CN117642990A - Radio wave propagation estimation device and radio wave propagation estimation method - Google Patents

Abstract

The radio wave propagation estimating device uses identification information capable of identifying an object including a structure and map data including the structure, and extracts a structure of the object which can be seen from both a direct or virtual transmission point of a radio wave and a direct or virtual reception point of the radio wave based on the identification information. The radio wave propagation estimating device estimates propagation characteristics of the radio wave using information indicating characteristics of the structure of the extracted object.

Description

Radio wave propagation estimation device and radio wave propagation estimation method

Technical Field

The present disclosure relates to a radio wave propagation estimation device and a radio wave propagation estimation method that estimate radio wave propagation characteristics in a wireless communication system.

Background

In order to estimate radio wave propagation characteristics in a radio communication system such as a mobile communication system, a ray tracing method based on a geometrical optical diffraction theory is widely used. The ray tracing method is a method of tracing a path of rays from a transmission point to a reception point, and an image method and a radiation emission method are known.

The imaging method is a method of extracting structures from which rays are reflected and diffracted in advance and searching for rays having the shortest path from a transmission point to a reception point through the structures. The image method has an advantage in that rays from a transmission point to a reception point can be precisely obtained, but has the following problems: that is, if the number of surfaces (building walls, etc.) and edges of the structure to be considered and the maximum number of reflections and diffractions increase, the amount of calculation processing increases exponentially.

Therefore, in order to reduce the amount of computation processing by the image method, it has been proposed and verified to calculate a combination of a surface and an edge to trace rays only for a structure in an ellipse having a transmission point and a reception point as focuses, or to search for a building visible from the transmission point or the reception point to trace rays (patent document 1).

In addition, in order to improve the accuracy of estimation of radio propagation characteristics, a method using a neural network or deep learning has been developed (non-patent documents 1 and 2).

Prior art literature

Patent literature

Patent document 1: japanese patent application laid-open No. 2019-122008

Non-patent literature

Non-patent document 1: jinjing Zhang, aucun, unfortunately, analyzer, "a study related to estimation of wave propagation using deep learning", IEICE technical report, AP2017-165, pp81-86, month 1 of 2018

Non-patent document 2, jinjing Zhang, shallow-well Miao, relation between map parameters and estimation accuracy, which are one kind of study-input based on electric wave propagation estimation by deep learning using CNN, to IEICE technical report, vol.118, no.246, AP2018-89, pp.1-6, 10 months in 2018

Disclosure of Invention

Problems to be solved by the invention

However, the image method still has the following problems. For example, in the case where the normal reflection point Ir does not belong to the plane of the building during the tracking of the primary reflected ray, when the ray Tx-Ir-Rx is discarded, an error occurs in the calculation of the power of the radio wave from the transmission point Tx to the reception point Rx.

In addition, in the tracking of the reflected rays, the incidence angle of the radio wave or the area of a visible portion of the surface of the building may be reduced by the 1 st fresnel zone. In this case, since the radio wave is scattered not only in the normal direction but also in various directions, an error occurs in calculation of the power of the radio wave coming to the reception point Rx.

On the other hand, in order to reduce such errors, it is conceivable to track the diffracted rays based on a plurality of edges of the target building and collect the power coming to the reception point Rx, but the calculation amount required for the calculation of the power of the reception point Rx increases.

Accordingly, the following disclosure has been made in view of such a situation, and an object thereof is to provide a radio wave propagation estimation device and a radio wave propagation estimation method capable of effectively reducing errors and calculation amounts of power calculation of a radio wave coming to a receiving point.

Means for solving the problems

One embodiment of the radio wave propagation estimation device includes: an extraction unit that extracts, from map data including a reception point of a radio wave transmitted from a transmission point and including identification information for an object including a structure, a structure of the object that can be seen from both the reception point and the transmission point based on the identification information; and an estimating unit that estimates propagation characteristics of the electric wave based on the structure extracted by the extracting unit and a structure that may be scattering of the electric wave in a surrounding environment.

Further, according to one embodiment of the radio wave propagation estimating apparatus, the radio wave propagation estimating apparatus includes: an extraction unit that extracts two-dimensional data from three-dimensional map data including a reception point of a radio wave transmitted from a transmission point; and an estimating unit that estimates propagation characteristics of the electric wave in a surrounding environment along with diffraction calculation of the electric wave based on the two-dimensional data extracted by the extracting unit.

One embodiment of the radio wave propagation estimation device includes: an extraction unit that applies a convolutional neural network to map data including a reception point of a radio wave transmitted from a transmission point and including at least one of identification information and distance information for an object including a structure, thereby extracting, from the map data, a 1 st parameter indicating a structure that may be scattering of the radio wave in a surrounding environment including the reception point; and an estimating unit that estimates propagation characteristics of the radio waves by applying a fully-connected neural network to the 1 st parameter extracted by the extracting unit and a 2 nd parameter indicating a configuration of a wireless communication system that performs wireless communication between the transmitting point and the receiving point.

One embodiment of the radio wave propagation estimation device includes: an extraction unit that uses identification information capable of identifying an object including a structure and map data including the object, and extracts a structure of the object that can be seen from both a direct or virtual transmission point of an electric wave and a direct or virtual reception point of the electric wave based on the identification information; and an estimating unit that estimates a propagation characteristic of the electric wave using information indicating the feature of the structure extracted by the extracting unit.

Drawings

Fig. 1 is a functional block diagram showing an example of the configuration of the radio wave propagation estimation device according to embodiment 1.

Fig. 2 is a diagram for explaining the problem of the conventional ray tracing method.

Fig. 3 is a diagram for explaining the problem of the conventional ray tracing method.

Fig. 4 is a diagram for explaining the problem of the conventional ray tracing method.

Fig. 5 is a diagram for explaining the problem of the conventional ray tracing method.

Fig. 6 is a diagram for explaining the problem of the conventional ray tracing method.

Fig. 7 is a diagram for explaining the problem of the conventional ray tracing method.

Fig. 8 is a diagram for explaining the problem of the conventional ray tracing method.

Fig. 9 is a diagram for explaining the problem of the conventional ray tracing method.

Fig. 10 is a diagram showing the positional relationship of the identification number, the receiving point, and the transmitting point for the building.

Fig. 11 is a diagram showing a two-dimensional image seen from the transmission point Tx and a two-dimensional image seen from the reception point Rx in fig. 10.

Fig. 12 is a diagram comparing the method of the present embodiment with a conventional calculation by a ray tracing method in three dimensions.

Fig. 13 is a functional block diagram of embodiment 1.

Fig. 14 is a diagram showing examples of "an image seen from a transmitting point" and "an image seen from a receiving point".

Fig. 15 is a diagram showing a structural example of DNN applied to the city structural parameter extracting unit and the propagation calculating unit 102f according to embodiment 1.

Fig. 16 is a diagram showing an example of an image seen from a transmitting point Tx and a receiving point Rx according to embodiment 2.

Fig. 17 is a view showing an image in which a wall surface can be seen in common from a transmitting point Tx and a receiving point Rx according to embodiment 2.

Fig. 18 is a diagram showing an example of grouping of radio wave scattering objects.

Fig. 19 is a diagram showing an example of assigning one color to one surface as a whole and an example of dividing the surface into a plurality of areas (grids) and assigning different colors for each grid.

Fig. 20 is a diagram showing an example in the case where an electric wave is once reflected between a transmission point Tx (antenna) and a reception point Rx (antenna), and an example in the case where an electric wave is repeatedly reflected between a transmission point Tx (antenna) and a reception point Rx (antenna).

Fig. 21A is a diagram showing an example of an image of a receiving point, the ground, a structure, and the like, as seen from a transmitting point.

Fig. 21B is a diagram showing an example of a determination of a line of sight based on a conventional ray tracing method.

Fig. 22 is a diagram showing an example of a calculation method of a diffraction wave.

Fig. 23A is a diagram showing a configuration example of the radio wave propagation estimating device according to embodiment 2.

Fig. 23B is a diagram showing a configuration example of the radio wave propagation estimating device according to embodiment 2.

Fig. 24 is a diagram showing an operation flow of the preprocessing unit 231 of the calculation unit 230.

Fig. 25 is a diagram showing an operation flow of the post-processing unit 232 in the case of using the result of the previous preprocessing.

Fig. 26A is a diagram showing the operation flow of the preprocessing unit 231 and the post-processing unit 232 in the case of directly from the initial calculation.

Fig. 26B is a diagram showing the operation flow of the preprocessing unit 231 and the post-processing unit 232 in the case of directly from the initial calculation.

Fig. 27 is a diagram showing an example of calculation of the received power and the propagation loss by the radio wave propagation estimating device 200.

Fig. 28 is a diagram showing an example of the calculation of the channel characteristics by the radio wave propagation estimating apparatus 200.

Fig. 29 is a diagram showing an operation flow of the radio wave propagation estimating device 200 for acquiring an object that is not present in the radio wave scattering contributing object DB.

Fig. 30A is a diagram showing an example of an object that does not exist in the electric wave scattering contributing object DB and an example of a car-human model.

Fig. 30B is a diagram showing an example of an object that does not exist in the electric wave scattering contributing object DB and an example of a car-human model.

Fig. 30C is a diagram showing an example of an object that does not exist in the electric wave scattering contributing object DB and an example of a car-human model.

Fig. 31A is a diagram showing a control image based on propagation characteristics in a reception point of an RIS or the like.

Fig. 31B is a diagram showing a control image based on propagation characteristics in a reception point of RIS or the like.

Fig. 32 is a block diagram showing an example of a hardware configuration of the radio wave propagation estimation device according to the embodiment.

Detailed Description

Several embodiments of the present invention are described in detail below with reference to the drawings. The following description of the embodiments is not limited to the inventions described in the claims.

The following embodiments can be arbitrarily combined and implemented.

[ embodiment 1 ]

1. Summary of the embodiments

In the field of radio communication such as mobile communication, an estimation expression of a propagation loss (sometimes referred to as "path loss") with high accuracy is required for efficient cell design. As a method for obtaining the estimated expression, there are, for example, a method based on electromagnetic field theory and a method based on measurement data (may also be referred to as "measured data" or "measured value"). Among these methods, the latter method (i.e., a method based on measurement data) is mostly used in the technical field of wireless communication.

For example, in a range assumed in a wireless communication system, a propagation loss is measured by changing a frequency, a base station antenna height, a distance between transmissions and receptions, a propagation environment, and the like, and a relationship between these parameters and data obtained by the measurement is analyzed to generate an estimated expression of radio wave propagation.

When multiple regression analysis is used for analysis of the relationship between the parameters and the measured data, the "input parameters" and the "functional form" constituting the estimated expression are determined. The "input parameters" represent information associated with the propagation environment, for example, if urban, information associated with urban construction. For convenience of explanation, "information related to a city structure" may be referred to as a city structure parameter or a city structure related parameter. Examples of city structure-related parameters include building occupation area ratio, road width, road angle, road side building height, average building height, and building height near a base station.

Wherein the decision of the parameters and the functional form of the estimated expression depends largely on the sensing of the person generating the estimated expression and the data used. Further, for example, the parameters and the functional form of the estimated expression need to be determined for each case where the relation between the surrounding building height and the antenna height of the base station and/or the mobile station is different. For example, parameters and functional forms used for the estimation formula may be changed for each expansion scene of the cell, as in the relation between the macro cell and the small cell. Therefore, it is not easy to generate a unified estimation formula by any of the methods described above.

Thus, in one embodiment of the present embodiment, a case of using a deep neural network (deep neural network; DNN) which is an example of a Neural Network (NN) is studied with respect to "deep learning" in a functional form and parameters used for determining an estimation formula of radio wave propagation.

