Radio wave strength estimation system
The radio field strength estimation system addresses the inefficiency of three-dimensional CNNs by employing two-dimensional data and additional information, enabling faster and more accurate radio wave strength estimation for optimal wireless device placement.
Patent Information
- Application Number
- JP2023085563
- Authority / Receiving Office
- JP · JP
- Patent Type
- Applications
- Current Assignee / Owner
- Filing Date
- 2023-05-24
- Publication Date
- 2024-12-05
- Estimated Expiration
- Not applicable · inactive patent
AI Technical Summary
Existing radio field strength estimation systems using three-dimensional CNNs require long learning times due to the need for three-dimensional data, which is inefficient.
A radio field strength estimation system that uses two-dimensional teacher data, including plane coordinates and additional information such as antenna heights and reflected wave information, to generate algorithms for estimating radio wave strength through machine learning, reducing learning time.
The system significantly shortens the learning time by utilizing two-dimensional data, allowing for quicker and more efficient estimation of radio wave strength, including considerations for height and shielding objects, facilitating optimal wireless communication device placement.
Smart Images

Figure 2024168686000001_ABST
Abstract
Description
[Technical field]
[0001] The present disclosure relates to a radio wave intensity estimation system. [Background technology]
[0002] When determining the placement position of at least one of the transmitting antenna and the receiving antenna, the placement position is determined so that the receiving antenna can receive the radio wave transmitted from the transmitting antenna. The stronger the strength of the radio wave transmitted from the transmitting antenna is at the position, the easier it is for the receiving antenna to receive the radio wave. The radio wave strength estimation system disclosed in Non-Patent Document 1 estimates the strength of the radio wave transmitted from the transmitting antenna using a model generated by machine learning. [Prior art documents] [Non-patent literature]
[0003] [Non-Patent Document 1] Ryohei Hayashi, Kaoru Sawai, Masaaki Tateiwa, Satoshi Tanaka, Katsuki Fujisawa, "A study on a radio wave propagation estimation model in urban environments using deep learning," Technical Report of the Institute of Electronics, Information and Communication Engineers, March 19, 2020 Summary of the Invention [Problem to be solved by the invention]
[0004] In Non-Patent Document 1, machine learning is performed using a three-dimensional CNN. When using a three-dimensional CNN, the data used for learning also needs to be three-dimensional. This tends to increase the time required for learning. [Means for solving the problem]
[0005] A radio wave strength estimation system that solves the above problem is a radio wave strength estimation system that estimates the strength of a radio wave transmitted from a transmitting antenna, and includes a generation unit that generates an algorithm for estimating the strength of the radio wave in the planar coordinates through machine learning by using data indicating the strength of the radio wave in the planar coordinates and additional information regarding the height direction as teacher data, and an estimation unit that estimates the strength of the radio wave in the planar coordinates in an area where the transmitting antenna is used by inputting the additional information into the algorithm generated by the generation unit.
[0006] The algorithm is generated by using data indicating the strength of radio waves in planar coordinates and additional information as training data. By using data indicating the strength of radio waves in planar coordinates, learning can be performed using two-dimensional training data. Therefore, the time required for learning can be shortened compared to when learning is performed using three-dimensional training data.
[0007] In the radio wave intensity estimation system, the additional information may include information indicating a height of the transmitting antenna and a height of a receiving antenna that receives the radio wave. In the radio wave intensity estimation system, the additional information may include information on a reflected wave that is reflected from the transmitting antenna by at least one of a floor and a ceiling and reaches the receiving antenna.
