Information processing device, program, and information processing method

Abstract

To provide an information processing device equipped with a generation unit that generates a network whose input is radio wave status information and whose output is altitude information, a program, and an information processing method.SOLUTION: An estimation device 100 for estimating an altitude at which a wireless communication terminal is located based on a radio wave status of the wireless communication terminal includes an information acquisition unit that acquires multiple pieces of terminal-related information from a storage unit that stores the terminal-related information including radio wave status information including cell identification information of a serving cell and neighbor cells identified by the wireless communication terminal and altitude information identified from air pressure measured by the wireless communication terminal, and a generation unit that generates a network whose input is the radio wave status information and whose output is the altitude information by supervised learning using the multiple pieces of terminal-related information as teacher data.SELECTED DRAWING: Figure 4

Description

[Technical Field]

[0001] The present invention relates to an information processing device, a program, and an information processing method. [Background technology]

[0002] Patent Document 1 describes a technique for determining whether a wireless communication terminal is located indoors or outdoors. [Prior art document] [Patent documents] [Patent Document 1] Patent No. 6739499 Summary of the Invention [Means for solving the problem]

[0003] According to one embodiment of the present invention, there is provided an information processing device. The information processing device may include an information acquisition unit that acquires a plurality of pieces of terminal-related information from a storage unit that stores terminal-related information, the plurality of pieces of terminal-related information including radio wave condition information including cell identification information of a serving cell and a neighbor cell identified by a wireless communication terminal, and altitude information identified from atmospheric pressure measured by the wireless communication terminal. The information processing device may include a generation unit that generates a learning model that uses radio wave condition information as input and altitude information as output by supervised learning using the plurality of pieces of terminal-related information as training data.

[0004] In the information processing device, the radio wave condition information may further include radio wave reception strength from the serving cell by the wireless communication terminal, radio wave reception strength from the neighbor cell by the wireless communication terminal, and timing advance of the serving cell.

[0005] In the information processing device, the generation unit may generate the learning model using a mixture density network (MDN).

[0006] Any of the information processing devices may further include an altitude position determination unit that uses the learning model generated by the generation unit to determine an altitude position corresponding to each of multiple radio wave condition information for which altitude is to be determined.

[0007] The altitude location identification unit may use the learning model to output multiple pieces of altitude information corresponding to the multiple pieces of radio wave condition information, and filter the multiple pieces of altitude information to identify the altitude location. The altitude location identification unit may input the radio wave condition information to the learning model and exclude from the target of identifying the altitude location, radio wave condition information whose reliability indicated by reliability information output from the learning model is lower than a predetermined threshold. The altitude location identification unit may exclude from the target of identifying the altitude location, radio wave condition information whose peak value of a probability density function in the MDN output from the learning model is lower than the predetermined threshold.

[0008] The altitude position specifying unit may adjust the multiple altitude positions specified for the multiple pieces of radio wave condition information corresponding to one wireless communication terminal based on a characteristic that the altitude position of the wireless communication terminal does not change continuously within a certain period of time. The altitude position specifying unit may adjust the multiple altitude positions using dynamic programming so that the altitude position does not change continuously within a certain period of time. The altitude position specifying unit may adjust the multiple altitude positions based on a majority vote so that the altitude position does not change continuously within a certain period of time.

[0009] The advanced position specifying unit may specify on which floor of a building each of the plurality of wireless communication terminals for which the serving cell and the neighbor cell have been specified is located.

[0010] According to one embodiment of the present invention, there is provided a program for causing a computer to function as the information processing device.

[0011] According to one embodiment of the present invention, there is provided an information processing method executed by a computer. The information processing method may include an information acquisition step of acquiring, from a storage unit that stores terminal-related information, a plurality of pieces of terminal-related information, the plurality of pieces of terminal-related information including radio wave condition information including cell identification information of a serving cell and a neighbor cell identified by a wireless communication terminal, and altitude information identified from atmospheric pressure measured by the wireless communication terminal. The information processing method may also include a generation step of generating a learning model that uses the plurality of pieces of terminal-related information as training data through supervised learning, in which the radio wave condition information is used as input and the altitude information is used as output.

[0012] The above summary of the invention does not list all of the necessary features of the present invention, and subcombinations of these features may also constitute inventions. [Brief explanation of the drawings]

[0013] [Figure 1] 1 illustrates an example of a communication environment of the estimation device 100. [Figure 2] 4 shows a schematic diagram of an example of an estimation algorithm 400 used by the estimation device 100. [Figure 3] FIG. 10 is an explanatory diagram for explaining Embedding_layer 404. [Figure 4] 1 shows an example of a functional configuration of an estimation device 100. [Figure 5] FIG. 10 is an explanatory diagram for explaining an MDN by a generation unit 108. [Figure 6] FIG. 10 is an explanatory diagram for explaining an MDN by a generation unit 108. [Figure 7] FIG. 10 is an explanatory diagram for explaining an MDN by a generation unit 108. [Figure 8] Here is a specific example of a prediction by MDN for a specific input x. [Figure 9] FIG. 10 is an explanatory diagram illustrating time-series_filtering. [Figure 10] FIG. 10 is an explanatory diagram illustrating time-series_filtering. [Figure 11] FIG. 10 is an explanatory diagram illustrating time-series_filtering. [Figure 12] FIG. 10 is an explanatory diagram illustrating time-series_filtering. [Figure 13] FIG. 10 is an explanatory diagram illustrating time-series_filtering. [Figure 14] FIG. 10 is an explanatory diagram illustrating time-series_filtering. [Figure 15] An example of the functional configuration of a computer 1200 that functions as the estimation device 100 is shown in schematic form. DETAILED DESCRIPTION OF THE INVENTION

[0014] The present invention will be described below through embodiments of the invention, but the following embodiments do not limit the scope of the invention according to the claims. Furthermore, not all of the combinations of features described in the embodiments are necessarily essential to the solution of the invention.