"deep learning" is one of the methodologies of machine learning (machine learning) using a multi-layer neural network (multilayered neural network). For example, by flexibly applying abundant computer resources and big data on the internet to the multi-layer neural network, processing performance such as voice recognition, image recognition, natural language processing, and the like can be improved. One of the reasons for improving the processing performance is that deep learning has the ability to learn the feature quantity itself for solving the problem from the input data.

As described above, since the DNN can learn the feature quantity itself required for solving the problem from the input data, it is possible to learn the city construction parameter itself used for radio wave propagation estimation by using map data such as a house map for the input data of the DNN. Similarly, DNN is also used for determining the "functional form" constituting the estimation expression, and an estimation expression having higher radio propagation estimation accuracy than the conventional estimation expression can be obtained.

In the learning of DNN, data transmitted (or reported) from each Mobile Station (MS) such as a mobile phone to a communication network can be used. As an example of data reported from an MS to a communication network, information indicating explicitly or implicitly (or directly or indirectly) the reception power of the MS, the location of the MS, and/or the radio propagation environment such as a Base Station (BS) to which the MS is connected (accessed) is cited. The MS is an example of a User Equipment (UE).

The data reported from the mobile station to the communication network may be collected in, for example, a base station and/or a higher-order station of the base station constituting the communication network. The server computer of the data center (which may also be referred to as a "cloud computer") may correspond to a superordinate station. For example, data collected in a data center may be stored and managed as so-called "big data" in a storage device of the data center.

Therefore, "use data reported from the MS in the learning of DNN" may be replaced with "use large data related to the radio wave propagation environment in the learning of DNN". In addition, regarding data that is not available from the network, for example, simulation results based on ray tracing may be used to supplement.

However, when map data such as a house map is used as input data of DNN and it is desired to learn city structure parameters or the like used for radio propagation estimation, there is a problem that many iterations are generated until target parameters are obtained, and convergence is difficult. For example, when a plurality of map data items viewed from a transmission point, a reception point, a space above or the like of radio waves are used as input data, there is a problem that convergence is difficult until learning results are obtained.

In one embodiment of the present embodiment, the map data includes identification information (color, identification number, or the like) and/or distance information for an object including a structure. More specifically, a two-dimensional image such as a color or an identification number is applied to a face (and/or an edge) of a building or the like, and a portion of the face and the edge of the building that can be seen from both a transmission point and a reception point is extracted and used for propagation estimation (calculation). Further, a two-dimensional image (map) in which distance information to the surface and the edge of the building visible from the transmission point and the reception point is calculated is also used flexibly to extract the distance to the portion of the surface and the edge visible from each other and to use the extracted distance for propagation estimation (calculation). In this way, even when a plurality of map data are used, since identification information or distance information is provided to the same structure, the iteration of learning is easy to converge, and a radio wave propagation estimation result with higher accuracy than in the past can be obtained as compared with the case of the same resource.

In addition, in one embodiment of the present invention, the map data includes identification information (color, identification number, or the like) for an object including a structure when performing radio wave propagation estimation calculation, not limited to the case of using the neural network described above. Then, based on the identification information, the structure (surface, edge, etc.) of the object that can be seen from both the reception point and the transmission point is extracted from the map data. Then, based on the extracted structure (surface, edge, etc.), a structure that may be scattering of electric waves in the surrounding environment is estimated, and propagation characteristics of the electric waves are estimated. Accordingly, the structure (surface, edge, etc.) of the object that can be seen from both the reception point and the transmission point (that is, the structure that may be scattering of the radio wave in the surrounding environment) can be estimated based on the identification information, and therefore the calculation amount can be reduced, the speed can be increased, and the radio wave propagation estimation result can be obtained with higher accuracy than in the conventional case of the same resource.

In addition, in one embodiment of the present invention, the propagation characteristics of the electric wave, which accompanies the diffraction calculation of the electric wave in the surrounding environment, are estimated by extracting two-dimensional data from the map data when the electric wave propagation estimation calculation is performed, not limited to the case of using the neural network described above. Therefore, the calculation amount can be significantly reduced as compared with the diffraction calculation in three dimensions. Further, since diffraction calculation can be simplified and calculation of reflection, transmission, and the like can be performed, as a whole propagation estimation calculation, it is expected that a radio wave propagation estimation result with higher accuracy than before can be obtained as compared with the case of the same resource.

2. Structural example of radio wave propagation estimation device

Fig. 1 is a functional block diagram showing an example of the configuration of a radio wave propagation estimation device 100 according to an embodiment. The "estimation device" may be replaced with a "simulator", "model", or "tool", etc.

As shown in fig. 1, the radio wave propagation estimating device 100 may be exemplarily provided with a control unit 102 for extracting parameters of a city structure or the like or performing propagation estimation calculation, a storage unit 106 including various parameters, a database, or the like, and an output unit 108 for outputting calculation results of propagation characteristics, or the like. As shown in the figure, the control unit 102 includes a setting unit 102a, a face edge recognition unit 102b, a transmission point view search unit 102c, a reception point view search unit 102d, a propagation calculation information extraction unit 102e, and a propagation calculation unit 102f.

The setting unit 102a is a setting means for setting various data and parameters used for estimation calculation of propagation characteristics. For example, the setting unit 102a may set identification information or distance information for an object including a structure in map data including a transmission point and a reception point of radio waves. For example, the setting unit 102a may set a calculation target region, a propagation calculation parameter (e.g., a frequency band of a radio wave to be estimated, etc.), a rule for setting identification information (e.g., identification color or identification number) of a surface or edge of a building, etc. The identification information or distance information for map data and the like may be manually set, or may be automatically added from a color image or a 3D image read by a camera, a distance sensor, or the like.

The face-edge recognition unit 102b is a recognition means for recognizing a face and/or an edge of a building or the like. Fig. 2 to 9 are diagrams for explaining the problem of the conventional ray tracing method.

Various methods for reducing the amount of computation have been developed in the past. For example, as shown in fig. 2 and 3, in order to reduce the amount of calculation by the image method, a method of tracing rays by determining a combination of a face and an edge only with respect to a structure within an ellipse having a transmission point and a reception point as focuses, a method of tracing rays by searching for a building visible from the transmission point or the reception point, or the like has been proposed and verified (ACE tool).

In addition, the conventional method has a problem that an error occurs in the estimation accuracy. For example, in the case of the ray tracing method, for example, as shown in fig. 4 and 5, in the case of primary reflection, after an image point Tx 'of a transmission point Tx is generated for a surface of a building, the image point Tx' is connected to a reception point Rx with a line, whereby a normal reflection point Ir of the surface is derived. In the case where the normal reflection point Ir does not belong to the plane, or in the case where the ray Tx-Ir-Rx is blocked by the plane of another building although the normal reflection point Ir belongs to the plane, the ray Tx-Ir-Rx is discarded.

However, in the real world, since the surfaces seen from both Tx and Tx are utilized, the radio waves can come to Rx by utilizing the waves scattered not only in the normal direction but also in the respective directions. Therefore, errors are generated in the calculation of the power of the wave coming to Rx due to the reject rays Tx-Ir-Rx.

In addition, in the ray tracing method, as shown in fig. 6, the reflection level in the normal reflection direction is calculated considering that the part of the surface of the building which is visible from both the transmission point Tx and the reception point Rx is larger than the 1 st fresnel zone.

[ number 1]

1 st Fresnel radius

However, in the real world, as shown in fig. 7, the incident angle and the area of the visible portion of the building surface may be smaller than the 1 st fresnel zone, and in this case, the scattering is not only in the normal direction but also in the respective directions, and the level of the scattering is smaller than the level of the reflection by the ray tracing method in the normal reflection direction, and thus an error may occur.

In addition, conventionally, as shown in fig. 8 and 9, when the normal reflection point Ir does not belong to the plane, or when the normal reflection point Ir belongs to the plane but the ray Tx-Ir-Rx is blocked by the plane of another building, the ray Tx-Ir-Rx is discarded, but in reality, there are radio waves scattered by the building plane, and the like, and therefore, calculation errors occur with this.

Therefore, conventionally, in order to reduce calculation errors, rays diffracted by a plurality of edges of a target building are tracked, and calculation of collecting power to a receiving point is performed, but there is a disadvantage in that the amount of calculation increases. Further, there are such problems as follows: the operation of determining the condition of the reflection point and tracking the ray takes time and effort on the basis of the generated image point and the reflection point, and similarly, the operation of obtaining the diffraction point and tracking the ray and the calculation of the complex expression based on the diffraction take time and effort.

As described above, in the present embodiment, the following estimation method is proposed, in which the calculation accuracy is considered, and the calculation amount (time) can be significantly reduced as compared with the calculation amount (time) of the conventional ray tracing method.

That is, the face-edge recognition unit 102b uses the recognition information (for example, color or number) and correlates (corresponds) the recognition information so that the faces and edges of the respective buildings can be recognized in the two-dimensional image. Fig. 10 is a diagram showing the positional relationship among the identification number, the reception point, and the transmission point for the building. As shown in fig. 10, in addition to the identification color (or identification number) of each of the faces (or edges) of the building or the like, in the processing described later, a two-dimensional image of the face (projected) of the building or the like seen from the transmission point Tx and the reception point Rx is generated, and if the colors (or numbers) of each element of the image from the transmission point Tx and the image from the reception point Rx are examined, the face (or edge) of the building seen from the transmission point Tx or the reception point Rx is known.

For example, the identification color may be a 16-ary number of RGB. In addition, the identification color is not limited to the case of coloring by a person. For example, the color of the building may be used from the topography of the storage unit 106, the stereo photograph data of the building stored in the structure database, or the like, or the color or number may be given to the stereo structure data of the building by an artificial intelligence or the like. The surface-edge recognition unit 102b stores a database of surfaces and edges of the object structure to which the identification information is given in the storage unit 106.

The transmission point view search unit 102c is a search means for searching the surface and edge of a building or the like visible from the transmission point. Here, as shown in the upper diagram of fig. 11, after a two-dimensional image seen from the transmission point Tx is depicted in fig. 10, the image is stored in the storage section 106. As shown in the lower left diagram of fig. 14, in fig. 10, an image (map) of the distance to the building seen from the transmission point Tx is also generated, and the image (map) is stored in the storage unit 106. Further, these two-dimensional images may be automatically acquired by a camera or a distance sensor, and stored in the storage unit 106. The colors (or numbers) of the elements of these images are examined and stored in a "database of faces and edges visible from the transmission point".

The receiving point view search unit 102d is search means for searching the surface and the edge of a building or the like visible from the receiving point. Here, as shown in the lower diagram of fig. 11, after a two-dimensional image seen from a receiving point Rx is depicted in fig. 10, the image is stored in the storage section 106. As shown in the lower right diagram of fig. 14, in fig. 10, an image (map) of the distance to the building seen from the reception point Rx is also generated, and the image (map) is stored in the storage unit 106. Further, a two-dimensional image may be automatically acquired by a camera or a distance sensor, and the image may be stored in the storage unit 106. The colors (or numbers) of the elements of these images are examined and stored in a "database of faces and edges visible from the receiving points".

The propagation calculation information extracting unit 102e extracts and processes combinations of visible surfaces and edges based on information on the visible surfaces (and edges) of the building from the transmission point Tx or the reception point Rx, in accordance with conditions and information required for propagation calculation. Therefore, the area of the visible surface and the visible edge required for the calculation of the scattered (reflected or diffracted) wave by the propagation calculating unit 102f described later can be specified. In addition, there is a method of calculating the area of each viewable portion using the sum of elements (the number of pixels) of the image corresponding to the portion and the positional relationship between the origin of the radio wave irradiation and the observation point.