[0008] In the radio wave intensity estimation system, the additional information may include information indicating a height of an obstruction. Effect of the Invention
[0009] According to the present invention, the time required for learning can be shortened. [Brief description of the drawings]
[0010] [Figure 1] FIG. 1 is a schematic diagram of a radio wave intensity estimation system. [Diagram 2] FIG. 2 is a diagram showing an area in which a transmitting antenna is used. [Diagram 3] Figure 3 shows the architecture of a DNN. [Figure 4] FIG. 4 is a diagram for explaining the radio wave intensity used in the free space model. [Diagram 5] FIG. 5 illustrates the algorithm. [Figure 6] FIG. 6 is a diagram showing an area in which a transmitting antenna is used. [Figure 7] FIG. 7 illustrates the algorithm. [Figure 8] FIG. 8 is a diagram for explaining the radio wave intensity used in the free space model. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS
[0011] [First embodiment] A first embodiment of a radio wave intensity estimation system will be described. <Radio wave intensity estimation system> As shown in FIG. 1, the radio wave intensity estimation system 10 includes a processor 11 and a storage unit 12. The processor 11 is, for example, a central processing unit (CPU), a graphics processing unit (GPU), or a digital signal processor (DSP). The storage unit 12 includes a random access memory (RAM) and a read only memory (ROM). The storage unit 12 may include an auxiliary storage device. The auxiliary storage device is, for example, a hard disk drive or a solid state drive. The storage unit 12 stores program code or instructions configured to cause the processor 11 to execute processing. The storage unit 12, i.e., the computer-readable medium, includes any available medium accessible by a general-purpose or dedicated computer. The radio wave intensity estimation system 10 may be configured by a hardware circuit such as an application specific integrated circuit (ASIC) or a field programmable gate array (FPGA). The radio wave intensity estimation system 10, which is a processing circuit, may include one or more processors operating according to a computer program, one or more hardware circuits such as an ASIC or an FPGA, or a combination thereof.
[0012] As shown in FIG. 2, the radio wave intensity estimation system 10 estimates the intensity of radio waves transmitted from the transmitting antenna 21 in association with the plane coordinates of the area R1 in which the transmitting antenna 21 is used. The transmitting antenna 21 is provided in, for example, a transmitter. The transmitter transmits radio waves including, for example, a detection result of a sensor from the transmitting antenna 21. The radio waves transmitted from the transmitting antenna 21 are received by a receiving antenna 22 provided in the area R1. The receiving antenna 22 is provided in, for example, a receiver that acquires the detection result of the sensor. In this case, the transmitter and the receiver constitute a wireless sensor network. In the following description, the intensity of radio waves transmitted from the transmitting antenna 21 is appropriately referred to as radio wave intensity.
[0013] The area R1 in which the transmitting antenna 21 is used may be indoors. Indoors may be, for example, in a factory, a house, a public facility, a commercial facility, or a warehouse. The area R1 in which the transmitting antenna 21 is used may be outdoors.
[0014] A user of the radio wave intensity estimation system 10 recognizes the radio wave intensity associated with the planar coordinates of the area R1 by the radio wave intensity estimation system 10. The user of the radio wave intensity estimation system 10 sets the arrangement positions of the transmitting antenna 21, the receiving antenna 22, or the transmitting antenna 21 and the receiving antenna 22, according to the radio wave intensity associated with the planar coordinates of the area R1. More specifically, the radio wave intensity estimation system 10 is used, for example, when considering the arrangement of wireless communication devices indoors. A machine learning algorithm is generated using teacher data measured in advance in multiple patterns, and the algorithm is later used to estimate the radio wave intensity in various arrangement patterns of wireless communication devices, and an appropriate arrangement of the wireless communication devices is considered.
[0015] <Generation part> 1, the radio wave intensity estimation system 10 includes a generation unit 13. The generation unit 13 generates an algorithm A1 for estimating the radio wave intensity in the planar coordinates of an area R1 by machine learning. The generation unit 13 is a functional unit that functions when the processor 11 executes a predetermined process.
[0016] The algorithm A1 is a trained model generated by machine learning using a DNN (Deep Neural Network). More specifically, the algorithm A1 is generated by machine learning using a forward propagation type DNN. The machine learning is, for example, supervised learning using training data or semi-supervised learning. In addition, for example, EM DeepRay may be used for radio wave strength estimation (see "EM DeepRay: An Expedient, Generalizable, and Realistic Data-Driven Indoor Propagation Model" IEEE TRANSACTIONS ON ANTENNAS AND PROPAGATION, VOL. 70, NO. 6, June 2022). This technology allows the use of two-dimensional algorithms.