[0015] FIG. 1 schematically illustrates an example of a communication environment of an estimation device 100. The estimation device 100 may be an example of an information processing device. The estimation device 100 according to this embodiment estimates the altitude at which the wireless communication terminal 300 is located based on the radio wave conditions of the wireless communication terminal 300. For example, the estimation device 100 estimates the altitude above ground of the wireless communication terminal 300. For example, the estimation device 100 estimates the altitude position at which the wireless communication terminal 300 is located. The altitude position may be a position corresponding to the altitude, such as the number of floors of a building 40.

[0016] In this embodiment, the information collection device 200 collects, from each of the multiple wireless communication terminals 300, measurement information measured by the wireless communication terminal 300. The measurement information includes radio wave condition information measured by the wireless communication terminal 300 and indicating radio wave conditions from the wireless base station 20 around the wireless communication terminal 300. Furthermore, if the wireless communication terminal 300 includes a barometric pressure sensor, the measurement information may include barometric pressure information indicating the barometric pressure measured by the barometric pressure sensor included in the wireless communication terminal 300.

[0017] The radio base station 20 and the radio communication terminal 300 may be compliant with a 5G (5th Generation) communication system. The radio base station 20 and the radio communication terminal 300 may be compliant with an LTE (Long Term Evolution) communication system. The radio base station 20 and the radio communication terminal 300 may be compliant with a 3G (3rd Generation) communication system. The radio base station 20 and the radio communication terminal 300 may be compliant with a mobile communication system such as a 6G (6th Generation) communication system or later. Here, an example will be described in which the radio base station 20 and the radio communication terminal 300 are compliant with a 5G communication system. Examples of the radio communication terminal 300 include mobile phones such as smartphones, tablet terminals, and wearable terminals.

[0018] The radio wave condition information may include a PCI (Physical Cell ID) of the serving cell. The PCI may be an example of cell identification information. The radio wave condition information may include a PCI of a neighbor cell. The radio wave condition information may include radio wave reception strength from the serving cell. The radio wave condition information may include, for example, a Reference Signal Received Power (RSRP) from the serving cell. The radio wave condition information may include radio wave reception strength from a neighbor cell. The radio wave condition information may include, for example, a RSRP from a neighbor cell. The radio wave condition information may include a timing advance of the serving cell.

[0019] The wireless communication terminal 300, for example, periodically transmits the measurement information to the information collection device 200. The wireless communication terminal 300 may transmit the measurement information to the information collection device 200 as an MR (Measurement Report). The wireless communication terminal 300 may transmit the measurement information to the information collection device 200 via a network 10. The network 10 includes a mobile communication system in which the wireless communication terminal 300 and the wireless base station 20 are compliant. The network 10 may also include the Internet.

[0020] The information collection device 200 stores the measurement information of the wireless communication terminal 300 in association with the call ID of the wireless communication terminal 300. The information collection device 200 stores the measurement information of a plurality of wireless communication terminals 300. The information stored by the information collection device 200 includes information in which a call ID, radio wave condition information, and atmospheric pressure information are associated with each other, and information in which a call ID and radio wave condition information are associated with each other but does not include atmospheric pressure information.

[0021] The estimating device 100 may receive information used for learning from the information collecting device 200. The estimating device 100 may receive information collected by the information collecting device 200 from the information collecting device 200. The estimating device 100 may communicate with the information collecting device 200 via the network 10.

[0022] For learning purposes, the estimation device 100 may receive information in which measurement information and atmospheric pressure information are associated with a call ID from the information collection device 200. The estimation device 100 calculates altitude information from the atmospheric pressure information and stores information including the call ID, measurement information, and altitude information (sometimes referred to as terminal-related information).

[0023] The estimation device 100 may calculate the altitude information from the atmospheric pressure information using any known technique. For example, the estimation device 100 calculates the altitude information from the atmospheric pressure information using the following equation 1.

[0024]

number

[0025] However, P b is static pressure (sea level pressure) [Pa], T b is the standard temperature (sea surface temperature) [K], L b is the standard temperature lapse rate [K / m], h is the height above sea level [m], hb is the height of the bottom of the atmospheric layer [m], R is the universal gas constant, 8.31432 [Nm / molK], g0 is the gravitational acceleration constant, 9.80665 [m / s 2 ], M is the molar mass of the Earth's atmosphere [kg / mol].

[0026] The information collection device 200 may calculate the altitude information from the atmospheric pressure information. In this case, the estimation device 100 may receive the call ID, the measurement information, and the altitude information from the information collection device 200.