Here, the origin and the observation point are irradiated with the plane projection wave perpendicular to the edge for the edge that can be seen, and the irradiation angle and the diffraction angle are calculated. In this way, in the diffraction calculation at any edge of the propagation calculation performed by the propagation calculation unit 102f, which will be described later, the obtained irradiation angle and diffraction angle are substituted into a simple relational expression determined in advance, and the attenuation amount by diffraction can be obtained. In order to perform high-speed calculation, the same attenuation amount may be given in advance without obtaining the irradiation angle and diffraction angle.

Among them, all or part of the setting unit 102a, the face-edge recognition unit 102b, the transmission-point-view search unit 102c, the reception-point-view search unit 102d, and the propagation-calculation information extraction unit 102e function as an "extraction unit".

The propagation calculating unit 102f functions as an "estimating unit" that estimates propagation characteristics of the electric wave based on the data extracted by the "extracting unit". For example, the propagation calculating unit 102f can estimate a structure that may be scattering of radio waves in the surrounding environment, based on the extracted structure (surface, edge, etc.), and can estimate propagation characteristics of the radio waves.

More specifically, the propagation calculating unit 102f can estimate propagation characteristics of the electric wave with respect to the extraction result of the combination of the visible surface and the edge that satisfies the propagation condition. The propagation calculating unit 102f can also simplify calculation of the irradiation angle and the scattering angle by reducing the area of the surface that can be seen only in common from the electric wave irradiation origin (transmission point) and the observation point (reception point) to perform scattered wave calculation. The propagation calculating unit 102f calculates an irradiation angle and a diffraction angle of an edge of the building, which can be seen in common from the electric wave irradiation origin (transmission point) and the observation point (reception point), in order to calculate the diffraction wave. The propagation calculating unit 102f calculates the received power of each ray (path) and calculates the total of the received powers.

Since the estimation method according to the present embodiment can simplify the calculation, it is unnecessary to request an image point during reflection, calculate a reflection point, and determine a reflection point. In addition, it is possible to eliminate the need to obtain diffraction points during diffraction and various conditions required for obtaining complex calculation formulas. Further, by adding a color (or identification number) to the surface or edge and processing only the projected image, conditions and information necessary for propagation calculation can be extracted at high speed. Further, the scattering level is calculated only by using a radar equation or the like for a surface that can be seen from both the origin (transmission point) and the observation point (reception point) of the radio wave irradiation, and the calculation of the reflection of the surface and the diffraction of the edge can be omitted. Therefore, the calculation amount can be reduced, and high-speed estimation can be realized.

Further, according to the estimation method of the present embodiment, the scattering level is calculated by using a radar equation or the like for a portion of the surface that can be seen from both the origin (transmission point) and the observation point (reception point) of the radio wave irradiation, and the scattering can be calculated not only in the normal direction but also in any direction. Even when the portion of the surface that can be seen is smaller than the 1 st fresnel radius, the scattering level can be calculated (in the case of the ray tracing method, the above condition cannot be calculated). Therefore, it is expected to obtain a radio wave propagation estimation result with higher accuracy than the conventional one.

The propagation calculating unit 102f may estimate propagation characteristics of the electric wave, which are calculated by diffraction of the electric wave in the surrounding environment, based on the extracted two-dimensional data (for example, data of a vertical plane of the edge which can be seen from both Tx and Rx). Fig. 12 is a diagram comparing the method of the present embodiment with a conventional calculation by a ray tracing method in three dimensions. As shown in fig. 12, in the conventional ray tracing method, the calculation of diffraction in the three-dimensional space is relatively complicated with respect to a building or the like from a transmission point to a reception point. In the present embodiment, as described above, the calculation of diffraction in the two-dimensional space is performed using map data in two dimensions from the vertical plane of the edge that can be seen in common from Tx and Rx, and therefore the calculation is simplified. That is, the attenuation is obtained by projecting the edge on the vertical plane and substituting the projected edge into a simple relational expression for determining the irradiation angle and the diffraction angle. Since the diffracted wave order is originally weak, the estimation accuracy can be maintained even if the simple relational expression described above is used. Further, the amount of calculation can be reduced, and high-speed estimation can be realized.

The propagation calculating unit 102f may estimate propagation characteristics of the electric wave using a neural network or the like. In particular, in the present embodiment, the propagation calculating unit 102f may function as an "estimating unit" that estimates propagation characteristics of the radio wave by applying the fully-connected neural network to the 1 st parameter extracted by the "extracting unit" and the 2 nd parameter indicating the configuration of the wireless communication system that performs wireless communication between the transmission point and the reception point.

The "extraction unit" is input with map data such as a house map, for example, and can extract city construction parameters for propagation estimation from the map data via a Neural Network (NN).

As a non-limiting example, the city construction parameters may include one or more parameters selected from the building occupation area ratio, the road width, the road angle, the road side building height, the average building height, and the building height near the base station.

For example, the city construction parameter may take different values according to the relationship of the location of the base station and/or the mobile station to the location of the surrounding building. In other words, the city structure parameter is an example of the 1 st parameter indicating a structure that may cause scattering of radio waves in the surrounding environment including the reception point (mobile station or base station) of the radio waves transmitted from the transmission point (base station or mobile station).

As a non-limiting example, convolutional NN (convolutional neural network; CNN) may be applied to NN.

For example, as described later in fig. 13, the CNN has a structure including a convolution layer for repeatedly locally extracting feature values from input data (for example, images) and a pooling layer for locally aggregating the feature values. The pooling layer (or sub-sampling layer), for example, compresses (sometimes also referred to as "downsampling") the input data from the convolutional layer into a form that is easy to process.

In CNN, parameters of a convolution filter (hereinafter, sometimes simply referred to as "filter") are shared in all places of an input image. Therefore, the number of parameters can be reduced as compared with a simple fully connected neural network (Fully Connected Neural Network; FNN).

Further, the number of parameters can be further reduced by the pooling layer, and invariance to parallel movement of the input image can be imparted stepwise.

CNNs can be conceptually interpreted as networks as follows: when the input data is image data, the co-occurrence of feature amounts adjacent to each other in different scales is calculated while the resolution of the image data is slightly reduced, and information effective for classification or recognition is selectively transferred to the next layer (higher layer).

The propagation calculating unit 102f receives, for example, the city structure parameter and the system parameter extracted by the city structure parameter extracting unit, and estimates radio wave propagation in the wireless communication system by NN using the two parameters. The system parameter is an example of the 2 nd parameter showing the configuration of the radio communication system for performing radio communication between the transmission point and the reception point. The "estimation" of the wave propagation may be replaced by "analysis", "analysis" or "simulation", etc.

For example, the system parameter may include information indicating one or more parameters selected from the distance between the transmission point and the reception point, that is, the inter-transmission/reception distance, the frequency of the radio wave to be estimated for the propagation of the radio wave, the antenna height of the base station, the antenna height of the mobile station, and the presence or absence of a line of sight (line of sight) between the transmission points.

The system parameter may be determined according to a system and/or a scenario as an estimation object of the propagation of the electric wave. For example, in the case where the macrocell environment is an estimation object, it can be assumed that the base station antenna is disposed at a position higher than the surrounding building (i.e., base station antenna height > surrounding building height).

The "frequency" is determined, for example, by the arrangement of the base station corresponding to any one of the transmission point and the reception point (in other words, the cell design), and therefore can be understood to be included in an example of the system parameter determined from the positional relationship between the transmission point and the reception point.

For example, the function of estimating the propagation loss from the input parameter is expressed by applying FNN to NN in the propagation calculating unit 102 f. In the FNN, the nodes of each layer are connected to all the nodes of the next layer. The FNN functions as a recognition unit that classifies a combination of feature amounts detected until reaching a lower layer as a specific prediction result. In addition, in NNs, the "nodes" in each layer are also sometimes referred to as "neurons" or "units".

The propagation calculating unit 102f may include an error calculating unit and a weight updating unit, which are not shown. The error calculation unit and the weight update unit are used, for example, to learn the weight coefficient (may also be referred to as "weight parameter") of the DNN formed by the CNN and the FNN.

In CNN, the weight parameter can be understood as, for example, a filter coefficient equivalent to a convolution filter applied in a convolution layer (C layer). In the FNN, the weight parameter can be understood as a coefficient corresponding to a weight combination between layers (in other words, between nodes of different layers) constituting the FNN. Illustratively, an error back propagation method (back propagation) may be applied in the learning of the weight parameters.

The error back propagation method is a method of calculating the update amount of the weight parameter of each layer using a value obtained from a higher layer side (output layer side), and propagating the value of the update amount of the weight parameter of each layer in a lower layer (input layer) direction while calculating the value of the update amount of the weight parameter of each layer.

For example, the error calculation unit calculates an error between an estimated value of the radio wave propagation characteristic (hereinafter, may be simply referred to as "propagation characteristic") obtained by the propagation calculation unit 102f and the radio wave propagation characteristic obtained from the actual measurement value (or may be a theoretical value). The weight parameter of DNN is repeatedly updated by the weight updating unit until the calculated error converges to the given convergence condition.

The error calculated by the error calculation unit can be represented by an error function using the weight parameter as an argument, for example. The error function is sometimes also referred to as a loss function, a cost function, or an (estimated) objective function, etc.

For example, a gradient descent method (Gradient Descent Method) may be applied in the updating of the weight parameters. In the gradient descent method, for example, the weight parameter of the DNN is repeatedly updated by bringing the gradient value of the error function close to zero, in other words, bringing the partial differential value of the error function close to zero.

The middle section of fig. 13 is a functional block diagram showing the functional block diagram illustrated by the upper section of fig. 13 more abstractly than the upper section of fig. 13. In the middle stage of fig. 13, the weight parameters of the FNN and the CNN can be updated sequentially from the higher layer toward the lower layer (right to left in the middle stage of fig. 13) for the weight parameters in each layer of the DNN by the error back propagation method described above.

In learning of weight parameters using the gradient descent method, batch learning (batch learning) may be applied, or small batch learning (mini-batch learning) may be applied. The small-lot learning performs learning by dividing data for learning (which may also be referred to as "learning data", "teacher data", or "training data") into a plurality of units.

According to the small-lot learning, it is possible to realize shortening of learning time, improvement of utilization efficiency of computer resources, reduction suppression of learning efficiency, and the like, as compared with the lot learning. For example, in the case where large data concerning the radio wave propagation environment described above is used for learning data, it is useful to apply small-lot learning. In addition, a unit obtained by dividing learning data is sometimes referred to as "training wheel number" (epoch).

The items described in japanese patent application laid-open No. 2019-122008 can be appropriately introduced with respect to a calculation method using an input data example, a neural network, or the like used in the radio wave propagation estimation device 100 of the present embodiment.

The leftmost map data in the lower stage of fig. 13 is a diagram showing an example of data (for example, map data) used for estimating propagation characteristics for a pair (pair) of a position of a certain BS and a position of a certain MS. In the figure, the location of the BS corresponds to a transmission point, and the location of the MS corresponds to a reception point.

As shown in the lower stage of fig. 13, map data of a specific area (for example, a 128m×128m rectangular area) including the position of the MS corresponding to the reception point is input to the city construction parameter extracting section.

The specific region including the position of the MS is not limited to a rectangular region, and may be circular, elliptical, or other shapes. The size of the specific region including the position of the MS may be appropriately adjusted according to, for example, the load expected by the calculation.

As shown in the lower stage of fig. 13, the map data input to the city structural parameter extracting unit may be divided into, for example, a mesh shape. The size of one mesh as a unit of division is, for example, 2m×2m. However, the unit of division is not limited to 2m×2m, and may be appropriately adjusted according to the load expected by the operation, for example.