[0017] As shown in Fig. 3, the DNN includes an input layer 31, a hidden layer 32, and an output layer 33. Teacher data 41 is input to the input layer 31. The hidden layer 32 extracts features of the teacher data 41. The output layer 33 outputs an inference result. Features extracted from the teacher data 41 by a CNN (Convolutional Neural Network) may be input to the input layer 31.
[0018] The teacher data 41 input to the input layer 31 is data in which the permittivity map 42, the conductivity map 43, the distance map 44, and the free space model 45 are associated with the radio wave intensity map 46 as correct answer data. The permittivity map 42, the conductivity map 43, the distance map 44, the free space model 45, and the radio wave intensity map 46 are maps of, for example, a planar view of an environment in which the transmitting antenna 21 may be used. These maps are divided into a plurality of grids. The grids correspond to the planar coordinates of the environment. The planar coordinates of the environment are coordinates that represent the horizontal position of the environment. The permittivity map 42, the conductivity map 43, the distance map 44, the free space model 45, and the radio wave intensity map 46 are heat maps. The permittivity map 42, the conductivity map 43, the distance map 44, the free space model 45, and the radio wave intensity map 46 are two-dimensional data that represent, in a plane, an environment in which the transmitting antenna 21 may be used.
[0019] The permittivity map 42 is a map that represents the permittivity (ε) when the environment in which the transmitting antenna 21 and the object exist is viewed in a plan view. The object is, for example, a structure such as a wall or a pillar that exists in the environment. The permittivity map 42 represents, for example, the permittivity corresponding to the planar coordinates of the environment by shading. This makes it possible to capture the permittivity of the environment in which the transmitting antenna 21 and the object exist as image data. In the permittivity map 42, for example, the higher the permittivity, the lighter the shading. The permittivity map 42 may represent the permittivity by color.
[0020] The conductivity map 43 is a map that represents the conductivity (σ) [S / m] when the environment in which the transmitting antenna 21 and the object exist is viewed in a plane. The position of the conductivity map 43 represents the plane coordinates of the environment. The conductivity map 43 represents, for example, the conductivity corresponding to the plane coordinates of the environment by shading. This makes it possible to capture the conductivity of the environment in which the transmitting antenna 21 and the object exist as image data. For example, the conductivity map 43 represents the conductivity by lighter shading at a position with higher conductivity. The conductivity map 43 may represent the conductivity by color.
[0021] Distance map 44 is a map that represents the distance (d) [m] between transmitting antenna 21 and each position in the environment when the environment in which transmitting antenna 21 exists is viewed in a plane. Distance map 44 represents the distance from transmitting antenna 21 by, for example, shading. This allows the distance from transmitting antenna 21 to be captured as image data. For example, the distance map 44 has a lighter shade the farther the position is from transmitting antenna 21. Distance map 44 may represent the distance from transmitting antenna 21 by color. In this embodiment, the position of transmitting antenna 21 is fixed, and there is one distance map 44.
[0022] The free space model 45 is a map showing radio wave strength (RSSI) [dB] assuming that the environment includes the transmitting antenna 21 and no objects are present. The position of the free space model 45 represents the planar coordinates of the environment. The free space model 45 represents, for example, the radio wave strength corresponding to the planar coordinates of the environment by shading. For example, the free space model 45 represents the higher the radio wave strength is at a position with a darker shade. The free space model 45 may represent the radio wave strength by color.
[0023] The radio wave intensity used in the free space model 45 is calculated by the Friis formula according to equations (1) to (3).