[0027] The estimation device 100 and the information collection device 200 may be integrated. That is, the estimation device 100 may have the functions of the information collection device 200.

[0028] The estimation device 100 generates a learning model using supervised learning with multiple pieces of device-related information as training data, with radio wave condition information as input and altitude information as output. Then, the estimation device 100 uses the learning model to identify altitude positions corresponding to multiple pieces of radio wave condition information for which altitude is to be identified. For example, the estimation device 100 may receive measurement information from the information collection device 200 that is not associated with atmospheric pressure information or altitude information as multiple pieces of radio wave condition information for which altitude is to be identified. Alternatively, the estimation device 100 may separately receive input of multiple pieces of radio wave condition information for which altitude is to be identified.

[0029] The estimation device 100 may use a mixture density network (MDN) as a learning algorithm. The estimation device 100 may generate a learning model using the MDN, with a plurality of pieces of terminal-related information as training data.

[0030] 2 shows an example of an estimation algorithm 400 used by the estimation device 100. Here, the description is given assuming that learning has been completed. For each of a plurality of cells, the estimation device 100 uses the estimation algorithm 400 to estimate the altitude position of the wireless communication terminal 300 from the radio wave condition information of the wireless communication terminal 300 that has each cell as its serving cell.

[0031] Neighbor_PCI_list 402 is a list of PCIs of neighboring cells. In this example, Neighbor_PCI_list 402 is input to Embedding_layer 404 and converted by Embedding_layer 404. PCIs are expressed as numbers, but the relationship between these numbers does not fundamentally represent the relative positions of the cells. For example, PCI:200 is twice as large as PCI:100 in terms of number, but there is no such relationship in terms of the relative positions of the cells. For example, PCI:199 and PCI:200 are close in terms of number, but the cell positions are not close. Embedding_layer 404 converts PCIs into values ​​that indicate the relative positions of the cells.

[0032] The RF_features 406 may include the signal strength received by the wireless communication terminal 300 from the serving cell. The RF_features 406 may include the signal strength received by the wireless communication terminal 300 from a neighbor cell. The RF_features 406 may include the timing advance of the serving cell. The output of the Embedding_layer 404 and the RF_features 406 are input to the MDN 408. The MDN 408 outputs altitude information of the wireless communication terminal 300.

[0033] Altitude_extraction 410 extracts an altitude position from the altitude information output by MDN 408. The estimation device 100 identifies the number of floors of the building 40 from the altitude information, for example, by using association information that associates the altitude information with the number of floors of the building 40.

[0034] Time-series_filtering 412 adjusts the multiple altitude positions output by Altitude_extraction 410 for each of the multiple wireless communication terminals 300 based on the characteristic that the altitude positions of the wireless communication terminals 300 do not change frequently. The information collection device 200 may identify the multiple altitude positions corresponding to one wireless communication terminal 300 by the call ID corresponding to the radio wave condition information.

[0035] The estimation device 100 uses the output of Time-series_filtering412 to identify the altitude position corresponding to each of the multiple pieces of radio wave condition information, i.e., the altitude position of the wireless communication terminal 300 when the measurement was performed (for example, the floor number of the building 40 on which the wireless communication terminal 300 was located when the measurement was performed).

[0036] 3 is an explanatory diagram for explaining the Embedding_layer 404. Here, a case will be described in which the radio base station 20 and the radio communication terminal 300 are compliant with the 5G communication system and 1008 different IDs are used as PCIs, but if the radio base station 20 and the radio communication terminal 300 are compliant with the LTE communication system, 504 different IDs are used as PCIs.

[0037] In the Embedding_layer 404, the PCI is represented by a one-hot vector that has been subjected to one-hot encoding processing, and is transformed by applying an embedding matrix.

[0038] The embedding matrix can be learned using multiple pieces of terminal-related information. For example, when a first altitude corresponding to the atmospheric pressure measured by a wireless communication terminal 300 located in a first cell is close to a second altitude corresponding to the atmospheric pressure measured by a wireless communication terminal 300 located in a second cell, the embedding matrix is ​​generated by learning so that the converted value of the PCI of the first cell is close to the converted value of the PCI of the second cell; when the first altitude and the second altitude are far from each other, the converted value of the PCI of the first cell is far from the converted value of the PCI of the second cell. While the example shown in FIG. 3 illustrates a case where the embedding matrix is ​​1008×2, the embedding matrix may be 1008×1, 1008×3, or greater. By using the Embedding_layer 404, the estimation device 100 can convert the PCI into a value meaningful for identifying the altitude of the wireless communication terminal 300.

[0039] 4 illustrates an example of the functional configuration of the estimation device 100. The estimation device 100 includes a storage unit 102, an information collection unit 104, an information acquisition unit 106, a generation unit 108, an altitude position identification unit 110, and an output control unit 112.

[0040] The storage unit 102 stores various types of information, including information collected by the information collection unit 104.

[0041] The information collection unit 104 collects various types of information. The information collection unit 104 may collect information about the building 40. The information about the building 40 may include the number of floors in the building 40 and the height of each floor.

[0042] The information collecting unit 104 may receive information from the information collecting device 200. For example, the information collecting unit 104 receives measurement information by the wireless communication terminal 300 from the information collecting device 200. When the measurement information includes atmospheric pressure information, the information collecting unit 104 may calculate altitude information from the atmospheric pressure information. Note that the process of calculating altitude information from atmospheric pressure information may be executed in the information collecting device 200. In this case, the information collecting unit 104 may receive the altitude information from the information collecting device 200.