The map data input to the city construction parameter extraction unit may include "building map data", "line-of-sight map data", "image seen from a transmitting point", and "image seen from a receiving point". Hereinafter, the "building map data" and the "sight line map data" may be simply referred to as "building map" and "sight line map", respectively.

The "building map" is, for example, "map data composed of height information of buildings in each grid", and a grid in which no building exists is defined as 0m. On the other hand, the "line-of-sight map" is, for example, "map data composed of information on the presence or absence of line-of-sight with respect to each mesh of the BS. Among them, fig. 14 is a diagram showing examples of "an image seen from a transmitting point" and "an image seen from a receiving point".

As shown in fig. 14, the map data (1) of the present embodiment may be an image viewed from a transmission point, to which a structure is provided with a recognition color, (2) may be an image viewed from a transmission point, to which a structure is provided with distance information, (3) may be an image viewed from a reception point, to which a structure is provided with a recognition color, and (4) may be an image viewed from a reception point, to which a structure is provided with distance information.

In the prior art, the estimation accuracy of the propagation characteristics (propagation loss) is poor (RMS error is about 10 dB). The reason for this is mainly: limiting the range processed in the calculation, not all buildings contributing to the scattered wave can be processed; and the exact scattered power cannot be calculated without using information on the size of the contributing building surface. According to the present embodiment, in order to improve the accuracy, as shown in fig. 15, the images (1) to (4) described below are input to the CNN, and after the feature quantity (the size of the wall and edge portions visible from Tx and Rx and the distance to the portion) is extracted by the CNN, the control unit 102 (extraction unit) inputs the feature quantity to the FNN.

(1) Two-dimensional image (colored image) of face and edge of building or the like seen from transmitting point Tx

(2) Distance from transmission point Tx to position corresponding to each element of the image of (1) (also sometimes average distance per surface or per edge)

(3) Two-dimensional image (colored image) of face and edge of building and the like seen from receiving point Rx

(4) Distance from receiving point Rx to position corresponding to each element of image of (3) (also sometimes average distance per surface or per edge)

3. Operational example

In the case where the operations of the radio wave propagation estimation device 100 are considered in the learning phase and the estimation phase after the completion of the learning, the big data described in the learning phase can be used for the learning of CNN and FNN. The term "stage" may be replaced by other terms such as "phase", "stage", "process", and "procedure".

The system parameters input to the propagation calculating unit 102f in the learning stage and the estimation stage are, for example, information indicating one or both of "distance between transmission and reception D" (meters) and "presence or absence of line of sight", which correspond to the distance between the BS and the MS.

Further, the logarithm (log D) of the inter-transmission distance may be used for the information indicating the "inter-transmission distance D (m)". In addition, a line of sight (LOS) flag may be used for information indicating "presence or absence of line of sight". The line-of-sight flag is an example of a parameter indicating the presence or absence of line of sight by each numerical value (for example, "1" and "0").

The input data described using fig. 3 is input data in the case where the MS corresponds to a reception point. In the case where the BS corresponds to the reception point, map data of a specific area including the BS, which is the location of the reception point, is also input to the city structure parameter extraction unit (see fig. 15, etc.), whereby city structure parameters for the surrounding environment including the BS, which is the reception point, can be extracted. Therefore, the radio wave propagation characteristics of the surrounding environment including the BS can be estimated.

< structural example of DNN >

Next, a structural example of DNN applied to the city structural parameter extracting unit and the propagation calculating unit 102f according to the present embodiment will be described with reference to fig. 15. As shown in fig. 15, CNN is applied to the city structure parameter extracting unit, and FNN is applied to the propagation calculating unit 102 f.

(structural example of CNN)

Illustratively, CNNs have a convolutional layer (layer C), a pooling layer (layer P), and a fully-connected layer (layer F).

For example, the layer C performs a convolution operation using a convolution filter on the input data, thereby extracting a feature amount of a local pattern in the input data.

The P-layer pools (in other words, aggregates) the values of small regions (sometimes referred to as "local regions") in the input data into one value. In pooling, a slight positional offset is allowed.

The F layer performs classification (or identification) based on the feature amounts obtained in the C layer and the P layer.

In fig. 15, as a non-limiting example, the city structure parameter extracting section has four C layers (C0 to C3), two P layers (P0 and P1), and two F layers (F0 and F1). However, the number of each of the C layer, the P layer, and the F layer is not limited to the number illustrated in fig. 15, and may be appropriately adjusted according to, for example, the load expected by the operation.

As shown in fig. 15, the city construction parameter extraction unit receives input data of, for example, "256×256" size. In addition, "256×256" indicates the number of samples for each map. Further, "4" which is additionally recorded on the input data of the "256×256" size indicates the map number. For example, in the case of a transmission point, it can be understood that the four maps correspond to "a map of an R component of an image color of a building seen from the transmission point", "a map of a G component of an image color of a building seen from the transmission point", "a B component of an image color of a building seen from the transmission point", and "a map of a distance to a building seen from the transmission point". In the case of the receiving point, it can be understood that it corresponds to "a map of R component of image color of building seen from the receiving point", "a map of G component of image color of building seen from the receiving point", "B component of image color of building seen from the receiving point", and "a map of distance to building seen from the receiving point".

In the C0 layer, the city structure parameter extraction unit performs, for example, a convolution operation of the filter size= "5×5×map number" and the filter number= "48" on the input data of the size "256×256×map number".

For example, the city construction parameter extraction section multiplies the element value of the filter represented by the three-dimensional array of "5×5×map number" (in other words, "matrix") by the element value of the input data represented by the matrix of "256×256×map number" (in other words, sample value) for each element value, and adds (sums) the multiplication results.

This multiplication and addition are performed while sliding the filter so as to cover the "256×256" range in the two-dimensional direction, thereby obtaining a convolution operation result for the two-dimensional array of "252×252". For example, map data having a size of "252×252" and a number equal to the filter number=48 can be obtained by the convolution operation.

In addition, a filter that slides with respect to input data is sometimes referred to as a "sliding window", "kernel", or "feature detector", or the like.

Next, the city structure parameter extraction unit performs a pooling operation of the size of "4×4" as the size of the local area, for example, on the output of the C0 layer (i.e., "48" pieces of map data having the size of "252×252") in the P0 layer.

By the pooling operation of the P0 layer, 48 pieces of map data in which the size of "252×252" is compressed to the size of "63×63" can be obtained. Further, any one of Max pooling, which selects the maximum value in the local area, and Average pooling, which obtains the Average value in the local area, may be applied to the pooling operation. Hereinafter, for convenience of explanation, an example in which Average pooling is applied will be described.

The city structure parameter extraction unit performs, for example, a convolution operation of the filter size= "4×4×48" and the filter number= "24" on the output of the P0 layer (that is, 48 map data having a size of "63×63") at the C1 layer. By the convolution operation of the C1 layer, map data having a size of "60×60" and a number equal to the filter number=24 can be obtained.

Next, the city structure parameter extraction unit performs a pooling operation of the size of "4×4" as the size of the local area, for example, on the output of the C1 layer (that is, 24 map data having the size of "60×60") in the P1 layer. By the pooling operation of the P1 layer, 24 pieces of map data in which the size of "60×60" is compressed to the size of "15×15" can be obtained.

In addition, the city structure parameter extraction unit performs, for example, a convolution operation of the filter size= "4×4×24" and the filter number= "12" on the output of the P1 layer (i.e., 24 map data having a size of "15×15") in the C2 layer. By the convolution operation of the C2 layer, map data having a size of "12×12" and a number equal to the filter number=12 can be obtained.

Next, the city structure parameter extraction unit performs, for example, a convolution operation of the filter size= "4×4×12" and the filter number= "6" on the output of the C2 layer (i.e., 12 pieces of map data having a size of "12×12") in the C3 layer. By the convolution operation of the C3 layer, map data having a size of "9×9" and a number equal to the filter number=6 can be obtained.

The city structure parameter extraction unit performs the convolution operation and the pooling operation described above on the 2 atlas (4 atlas) of "the map of the structure seen from the transmitting point Tx" and "the map of the structure seen from the receiving point Rx", and then sums up 486 (9×9×6) samples of the output of the C3 layer to obtain 972 (9×9×6×2) data.

Then, the city construction parameter extraction unit inputs 972 pieces of sample data to the F0 layer fully connected to 972F 1 layers, and outputs 972 pieces of data (in other words, 972 pieces of data) from the F1 layers as a final output based on CNN.

As described above, the city structural parameter extracting unit can obtain 972 kinds of city structural parameter values from the map data of the set of the number of samples=256×256×map number×2.

In addition, the input data size for CNN, and the filter sizes and/or the filter numbers applied to the C layers (C0 to C3), the P layers (P0 and P1), and the F layers (F0 and F1), respectively, are not limited to the above-described values. For example, these values may be appropriately adjusted according to the load expected by the operation or the like.

(structural example of FNN)

On the other hand, as illustrated in fig. 15, the FNN applied to the propagation calculating section 102f may be constituted by three layers of an input layer, an intermediate layer, and an output layer.

The city structure parameters obtained by the city structure parameter extraction unit and the system parameters (for example, the inter-transmission/reception distance: log d and the presence or absence of line of sight: LOS) described above are input to the input layer. Therefore, the number of input parameters for the FNN constituting the propagation calculating section 102f is "974". In addition, the number of nodes of the intermediate layer and the output layer are "3" and "1", respectively.

In this FNN structure, an estimated value of propagation characteristics (e.g., propagation loss) can be obtained by two full-connection NN based on two F layers (F2 and F3).

(activation function)

Next, an activation function (activation function) F (x) applied to the C layer and the F layer is described. In addition, "x" represents the index of the local area. For example, the ReLU function may be applied to the C layers (C0-C3), respectively. In addition, "ReLU" is an abbreviation of "Rectified Linear Unit (rectifying linear unit)".

In contrast, for example, the Sigmoid function may be applied to F0 to F2 layers (F0 to F3), respectively.

Such as identity functions: f (x) =x may be applied to the F3 layer.

< learning method of DNN >

As described above, the error back propagation method can be applied in the learning of the weight parameters. Among them, there is a tendency that the larger the scale of DNN and the larger the amount of learning data, the longer the time taken for learning of DNN.

Thus, by applying the small-batch gradient descent method to the learning of DNN, the learning time of DNN can be shortened. The small-lot gradient descent method is a method of updating weight parameters by a gradient descent method every N (< N) samples (in other words, small lot) of learning data of the number of pointers to the number N of samples. To improve the convergence of the weight parameters Adam (adaptive moment estimation: adaptive moment estimation), one of the optimization methods, can be used in combination.

< pooling operation of P layer >

As described above, the P layer downsamples and compresses information in order to deform the input data into an easy-to-process form. By compressing the information, effects such as an improvement in robustness against a minute positional change, suppression of overdriving, and/or reduction in computation cost can be expected.

In MAX pooling, the maximum value is selected for the local region. For example, in DNN, the value of a node having the largest value among a plurality of nodes of a lower layer connected to a node(s) of a higher layer becomes the output value of the node of the higher layer.

In other words, in MAX pooling, feature amounts of a plurality of nodes of a lower layer are processed as "maximum values" among the plurality of nodes, and sub-sampling is performed. From the viewpoint of the neuron pool, MAX pooling can be said to be a structure in which the maximum value possessed by the pool is "intensively" output from the neuron pool composed of a plurality of neurons of a lower layer, and the concentration of the neuron pool is reproduced.

For example, if only a part of the data set having a characteristic of having a valid value is an object, MAX pooling can efficiently separate useless inputs as compared with average pooling, and the value of the node at the lower layer is substantially "0". In other words, MAX pooling can be said to have higher performance as a noise filter than average pooling.

