating, for each of the measurement results, based on the respective distance and the respective value of the received signal strength, a respective preliminarily estimated physical configuration of the first antenna; averaging the preliminarily estimated physical configurations for the plurality of measurement results to obtain an estimated physical configuration of the first antenna, wherein, if the first antenna is a directional antenna, the physical configuration comprises at least one of a direction of the first antenna, a downtilt of the first antenna, and a height of the first antenna; and if the first antenna is an omnidirectional antenna, the physical configuration comprises at least one of the downtilt of the first antenna, and the height of the first antenna.

[0009] For at least one of the measurement results, the indication of the respective distance may comprise an indication of the location where the respective measurement is performed, and the method may further comprise calculating, for the at least one of the measurement results, the distance between the first antenna and the location where the respective measurement is performed based on the indication of the location where the respective measurement is performed and a location of the first antenna.

[0010] The method may further comprise receiving, for each of the measurement results, an indication of a respective transmit power of a signal transmitted by the first antenna, wherein, for each of the measurement results the signal strength may be based on the signal transmitted with the respective transmit power; wherein, for each of the measurement results, the respective preliminarily estimated physical configuration of the first antenna may be estimated additionally based on the respective transmit power.

[0011] For each of the measurement results, the respective preliminarily estimated physical configuration of the first antenna may be estimated using a trained neural network. The method may further comprise training the neural network based on a set of plural training data for one or more second antennas, wherein for each of the one or more second antennas, each of the training data comprises an indication of a respective value of the signal strength received from the respective second antenna, an indication of a respective distance between the respective second antenna and a location where the respective value of the signal strength is received from the respective second antenna, the downtilt of the respective second antenna at the time when the respective value of the signal strength is received, the height of the respective second antenna at the time when the respective value of the signal strength is received, and, if the respective second antenna is a directional antenna, the direction of the respective second antenna at a time when the respective value of the signal strength is received.

[0012] The set of training data may comprise training data for more than one second antenna; the set of training data may comprise, for each of the second antennas, an indication of a directional characteristics of the respective second antenna; and the method may comprise receiving a directional characteristics of the first antenna; checking, for each of the second antennas, whether the directional characteristics of the first antenna differs from the directional characteristics of the respective second antenna by more than a predefined threshold value; excluding, for each of the second antennas, the plural training data for the respective second antenna from the training if the directional characteristics of the first antenna differs from the directional characteristics of the respective second antenna by more than the predefined threshold value.

[0013] The method may further comprise, for each of the one or more second antennas: if the respective second antenna is a directional antenna, determining the direction of the respective second antenna; determining the downtilt of the respective second antenna; determining the height of the respective second antenna, and for each of the training data for the respective second antenna: measuring the respective value of the signal strength received from the respective second antenna at the location where the respective value of the signal strength is received from the respective second antenna; determining the location where the respective value of the signal strength received from the respective second antenna is measured.

[0014] The preliminarily estimated configurations of the first antenna for the plurality of measurement results may be averaged by calculating an arithmetic mean value of the preliminarily estimated configurations of the first antenna.

[0015] The method may further comprise, for each of the measurement results, measuring the respective value of the signal strength received from the first antenna at the location where the respective measurement is performed; determining the location where the respective value of the received signal strength is measured.

[0016] According to a second aspect, there is provided an apparatus, comprising first means for receiving a plurality of measurement results each comprising an indication of a respective distance between a first antenna and a location where the respective measurement is performed and an indication of a respective value of a signal strength received from the first antenna at the location where the respective measurement is performed; means for estimating, for each of the measurement results, based on the respective distance and the respective value of the received signal strength, a respective preliminarily estimated physical configuration of the first antenna; means for averaging the preliminarily estimated physical configurations for the plurality of measurement results to obtain an estimated physical configuration of the first antenna, wherein, if the first antenna is a directional antenna, the physical configuration comprises at least one of a direction of the first antenna, a downtilt of the first antenna, and a height of the first antenna; and if the first antenna is an omnidirectional antenna, the physical configuration comprises at least one of the downtilt of the first antenna, and the height of the first antenna. For at least one of the measurement results, the indication of the respective distance may comprise an indication of the location where the respective measurement is performed, and the apparatus may further comprise means for calculating, for the at least one of the measurement results, the distance between the first antenna and the location where the respective measurement is performed based on the indication of the location where the respective measurement is performed and a location of the first antenna.