[0024]
number
[0025] As shown in Figure 4, the distance d 1 ,d 2 is the height h of the transmitting antenna 21 t , the height h of the receiving antenna 22 r , and the distance d between the transmitting antenna 21 and the receiving antenna 22 in the horizontal direction 3 The height h of the transmitting antenna 21 can be calculated from t is the height from the floor to the transmitting antenna 21. The height h of the receiving antenna 22 r is the height from the floor to the receiving antenna 22.
[0026] In this way, the free space model 45 is t , and the height h of the receiving antenna 22 r In detail, the free space model 45 is created using the radio wave intensity calculated using additional information on the height h of the transmitting antenna 21. t , and the height h of the receiving antenna 22 r and additional information about a reflected wave that is reflected by the floor from the transmitting antenna 21 and reaches the receiving antenna 22. By using the free space model 45 created in this way as the teacher data 41, the teacher data 41 includes additional information about the height direction in the region R1.
[0027] The radio wave intensity map 46, which is the correct data, is a map that represents the radio wave intensity when the environment in which the transmitting antenna 21 and the object are present is viewed in a plane. The radio wave intensity map 46 is data that indicates the radio wave intensity in plane coordinates. The position of the radio wave intensity map 46 represents the plane coordinates of the environment. The radio wave intensity map 46 represents, for example, the radio wave intensity corresponding to the plane coordinates of the environment by shading. This makes it possible to capture the radio wave intensity of the environment in which the transmitting antenna 21 and the object are present as image data. For example, the radio wave intensity map 46 is darker in positions where the radio wave intensity is higher. The radio wave intensity map 46 may represent the radio wave intensity by color.
[0028] The teacher data 41 is data in which a permittivity map 42, a conductivity map 43, a distance map 44, and a free space model 45 obtained when the environmental conditions are changed are associated with a radio wave intensity map 46 for the corresponding conditions as correct answer data. That is, the teacher data 41 is a plurality of data corresponding to various conditions changed. The conditions are the height h of the transmitting antenna 21, t , the position of the receiving antenna 22, and the height h of the receiving antenna 22. r , the positions of objects present in the environment, the types of objects present in the environment, and the sizes of objects present in the environment.
[0029] In this embodiment, the additional information regarding the height direction is the height h of the transmitting antenna 21. t , and the height h of the receiving antenna 22 r In this embodiment, a plurality of free space models 45 and a radio wave intensity map 46 linked to the free space models 45 as correct answer data are prepared. During measurements when creating the teacher data 41, the radio wave intensity of each plane coordinate is measured each time the antenna height is changed in various ways, and a plurality of sets of two-dimensional data free space models 45 and two-dimensional data radio wave intensity maps 46 linked to the free space models 45 are prepared. Since both are two-dimensional data, a two-dimensional algorithm can be used in machine learning.
[0030] A trained model is obtained as an algorithm A1 by inputting the above-mentioned teacher data 41 to the input layer 31. The algorithm A1 is stored in the storage unit 12, for example. <Estimation part> 1, the radio wave intensity estimation system 10 includes an estimation unit 14. The estimation unit 14 is a functional unit that functions when the processor 11 executes a predetermined process.
[0031] As shown in Figure 5, the estimation unit 14 inputs input data 51 to the algorithm A1 generated by the generation unit 13, and estimates the radio wave intensity transmitted from the transmitting antenna 21 in correspondence with the planar coordinates of the area R1 in which the transmitting antenna 21 is used.
[0032] The user of the radio wave intensity estimation system 10 determines the height h of the transmitting antenna 21 in the area R1 in which the transmitting antenna 21 is used. t , the position of the receiving antenna 22, and the height h of the receiving antenna 22. r Then, the height h of the transmitting antenna 21 is determined. t , the position of the receiving antenna 22, and the height h of the receiving antenna 22. r , and generates a permittivity map 52, a conductivity map 53, a distance map 54, and a free space model 55 according to the objects present in the region R1. By inputting these as input data 51 into an algorithm A1, a radio wave intensity map 56 corresponding to the region R1 in which the transmitting antenna 21 is used is output. This radio wave intensity map 56 is a map in which radio wave intensity is associated with the planar coordinates of the region R1 in which the transmitting antenna 21 is used.