[0043] The storage unit 102 may manage the information collected by the information collection unit 104 using a database. The storage unit 102 may store in the database a measurement information ID that identifies the measurement information, the PCI of the serving cell, the call ID, a timestamp indicating the time when the wireless communication terminal 300 performed the measurement, the radio wave reception strength from the serving cell, the PCI of the neighbor cell, the radio wave reception strength from the neighbor cell, and the timing advance of the serving cell in association with each other. If the measurement information includes atmospheric pressure information, the storage unit 102 may further store in the database the atmospheric pressure information and altitude information in association with each other. The information managed by the storage unit 102 may be referred to as terminal-related information.

[0044] The information acquisition unit 106 acquires terminal-related information to be used for learning from the storage unit 102. The information acquisition unit 106 acquires from the storage unit 102 a plurality of pieces of terminal-related information including altitude information.

[0045] The generation unit 108 generates a learning model that uses radio wave condition information as input and altitude information as output by supervised learning using the plurality of pieces of terminal-related information acquired by the information acquisition unit 106 as training data. The generation unit 108 may generate the learning model using MDN. The generation unit 108 stores the generated learning model in the storage unit 102.

[0046] The altitude position specifying unit 110 uses the learning model generated by the generation unit 108 to specify an altitude position corresponding to each of the multiple pieces of radio wave condition information that are the targets of altitude specification. The altitude position specifying unit 110 acquires, for example, multiple pieces of terminal-related information that do not include altitude information from the storage unit 102, and specifies the altitude position corresponding to each piece of information. The altitude position specifying unit 110 may, for example, newly acquire, from the outside, multiple pieces of radio wave condition information that are the targets of altitude specification. The altitude position specifying unit 110 stores the specification results in the storage unit 102.

[0047] The advanced position specifying unit 110 specifies, for example, on which floor in one building 40 each of the multiple wireless communication terminals 300 for which the serving and neighbor cells have been specified is located.

[0048] The altitude position specifying unit 110 may use a learning model to output a plurality of altitude information corresponding to a plurality of pieces of radio wave condition information, and may specify the altitude position by filtering the plurality of altitude information.

[0049] For example, the altitude location identification unit 110 inputs radio wave condition information into a learning model and excludes from the target of altitude location identification radio wave condition information in which the reliability indicated by the reliability information output from the learning model is lower than a predetermined threshold. For example, the altitude location identification unit 110 excludes from the target of altitude location identification radio wave condition information in which the peak value of the probability density function in the MDN output from the learning model is lower than a predetermined threshold. This makes it possible to avoid identifying the altitude location when it is expected that the accuracy of identifying the altitude location will be low, thereby increasing the reliability of the identification result.

[0050] The altitude position specifying unit 110 adjusts the multiple pieces of altitude information specified for the multiple pieces of radio wave condition information corresponding to one wireless communication terminal 300, based on the characteristic that the altitude position of the wireless communication terminal 300 does not change consecutively within a certain period of time. This allows correction when the multiple pieces of altitude information indicate an event that is unlikely to actually occur, such as the floor on which the wireless communication terminal 300 is located changing within a short period of time, and makes it possible to increase the reliability of the specification result.

[0051] The altitude position specifying unit 110 adjusts the plurality of altitude information pieces by using, for example, dynamic programming, so that the altitude position does not change continuously within a certain period of time. This reduces the calculation load on the estimation device 100. The altitude position specifying unit 110 adjusts the altitude position for the plurality of altitude information pieces by, for example, majority vote, so that the altitude position does not change continuously within a certain period of time.

[0052] The output control unit 112 controls the output of information stored in the storage unit 102. The output control unit 112 controls the output of, for example, the identification result by the altitude position identification unit 110. The output control unit 112 may display and output the identification result on a display provided in the estimation device 100. The output control unit 112 may transmit the identification result to an external device. The output control unit 112 may transmit the learning model generated by the generation unit 108 to an external device.

[0053] 5 to 7 are explanatory diagrams for explaining MDN by the generation unit 108. There is a high possibility that the radio wave conditions of radio communication terminals 300 located at the same altitude will exhibit the same characteristics. For example, there is a high possibility that radio communication terminals 300 located on the same floor in building 40 will be in the same serving cell and will detect the same neighbor cell, and there is a high possibility that the radio wave reception strength from the serving cell, the radio wave reception strength from the neighbor cell, and the timing advance of the serving cell will exhibit similar values.

[0054] The generation unit 108 generates a distribution of the altitude of the wireless communication terminal 300 for each combination x of the serving cell and neighbor cell of the wireless communication terminal 300, the radio wave reception strength from the serving cell, the radio wave reception strength from the neighbor cell, and the timing advance of the serving cell. Fig. 5 shows the distribution of the altitude of the wireless communication terminal 300 for a particular combination x. The horizontal axis represents the altitude (number of floors in the building 40), and the vertical axis represents the quantity of the wireless communication terminal 300.

