Estimation device, estimation system, estimation method, and program

Abstract

An estimation device for estimating a wireless communication quality value, the estimation device comprising: a quality information holding unit that stores position information and a wireless communication quality value for each observation point among a plurality of observation points; and an estimation processing unit that estimates a wireless communication quality value at an estimation point by performing spatial interpolation in consideration of spatial characteristics on the basis of the position information and wireless communication quality values of the observation points stored by the quality information holding unit.

Description

Estimation device, estimation system, estimation method, and program

[0001] The present invention relates to a technique for estimating a wireless communication quality value.

[0002] As a conventional technique related to estimation of a wireless communication quality value, there is a technique for predicting a wireless communication quality value at an arbitrary position based on a previously acquired actual value of wireless communication quality (for example, Non-Patent Document 1).

[0003] NTT Technical Journal, April 2020 issue, "Quality Prediction Technology for Optimal Use of Multiple Wireless Access Points," Keisuke Wakao, Kenichi Kawamura, Takatsune Moriyama, https: / / www.ntt.co.jp / journal / 2004 / JN20200411_h.html?_ga=2.187750880.999492446.1632821976-1982013439.1586410872, retrieved December 24, 2023. How Kriging Works, https: / / pro.arcgis.com / ja / pro-app / latest / tool-reference / spatial-analyst / how-kriging-works.htm, retrieved December 24, 2023.

[0004] To predict the wireless communication quality value as described above, it is necessary to obtain actual values ​​at many positions. However, obtaining actual values ​​at many positions takes a lot of time and is inefficient.

[0005] The present invention has been made in consideration of the above points, and aims to provide a technique for estimating a wireless communication quality value at an estimation point where the wireless communication quality value is unknown, based on a wireless communication quality value observed at an observation point.

[0006] According to the disclosed technology, there is provided an estimation device that estimates a wireless communication quality value, comprising: a quality information storage unit that stores position information and a wireless communication quality value of each observation point among a plurality of observation points; and an estimation processing unit that estimates the wireless communication quality value at an estimation point by performing spatial interpolation taking spatial characteristics into consideration, based on the position information and wireless communication quality value of each observation point stored in the quality information storage unit.

[0007] According to the disclosed technology, it is possible to estimate a wireless communication quality value at an estimation point where the wireless communication quality value is unknown, based on a wireless communication quality value observed at an observation point.

[0008] FIG. 1 is a diagram illustrating an example configuration of an estimation system according to an embodiment of the present invention. FIG. 2 is a diagram illustrating a normal interpolation method. FIG. 3 is a diagram illustrating spatial characteristics. FIG. 4 is a diagram illustrating differences in spatial correlation based on distance from a transmitting station. FIG. 5 is a diagram illustrating differences in spatial correlation based on direction from a transmitting station. FIG. 6 is a diagram illustrating an rθ plane. FIG. 7 is a diagram illustrating an image when an environmental variable e is applied. FIG. 8 is a diagram illustrating an example configuration 1 of the estimation device 100. FIG. 9 is a flowchart illustrating the operation of the estimation device 100. FIG. 10 is a diagram illustrating an output of the estimation device 100. FIG. 11 is a diagram illustrating an example configuration 2 of the estimation device 100. FIG. 12 is a diagram illustrating an example configuration 3 of the estimation device 100. FIG. 13 is a diagram illustrating an example hardware configuration of the estimation device 100.

[0009] Hereinafter, an embodiment of the present invention will be described with reference to the drawings. The embodiment described below is merely an example, and the embodiment to which the present invention is applied is not limited to the following embodiment.

[0010] Hereinafter, a point where a wireless communication quality value is observed is referred to as an observation point, and a point where a wireless communication quality value is estimated is referred to as an estimation point.

[0011] (System Configuration Example) Fig. 1 shows a configuration example of an estimation system according to this embodiment. As shown in Fig. 1, the estimation system includes a wireless network 10 and an estimation device 100. The wireless network 10 includes a plurality of wireless nodes such as terminals and base stations. The estimation device 100 is capable of communicating with each wireless node in the wireless network 10 using, for example, a control line.