[0017] The apparatus may further comprise second means for receiving, for each of the measurement results, an indication of a respective transmit power of a signal transmitted by the first antenna, wherein, for each of the measurement results the signal strength may be based on the signal transmitted with the respective transmit power; and the means for estimating may be configured to estimate, for each of the measurement results, the respective preliminarily estimated physical configuration of the first antenna additionally based on the respective transmit power.

[0018] The means for estimating may be configured to estimate, for each of the measurement results, the respective preliminarily estimated physical configuration of the first antenna using a trained neural network.

[0019] The apparatus may further comprise means for training the neural network based on a set of plural training data for one or more second antennas, wherein for each of the one or more second antennas, each of the training data comprises an indication of a respective value of the signal strength received from the respective second antenna, an indication of a respective distance between the respective second antenna and a location where the respective value of the signal strength is received from the respective second antenna, the downtilt of the respective second antenna at the time when the respective value of the signal strength is received, the height of the respective second antenna at the time when the respective value of the signal strength is received, and, if the respective second antenna is a directional antenna, the direction of the respective second antenna at a time when the respective value of the signal strength is received.

[0020] The set of training data may comprise training data for more than one second antenna; the set of training data may comprise, for each of the second antennas, an indication of a directional characteristics of the respective second antenna; and the apparatus may comprise third means for receiving a directional characteristics of the first antenna; means for checking, for each of the second antennas, whether the directional characteristics of the first antenna differs from the directional characteristics of the respective second antenna by more than a predefined threshold value; means for excluding, for each of the second antennas, the plural training data for the respective second antenna from the training if the directional characteristics of the first antenna differs from the directional characteristics of the respective second antenna by more than the predefined threshold value.

[0021] The means for averaging may be configured to average the preliminarily estimated configurations of the first antenna for the plurality of measurement results by calculating an arithmetic mean value of the preliminarily estimated configurations of the first antenna.

[0022] According to a third aspect, there is provided a computer program product comprising a set of instructions which, when executed on an apparatus, is configured to cause the apparatus to carry out the method according to the first aspect. The computer program product may be embodied as a computer-readable medium or directly loadable into a computer.

[0023] According to some embodiments, it is not necessary to visit the antenna site in order to determine the physical configuration of the antenna. Thus, time and energy may be saved.

[0024] BRIEF DESCRIPTION OF THE FIGURES Further details, features, objects, and advantages are apparent from the following detailed description of preferred embodiments, which is to be taken in conjunction with the appended drawings, wherein:

[0025] Fig. 1 illustrates a method according to some embodiments;

[0026] Fig. 2 illustrates preliminarily estimated downtilt of an antenna vs. its real downtilt;

[0027] Fig. 3 illustrates preliminarily estimated azimuths of an antenna vs. its real azimuths;

[0028] Fig. 4 illustrates preliminarily estimated heights of the antenna vs. its real heights;

[0029] Fig. 5 shows an apparatus according to an embodiment;

[0030] Fig. 6 shows a method according to an embodiment;

[0031] Fig. 7 shows an apparatus according to an embodiment.

[0032] DETAILED DESCRIPTION OF CERTAIN EMBODIMENTS

[0033] Herein below, certain embodiments are described in detail with reference to the accompanying drawings, wherein the features of the embodiments can be freely combined with each other, unless otherwise described. However, it is to be expressly understood that the description of certain embodiments is given by way of example only, and that is by no way intended to be understood as limiting the disclosure of the disclosed details.

[0034] Moreover, it is to be understood that the apparatus is configured to perform the corresponding method, although in some cases only the apparatus or only the method are described. The signal strength received by a user equipment (UE) from an antenna depends on the radio channel conditions. For the signal strength received from the same antenna at different locations, the channel conditions may be different due to different distances, fast and slow fading, knife edge diffraction, attenuation by vegetation and multipath effect, for example. Therefore, it is not possible to determine a physical configuration of the antenna from a single measurement of a received signal strength and the location of the UE, when the location of the antenna is known.