[0055] 5, the example includes a Gaussian component (Gaussian component 1 in FIG. 5) indicating that the wireless communication terminal 300 was located around the 10th floor, and a Gaussian distribution (Gaussian component 2 in FIG. 5) indicating that the wireless communication terminal 300 was located around the 5th floor. That is, this combination indicates that the wireless communication terminal 300 was located around the 10th floor in many cases, and the wireless communication terminal 300 was located around the 5th floor in the next most cases.

[0056] The generation unit 108 uses MDN to learn a mapping from x to θ, assuming that a parameter θ of a distribution such as that shown in FIG. 5 is a function of the combination x.

[0057] As shown in Figure 6, θ is represented by three vectors μ, σ, and P. μ denotes the mean of the Gaussian components, σ ​​denotes the standard deviation of the Gaussian components, and P denotes the mixture probability of the Gaussian components.

[0058] As shown in Figure 7, x is the combination of the value of the PCI vector transformed by the Embedding_layer and the RF_features, and the MDN 408 is trained in an end-to-end network using a negative log likelihood loss function.

[0059] FIG. 8 shows a specific example of a prediction by MDN for a specific input x. MDN predicts a conditional distribution of altitude, rather than the altitude itself. In this example, the altitude position identification unit 110 uses the average of the components with the highest PDF (Probability Density Function) peaks as the predicted altitude, as shown in FIG.

[0060] 8 , the first component has P=0.6124, σ=1.142, and μ=1.2312, the second component has P=0.2535, σ=0.515, and μ=3.2215, and the third component has P=0.1341, σ=1.523, and μ=7.0942. The peak PDF of the first component is 0.536, the peak PDF of the second component is 0.492, and the peak PDF of the third component is 0.088. Because the peak PDF of the first component is the highest, the altitude position specifying unit 110 sets the predicted altitude to μ=1.2312 of the first component.

[0061] The altitude position specifying unit 110 may exclude radio wave condition information whose peak PDF is lower than a predetermined threshold from the target for specifying the altitude position. A low peak PDF value means that the reliability of the altitude is low, so this exclusion can increase the reliability of the specification result.

[0062] The threshold may be arbitrarily settable and may be changeable. For example, the threshold may be manually set by a user of the estimation device 100 or the like. If the threshold is set high, the accuracy of the altitude position output as a result increases, but more radio wave condition information is excluded. If the threshold is set low, more radio wave condition information can be output as a result, but the accuracy of the altitude position decreases. The user of the estimation device 100 or the like may set the threshold in consideration of the balance between the accuracy of the altitude position and the number of results output.

[0063] 9 to 14 are explanatory diagrams for explaining time-series filtering. The estimation device 100 may divide the radio wave condition information into subsets for each wireless communication terminal 300 using the call ID, identify multiple pieces of radio wave condition information within a predetermined period, such as 10 minutes, using timestamps, and perform time-series filtering. The predetermined period is not limited to 10 minutes and may be arbitrarily set or changeable.

[0064] The altitude position determination unit 110 adjusts the multiple altitude positions determined for multiple pieces of radio wave condition information corresponding to one wireless communication terminal 300 based on the characteristic that the altitude position of the wireless communication terminal 300 does not change continuously within a certain period of time.

[0065] 9 to 13 are explanatory diagrams for explaining time-series filtering using dynamic programming. The altitude position specifying unit 110 receives as input a plurality of Gaussian components (P, μ, σ) included in the subset that change every T time steps and the cost of the path from time t to time t+1, outputs the shortest path (the path with the lowest cost) from time 0 to time T-1, and adjusts the corresponding altitude position.

[0066] The altitude position specifying unit 110 calculates the cost of the route using the following Equation 2.

[0067]

number

[0068] The components at time 0 and time T-1 are the starting and ending costs, σ i0 / P i0 and σ i(T-1) / P i(T-1) It has the following.

[0069] As a specific example, μ 2t is 2.1, P 2(t+1) is 0.65, μ 2(t+1) is 1.75, σ 2(t+1) is 1.1, the cost of the route is 0.59 according to the above formula 2.

[0070] The advanced position specifying unit 110 uses dynamic programming to specify the shortest route from time 0 to time T-1. As an example, Fig. 10 illustrates a flow for specifying the shortest route between time 0 and time 2. In Fig. 10, the numbers in nodes 510, 520, 512, and 522 represent starting and ending costs, and the numbers placed next to the arrows represent transition costs.

[0071] In this dynamic programming method, we consider two separate problems. The first problem is to identify the shortest path from time 1 to time 2 for all nodes, as shown in Figure 11. The second problem is to identify the shortest path from time 0 to time 2 for all nodes, as shown in Figure 12.

[0072] As shown in FIG. 11, in this specific example, the cost of the path from node 511 to node 522 is 1.5, the cost of the path from node 511 to node 512 is 5.3, the cost of the path from node 521 to node 522 is 1.5, and the cost of the path from node 521 to node 512 is 2.3, and the shortest paths from time 1 to time 2 are from node 511 to node 522 and from node 521 to node 522, with costs of 1.5 and 2, respectively.

[0073] To solve the problem shown in Fig. 12, it is necessary to solve the problem shown in Fig. 11. Therefore, the altitude position specifying unit 110 reduces the overall amount of calculation by making the shortest route shown in Fig. 11 essential. That is, when solving the problem shown in Fig. 12, the route to node 512 is deleted and calculation is performed as shown in Fig. 13.