[0012] The estimation device 100 acquires wireless communication quality values ​​(observation values) at various observation points within a target area where the wireless network 10 exists, and estimates wireless communication quality values ​​of estimation points located at positions other than the observation points by spatial interpolation based on the position information of the observation points and the observation values. Note that the estimation device 100 may be a terminal or a base station within the wireless network 10.

[0013] For example, the estimation device 100 (or a device other than the estimation device 100) can learn a machine learning model that predicts future wireless communication quality values ​​from location information using wireless communication quality values ​​at observation points and estimation points. By using such a machine learning model, for example, a terminal using wireless access can obtain a future predicted value of wireless communication quality by using location information to make an inquiry to a device having the machine learning model.

[0014] It should be noted that the "wireless communication quality" in this embodiment is not limited to a specific one, but may be, for example, radio wave propagation loss, radio wave reception strength, throughput, delay, jitter, or packet loss.

[0015] (Regarding the Interpolation Method) The estimation device 100 estimates a wireless communication quality value at an estimation point where the wireless communication quality value is unknown, from observation values ​​at a plurality of sparsely arranged observation points, using spatial interpolation (which may also be called spatial interpolation estimation).

[0016] In general, spatial interpolation uses weights that assume that the smaller the distance between the observation point and the estimation point, the smaller the difference between the observed value and the value at the estimation point. An illustration of this calculation is shown in Figure 2.

[0017] However, when it comes to wireless communication quality, the degree of similarity between the observed value and the wireless communication quality value at the estimation point cannot be determined solely by the distance between the observation point and the estimation point. Specifically, the degree of similarity between the observed value and the value at the estimation point changes depending on physical conditions such as the distance from the transmitting station that emits radio waves to the estimation point and the surrounding environment of the estimation point. Such physical conditions are called spatial characteristics. An image of such spatial characteristics is shown in Figure 3.

[0018] In this embodiment, the estimation device 100 estimates the wireless communication quality value of the estimation point by spatial interpolation that takes spatial characteristics into consideration, based on the position information and observation values ​​(wireless communication quality values) of multiple sparsely arranged observation points. By performing spatial interpolation that takes spatial characteristics into consideration, the wireless communication quality value can be estimated with high accuracy.

[0019] In this embodiment, Kriging is used as a spatial interpolation method that takes spatial characteristics into consideration. However, using Kriging as a spatial interpolation method that takes spatial characteristics into consideration is just one example, and methods other than Kriging may also be used.

[0020] In kriging, data is treated as the realization of a random phenomenon depending on the location. However, the variability of the data (spatial variation) reflects the random phenomenon of the spatial characteristics of the phenomenon, and it is expected that there will be a statistical tendency (called spatial correlation) in which values ​​tend to be closer the closer the location (although this does not necessarily mean that they will be closer).

[0021] There are various types of Kriging, such as simple Kriging, ordinary Kriging, universal Kriging, exogenous drift Kriging, cokriging, block Kriging, and indicator Kriging, and any type of Kriging may be used in this embodiment.

[0022] More specifically, in kriging, observation points s distributed in the target area are ｉ (i = 1, 2, ... N) ｉ ) is obtained, the estimated point s ０ The estimated value ^Z(s ０ ) is calculated as a linear estimator of the observed value as follows: Note that "^Z" is intended to have the "^" at the beginning of "Z".

[0023] ^Z(s ０ ) = Σλ ｉ Z (s ｉ ) The above Σ is the sum from i=1 to N. λ ｉ are weights according to the observations, and are called Kriging coefficients.

[0024] For example, in simple kriging, λ is calculated by solving simultaneous linear equations with N unknowns, whose coefficients are the values ​​of the covariance function for the distance between observation points and the distance between the observation point and the estimation point. ｉ In addition, for example, in ordinary kriging, λ can be calculated by solving simultaneous linear equations with N+1 unknowns, which have coefficients of variogram values ​​for the distance between observation points and the distance between the observation point and the estimation point, and Lagrange multiplier μ. ｉ can be calculated.

[0025] In addition, in kriging, when spatial correlation does not depend on the direction (orientation) of the vector from one point to another, but only on the distance between points, it is called isotropy, and when it depends on both distance and direction, it is called anisotropy.The spatial interpolation method using kriging itself is an existing technology, and kriging calculation tools are often included in software for analyzing spatial data.