[0035] However, if a lot of measurements are performed at different locations, it is possible to estimate the physical configuration of the antenna. Namely, in this case, all the samples may be considered as coming from one single stochastic process. Therefore, averaging will smooth out the different radio channel conditions and one may obtain an accurate estimation of the physical configuration of the antenna, such as the antenna's direction (azimuth), the antenna's height, and / or the antenna's tilt. The distribution of the samples can be used for further analysis.

[0036] The measurements may be performed with special equipment and / or with conventional user equipment. For example, 3GPP has specified that the UE has to measure the signal strength of some downlink reference signals and to report the measurement result to the network. As another example, 3GPP has specified the feature "Minimization of Drive Tests" (MDT). By means of MDT, the UE may measure the signal strength at certain locations and report it back to the network. For the signal to be measured, one may typically use downlink reference signals, such as cell-specific reference signals or UE-specific reference signals (e.g. Demodulation Reference Signal, DMRS). However, other reference signals or nonreference signals may be used as long as the transmit power of the respective reference signal (or signal) is known.

[0037] Fig. 1 illustrates a method according to an embodiment. The method comprises a training phase (actions 1 to 3), and an inference phase (actions 4 to 8). The method may be performed with any regression model such as a neural network. In one example, a decision tree regression is used. For the training phase, drive test samples are collected. The drive test samples comprise for one antenna, an indication of the distance between the measurement device and the antenna, and the power (signal strength) of the specified (reference) signal received from the antenna by the measurement device. The measurement device may be a UE or a dedicated measurement device. Furthermore, each of the training data comprises the physical configuration of the antenna (antenna height, antenna tilt (typically expressed as downtilt), and the bearing from the measurement point to the antenna (unless the antenna is an omnidirectional antenna). Downtilt means total antenna tilt value (e.g. angle) that combines mechanical tilt value and electronic tilt (e-tilt) value. Typically mechanical tilt value is set (and fixed) when the antenna is installed or reconfigured by service personnel. Electronic tilt (e-tilt) value may be adjusted remotely (remote tilt) and even automatically by service provider automation system, for example.

[0038] Table 1: Training data samples (measurement results used for training)

[0039] Table 1 shows an example of training data samples. Each of the samples comprises the latitude and the longitude of the position of the measurement device (e.g. UE). Thus, the position of the measurement device is determined. Furthermore, each sample comprises the cell ID, which is an identifier of the antenna, for which the physical configuration is estimated. The column "rs power" indicates the received signal strength of the reference signal used for the measurement. Finally, each of the training data samples comprises the measurement time.

[0040] For the training data set, the physical configuration (azimuth (in case of a directional antenna), height, and downtilt) is known. Also, the R.F frequency, on which the reference signal is transmitted, is known.

[0041] In some examples, the drive test samples may comprise only the location of the measurement device when measuring the received power (signal strength). In this case, in action 2, the drive test samples may be processed. Namely, from the location of the measurement device and the location of the antenna, the distance between the measurement device and the antenna at the time of measuring the received signal strength may be determined. Furthermore, from the known azimuth of the antenna and the position of the measurement device, one may determine the bearing to the antenna.

[0042] Table 2: Processed training data samples (In Table 2, instead of downtilt, e-tilt is indicated because it is assumed that the mechanical tilt is not modified during the present measurement; thus, downtilt may be derived from e-tilt by a fixed offset)

[0043] Table 2 shows the training data after being processed in action 2 of Fig. 1. As may be seen in Table 2, the distance to the antenna was calculated based on the latitude and longitude of the measurement device and the known position of the antenna. Also, the bearing to the antenna was calculated based on the position and the azimuth of the antenna and the position of the measurement device. Furthermore, the height, transmission frequency and downtilt of the antenna were added to each sample. The bearing is the forward azimuth between the coordinates of the antenna location and the measurement device.