[0074] The cost of the route between nodes 510, 511, and 522 is 0.6 + 3.8 + 1.5 = 5.9, the cost of the route between nodes 510, 521, and 522 is 0.6 + 0.9 + 2 = 3.5, the cost of the route between nodes 520, 511, and 522 is 0.1 + 0.7 + 1.5 = 2.3, and the cost of the route between nodes 520, 521, and 522 is 0.1 + 1.0 + 2.0 = 3.1. Because 2.3 is smaller than 3.5, the advanced position specifying unit 110 determines that the route between nodes 520, 511, and 522, which depart from component 2 at time 0, is the shortest route.

[0075] 14 is a flow chart showing the flow of the time-series filtering process based on majority vote. The flow shown in FIG.

[0076] The altitude position identification unit 110 sequentially examines the altitude position evaluations for each T time step, and if an anomaly is detected, starts voting to determine whether the next reliable evaluation is in agreement or disagreement with the current state. Then, the voting is terminated by replacing all evaluations with the average of the winning side (agreement or disagreement). The input of the time series filtering based on majority voting may be the time series altitude position evaluations and their respective confidence levels (e.g., peak PDFs), and the output may be the adjusted altitude position evaluations.

[0077] In FIG. 14, the evaluation of the altitude position is described as sample s. thresh is a threshold value for determining whether or not to start voting. If you want to start voting when the altitude position of the wireless communication terminal 300 changes by at least one floor, d thresh is set to about 1. R thresh is the threshold for deciding whether to reject sample s during voting. votes is the number of Near / Far samples that terminates voting. α is a weighting coefficient. g(e, s) = αe + (1-α)s, and R s =P s / σ s is.

[0078] The altitude position determination unit 110 first initializes e by averaging the first few samples. In step (sometimes abbreviated as S) 102, the next sample S is obtained. In S104, it is determined whether voting is continuing. If not, the process proceeds to S106; if so, the process proceeds to S112.

[0079] In S106, |se| is d thresh It is determined whether or not it is greater. If it is not greater, the process proceeds to S108, and if it is greater, the process proceeds to S110. In S108, the result of g(e, s) is inserted into e, and the process returns to S102.

[0080] In S110, voting begins. In S112, Rs of sample s is R threshIf it is greater, the process proceeds to S114, and if it is not greater, the process returns to S102.

[0081] In S114, |se| is d thresh It is determined whether s is greater than 0. If it is not greater, the process proceeds to S116, and if it is greater, the process proceeds to S118. In S116, s is inserted into near_buffer. In S118, s is inserted into far_buffer.

[0082] In S120, it is determined whether any of the buffers is full. votes If the storage capacity reaches 100%, it is determined to be full. If the storage capacity is not full, the process returns to S102, and if the storage capacity is full, the process proceeds to S122.

[0083] In S122, the altitude position estimate corresponding to the voting sample s is set to the average of the full winner's buffer, and the average of the full winner's buffer is inserted into e.

[0084] In S124, the near_buffer and far_buffer are flushed, voting ends, and the process returns to S102.

[0085] In this way, the altitude location unit 110 maintains the average e of the sample s and adjusts the sample s when the voting is complete. The altitude location unit 110 determines whether the current sample s is equal to the average e of the sample s and the average d of the sample s. thresh If the values ​​are different from each other, that is, if the altitude of the wireless communication terminal 300 has changed, voting begins. The altitude position specifying unit 110 maintains e during voting, and the reliability is R thresh Discard samples smaller than s and the difference between samples s and e is d. thresh The sample s is classified into a near buffer or a far buffer depending on whether it is larger than n votes When the voting is finished, the advanced position determination unit 110 replaces all samples s during the voting with the average of the winning side's buffer and inserts the average into e.

[0086] By the above processing, when the altitude position changes, it is possible to determine whether the change is temporary (i.e., an erroneous determination) from the average before and after the change, and if it is an erroneous determination, it can be corrected.

[0087] 15 schematically illustrates an example of the hardware configuration of a computer 1200 that functions as the estimation apparatus 100. A program installed on the computer 1200 can cause the computer 1200 to function as one or more "parts" of the apparatus according to the present embodiment, or can cause the computer 1200 to perform operations associated with the apparatus according to the present embodiment or one or more "parts" thereof, and / or can cause the computer 1200 to perform a process according to the present embodiment or steps of the process. Such a program can be executed by the CPU 1212 to cause the computer 1200 to perform specific operations associated with some or all of the blocks in the flowcharts and block diagrams described herein.

[0088] The computer 1200 according to this embodiment includes a CPU 1212, a RAM 1214, and a graphics controller 1216, which are interconnected by a host controller 1210. The computer 1200 also includes input / output units such as a communications interface 1222, a storage device 1224, a DVD drive 1226, and an IC card drive, which are connected to the host controller 1210 via an input / output controller 1220. The DVD drive 1226 may be a DVD-ROM drive, a DVD-RAM drive, or the like. The storage device 1224 may be a hard disk drive, a solid-state drive, or the like. The computer 1200 also includes a ROM 1230 and legacy input / output units such as a keyboard, which are connected to the input / output controller 1220 via an input / output chip 1240.