[0033] The permittivity map 52 was obtained by mapping the permittivity of the region R1 using a method similar to that for the permittivity map 42 used as the training data 41. In the example shown in Fig. 5, the permittivity changes due to objects such as walls.
[0034] The conductivity map 53 was obtained by mapping the conductivity of the region R1 using a method similar to that of the conductivity map 43 used as the training data 41. In the example shown in Fig. 5, the conductivity changes due to objects such as walls.
[0035] The distance map 54 was obtained by mapping the distance from the transmitting antenna 21 to the region R1 using a method similar to that of the distance map 44 used as the training data 41. In the distance map 54 shown in FIG. 5, the shading becomes lighter as the distance from the transmitting antenna 21 increases.
[0036] The free space model 55 was obtained by mapping the radio wave intensity in the region R1 using the same method as the free space model 45 used as the teacher data 41. In other words, the free space model 55 includes additional information. In the example shown in FIG. 5, the radio wave intensity varies due to the influence of reflected waves. The free space model 55 used in the estimation unit 14 is based on the height h of the transmitting antenna 21 to be actually used. t , and the height h of the receiving antenna 22 r It is a two-dimensional map using the values of
[0037] The radio wave intensity map 56 is an estimation result of the radio wave intensity in planar coordinates in the region R1 in which the transmitting antenna 21 is used. The user can recognize the radio wave intensity in the region R1 from the radio wave intensity map 56.
[0038] [Effects of the first embodiment] (1-1) Algorithm A1 is generated by using data indicating radio wave intensity in planar coordinates and additional information as teacher data 41. By using data indicating radio wave intensity in planar coordinates, learning can be performed with a two-dimensional algorithm using two-dimensional teacher data 41. Therefore, the time required for learning can be shortened compared to the case where learning is performed with a three-dimensional algorithm using three-dimensional teacher data.
[0039] (1-2) The teacher data 41 includes additional information about the height direction in the region R1. Therefore, it is possible to obtain a radio wave intensity map 56 that takes into account the height direction element in the region R1.
[0040] When the training data does not include additional information on the height direction in the region R1, the height h of the transmitting antenna 21 is t , the height h of the receiving antenna 22 r The value of is a fixed value, and a free space model is created using the radio wave intensity calculated using the Friis formula. The algorithm that learned using this free space model as training data is t and the height h of the receiving antenna 22 r Even if the height h of the transmitting antenna 21 is changed, the same radio wave intensity map is output. t and the height h of the receiving antenna 22 r If the height h of the transmitting antenna 21 is changed from the fixed value used in the training data, the radio wave intensity map 56 that takes this change into account cannot be output. t and the height h of the receiving antenna 22 r To create the radio wave intensity map 56 that takes into account the height h of the transmitting antenna 21, t and the height h of the receiving antenna 22 r Once this has been determined, training data must be created and an algorithm must be generated.
[0041] In contrast, in the embodiment, the height h of the transmitting antenna 21 is used as additional information regarding the height direction in the region R1. t , and the height h of the receiving antenna 22 r For this reason, the height h of the transmitting antenna 21 is t , and the height h of the receiving antenna 22 r When the height h of the transmitting antenna 21 is changed, t , and the height h of the receiving antenna 22 r As a result, a radio wave intensity map 56 corresponding to the height h of the transmitting antenna 21 can be obtained. t , and the height h of the receiving antenna 22 rEven in situations where the placement of wireless communication devices has not been decided, training data can be created and an algorithm can be generated, making it possible to quickly consider the placement of wireless communication devices.
[0042] [Second embodiment] A second embodiment of the radio wave intensity estimation system will be described below. Descriptions of the same parts as those in the first embodiment will be omitted.