[0026] Hereinafter, examples 1 and 2 will be described as specific examples of spatial interpolation by the estimation device 100.

[0027] Example 1 In Example 1, the estimation device 100 estimates radio wave propagation loss as an example of a wireless communication quality value. Also, in Example 1, it is assumed that a transmitting station (e.g., a base station) that outputs radio waves is present in a target area, and the estimation device 100 estimates radio wave propagation loss from the transmitting station to a point where the radio waves are received (e.g., a point where a terminal is present). The estimation uses the above-described kriging as a spatial interpolation method that takes spatial characteristics into consideration. In kriging, the estimation device 100 calculates spatial correlation (specifically, a covariance function, a variogram, etc.) from the distance between observation points, and calculates the value of the estimation point by spatial interpolation using the spatial correlation.

[0028] Radio wave propagation loss attenuates radially from the position of the transmitting station. Therefore, the degree of similarity of values ​​between two points varies depending on the distance of the two points from the transmitting station. For example, if a transmitting station uses an omnidirectional antenna to output radio waves evenly in 360 degrees and there are no obstacles such as buildings, in FIG. 4 , which shows a target area viewed from above, the similarity of radio wave propagation loss between points A and B will be greater than the similarity between points A and C. In other words, the distance on the xy plane between points A and B is greater than that between points A and C, but the difference in radio wave propagation loss values ​​between points A and B is smaller.

[0029] Furthermore, when a transmitting station uses a directional antenna to output strong radio waves in a specific direction, the distance between the transmitting station and a point and the direction from the transmitting station to the point affect the similarity of values ​​between the two points. For example, as shown in Figure 5, when a transmitting station has directionality, the similarity of radio wave propagation loss between point A and point B is greater than the similarity between point B and point C. Generally, in wireless communications, the antenna of a transmitting station has directionality.

[0030] From the above perspective, in this embodiment, focusing on the difference between the spatial correlation in the distance r direction from the transmitting station and the spatial correlation in the angle θ direction from the transmitting station, the estimation device 100 converts the coordinates of the observation point from coordinates on the xy plane to coordinates on the rθ plane, and performs spatial interpolation using a variogram using anisotropy (different correlation distances) between the r direction and the θ direction. A parameter specifying the degree of anisotropy is specified in advance. By using anisotropy in this way, it is possible to estimate a wireless communication quality value with higher accuracy.

[0031] Fig. 6 shows an image of coordinate transformation from the xy plane to the rθ plane. As shown in Fig. 6, the rθ plane is a plane in which r and θ in the polar coordinate plane are the vertical and horizontal axes, respectively. Note that in Fig. 6, the vertical axis is r and the horizontal axis is θ, but the vertical axis may also be θ and the horizontal axis may also be r.

[0032] That is, the estimation device 100 converts the position information of each observation point into position information on a plane having a first coordinate axis representing the distance from the transmitting station and a second coordinate axis representing the angle from a reference direction with the transmitting station as the origin, and performs spatial interpolation based on the converted position information.

[0033] Note that coordinate transformation onto the rθ plane is not essential, and coordinate transformation onto the rθ plane may not be performed.

[0034] Second Embodiment Next, a second embodiment will be described. In the second embodiment, the estimation device 100 also estimates radio wave propagation loss. Differences from the first embodiment will be described below.

[0035] In the second embodiment, the estimation device 100 sets the environmental variable e as the z-axis in the rθ plane, and performs spatial interpolation by weighting the z-axis direction differently from the rθ direction. FIG. 7 illustrates an example of applying the environmental variable e. The Kriging calculation method in the second embodiment corresponds to an extension of the calculation method using two variables, r and θ, to a calculation method using three variables, r, θ, and e. The Kriging calculation method using three variables is itself an existing technology. Furthermore, the number of environmental variables is not limited to one, and multiple environmental variables may be used. For example, when two environmental variables are used, Kriging calculation using four variables is performed.

[0036] The value of the environmental variable e at a certain point (which may also be called environmental information) may be, for example, one or more of the "building occupancy rate, fixture occupancy rate, average building height, and pedestrian flow density" around the point. The environmental variable e may be applied whether the target point is outdoors or indoors. Furthermore, whether the target point is inside or outside a building may be reflected in the environmental variable and applied to spatial interpolation. The environmental variable e may also be applied to the xy plane.