[0089] The CPU 1212 operates according to programs stored in the ROM 1230 and the RAM 1214, thereby controlling each unit. The graphics controller 1216 acquires image data generated by the CPU 1212 into a frame buffer or the like provided in the RAM 1214 or into the graphics controller itself, and causes the image data to be displayed on the display device 1218.

[0090] The communication interface 1222 communicates with other electronic devices via a network. The storage device 1224 stores programs and data used by the CPU 1212 in the computer 1200. The DVD drive 1226 reads programs or data from a DVD-ROM 1227 or the like and provides them to the storage device 1224. The IC card drive reads programs and data from an IC card and / or writes programs and data to an IC card.

[0091] The ROM 1230 stores therein a boot program or the like that is executed by the computer 1200 upon activation, and / or programs that depend on the hardware of the computer 1200. The input / output chip 1240 may also connect various input / output units to the input / output controller 1220 via a USB port, a parallel port, a serial port, a keyboard port, a mouse port, etc.

[0092] The programs are provided by a computer-readable storage medium such as a DVD-ROM 1227 or an IC card. The programs are read from the computer-readable storage medium, installed in the storage device 1224, RAM 1214, or ROM 1230, which are also examples of computer-readable storage media, and executed by the CPU 1212. Information processing described in these programs is read by the computer 1200, and causes cooperation between the programs and the various types of hardware resources described above. An apparatus or method may be configured by implementing operations or processing of information in accordance with the use of the computer 1200.

[0093] For example, when communication is performed between the computer 1200 and an external device, the CPU 1212 may execute a communication program loaded into the RAM 1214 and instruct the communication interface 1222 to perform communication processing based on the processing described in the communication program. Under the control of the CPU 1212, the communication interface 1222 reads transmission data stored in a transmission buffer area provided in the RAM 1214, the storage device 1224, the DVD-ROM 1227, or a recording medium such as an IC card, and transmits the read transmission data to the network, or writes reception data received from the network to a reception buffer area or the like provided on the recording medium.

[0094] Furthermore, the CPU 1212 may cause all or a necessary portion of a file or database stored in an external recording medium such as the storage device 1224, the DVD drive 1226 (DVD-ROM 1227), an IC card, etc. to be read into the RAM 1214, and may perform various types of processing on the data on the RAM 1214. The CPU 1212 may then write back the processed data to the external recording medium.

[0095] Various types of information, such as various types of programs, data, tables, and databases, may be stored on the recording medium and may undergo information processing. The CPU 1212 may perform various types of processing on data read from the RAM 1214, including various types of operations, information processing, conditional judgment, conditional branching, unconditional branching, information search / replacement, etc., as described throughout this disclosure and specified by the instruction sequences of the programs, and write the results back to the RAM 1214. The CPU 1212 may also search for information in a file, database, etc. on the recording medium. For example, if multiple entries, each having an attribute value of a first attribute associated with an attribute value of a second attribute, are stored on the recording medium, the CPU 1212 may search for an entry whose attribute value of the first attribute matches a specified condition from among the multiple entries, read the attribute value of the second attribute stored in the entry, and thereby obtain the attribute value of the second attribute associated with the first attribute that satisfies a predetermined condition.

[0096] The above-described programs or software modules may be stored in a computer-readable storage medium on or near the computer 1200. A recording medium such as a hard disk or RAM provided in a server system connected to a dedicated communication network or the Internet can also be used as a computer-readable storage medium, thereby providing the programs to the computer 1200 via the network.

[0097] The blocks in the flowcharts and block diagrams in the present embodiments may represent stages of a process in which an operation is performed or "parts" of an apparatus responsible for performing the operation. Particular stages and "parts" may be implemented by dedicated circuitry, programmable circuitry provided with computer-readable instructions stored on a computer-readable storage medium, and / or a processor provided with computer-readable instructions stored on a computer-readable storage medium. The dedicated circuitry may include digital and / or analog hardware circuits, including integrated circuits (ICs) and / or discrete circuits. The programmable circuitry may include reconfigurable hardware circuits, such as field programmable gate arrays (FPGAs) and programmable logic arrays (PLAs), including AND, OR, XOR, NAND, NOR, and other logical operations, flip-flops, registers, and memory elements.

[0098] A computer-readable storage medium may include any tangible device capable of storing instructions that are executed by an appropriate device, such that a computer-readable storage medium having instructions stored thereon comprises an article of manufacture, including instructions that can be executed to create means for performing the operations specified in the flowcharts or block diagrams. Examples of computer-readable storage media may include electronic storage media, magnetic storage media, optical storage media, electromagnetic storage media, semiconductor storage media, etc. More specific examples of computer-readable storage media may include floppy disks, diskettes, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), electrically erasable programmable read-only memory (EEPROM), static random access memory (SRAM), compact disc read-only memory (CD-ROM), digital versatile disc (DVD), Blu-ray disc, memory stick, integrated circuit card, etc.

[0099] The computer readable instructions may include either assembler instructions, instruction set architecture (ISA) instructions, machine instructions, machine-dependent instructions, microcode, firmware instructions, state-setting data, or source or object code written in any combination of one or more programming languages, including object-oriented programming languages ​​such as Smalltalk®, JAVA®, C++, etc., and conventional procedural programming languages ​​such as the “C” programming language or similar programming languages.