[0044] Using these data, the regression model may be trained. Typically, the available training data are split into two parts, one part for training the model, and another part for validation of the trained model. However, in some embodiments, validation may be omitted.

[0045] Once the regression model is trained, it may be used for a lot of different measurement campaigns, each providing a lot of measurement results, without performing the training again.

[0046] The trained (and potentially validated) regression model may be used in the inference phase to estimate the actual physical configuration of an antenna. As shown in Fig. 1, in action 4, new drive test samples may arrive. As already discussed for the action 1, these drive test samples may come from dedicated measurement equipment or from conventional UEs, for example by the MDT feature. In action 5, the new drive test samples may be processed similarly as described for the action 2.

[0047] The potentially processed drive test data may then, in action 6, be input into the trained regression model. Thus, for each of the samples, the trained regression model may preliminarily estimate a value for the physical configuration of the antenna (action 7). As explained hereinabove, this preliminary estimation might differ significantly from the actual physical configuration of the antenna. Therefore, in action 8, the preliminary estimations are averaged. This average (denoted "E(N samples)") provides a good estimation of the actual physical configuration of the antenna because the different channel conditions are basically averaged out.

[0048] The averaging in action 8 may be simply calculating a mean value of the respective parameter of the physical configuration of the antenna obtained in the preliminary estimations, without any weighting. However, in some embodiments, the preliminary estimations may be weighted. For example, measurements very close to the antenna locations depend strongly on the distance to the antenna. That is, small deviations in the determination of the location may result in very different preliminary estimations. Therefore, in some embodiments, measurements obtained very close to the antenna may be weighted with a lower weight.

[0049] On the other hand, measurements obtained very far away from the antenna may suffer from a low signal strength and interference from other cells. Therefore, these measurement results of the signal strength might not be too reliable. Therefore, these measurements may be weighted with a lower weight. However, other weighting or other ways of averaging than discussed here are feasible, too.

[0050] In some embodiments, the training data and the measurement data are obtained for the same antenna. However, this is not necessary. For example, the training data may be obtained for one antenna of a certain type, and the measurement data may be obtained for another antenna of the same type. The type of the antenna may define, for example, the radiation characteristics of the antenna. In some embodiments, the same type may even mean the same model of a certain manufacturer.

[0051] The training data may be obtained for more than one antenna. If the antennas have different directional characteristics (e.g. some antennas are omnidirectional antennas and some antennas are directional antennas, or directional antennas have different directional characteristics) the training data may comprise an indication of the respective directional characteristics. In some embodiments, training data from an antenna having a directional characteristics which differs from that of the antenna for which the measurement data are obtained by more than a threshold may be excluded from the training.

[0052] The apparatus performing the estimation may receive the directional characteristics of the antenna for which the measurement data are obtained either directly (e.g. by operator's input), from the respective base station, or the directional characteristics of the antenna may be included in the measurement data such that the apparatus performing the estimation receives the directional characteristics of the antenna within the measurement data.

[0053] However, in many 4G and 5G networks, antennas of a same frequency band have the same or similar directional characteristics. Therefore, in some example embodiments, it is sufficient to restrict the measurements for both the training data and the measurement data to a certain frequency band, not taking into account specific directional characteristics of the antenna.

[0054] The number of the samples in the training data set is preferably very large. Namely, the number of the samples in the training data set may be larger than 100, preferably larger than 500, more preferably larger than 1,000, still more preferably larger than 3,000 and even more preferably larger than 5,000. For the measurement data, preferably, the number of data is as large as for the training data set. However, in some embodiments, the number of training data may be different from the number of measurement data. For example, the number of measurement samples may be just 10 or more, more preferably 50 or more, still more preferably 100 or more and even more preferably 500 or more. That is, for practical reasons, the number of measurement samples may be lower than the number of training samples.