[0037] An example of the device configuration common to the first and second embodiments and the processing procedure thereof will be described below.

[0038] (Device Configuration Example 1) Fig. 8 shows Device Configuration Example 1. As shown in Fig. 8, an estimation device 100 in Device Configuration Example 1 includes a quality information holding unit 110, a position information holding unit 120, an environmental information holding unit 130, an estimation processing unit 140, an input unit 160, and an output unit 170.

[0039] The quality information storage unit 110 stores the position information and wireless communication quality value of each observation point in the target area. The position information storage unit 120 stores the position information of the transmitting station. The environmental information storage unit 130 stores environmental information such as the building occupancy rate, furniture occupancy rate, average building height, and pedestrian flow density described in the second embodiment. Note that the quality information storage unit 110 may store the environmental information.

[0040] The estimation processing unit 140 performs spatial interpolation using Kriging, as described in Examples 1 and 2. The input unit 160 receives input of information. The output unit 170 outputs the estimation result obtained by the estimation processing unit 140.

[0041] An example of the operation of the estimation device 100 will be described below in accordance with the procedure shown in Fig. 9. Here, the example of the operation will be described assuming that a terminal inquires of the estimation device 100 about an estimated value at a certain position, and the terminal obtains an estimation result from the estimation device 100.

[0042] In S101, the position information of the estimation target is input from the input unit 160. The position information is, for example, x and y coordinate values ​​in the target area = (x ０ , y ０ ) When the operation of the second embodiment is performed, (x ０ , y ０ ) together with (x ０ , y ０ Environmental information in ０ may be input.

[0043] In S102, the estimation processing unit 140 performs spatial interpolation using the position information and wireless communication quality values ​​of each observation point stored in the quality information storage unit 110 to obtain (x ０ , y ０When the operation of the second embodiment is performed, the environment information of each observation point is obtained from the information stored in the environment information storage unit 130, and the wireless communication quality value at each observation point is estimated using the position information, wireless communication quality value, and environment information. ０ , y ０ ) and estimate the wireless communication quality value at

[0044] In S103, the output unit 170 outputs (x ０ , y ０ The wireless communication quality value is transmitted to the terminal that made the inquiry, for example.

[0045] Next, an example of operation will be described, which assumes that an estimated value for the entire area is inquired and an estimated result is obtained. Again, this will be described with reference to FIG.

[0046] In S101, a wireless communication quality value for each observation point is input from the input unit 160. For example, if there are three observation points, (x １ , y １ ) and its value z １ , (x ２ , y ２ ) and its value z ２ , (x ３ , y ３ ) and its value z ３ When the operation of the second embodiment is performed, environmental information at the position may be input together with the position information. The information input in S101 is stored in the quality information holding unit 110.

[0047] In addition, if the necessary information is already stored in the quality information storage unit 110 and the environmental information storage unit 130, in S101, only an instruction to estimate the entire area may be input from the input unit 160.

[0048] In S102, the estimation processing unit 140 estimates wireless communication quality values ​​at a plurality of estimation points, which are points other than the observation points, in the target area by performing spatial interpolation using the position information and wireless communication quality values ​​of each observation point stored in the quality information storage unit 110. The estimation points may be determined in advance or may be input in S101.

[0049] In S103, the output unit 170 outputs a map representing wireless communication quality values ​​at the observation points and the estimation points. Fig. 10 shows (a) an image of the observation points at the input and (b) an image of the observation points and the estimation points at the output. At each point in (b), the wireless communication quality value may be represented by a numerical value, or by a color and its shading.

[0050] (Device Configuration Example 2) When performing spatial interpolation taking spatial characteristics into consideration, coordinate transformation onto the rθ plane described in Example 1 or use of environmental information described in Example 2 are not essential, and these may not be performed. In that case, the estimation device 100 may not include the position information holding unit 120 and the environmental information holding unit 130, as shown in FIG. 11 .