[0100] Computer-readable instructions may be provided locally or over a wide area network (WAN) such as a local area network (LAN), the Internet, etc. to a processor of a general purpose computer, special purpose computer, or other programmable data processing apparatus, or programmable circuitry, such that the processor or programmable circuitry executes the computer-readable instructions to generate means for performing the operations specified in the flowcharts or block diagrams. Examples of processors include computer processors, processing units, microprocessors, digital signal processors, controllers, microcontrollers, etc.

[0101] Although the present invention has been described above using embodiments, the technical scope of the present invention is not limited to the scope described in the above embodiments. It will be apparent to those skilled in the art that various modifications and improvements can be made to the above embodiments. It is clear from the claims that such modifications and improvements can also be included within the technical scope of the present invention.

[0102] It should be noted that the order of execution of each process, such as operations, procedures, steps, and stages, in the devices, systems, programs, and methods shown in the claims, specifications, and drawings is not specifically stated as "before," "prior to," etc., and that the processes can be performed in any order unless the output of a previous process is used in a subsequent process. Even if the operational flow in the claims, specifications, and drawings is described using "first," "next," etc. for convenience, this does not mean that the processes must be performed in this order. [Explanation of symbols]

[0103] 10 Network, 20 Wireless base station, 40 Building, 100 Estimation device, 102 Memory unit, 104 Information collection unit, 106 Information acquisition unit, 108 Generation unit, 110 Altitude position determination unit, 112 Output control unit, 200 Information collection device, 300 Wireless communication terminal, 400 Estimation algorithm, 402 Neighbor_PCI_list, 404 Embedding_layer, 406 RF_features, 408 MDN, 410 Altitude_extraction, 412 Time-series_filtering, 510, 511, 512 Node, 520, 521, 522 Node, 1200 Computer, 1210 Host controller, 1212 CPU, 1214 RAM, 1216 Graphics controller, 1218 Display device, 1220 Input / output controller, 1222 Communication interface, 1224 Storage device, 1226 DVD drive, 1227 DVD-ROM, 1230 ROM, 1240 I / O chip

Claims

1. an information acquisition unit that acquires a plurality of pieces of terminal-related information from a storage unit that stores radio wave condition information including cell identification information of a serving cell and a neighbor cell identified by the wireless communication terminal, and terminal-related information including altitude information identified from atmospheric pressure measured by the wireless communication terminal; a generation unit that generates a learning model that uses radio wave condition information as input and advanced information as output by supervised learning using the plurality of pieces of terminal-related information as training data; An information processing device comprising:

2. The information processing device according to claim 1 , wherein the radio wave condition information further includes a radio wave reception strength from the serving cell by the wireless communication terminal, a radio wave reception strength from the neighbor cell by the wireless communication terminal, and a timing advance of the serving cell.

3. The information processing device according to claim 1 , wherein the generation unit generates the learning model using a Mixture Density Network (MDN).

4. an altitude position specifying unit that specifies an altitude position corresponding to each of a plurality of pieces of radio wave condition information that are targets for altitude specification, using the learning model generated by the generation unit; The information processing device according to claim 3 , further comprising:

5. The information processing device according to claim 4 , wherein the altitude position identification unit uses the learning model to output a plurality of pieces of altitude information corresponding to the plurality of pieces of radio wave condition information, performs filtering on the plurality of pieces of altitude information, and identifies the altitude position.

6. The information processing device described in claim 5, wherein the altitude position identification unit inputs the radio wave condition information into the learning model and excludes the radio wave condition information whose reliability indicated by the reliability information output from the learning model is lower than a predetermined threshold from the altitude position identification target.

7. The information processing device according to claim 6, wherein the altitude position identification unit excludes the radio wave condition information in which a peak value of a probability density function in the MDN output from the learning model is lower than the predetermined threshold from the altitude position identification target.

8. 5. The information processing device according to claim 4, wherein the altitude position determination unit adjusts the altitude positions determined for the plurality of pieces of radio wave condition information corresponding to a single wireless communication terminal based on a characteristic that the altitude position of the wireless communication terminal does not change continuously within a certain period of time.

9. The information processing device according to claim 8 , wherein the altitude position specifying unit adjusts the plurality of altitude positions using dynamic programming so that the altitude positions do not change continuously within a certain period of time.

10. The information processing device according to claim 8 , wherein the altitude position specifying unit adjusts the plurality of altitude positions based on a majority vote so that the altitude positions do not change consecutively within a certain period of time.

11. The information processing device according to claim 5, wherein the altitude position specifying unit specifies on which floor within a building each of the plurality of wireless communication terminals that have specified the serving cell and the neighbor cell is located.

12. A program for causing a computer to function as the information processing device according to any one of claims 1 to 11.

13. 1. A computer-implemented information processing method, comprising: an information acquisition step of acquiring a plurality of pieces of terminal-related information from a storage unit that stores radio wave condition information including cell identification information of a serving cell and a neighbor cell identified by the wireless communication terminal, and terminal-related information including altitude information identified from atmospheric pressure measured by the wireless communication terminal; a generation step of generating a learning model that uses radio wave condition information as input and advanced information as output by supervised learning using the plurality of pieces of terminal-related information as training data; An information processing method comprising:

Images