[0043] 6, a shield 23 may exist in an area R1 in which the transmitting antenna 21 is used. The shield 23 is located between the transmitting antenna 21 and the receiving antenna 22, for example.
[0044] As shown in FIG. 3, the teacher data 41 of the second embodiment includes a height map 47 in addition to the teacher data 41 of the first embodiment. That is, the teacher data 41 is data in which a radio wave intensity map 46 is associated with a permittivity map 42, a conductivity map 43, a distance map 44, a free space model 45, and a height map 47 as correct answer data. The height map 47 is divided into a plurality of grids. The grids correspond to the planar coordinates of the environment. The height map 47 is a heat map. The height map 47 is two-dimensional data that represents, in a plane, the environment in which the transmitting antenna 21 may be used.
[0045] The height map 47 is a map that represents the height of the obstruction 23 when the environment in which the transmitting antenna 21 and the object exist is viewed in a planar view. The position of the height map 47 represents the plane coordinates of the environment. The height map 47 represents, for example, the height of the obstruction 23 corresponding to the plane coordinates of the environment by shading. This allows the height of the obstruction 23 in the environment in which the transmitting antenna 21 and the object exist to be captured as image data. For example, the higher the obstruction 23, the lighter the shading of the height map 47. The height map 47 may represent the height of the obstruction 23 by color. The height map 47 is additional information that indicates the height of the obstruction 23 that exists in the environment.
[0046] The teacher data 41 is data in which a permittivity map 42, a conductivity map 43, a distance map 44, a free space model 45, and a height map 47 obtained when a condition is changed are associated with a radio wave intensity map 46 for the corresponding condition as correct answer data. The conditions include the position of the shielding 23, the size of the shielding 23, the shape of the shielding 23, and the height of the shielding 23 in addition to the elements listed in the first embodiment.
[0047] In this embodiment, the additional information in the height direction can be said to be a plurality of height maps 47 in addition to a plurality of free space models 45 as in the first embodiment. In this embodiment, a plurality of height maps 47 and a radio wave intensity map 46 as answer data linked thereto are prepared for each free space model 45. During measurements for creating the teacher data 41, the antenna height is varied as in the first embodiment, and the radio wave intensity of each plane coordinate is measured each time the height of the obstruction is varied, and a plurality of sets of two-dimensional data height maps 47 and two-dimensional data radio wave intensity maps 46 linked thereto are prepared. Since both are two-dimensional data, a two-dimensional algorithm can be used in machine learning.
[0048] A trained model is obtained as an algorithm A1 by inputting the above-mentioned teacher data 41 to the input layer 31. The algorithm A1 is stored in the storage unit 12, for example. As shown in Figure 7, the estimation unit 14 inputs input data 51 to an algorithm A1 generated by the above-mentioned teacher data 41, and estimates the radio wave intensity transmitted from the transmitting antenna 21 in correspondence with the planar coordinates of the area R1 in which the transmitting antenna 21 is used.
[0049] The user of the radio wave intensity estimation system 10 determines the height h of the transmitting antenna 21 in the area R1 in which the transmitting antenna 21 is used. t , the position of the receiving antenna 22, and the height h of the receiving antenna 22. r , and the height of the shielding object 23 are determined. Then, the position of the transmitting antenna 21, the height h t , the position of the receiving antenna 22, and the height h of the receiving antenna 22. r, and the height of the shielding object 23. By inputting these as input data 51 to algorithm A1, a radio wave intensity map 56 corresponding to region R1 in which the transmitting antenna 21 is used is output.
[0050] The height map 57 is obtained by mapping the heights of the obstructions 23 in the region R1 using the same method as the height map used as the teacher data 41. The height map 57 used by the estimation unit 14 is a two-dimensional map using the heights of the obstructions 23 that are actually to be used. In the example shown in Fig. 7, the height map 57 has shading depending on the height of the obstructions 23.
[0051] [Effects of the second embodiment] According to the second embodiment, in addition to the effects of the first embodiment, the following effects can be obtained.