[0055] Figs. 2 to 4 show preliminarily estimated values of the parameters of the physical configuration of the antenna. Namely, Fig. 2 shows different preliminary estimations for the downtilt vs. the actual (real) downtilt; Fig. 3 shows preliminary estimations of the azimuth vs. the actual (real) azimuth, and Fig. 4 shows preliminary estimations of the height of the antenna vs. the actual (real) height of the antenna. Each of the points in Figs. 2 to 4 is obtained from a single sample (measurement) of the received signal strength, and the location of the measurement device. As can be seen from Figs. 2 to 4, the preliminary estimations vary a lot such that from a preliminary estimation, a reliable value for the actual configuration cannot be obtained. However, as may also be seen from these figures, the preliminarily estimated values have a tendency to increase with the real values. Therefore, from averaging the preliminary estimations of the parameters, one may obtain a reliable estimation of the actual parameter. In Figs. 2 to 4, only one parameter (tilt, azimuth, or height) of the antenna configuration is estimated, and the other parameters are assumed to be fixed (determined on-site, or as planned). However, if the set of training data and the measurement data comprise a sufficient large number of data, two or even three of these parameters may be estimated simultaneously, without fixing them.

[0056] In some embodiments, it is assumed that cells of the same frequency band have very similar or the same transmit powers. This is usual in many 4G and 5G networks. Therefore, in some embodiments, the transmit power of the signal by the antenna may be ignored as a parameter.

[0057] However, in some embodiments, the training data and / or the measurement data may comprise an indication of the transmit power. In this case, for each of the measurement data and / or training data, the received signal strength may be normalized by the respective transmit power.

[0058] The method increases the chances to detect installation errors or mechanical failures caused by extreme weather conditions. From a comparison of antennas of the same type, the method may also reveal a poor manufacturing quality or even a faulty antenna element.

[0059] The method may be quite precise. While conventional methods just may analyze whether an antenna points to a completely wrong direction (e.g. deviation larger than 120°) some embodiments may estimate the direction within ±5° or less or within ±10° or less. Correspondingly, in some embodiments, the tilt may be estimated quite precisely within a few degrees (e.g. within ±5° or less or within ±10° or less).

[0060] Fig. 5 shows an apparatus according to an embodiment. The apparatus may be a computer or an element thereof. Fig. 6 shows a method according to an embodiment. The apparatus according to Fig. 5 may perform the method of Fig. 6 but is not limited to this method. The method of Fig. 6 may be performed by the apparatus of Fig. 5, but is not limited to be performed by this apparatus. The apparatus comprises means for receiving 110, means for estimating 120 and means for averaging 130. The means for receiving 110, means for estimating 120, and means for averaging 130 may be a receiving means, estimating means, and averaging means, respectively. The means for receiving 110, means for estimating 120, and means for averaging 130 may be a receiver, an estimator, and an averager, respectively. The means for receiving 110, means for estimating 120, and means for averaging 130 may be a receiving processor, an estimating processor and an averaging processor, respectively.

[0061] The means for receiving 110 receives a plurality of measurement results (S110). Each of the measurement results comprises an indication of a distance between an antenna and a location where the respective measurement is performed. The indication of the distance may be given in terms of the location of the measurement device. Furthermore, each of the measurement results comprises an indication of a value of the signal strength received from the antenna at the location.

[0062] The means for estimating 120 estimates, for each of the measurement results, a respective preliminarily estimated physical configuration of the antenna (S120). The physical configuration comprises at least one of a direction of the antenna, a downtilt of the antenna, and a height of the antenna. This preliminary estimation is performed based on the respective distance and the respective value of the received strength for the measurement result. Thus, the means for estimating 120 performs the estimating individually for each measurement result. The means for estimating may perform the estimating based on a trained regression model, such as a trained neural network.

[0063] The means for averaging 130 averages the preliminarily estimated physical configurations for the plurality of measurement results obtained by the means for estimating 120 (S130). Thus, the means for averaging obtains an estimated physical configuration of the antenna.

[0064] Fig. 7 shows an apparatus according to an embodiment. The apparatus comprises at least one processor 810 and at least one memory 820 storing instructions that, when executed by the at least one processor 810, cause the apparatus at least to perform the method according to at least Fig. 6 and related description.