[0051] (Device Configuration Example 3) The estimation device 100 may be a terminal that communicates with a base station in the wireless network 10. An example configuration of the estimation device 100 in this case is shown in Fig. 12. As shown in Fig. 12, this configuration corresponds to the device configuration example 1 in which the input unit 160 and the output unit 170 are replaced with a quality acquisition unit 150.

[0052] In this configuration, the quality acquisition unit 150 acquires a wireless communication quality value for each position and stores it in the quality information storage unit 110. The estimation processing unit 140 estimates a wireless communication quality value at a desired position (e.g., the position of the terminal itself t seconds later) by spatial interpolation using the information stored in the quality information storage unit 110. For example, the terminal (estimation device 100) can perform an operation such as switching the communication method from method A (e.g., 5G) to method B (e.g., wireless LAN) according to the estimated wireless communication quality value.

[0053] (Hardware Configuration Example) The estimation device 100 described in this embodiment can be realized, for example, by causing a computer to execute a program. This computer may be a physical computer or a virtual machine on the cloud.

[0054] That is, the estimation device 100 can be realized by using hardware resources such as a CPU and memory built into a computer to execute a program corresponding to the processing performed by the estimation device 100. The program can be recorded on a computer-readable recording medium (such as a portable memory) and can be saved or distributed. The program can also be provided via a network such as the Internet or email.

[0055] Fig. 13 is a diagram showing an example of the hardware configuration of the computer. The computer in Fig. 13 includes a drive device 1000, an auxiliary storage device 1002, a memory device 1003, a CPU 1004, an interface device 1005, a display device 1006, an input device 1007, an output device 1008, and the like, all of which are interconnected via a bus B. The computer may further include a GPU.

[0056] The program that realizes the processing on the computer is provided by a recording medium 1001, such as a CD-ROM or a memory card. When the recording medium 1001 storing the program is set in the drive device 1000, the program is installed from the recording medium 1001 to the auxiliary storage device 1002 via the drive device 1000. However, the program does not necessarily have to be installed from the recording medium 1001, but may be downloaded from another computer via a network. The auxiliary storage device 1002 stores the installed program as well as necessary files, data, etc.

[0057] The memory device 1003 reads and stores the program from the auxiliary storage device 1002 when an instruction to start the program is received. The CPU 1004 realizes functions related to the estimation device 100 in accordance with the program stored in the memory device 1003. The interface device 1005 is used as an interface for connecting to a network, etc. The display device 1006 displays a GUI (Graphical User Interface) or the like according to the program. The input device 1007 is composed of a keyboard, mouse, buttons, a touch panel, etc., and is used to input various operation instructions. The output device 1008 outputs calculation results.

[0058] (Effects of the embodiment, etc.) As described above, the technology described in this embodiment makes it possible to estimate a wireless communication quality value at an estimation point where the wireless communication quality value is unknown, based on a wireless communication quality value observed at an observation point.

[0059] This eliminates the need to obtain actual values ​​at estimation points, making it possible to obtain wireless communication quality values ​​at a wide range of points.

[0060] In addition, the estimation results can be used for route selection for self-driving cars, video codec control for remote control images, and network control (priority allocation of RBs and antenna tilt control).

[0061] Furthermore, by using the estimation results as input data for machine learning models, it is possible to increase the amount of training data required for training.Furthermore, by applying this technology to experimental data such as radio wave propagation loss and received power, it is possible to increase the amount of experimental data needed to build radio wave propagation models and improve the accuracy of the models.

[0062] The following additional notes are provided regarding the above-described embodiments.