On the other hand, in average pooling, the values output from the plurality of nodes of the lower layer to the nodes of the higher layer are average values, not maximum values. From the viewpoint of the neuron pool, if the sum of neurons of a lower layer does not fluctuate or fluctuates to a negligible extent, it sufficiently functions as a noise filter.

In the average pooling, the value of the neuron having a smaller value than the maximum value is not ignored as in the MAX pooling, and the possibility of trapping the local solution problem (local minimum problem) can be reduced as compared with the MAX pooling, for example. The "local solution problem" refers to a phenomenon in which the system converges on a partial optimal solution and cannot reach a true optimal solution, for example.

The operation of the radio wave propagation estimation device 100 having the above-described configuration, particularly, learning, estimation calculation, and the like, can be performed by a known method described in japanese patent application laid-open No. 2019-122008, and the like.

As an example, the radio wave propagation estimating device 100 calculates the city structure parameter from the acquired map data by, for example, the city structure parameter extracting unit, and calculates the propagation characteristic by the propagation calculating unit 102f using the city structure parameter, the system parameter, and the like in fig. 15 as described with reference to fig. 15.

When the propagation characteristics are calculated, the radio wave propagation estimating device 100 may calculate, for example, an error between an estimated value of the calculated propagation characteristics (for example, propagation loss) and the propagation characteristics obtained from the actual measurement values (or may be theoretical values) related to the propagation characteristics by an error calculating unit. The actual measurement value (or theoretical value) related to the propagation characteristics can be obtained from, for example, information (in other words, large data) related to the radio propagation environment between the BS and the MS stored in the center DB 400.

Then, the error calculation unit determines whether or not the calculated error converges on the given convergence condition. If the calculated error does not converge on the given convergence condition, for example, the weight updating unit updates the weight parameters of the FNN and CNN constituting the DNN by the error back propagation method.

After the weight parameter is updated, the city structure parameter extraction unit and the propagation calculation unit 102f recalculate the city structure parameter and the propagation characteristic, respectively, and the error calculation unit recalculates the error to determine the convergence condition.

The above processing is repeatedly performed until it is determined that the convergence condition is satisfied, whereby the weight parameters of the FNN and CNN constituting the DNN are updated. When it is determined that the convergence condition is satisfied, the radio wave propagation estimation device 100 may end the learning stage process.

In the estimation stage, the radio wave propagation estimation device 100 sets, for example, an estimation condition of radio wave propagation. Illustratively, the setting of the estimation condition may be performed in the city construction parameter extraction unit.

Illustratively, the setting data of the estimation condition may include information corresponding to a transmission point of the BS to which the MS can connect and information corresponding to a reception point of the MS.

Further, for example, the radio wave propagation estimating device 100 acquires map data (for example, a building map and a line-of-sight map) input to the CNN from the DB of map data as illustrated in fig. 15.

The radio propagation estimating device 100 sets parameters (for example, weight parameters) for radio propagation estimation for each layer constituting the DNN, for example. Illustratively, the setting may be performed in the city construction parameter extraction unit.

In other words, the city structural parameter extraction unit may have a function as a setting unit that sets parameters for radio wave propagation estimation.

Next, the radio wave propagation estimation device 100 calculates city structure parameters from the acquired map data by, for example, the city structure parameter extraction unit that has already been learned in the learning stage, as described with reference to fig. 15.

When calculating the city structure parameter, the radio wave propagation estimating device 100 calculates the propagation characteristics using the city structure parameter, the system parameter, and the like, for example, by the propagation calculating section 102f that has been learned in the learning stage, as described in fig. 15.

As described above, according to the radio wave propagation estimation device 100 according to one embodiment, the city structure parameter extraction unit using CNN can extract and determine the city structure parameter contributing to the radio wave propagation estimation by limiting the city structure parameter extraction unit to a structure or the like visible from the reception point and/or the transmission point based on map data such as a house map. Therefore, the determination of the city structure parameters to be used for the radio wave propagation estimation can be simplified, and the time taken for the radio wave propagation estimation can be shortened, so that the radio wave propagation estimation can be estimated with higher accuracy than the calculation in the same resource.

Further, by using information (e.g., big data) about the radio propagation environment reported from the MS to the network in learning DNNs (CNN and FNN) for the city structural parameter extraction unit and the propagation calculation unit 102f, for example, it is possible to improve the accuracy of learning DNNs. Since the learning accuracy of DNN can be improved, the estimation accuracy of propagation characteristics can also be improved.

[ embodiment 2 ]

Next, an embodiment in which identification information (identification number or color) is given to the surface (and edge) of the structure to be used for propagation calculation will be described further. Hereinafter, the portions different from embodiment 1 will be mainly described, and the same portions will be omitted.

Specifically, a radio wave propagation estimating apparatus (and a radio wave propagation estimating method) capable of calculating reflected (scattered) waves in any direction on a wall surface of a structure or the like will be described.

(feature 1)

In order to search for the wall surface of a structure or the like in which electric waves are reflected (scattered) in any direction, different identification numbers or colors are given to the surfaces of all objects including the structure (which may include road signs, utility poles, street lamps, daily necessities).

The radio wave propagation estimating device according to the present embodiment draws an image of a structure or the like seen from both a radio wave transmitting point and a radio wave receiving point, and counts the number of pixels of each color (corresponding to a wall surface) that can be seen in common. The counted number of pixels corresponds to the area of the wall surface portion that can be seen in common from the transmission point Tx and the reception point Rx, and changes according to the distance between the transmission point Tx and the reception point Rx and the wall surface and the angular relationship. Therefore, the result obtained by multiplying the numbers of pixels of the images of both the transmission point Tx and the reception point Rx is proportional to the intensity (reception power) of the reception electric wave in the wall surface.

Fig. 16 shows an example of an image seen from the transmission point Tx and the reception point Rx. Fig. 17 shows an image of a wall surface that can be seen in common from the transmission point Tx and the reception point Rx. The image may be generated using a computer, but may be a still image or a moving image of an object such as a structure obtained using a camera or the like.

As shown in fig. 16 and 17, the radio wave propagation estimating device detects a color (here, a wall surface of a structure such as a building is assumed) that can be seen in common from both the transmission point Tx and the reception point Rx, and counts the number of pixels of the detected color that can be seen in common. The number of pixels seen from the transmission point Tx and the reception point Rx (the area corresponding to the viewable portion) can be expressed as follows.

The case seen from the transmitting point Tx: tx_pixels

Seen from the reception point Rx: rx_pixels

As shown in fig. 17, the surfaces of structures to which the same color is applied (wall surfaces of buildings, etc.) can be seen in common from the transmitting point Tx and the receiving point Rx. The intensity of the electric wave received in the reception point Rx via such a wall surface (which can be interpreted as reception power) can be expressed as follows.

Intensity of received electric wave ∈tx_pixels × rx_pixels

Alternatively, the intensity of the received electric wave may be expressed as follows.

Intensity pr=k tx_pixels rx_pixels of received electric wave

In addition, the proportionality coefficient (k) is not limited to a constant. The scaling factor may be a function of antenna gain, distance, angle, constant of the material of the scattering object, frequency, etc.

Among them, a method of calculating (estimating) the intensity of the received electric wave using a color of a surface of a structure that can be seen in common from the transmission point Tx and the reception point Rx, which can be uniquely identified, is called a color image method.

The identification information for identifying the object (specifically, the surface of the object) including the structure may be color or number. Furthermore, RBG may typically be used for expression forms of color, but other forms (e.g., CYMK) may also be used.

Further, the structure may typically include, but is not limited to, buildings (buildings) including buildings such as buildings. Specifically, all objects contributing to reflection (scattering) of the electric wave may be contained, and for example, road signs, utility poles, street lamps, daily necessities, and the like may be contained. In addition, not only stationary objects but also moving objects (e.g., vehicles, etc.) may be included.

Fig. 18 shows an example of grouping of radio wave scattering objects. As shown in fig. 18, the identification information (number or set of colors) is distinguished into a ground and an electric wave scattering object, and the electric wave scattering object can be distinguished into a stationary object and a moving object.

The stationary object may include the above-described structure, utility pole, street lamp, daily appliance, etc., and the moving object may include a vehicle, etc. (may include a train).

As a method of assigning identification numbers or colors (identification information), a method of randomly assigning each surface is exemplified. Alternatively, the objects belonging to the same group may be randomly allocated to the surfaces of the objects, based on the classification of the radio scattering objects, or the like. In addition, the allocation of the identification information may not necessarily be random.

In addition, one identification number or color is generally assigned to the entire one surface, but in some cases, one surface may be divided into a plurality of areas (for example, a mesh shape), and the identification number or color may be assigned for each mesh.

Fig. 19 shows an example of assigning one color to one surface as a whole, and an example of dividing the surface into a plurality of areas (grids) and assigning different colors for each grid.

As shown in fig. 19 (left side), a specific color can be generally assigned to one surface as a whole. On the other hand, as shown in fig. 19 (right side), the surface may be divided into a plurality of areas (grids), and different colors may be assigned to each grid.

When a specific color is assigned to the whole of one surface, the whole of the surface is colored, and therefore, even when the portions actually seen from the transmission point Tx and the reception point Rx do not overlap, it is determined that the whole of the surface can be seen in common, and an error occurs in calculation of the radio wave scattering level.

On the other hand, when the surface is divided into a plurality of areas (grids), there is no grid (right side in fig. 19) that can be seen in common from both the transmission point Tx and the reception point Rx when viewed in terms of colors in units of grids, and therefore, errors do not occur in the calculation of the above-described radio wave scattering level.

Typically, the transmitting point Tx and the receiving point Rx refer to the set position of the transmitting antenna and the set position of the receiving antenna, but are not necessarily limited to the set positions of the antennas. For example, when the radio wave is reflected (scattered) a plurality of times, virtual transmission points and virtual reception points may be included.

Fig. 20 shows an example of a case where an electric wave is once reflected between a transmission point Tx (antenna) and a reception point Rx (antenna), and an example of a case where an electric wave is repeatedly reflected between a transmission point Tx (antenna) and a reception point Rx (antenna).

As shown in fig. 20, when an electric wave (alternatively, a wireless signal) reaches a reception point Rx (antenna) from a transmission point Tx (antenna) via a plurality of walls, the wall 1, the wall N-1, and the like can be interpreted as virtual reception points and/or transmission points.

(feature 2)

In order to determine the line-of-sight state between the transmitting point Tx and the receiving point Rx at high speed, the ground may be given an identification number or a color. Specifically, the radio wave propagation estimating device generates an image of the reception point Rx, the ground, the structure, and the like seen from the transmission point Tx, and can determine that the line-of-sight state is between the transmission point Tx and the reception point Rx when the color (or identification number) of the pixel to which the reception point Rx belongs matches the color (or identification number) of the ground that can be seen.

Fig. 21A shows an example of an image of a receiving point, the ground, a structure, and the like, as seen from a transmitting point. Fig. 21B shows an example of a determination of a line of sight based on a conventional ray tracing method.

The ground may be divided by grids, and the identification number or color may be assigned to each grid. Thus, only one image viewed from the transmission point is analyzed, and the line-of-sight state of the reception point in the entire evaluation target area can be determined.

As shown in fig. 21A, the ground may be divided into a plurality of areas, and a color (or an identification number) capable of uniquely identifying each area is assigned. In fig. 21A, the structure and the background are assigned the same color (here, black) different from the ground.

Fig. 21A shows an image (a topographic map) seen from a transmission point, and the reception point shown in fig. 21A is determined to be in a line-of-sight state with the transmission point.