[0065] The method may be performed for an antenna transmitting signals of an arbitrary radio standard. For example, some radio standards are specified by 3GPP, e.g. LTE, LTE-A, and UMTS.

[0066] One piece of information may be transmitted in one or plural messages from one entity to another entity. Each of these messages may comprise further (different) pieces of information.

[0067] Names of network elements, network functions, protocols, and methods are based on current standards. In other versions or other technologies, the names of these network elements and / or network functions and / or protocols and / or methods may be different, as long as they provide a corresponding functionality. The same applies correspondingly to the terminal.

[0068] If not otherwise stated or otherwise made clear from the context, the statement that two entities are different means that they perform different functionalities. It does not necessarily mean that they are based on different hardware. That is, each of the entities described in the present description may be based on a different hardware, or some or all of the entities may be based on the same hardware. It does not necessarily mean that they are based on different software. That is, each of the entities described in the present description may be based on different software, or some or all of the entities may be based on the same software. Each of the entities described in the present description may be deployed in the cloud.

[0069] According to the above description, it should thus be apparent that example embodiments provide, for example, a regression model (such as a neural network), or an element thereof, an apparatus embodying the same, a method for controlling and / or operating the same, and computer program(s) controlling and / or operating the same as well as mediums carrying such computer program(s) and forming computer program product(s). Implementations of any of the above described blocks, apparatuses, systems, techniques or methods include, as non-limiting examples, implementations as hardware, software, firmware, special purpose circuits or logic, general purpose hardware or controller or other computing devices, or some combination thereof. Each of the entities described in the present description may be embodied in the cloud.

[0070] It is to be understood that what is described above is what is presently considered the preferred embodiments. However, it should be noted that the description of the preferred embodiments is given by way of example only and that various modifications may be made without departing from the scope of the disclosure as defined by the appended claims.

[0071] The terms "first X" and "second X" include the options that "first X" is the same as "second X" and that "first X" is different from "second X", unless otherwise specified.

Claims

Claims:

1. Computer-implemented method, comprising receiving a plurality of measurement results each comprising an indication of a respective distance between a first antenna and a location where the respective measurement is performed and an indication of a respective value of a signal strength received from the first antenna at the location where the respective measurement is performed; estimating, for each of the measurement results, based on the respective distance and the respective value of the received signal strength, a respective preliminarily estimated physical configuration of the first antenna; averaging the preliminarily estimated physical configurations for the plurality of measurement results to obtain an estimated physical configuration of the first antenna, wherein, if the first antenna is a directional antenna, the physical configuration comprises at least one of a direction of the first antenna, a downtilt of the first antenna, and a height of the first antenna; and if the first antenna is an omnidirectional antenna, the physical configuration comprises at least one of the downtilt of the first antenna, and the height of the first antenna.

2. The method according to claim 1, wherein, for at least one of the measurement results, the indication of the respective distance comprises an indication of the location where the respective measurement is performed, and the method further comprises calculating, for the at least one of the measurement results, the distance between the first antenna and the location where the respective measurement is performed based on the indication of the location where the respective measurement is performed and a location of the first antenna.

3. The method according to any of claims 1 and 2, further comprising receiving, for each of the measurement results, an indication of a respective transmit power of a signal transmitted by the first antenna, wherein,for each of the measurement results the signal strength is based on the signal transmitted with the respective transmit power; wherein, for each of the measurement results, the respective preliminarily estimated physical configuration of the first antenna is estimated additionally based on the respective transmit power.

4. The method according to any of claims 1 to 3, wherein, for each of the measurement results, the respective preliminarily estimated physical configuration of the first antenna is estimated using a trained neural network.

5. The method according to claim 4, further comprising training the neural network based on a set of plural training data for one or more second antennas, wherein for each of the one or more second antennas, each of the training data comprises an indication of a respective value of the signal strength received from the respective second antenna, an indication of a respective distance between the respective second antenna and a location where the respective value of the signal strength is received from the respective second antenna, the downtilt of the respective second antenna at the time when the respective value of the signal strength is received, the height of the respective second antenna at the time when the respective value of the signal strength is received, and, if the respective second antenna is a directional antenna, the direction of the respective second antenna at a time when the respective value of the signal strength is received.