[0063] <Supplementary Notes> (Supplementary Item 1) An estimation device for estimating a wireless communication quality value, comprising: a memory; and at least one processor connected to the memory, wherein the memory stores position information and a wireless communication quality value of each of a plurality of observation points, and the processor estimates the wireless communication quality value at the estimation point by performing spatial interpolation taking spatial characteristics into consideration, based on the position information and wireless communication quality value of each observation point stored in the memory. (Supplementary Item 2) The estimation device according to Supplementary Item 1, wherein the estimation processing unit converts the position information of each observation point into position information on a plane having a first coordinate axis representing a distance from a transmitting station and a second coordinate axis representing an angle from a reference direction having the transmitting station as its origin, and performs the spatial interpolation based on the converted position information. (Supplementary Item 3) The estimation device according to Supplementary Item 2, wherein the processor performs the spatial interpolation using anisotropy, which indicates that spatial correlation in the direction of the first coordinate axis is different from spatial correlation in the direction of the second coordinate axis. (Supplementary Item 4) The processor performs the spatial interpolation by further using environmental information at each observation point. The estimation device according to any one of Supplementary Items 1 to 3. (Supplementary Item 5) An estimation system having an estimation device that estimates a wireless communication quality value and a terminal, wherein the estimation device comprises: an input unit that inputs a query from the terminal, a quality information storage unit that stores position information and a wireless communication quality value of each observation point among a plurality of observation points, an estimation processing unit that estimates the wireless communication quality value at the estimation point related to the query by performing spatial interpolation taking spatial characteristics into consideration, based on the position information and wireless communication quality value of each observation point stored in the quality information storage unit, and an output unit that outputs the wireless communication quality value estimated by the estimation processing unit to the terminal. (Supplementary Item 6) An estimation method executed by an estimation device that estimates a wireless communication quality value, wherein the estimation device includes a quality information storage unit that stores position information and a wireless communication quality value of each observation point among a plurality of observation points, and the estimation method includes an estimation processing step that estimates the wireless communication quality value at the estimation point by performing spatial interpolation taking spatial characteristics into consideration, based on the position information and wireless communication quality value of each observation point stored in the quality information storage unit.(Supplementary Item 7) A non-transitory storage medium storing a program for causing a computer to function as each unit in the estimation device according to any one of Supplementary Items 1 to 4.

[0064] Although the present embodiment has been described above, the present invention is not limited to such a specific embodiment, and various modifications and changes are possible within the scope of the gist of the present invention described in the claims.

[0065] 10 Wireless network 100 Estimation device 110 Quality information storage unit 120 Location information storage unit 130 Environmental information storage unit 140 Estimation processing unit 150 Quality acquisition unit 160 Input unit 170 Output unit 1000 Drive device 1001 Recording medium 1002 Auxiliary storage device 1003 Memory device 1004 CPU 1005 Interface device 1006 Display device 1007 Input device 1008 Output device

Claims

1. An estimation device for estimating a wireless communication quality value, comprising: a quality information holding unit that stores position information and wireless communication quality values of each observation point at a plurality of observation points; and an estimation processing unit that estimates the wireless communication quality value at a estimation point by performing spatial interpolation considering spatial characteristics based on the position information and wireless communication quality values of each observation point stored in the quality information holding unit.

2. The estimation device according to claim 1, wherein the estimation processing unit converts the position information of each observation point into position information in a plane having a first coordinate axis representing the distance from a transmission station and a second coordinate axis representing an angle from a reference direction with the transmission station as the origin, and performs the spatial interpolation based on the converted position information.

3. The estimation device according to claim 2, wherein the estimation processing unit performs the spatial interpolation using anisotropy indicating that the spatial correlation in the direction of the first coordinate axis is different from the spatial correlation in the direction of the second coordinate axis.

4. The estimation device according to claim 1, wherein the estimation processing unit further performs the spatial interpolation using the environmental information at each observation point.

5. An estimation system having an estimation device for estimating a wireless communication quality value and a terminal, wherein the estimation device comprises: an input unit that inputs an inquiry from the terminal; a quality information holding unit that stores position information and wireless communication quality values of each observation point at a plurality of observation points; an estimation processing unit that estimates the wireless communication quality value at an estimation point related to the inquiry by performing spatial interpolation considering spatial characteristics based on the position information and wireless communication quality values of each observation point stored in the quality information holding unit; and an output unit that outputs the wireless communication quality value estimated by the estimation processing unit to the terminal.

6. An estimation method executed by an estimation device for estimating a wireless communication quality value, wherein the estimation device comprises a quality information holding unit that stores position information and wireless communication quality values of each observation point at a plurality of observation points, and an estimation processing step of estimating the wireless communication quality value at an estimation point by performing spatial interpolation considering spatial characteristics based on the position information and wireless communication quality values of each observation point stored in the quality information holding unit.

7. A program for causing a computer to function as each unit in the estimation device according to any one of claims 1 to 4.

Images