On the other hand, as shown in fig. 21B, in the line-of-sight determination by the conventional ray tracing method, it takes a relatively long time until the determination is made because it is sequentially checked whether or not the line is blocked by each structure for each receiving point (RX 1, 2, 3) and the transmitting point.

(feature 3)

If time or the like is taken into consideration, there is naturally a limit to the maximum number of reflections (scattering) that can be calculated by the radio wave propagation estimating apparatus (maximum N times). Therefore, it is assumed that even N reflections (scattering) cannot calculate a reception point of the reception intensity, that is, a so-called plane of an object that can be seen in common is not found. In order to cope with such a case, the radio wave propagation estimating apparatus may calculate propagation characteristics such as reception intensity using a complementary calculation method.

Specifically, the radio wave propagation estimation device may use the following complementary calculation method.

(1) Setting a default value of the reception intensity

(2) Changing (increasing) the height of the transmission point so that the reception point of the object can be calculated, obtaining statistics of differences between calculation results in other reception points that can be calculated, that is, so-called average value and standard deviation value, and calculating (estimating) the reception intensity before the change using the average value and standard deviation value and the calculation result after the change in the reception point

(3) Calculating diffraction waves

(4) Using ray tracing

(5) Calculating power in receiving points of an object by using other receiving points capable of being calculated as virtual transmitting points

Fig. 22 shows an example of a calculation method of the diffraction wave. Since the power of the diffracted wave is originally weak, a simple calculation method is only required. As shown in fig. 22, the attenuation amount may be obtained by projecting the attenuation amount onto a vertical plane at the edge of the structure and substituting the attenuation amount into a simple relational expression in which the irradiation angle (θt) and the diffraction angle (θr) of the radio wave are determined.

(configuration example of radio propagation estimation device)

Fig. 23A and 23B show a configuration example of the radio wave propagation estimating device according to the present embodiment. As shown in fig. 23A and 23B, the radio wave propagation estimating device 200 includes an external input unit 210, a storage unit 220, a calculation unit 230, and an output unit 240.

The external input unit 210 can input information not stored (held) in the storage unit 220 to the storage unit 220 or the like from the outside. For example, the external input unit 21 can input image data acquired by a camera or a laser scanner to the storage unit 220 or the like.

The storage unit 220 can store various Databases (DBs) such as map data. For example, the storage unit 220 may store geographic information DB, radio wave scattering contributing object DB, identification number/color DB, reception point DB, transmission point DB, and the like.

The calculation unit 230 includes a preprocessing unit 231 and a post-processing unit 232. The preprocessing unit 231 sets the reception points within a specified range, and performs generation and analysis of images of the reception points in advance. Specifically, the preprocessing unit 231 may have the following functions.

(1) Specifying conditions such as an evaluation area and an interval between receiving points and using DB

(2) Using a receiving point acquisition DB (e.g., a road network), receiving points are calculated from the evaluation area and the intervals of the receiving points

(3) Coloring or associating identification numbers with surfaces of objects contributing to radio wave scattering

(4) Generating an image seen from a receiving point

(5) Analysis of images seen from a receiving point

(6) Storing the image seen from the receiving point and the analysis result in the DB

In the present embodiment, the preprocessing unit 231 may function as a part of the extraction unit. Specifically, as described above, the preprocessing unit 231 can extract the structure of the object that can be seen from the direct or virtual receiving point of the radio wave based on the identification information, using the identification information that can identify the object including the structure and the map data including the structure.

The structure of the object may be an external shape of the structure, and in particular, may include a surface structure of a building or the like composed of a plurality of surfaces (may be a plane surface, but may also be a curved surface).

As described above, the preprocessing section 231 (extraction section) can extract the structure (face) of the object based on the color (or number) of the face of the recognition object. Further, as shown in fig. 19, the surface of the object may be divided into a plurality of areas (meshes). Different identification information (number or color) may be assigned for each region (grid).

In addition, in a broad sense, the structure of the object may include the surface of the surrounding structure. The surface may be replaced with the ground. The ground surface can be interpreted as an area where a structure such as a building is not provided.

The post-processing unit 232 can perform setting of a transmission point, generation/analysis of an image seen from the transmission point, and the like. The post-processing unit 232 compares and combines the analysis results of the images of the reception points, and calculates and synthesizes the radio wave scattering levels on the respective surfaces of the structure. Specifically, the post-processing unit 232 may have the following functions.

(1) Conditions for post-processing (setting of scattering times, various settings of transmission points, antenna pattern, etc.) are set

(2) Generating an image seen from a transmission point

(3) Analyzing images seen from a transmission point

(4) Comparing and combining the analysis result of the image seen from the transmitting point with the analysis result of the image seen from the receiving point (counting the number of pixels of a common plane seen from the transmitting point and the receiving point, etc.)

(5) Calculation and synthesis of radio wave scattering level in each surface of structure

In the present embodiment, the functions (1) to (4) of the post-processing unit 232 may function as a part of the extracting unit. Specifically, as described above, the functions (1) to (4) of the post-processing unit 232 can extract the structure of the object that can be seen from the direct or virtual transmission point of the radio wave based on the identification information, using the identification information that can identify the object including the structure and the map data including the structure. Therefore, the functions (1) to (4) of the post-processing unit 232 and the preprocessing unit 231 can function as an extracting unit. The propagation calculating section of the post-processing section 232 may function as an estimating section. Specifically, the propagation calculating unit can estimate propagation characteristics of the electric wave using information indicating characteristics of structures of objects such as structures extracted in the functions (1) to (4) (extracting unit) of the preprocessing unit 231 and the post-processing unit 232. The structural features may include the type of object, the structure of the surface in appearance (number, size, direction, position (distance), area of the portion that can be seen, etc.), the material of the surface, and the like.

As described above, the propagation calculating unit (estimating unit) may estimate the propagation characteristics using the number of pixels (corresponding to the area of the visible portion) constituting the surface of the object visible from both the transmission point and the reception point. Specifically, as shown in fig. 16 and 17, the propagation calculating unit may calculate the reception intensity from the number of pixels of the image of the surface that can be seen in common from the transmission point and the reception point, and estimate the propagation characteristics.

The propagation calculating unit may change the height of the transmission point when the object visible from both the transmission point and the reception point cannot be found. In general, it is effective to increase the height for the change of the height, but it may be possible to include a case of decreasing the height.

As described above, the propagation calculating unit may estimate the propagation characteristic before the change in the altitude from the difference between the calculation results of the reception characteristics at all other reception points before and after the change in the altitude. Specifically, the height of the transmission point may be changed (increased) so that the reception point of the object can be calculated, statistics of differences between calculation results in other reception points that can be calculated by both before and after the change, that is, so-called average value and standard deviation value, may be obtained, and the reception intensity before the change may be calculated (estimated) using the average value, the standard deviation value, and calculation results after the change in the reception point.

In addition, when an object that can be seen from both the transmission point and the reception point cannot be found, the propagation calculating unit may estimate the propagation characteristics by setting a default value of the reception intensity at the reception point, calculating the diffracted wave of the electric wave, or using a ray tracing method, as described above. Alternatively, when an object that can be seen from both the transmission point and the reception point cannot be found, the propagation calculating unit may estimate the propagation characteristics at the reception point of the estimation object using the other reception point that can be calculated as a virtual transmission point.

The propagation calculating unit may estimate at least any one of the received power in the reception point, the propagation loss between the transmission point and the reception point, and the channel characteristics between the transmission point and the reception point. The channel characteristics may include power, delay time, angle of arrival, etc. of each ray. The specific estimation method will be described further below.

The propagation calculating unit may acquire information of an object contributing to scattering of the radio wave (radio wave scattering contributing object) from the outside. Specifically, the propagation calculating unit may acquire information of the radio wave scattering contributing object via the external input unit 210. The propagation calculating unit may estimate the propagation characteristics using the acquired information.

The object may include a structure, but may include a reflective plate capable of dynamically controlling the reflected wave. Such a reflective panel may be referred to as a reconfigurable intelligent surface (Reconfigurable Intelligent Surface; RIS) or the like. The propagation calculating unit may control the characteristics of the reflection plate to estimate the propagation characteristics.

The output unit 240 can display the calculation result (reception level) based on the calculation unit 230 on a map in the form of a color map, or can output the calculation result (reception level) based on the calculation unit 230 to a file. The output unit 240 can display or output the image data shown in fig. 16, 21, and the like.

(working example)

Next, an operation example of the radio wave propagation estimating device 200 according to the present embodiment will be described. Fig. 24 shows an operation flow of the preprocessing unit 231 of the calculation unit 230. As shown in fig. 24, the preprocessing unit 231 performs setting of calculation conditions, processing of topographic information, calculation of reception points, association of colors or identification numbers, generation and analysis of images seen from the reception points.

Fig. 25 shows an operation flow of the post-processing unit 232 in the case of using the result of the previous preprocessing. As shown in fig. 25, the post-processing unit 232 uses the result of the previous preprocessing to simultaneously perform setting of the post-processing conditions, generation of an image seen from a transmission point, analysis of an image seen from a transmission point, comparison and combination of the result of analysis of an image seen from a transmission point and the result of analysis of an image seen from a reception point, calculation and combination of the radio wave scattering level in each surface of the structure, and the output unit 240 outputs the calculation result for each reception point.

Fig. 26A and 26B show operation flows of the preprocessing unit 231 and the post-processing unit 232 in the case of directly from the initial calculation. As shown in fig. 26A and 26B, the preprocessing unit 231 may perform an operation substantially similar to the operation shown in fig. 24, and the post-processing unit 232 may perform an operation substantially similar to the operation shown in fig. 25.

(available propagation characteristics)

Next, an example of the propagation characteristics that can be obtained by the radio wave propagation estimation device 200 according to the present embodiment will be described.

Fig. 27 shows an example of calculation of the received power and propagation loss by the radio wave propagation estimating device 200. As shown in fig. 27, when the scattered received power in each plane i is Pri, the total Pr of the received powers at the receiving points can be expressed by the expression shown in fig. 27.

Fig. 28 shows an example of the calculation of the channel characteristics by the radio wave propagation estimating apparatus 200. As described above, the channel characteristics may include the power, delay time, arrival angle, and the like of each ray.

Since the ray i (path) in each plane i has the information described below, the channel characteristics at the reception point can be calculated.

Received power Pri

Emission direction θti, arrival angle θti, delay time Ti

The radio wave propagation estimation device 200 has the external input unit 210, and can flexibly use external input as described above. The purpose of the external input is to acquire an object (e.g., a vehicle, a telephone booth, a human body, etc.) that is not present in the electric wave scattering contributing object DB for calculation of propagation characteristics.

Fig. 29 shows an operation flow of the radio wave propagation estimating device 200 for acquiring an object that is not present in the radio wave scattering contributing object DB. Fig. 30A to 30C show an example of an object that does not exist in the radio wave scattering contributing object DB, and an example of a car-human model.

As shown in fig. 29, when acquiring an object that is not present in the radio wave scattering contribution object DB, the radio wave propagation estimation device 200 may use a rough model (see fig. 30B) as shown in methods 1 to 3, or may use a detailed model (see fig. 30C) as shown in methods 2 and 3.

The radio wave propagation estimation device 200 can generate a model of the radio wave scattering contributing object using such a model, and store the model in the radio wave scattering contributing object DB. This can truly reproduce the surrounding environment, improve the estimation accuracy, and evaluate the influence of the fluctuation of the surrounding environment due to the moving object on the channel characteristics.

(modification)

The wave propagation estimation apparatus 200 may be used in combination with a Reconfigurable Intelligent Surface (RIS) or the like.

Fig. 31A and 31B show control images based on propagation characteristics in reception points of RIS or the like. In the case of combining RIS, the radio wave propagation estimation device 200 can operate as follows.