6. The method according to claim 5, wherein the set of training data comprises training data for more than one second antenna; the set of training data comprises, for each of the second antennas, an indication of a directional characteristics of the respective second antenna; and the method comprises receiving a directional characteristics of the first antenna; checking, for each of the second antennas, whether the directional characteristics of the first antenna differs from the directional characteristics of the respective second antenna by more than a predefined threshold value;excluding, for each of the second antennas, the plural training data for the respective second antenna from the training if the directional characteristics of the first antenna differs from the directional characteristics of the respective second antenna by more than the predefined threshold value.

7. The method according to any of claims 5 and 6, further comprising, for each of the one or more second antennas: if the respective second antenna is a directional antenna, determining the direction of the respective second antenna; determining the downtilt of the respective second antenna; determining the height of the respective second antenna, and for each of the training data for the respective second antenna: measuring the respective value of the signal strength received from the respective second antenna at the location where the respective value of the signal strength is received from the respective second antenna; determining the location where the respective value of the signal strength received from the respective second antenna is measured.

8. The method according to any of claims 1 to 7, wherein the preliminarily estimated configurations of the first antenna for the plurality of measurement results are averaged by calculating an arithmetic mean value of the preliminarily estimated configurations of the first antenna.

9. The method according to any of claims 1 to 8, further comprising, for each of the measurement results, measuring the respective value of the signal strength received from the first antenna at the location where the respective measurement is performed; determining the location where the respective value of the received signal strength is measured.

10. Apparatus, comprising first means for receiving a plurality of measurement results each comprising an indication of a respective distance between a first antenna and a location where the respective measurement is performed and an indication of a respective valueof a signal strength received from the first antenna at the location where the respective measurement is performed; means for estimating, for each of the measurement results, based on the respective distance and the respective value of the received signal strength, a respective preliminarily estimated physical configuration of the first antenna; means for averaging the preliminarily estimated physical configurations for the plurality of measurement results to obtain an estimated physical configuration of the first antenna, wherein, if the first antenna is a directional antenna, the physical configuration comprises at least one of a direction of the first antenna, a downtilt of the first antenna, and a height of the first antenna; and if the first antenna is an omnidirectional antenna, the physical configuration comprises at least one of the downtilt of the first antenna, and the height of the first antenna.

11. The apparatus according to claim 10, wherein, for at least one of the measurement results, the indication of the respective distance comprises an indication of the location where the respective measurement is performed, and the apparatus further comprises means for calculating, for the at least one of the measurement results, the distance between the first antenna and the location where the respective measurement is performed based on the indication of the location where the respective measurement is performed and a location of the first antenna.

12. The apparatus according to any of claims 10 and 11, further comprising second means for receiving, for each of the measurement results, an indication of a respective transmit power of a signal transmitted by the first antenna, wherein, for each of the measurement results the signal strength is based on the signal transmitted with the respective transmit power; and the means for estimating is configured to estimate, for each of the measurement results, the respective preliminarily estimated physical configuration of the first antenna additionally based on the respective transmit power.

13. The apparatus according to any of claims 10 to 12, wherein the means for estimating is configured to estimate, for each of the measurement results, the respective preliminarily estimated physical configuration of the first antenna using a trained neural network.

14. The apparatus according to claim 13, further comprising means for training the neural network based on a set of plural training data for one or more second antennas, wherein for each of the one or more second antennas, each of the training data comprises an indication of a respective value of the signal strength received from the respective second antenna, an indication of a respective distance between the respective second antenna and a location where the respective value of the signal strength is received from the respective second antenna, the downtilt of the respective second antenna at the time when the respective value of the signal strength is received, the height of the respective second antenna at the time when the respective value of the signal strength is received, and, if the respective second antenna is a directional antenna, the direction of the respective second antenna at a time when the respective value of the signal strength is received.

15. A computer program product comprising a set of instructions which, when executed on an apparatus, is configured to cause the apparatus to carry out the method according to any of claims 1 to 9.

Images