(1) The RIS is given an identification number or color

(2) Generating and analyzing images from the transmitting point and the receiving point, and counting the number of pixels visible from both sides

(3) Calculating the intensity of the received electric wave (pr=k tx_pixels rx_pixels)

(4) Adjusting and controlling the incidence angle, reflection angle of RIS, etc. to maximize the received power

(action and Effect)

According to the radio wave propagation estimation device 200 of the present embodiment, the following actions and effects can be obtained. Specifically, the radio wave propagation estimating device 200 uses identification information (color or number) capable of identifying an object including a structure or the like and map data including the structure, and extracts a structure (an apparent surface or the like) of the object that can be seen from both a direct or virtual transmission point of a radio wave and a direct or virtual reception point of the radio wave based on the identification information. The radio wave propagation estimating device 200 estimates propagation characteristics of the radio wave using information indicating the extracted characteristics of the structure.

Therefore, even when the normal reflection point Ir does not belong to the building, the error and the calculation amount of the power calculation of the radio wave coming to the reception point can be effectively reduced. Further, since the radio wave is scattered not only in the normal direction but also in various directions, the radio wave propagation estimating device 200 can effectively reduce the error and the calculation amount of the power calculation of the radio wave coming to the receiving point.

In the present embodiment, the radio wave propagation estimating device 200 can extract the above-described structure from the color of the surface of the recognition object, and estimate the propagation characteristics using the number of pixels constituting the surface of the object that can be seen from both the transmission point and the reception point. Therefore, the propagation characteristics can be estimated with higher accuracy while suppressing the processing load.

In the present embodiment, the surface of the object may be divided into a plurality of areas (grids), and different identification information is assigned to each area. This makes it possible to determine the surface visible from both the transmission point and the reception point with higher accuracy, and thus to estimate the propagation characteristics with higher accuracy.

In this embodiment, the object may include a ground surface around the structure. Accordingly, the line-of-sight state of the receiving point in the entire evaluation target region can be determined by analyzing only one image seen from the transmitting point, and the propagation characteristics can be estimated efficiently.

In the present embodiment, the radio wave propagation estimation device 200 can estimate at least one of the reception power at the reception point, the propagation loss between the transmission point and the reception point, and the channel characteristics between the transmission point and the reception point. Therefore, a desired propagation characteristic can be easily provided.

In the present embodiment, the radio wave propagation estimating device 200 can acquire information of a radio wave scattering contributing object from the outside and estimate propagation characteristics using the information. Thus, even when an object that is not present in the radio wave scattering contributing object DB is included, the propagation characteristics can be estimated with high accuracy.

In the present embodiment, the object may include a reflection plate (RIS or the like) capable of dynamically controlling the reflected wave, and the radio wave propagation estimating device 200 may estimate the propagation characteristics by controlling the characteristics of the reflection plate. Therefore, the propagation characteristics at the reception points of high accuracy of the combined RIS can be controlled.

Other embodiments

While the present invention has been described with reference to the embodiments, it will be apparent to those skilled in the art that the present invention is not limited to these descriptions, but various modifications and improvements can be made.

The functional blocks (constituent elements) in the radio wave propagation estimation device 100 illustrated in fig. 1 and the radio wave propagation estimation device 200 illustrated in fig. 23A and 23B may be realized by any combination of hardware and/or software. The implementation means of each functional block is not limited to a specific means. For example, functional blocks may be implemented using one device physically and/or logically combined, or a plurality of devices physically and/or logically separated may be directly and/or indirectly (e.g., via wired and/or wireless, etc.) connected.

Fig. 32 shows an example of a hardware configuration of the radio wave propagation estimation device 100 and the radio wave propagation estimation device 200. The radio wave propagation estimation device 100 may be configured by a computer such as a Personal Computer (PC) or a server computer. A computer is an example of an information processing apparatus.

The radio wave propagation estimating device 100 (and the radio wave propagation estimating device 200, which will be described later) may be configured by 1 computer or may be configured by a plurality of computers. The load distribution of the processing to be executed by the radio wave propagation estimation apparatus 100 can be realized by a plurality of computers.

As shown in fig. 16, the radio wave propagation estimating device 100 may have a processor 1001, a memory 1002, a storage 1003, a communication device 1004, an input device 1005, an output device 1006, and a bus 1007, for example.

In addition, in the following description, the term "means" may be replaced with "circuit", "device", "unit", or the like. The hardware configuration of the radio wave propagation estimation device 100 may be configured to include one or more of the devices illustrated in fig. 10, or may be configured to include no part of the devices.

For example, only one processor 1001 is illustrated in fig. 32, but the radio wave propagation estimating apparatus 100 may have a plurality of processors. The processing in the radio wave propagation estimation apparatus 100 may be executed by one processor 1001 or may be executed by a plurality of processors. In one or more processors, multiple processes may be performed simultaneously, in parallel, or sequentially, or may be performed by other methods. The processor 1001 may be a single-core processor or a multi-core processor. The processor 1001 may be mounted using more than one chip.

One or more functions of the radio wave propagation estimation apparatus 100 may be realized by reading hardware such as the processor 1001 and the memory 1002 into predetermined software. In addition, "software" may also be referred to as a "program," application, "or" software module.

For example, the processor 1001 reads a program by controlling one or both of reading and writing of data stored in one or both of the memory 1002 and the memory 1003, and executes the program. The program may be transmitted from the network by communication via a telecommunication line by the communication device 1004.

The program may be a program that causes a computer to execute all or a part of the processing in the radio wave propagation estimation device 100. One or more functions of the radio wave propagation estimation device 100 are realized by execution of the program code included in the program. All or a part of the program code may be stored in the memory 1002 or the memory 1003, or may be described as a part of an Operating System (OS).

For example, the program may include program code for implementing the functional blocks illustrated in fig. 1 and the like, and may include program code for executing the illustrated flow. A program including such a program code may be referred to as a "wave propagation estimation program".

The processor 1001 is an example of a processing unit, and controls the entire computer by operating the OS, for example. The processor 1001 may be configured using a central processing unit (CPU: central Processing Unit) including an interface with peripheral devices, a control device, an arithmetic device, a register, and the like.

The processor 1001 reads one or both of a program and data from one or both of the memory 1003 and the communication device 1004 and the bidirectional memory 1002, for example, and executes various processes according to the read program and/or data.

The memory 1002 is an example of a computer-readable recording medium, and may be configured using at least one of ROM, EPROM, EEPROM, RAM, SSD and the like, for example. In addition, "ROM" is an abbreviation of "Read Only Memory", "EPROM" is an abbreviation of "Erasable Programmable ROM (erasable programmable ROM)", "EEPROM" is an abbreviation of "Electrically Erasable Programmable ROM (electrically erasable programmable ROM)", "RAM" is an abbreviation of "Random Access Memory (random access Memory)", and "SSD" is an abbreviation of "Solid State Drive (solid state drive)".

The memory 1002 may be referred to as registers, cache, main memory, working memory, main storage, etc. The memory 1002 stores a program executable to implement the radio wave propagation estimation method according to one embodiment of the present invention.

The memory 1003 is an example of a computer-readable recording medium, and may be configured using at least one of an optical disk such as a CD-ROM (Compact Disc ROM), a Hard Disk Drive (HDD), a floppy disk, a magneto-optical disk (for example, a compact disk, a digital versatile disk, a Blu-ray (registered trademark) disk, a smart card, a flash memory (for example, a card, a stick, a Key drive), a floppy disk, a magnetic stripe, and the like).

The communication device 1004 is an example of hardware (may also be referred to as a "transmitting/receiving device") for performing communication between computers via one or both of a wired network and a wireless network. The "communication apparatus" may also be referred to as a network device, a network controller, a network card, a communication module, etc.

The input device 1005 is an example of an input apparatus that receives an input from the outside of the radio wave propagation estimation device 100, and the input device 1005 may include one or more of a keyboard, a mouse, a microphone, a switch, a button, and a sensor, for example.

The output device 1006 is an example of an output apparatus (output unit 108) that outputs to the outside of the radio wave propagation estimation device 100. Illustratively, the output device 1006 may include one or more displays, speakers, LED lights, and the like. In addition, "LED" is an abbreviation of "light emitting diode (Light emitting diode)".

The input device 1005 and the output device 1006 may be separate structures or may be integrated structures such as a touch panel.

The processor 1001 and the memory 1002 may be communicably connected via a bus 1007. Devices may be connected using a single bus 1007 or may be connected using a plurality of buses.

The radio wave propagation estimation device 100 may be configured to include hardware such as a microprocessor and DSP, ASIC, PLD, FPGA. For example, the processor 1001 may be installed by at least one of these hardware. By this hardware, part or all of the functional blocks illustrated in fig. 1 and 2 can be realized.

In addition, "DSP" is an abbreviation of "Digital Signal Processor (digital signal processor)", "ASIC" is an abbreviation of "Application Specific Integrated Circuit (application specific integrated circuit)", "PLD" is an abbreviation of "Programmable Logic Device (programmable logic device)", and "FPGA" is an abbreviation of "Field Programmable Gate Array (field programmable gate array)".

(applicable System)

The embodiments described in this specification can also be applied to systems using LTE (Long Term Evolution: long term evolution), LTE-Advanced (LTE-a), SUPER 3G, IMT-Advanced, 4G, 5G, FRA (Future Radio Access: future wireless access), W-CDMA (registered trademark), GSM (registered trademark), CDMA2000, UMB (Ultra Mobile Broadband: ultra mobile broadband), IEEE 802.11 (Wi-Fi), IEEE 802.16 (WiMAX), IEEE 802.20, UWB (Ultra-wide band), bluetooth (registered trademark), other suitable systems, and/or next generation systems extended accordingly.

(treatment Process, etc.)

The processing procedures, timings, flows, and the like of each form/embodiment described in the present specification can be replaced in order without contradiction. For example, for the methods described in this specification, elements of the various steps are presented using the illustrated order, but are not limited to the particular order presented.

(processing of input or output information and the like)

The input or output information may be stored in a specific location (e.g., a memory), or may be managed by a management table. Information input or output, etc. may be rewritten, updated, or recorded. The outputted information and the like may also be deleted. The input information and the like may also be transmitted to other devices.

(determination method)

The determination may be performed by a value (0 or 1) represented by 1 bit, may be performed by a Boolean value (true or false), or may be performed by a comparison of values (e.g., a comparison with a predetermined value).

(software)

With respect to software, whether referred to as software, firmware, middleware, microcode, hardware description language, or by other names, should be broadly interpreted to refer to a command, a set of commands, code, a code segment, program code, a program (program), a subroutine, a software module, an application, a software package, a routine, an object, an executable, a thread of execution, a procedure, a function, or the like.

In addition, software, commands, etc. may be transmitted and received via a transmission medium. For example, where software is transmitted from a web page, server, or other remote source using a wireline technology, such as coaxial cable, fiber optic cable, twisted pair, digital Subscriber Line (DSL), and/or infrared, wireless, and microwave wireless technologies, the wireline and/or wireless technologies are included in the definition of transmission medium.

(information, signal)

Information, signals, etc. described in this specification may also be represented using any of a variety of different technologies. For example, data, commands, instructions, information, signals, bits, symbols, chips, and the like may be represented by voltages, currents, electromagnetic waves, magnetic fields or particles, optical fields or photons, or any combination thereof.

The terms described in the present specification and/or terms that contribute to the understanding of the present invention may be replaced with terms having the same or similar meaning.

(parameters)

The information, parameters, and the like described in this specification may be expressed by absolute values, relative values to predetermined values, or other information.

The names used for the above parameters are non-limiting in any respect. Further, the numerical formulas and the like using these parameters may also be different from those explicitly disclosed in the present specification.

(base station)

A base station can accommodate one or more (e.g., three) cells (also referred to as sectors). In the case of a base station accommodating a plurality of cells, the coverage area of the base station can be divided into a plurality of smaller areas, and each of the smaller areas can also be provided with communication services by the base station subsystem (e.g., indoor small base station (Remote Radio Head) (remote radio head): RRH), "cell" or "sector" which means a part or the whole of the coverage area of the base station and/or the base station subsystem providing communication services in the coverage area.

(UE)

For a UE, those skilled in the art are sometimes referred to by the following terms: a user terminal, mobile station, subscriber station, mobile unit (mobile unit), subscriber unit, wireless unit, remote unit, mobile device, wireless communication device, remote device, mobile subscriber station, access terminal, mobile terminal, wireless terminal, remote terminal, handset, user agent, mobile client, UE, or some other suitable terminology. The UE may be a fixed terminal whose location does not change, or may be a mobile terminal whose location changes. As a non-limiting example, the UE may be a mobile terminal such as a mobile phone or a smart phone, a tablet terminal, or the like. Further, the UE may be an internet of things (Internet of Things: ioT) terminal. Communication functions can be carried over IoT for a wide variety of "things". Various "objects" equipped with a communication function can communicate by being connected to the internet, a wireless access network, or the like. For example, ioT terminals may contain sensor devices or meters (gauges) or the like that are capable of wireless communication. Any monitoring device such as a monitoring camera or a fire alarm equipped with a sensor device or a meter may be used as the terminal. Wireless communication between a UE, which is an IoT terminal such as a monitoring device, and a base station is sometimes referred to as MTC (Machine Type Communications: machine type communication). Thus, UEs that are IoT terminals are sometimes referred to as "MTC devices.

The functions include, but are not limited to, judgment, decision, judgment, calculation, processing, derivation, investigation, search, confirmation, reception, transmission, output, access, resolution, selection, establishment, comparison, assumption, view, broadcast (broadcast), notification (notification), communication (communication), forwarding (forwarding), configuration (reconfiguration), reconfiguration (allocating, mapping), assignment (assignment), and the like. For example, a functional block (configuration unit) that causes transmission to function is referred to as a transmitter (transmitting unit) or a transmitter (transmitter). In short, the implementation method is not particularly limited as described above.

The specific actions performed by the base station in the present disclosure are sometimes performed by its upper node (upper node) according to circumstances. In a network comprising one or more network nodes (network nodes) having a base station, it is apparent that various operations performed for communication with a terminal may be performed by the base station and at least one of the other network nodes (for example, MME or S-GW, etc. are considered but not limited thereto) other than the base station. In the above, the case where one other network node other than the base station is illustrated, but the other network node may be a combination of a plurality of other network nodes (for example, MME and S-GW).

Information, signals (information, etc.) can be output from a higher layer (or lower layer) to a lower layer (or higher layer). Or may be input or output via a plurality of network nodes.

Each of the modes and embodiments described in the present disclosure may be used alone, in combination, or may be used in combination with execution. Note that the notification of the predetermined information is not limited to being explicitly performed (for example, notification of "yes" or "X"), and may be performed implicitly (for example, notification of the predetermined information is not performed).

The terms "system" and "network" as used in this disclosure may be used interchangeably.

The terms "connected," "coupled," or any variation of these terms are intended to refer to any direct or indirect connection or coupling between two or more elements, including the case where one or more intervening elements may be present between two elements that are "connected" or "coupled" to each other. The combination or connection of the elements may be physical, logical, or a combination of these. For example, "connection" may be replaced with "Access". As used in this disclosure, it is considered that for two elements, the interconnection "or" bonding "is made by using at least one of one or more wires, cables, and printed electrical connections, and as some non-limiting and non-inclusive examples, by using electromagnetic energy or the like having wavelengths in the wireless frequency domain, the microwave region, and the optical (including both visible and invisible) region.

The reference signal may be simply RS (Reference Signal) or may be called Pilot (Pilot) depending on the standard applied.

As used in this disclosure, the recitation of "according to" is not intended to mean "according to" unless explicitly recited otherwise. In other words, the term "according to" means "according to only" and "according to at least" both.

Any reference to elements referred to using "1 st", "2 nd", etc. as used in this disclosure, is not intended to limit the number and order of such elements in any way. These designations are used in this disclosure as a convenient way of distinguishing between two or more elements. Thus, references to elements 1 and 2 do not indicate that only two elements can be taken herein or that in any form element 1 must precede element 2.

Where the terms "include", "comprising" and variations thereof are used in this disclosure, these terms are intended to be inclusive in the same sense as the term "comprising". Also, the term "or" as used in this disclosure does not refer to exclusive or.

In the present disclosure, for example, where an article is added by translation as in a, an, and the in english, the present disclosure also includes a case where a noun following the article is in plural.

The terms "determining" and "determining" used in the present disclosure may include various operations. The "judgment" and "determination" may include, for example, a matter in which judgment (determination), calculation (calculation), processing (processing), derivation (development), investigation (investigation), search (lookup up, search, inquiry) (for example, search in a table, database, or other data structure), confirmation (evaluation), or the like are regarded as a matter in which "judgment" and "determination" are performed. Further, "determining" and "deciding" may include a matter in which reception (e.g., reception of information), transmission (e.g., transmission of information), input (input), output (output), access (e.g., access of data in a memory) is performed as a matter in which "determining" and "deciding" are performed. Further, "judging" and "determining" may include matters of solving (resolving), selecting (selecting), selecting (setting), establishing (establishing), comparing (comparing), and the like as matters of judging and determining. That is, the terms "determine" and "determining" may include what is considered to be any action. The "judgment (decision)" may be replaced by "assumption", "expectation", "consider", or the like.

In the present disclosure, the term "a and B are different" may also mean that "a and B are different from each other". The term "a and B are different from C" may also be used. The terms "separate," coupled, "and the like may also be construed as" different.

The present disclosure has been described in detail above, but it should be clear to those skilled in the art that the present disclosure is not limited to the embodiments described in the present disclosure. The present disclosure can be implemented as modifications and variations without departing from the spirit and scope of the present disclosure as defined by the claims. Accordingly, the description of the present disclosure is intended to be illustrative, and not in any limiting sense.

Description of the reference numerals

100: radio wave propagation estimating device

102: control unit

102a: setting part

102b: face edge recognition part

102c: transmitting point view searching unit

102d: receiving point view search unit

102e: information extraction unit for propagation calculation

102f: propagation calculating unit

106: storage unit

108: output unit

200: radio wave propagation estimating device

210: external input part

220: storage unit

230: calculation unit

231: pretreatment unit

232: post-treatment part

240: output unit

1001: processor and method for controlling the same

1002: memory

1003: memory device

1004: communication device

1005: input device

1006: output device

1007: bus line

Claims (15)

1. An electric wave propagation estimation device, wherein the electric wave propagation estimation device has:

an extraction unit that uses identification information capable of identifying an object including a structure and map data including the object, and extracts a structure of the object that can be seen from both a direct or virtual transmission point of an electric wave and a direct or virtual reception point of the electric wave based on the identification information; and

an estimating unit that estimates propagation characteristics of the electric wave using information indicating characteristics of the structure of the object extracted by the extracting unit.

2. The radio wave propagation estimating apparatus according to claim 1, wherein,

the extraction section extracts the configuration based on a color or an identification number identifying a face of the object,

the estimating unit estimates the propagation characteristic using the number of pixels constituting the surface of the object visible from both the transmission point and the reception point.

3. The radio wave propagation estimating apparatus according to claim 1, wherein,

the face of the object is divided into a plurality of regions,

And assigning different identification information to each region.

4. The radio wave propagation estimating apparatus according to claim 1, wherein,

the object comprises a surface of the perimeter of the structure.

5. The radio wave propagation estimating apparatus according to claim 1, wherein,

the estimating unit changes the height of the transmitting point when the object visible from both the transmitting point and the receiving point is not found,

the estimating unit estimates the propagation characteristics before the change in the altitude based on differences in propagation characteristics in other reception points that can be calculated both before and after the change in the altitude and on propagation characteristics in the reception points after the change in the altitude.

6. The radio wave propagation estimating apparatus according to claim 1, wherein,

the estimating unit estimates the propagation characteristic by, if the object visible from both the transmitting point and the receiving point cannot be found, any one of the following three: setting a default value of the receiving intensity in the receiving point; calculating a diffraction wave of the electric wave; ray tracing is used.

7. The radio wave propagation estimating apparatus according to claim 1, wherein,

The estimating unit estimates the propagation characteristics in the receiving points of the object by using the other receiving points which can be calculated as virtual transmitting points when the object which can be seen from both the transmitting points and the receiving points cannot be found.

8. The radio wave propagation estimating apparatus according to claim 1, wherein,

the estimating unit estimates at least one of a received power at the receiving point, a propagation loss between the transmitting point and the receiving point, and a channel characteristic between the transmitting point and the receiving point.

9. The radio wave propagation estimating apparatus according to claim 1, wherein,

the estimating unit obtains information of the object contributing to scattering of the electric wave from outside, and estimates the propagation characteristic using the information.

10. The radio wave propagation estimating apparatus according to claim 1, wherein,

the object comprises a reflecting plate capable of dynamically controlling reflected waves,

the estimating section controls the characteristics of the reflection plate to estimate the propagation characteristics.

11. An electric wave propagation estimation method, wherein the electric wave propagation estimation method comprises the steps of:

using identification information capable of identifying an object including a structure and map data including the structure, and extracting a structure of the object that is visible from both a direct or virtual transmission point of an electric wave and a direct or virtual reception point of the electric wave based on the identification information; and

Using information indicating the extracted feature of the structure, propagation characteristics of the electric wave are estimated.

12. An electric wave propagation estimation device, wherein the electric wave propagation estimation device has:

an extraction unit that extracts two-dimensional data from three-dimensional map data including a reception point of a radio wave transmitted from a transmission point; and

an estimating unit that estimates a propagation characteristic of the electric wave, which accompanies diffraction calculation of the electric wave in a surrounding environment, based on the two-dimensional data extracted by the extracting unit.

13. An electric wave propagation estimation method, wherein the electric wave propagation estimation method comprises the steps of:

extracting two-dimensional data from three-dimensional map data including a reception point of an electric wave transmitted from a transmission point; and

estimating propagation characteristics of the electric wave calculated by diffraction of the electric wave in the surrounding environment based on the extracted two-dimensional data.

14. An electric wave propagation estimation device, wherein the electric wave propagation estimation device has:

an extraction unit that extracts, from map data, a 1 st parameter indicating a structure that may become an obstacle to a radio wave in a surrounding environment including a reception point, by applying a convolutional neural network to the map data including the reception point of the radio wave transmitted from a transmission point and including at least one of identification information and distance information of a structure; and

An estimating unit that estimates propagation characteristics of the radio waves by applying a fully-connected neural network to the 1 st parameter extracted by the extracting unit and a 2 nd parameter indicating a configuration of a wireless communication system that performs wireless communication between the transmitting point and the receiving point.

15. An electric wave propagation estimation method, wherein the electric wave propagation estimation method comprises the steps of:

by applying a convolutional neural network to map data including a reception point of an electric wave transmitted from a transmission point and including at least one of identification information and distance information of a structure, a 1 st parameter indicating a structure that may become an obstacle of the electric wave in a surrounding environment including the reception point is extracted from the map data; and

the propagation characteristics of the electric wave are estimated by applying a fully connected neural network to the extracted 1 st parameter and a 2 nd parameter representing a structure of a wireless communication system that performs wireless communication between the transmission point and the reception point.