[0052] (2-1) The additional information includes information indicating the height of the obstruction 23. Therefore, when the height of the obstruction 23 is changed, by inputting the input data 51 including the height map 57 according to the height of the obstruction 23 into the algorithm A1, it is possible to obtain a radio wave intensity map 56 according to the height of the obstruction 23. Therefore, even in a situation where the height of the obstruction 23 has not been determined, it is possible to create training data and generate an algorithm, and it is possible to quickly consider the placement of wireless communication devices.
[0053] [Example of change] Each embodiment can be modified as follows: Each embodiment and the following modified examples can be combined with each other to the extent that there is no technical contradiction.
[0054] In each embodiment, the additional information may include information regarding a reflected wave that is reflected from the transmitting antenna 21 on the ceiling and reaches the receiving antenna 22. Even in this case, the radio wave intensity used to create the free space model 45 can be calculated using equations (1) to (3).
[0055] As shown in Figure 8, d 2 is the distance that the radio wave reflected by the ceiling travels from the transmitting antenna 21 to the receiving antenna 22. 2 is the height h of the transmitting antenna 21 t , the height h of the receiving antenna 22 r , the ceiling height h, and the horizontal distance d between the transmitting antenna 21 and the receiving antenna 22. 3 The ceiling height h can be calculated from the floor to the ceiling.
[0056] The additional information may include information on the reflected wave that is reflected from the transmitting antenna 21 by the floor and reaches the receiving antenna 22, and information on the reflected wave that is reflected from the transmitting antenna 21 by the ceiling and reaches the receiving antenna 22. In this case, the radio wave intensity used to create the free space model 45 may be the radio wave intensity of the reflected wave that is reflected from the transmitting antenna 21 by the floor and reaches the receiving antenna 22, the reflected wave that is reflected from the transmitting antenna 21 by the ceiling and reaches the receiving antenna 22, and the direct wave.
[0057] In each embodiment, the teacher data 41 may include at least the additional information and the radio wave intensity map 46. In each embodiment, the generation unit 13 and the estimation unit 14 may function using different hardware. For example, an algorithm A1 obtained by performing machine learning on a first computer may be executed on a second computer. In this case, the first computer is the generation unit 13, and the second computer is the estimation unit 14. The radio wave intensity estimation system 10 includes a first computer and a second computer.
[0058] In each embodiment, the teacher data is created using actual measured values. However, the teacher data may be generated using any simulation means. In each embodiment, the distance map 44 is prepared with the position of the transmitting antenna 21 fixed, but multiple distance maps 44 may be prepared by changing the position of the transmitting antenna 21. In this case, a free space model 45, a radio wave intensity map 46, and a height map 47 may be prepared for each distance map 44.
[0059] In the first embodiment, the radio wave intensity used to create the free space model 45 is P r However, the reflected wave was not taken into account. 1 Only the above may be used. In the second embodiment, the radio wave intensity used to create the free space model 45 may be calculated by the Friis rule. In each embodiment, the free space model may be created by any method.
[0060] In the second embodiment, the height h of the transmitting antenna 21 t , the height h of the receiving antenna 22 r Alternatively, a free space model 45 may be prepared in which the value of is a fixed value. [Explanation of symbols]
[0061] A1...algorithm, R1...area, 10...radio signal strength estimation system, 13...generation unit, 14...estimation unit, 21...transmitting antenna, 22...receiving antenna.
Claims
1. A radio wave intensity estimation system for estimating the intensity of a radio wave transmitted from a transmitting antenna, comprising: a generation unit that generates an algorithm for estimating the intensity of the radio wave in the plane coordinates by machine learning using data indicating the intensity of the radio wave in the plane coordinates and additional information related to a height direction as teacher data; an estimation unit that estimates the radio wave strength at the plane coordinates in an area in which the transmitting antenna is used by inputting the additional information into the algorithm generated by the generation unit.
2. The radio wave intensity estimation system according to claim 1 , wherein the additional information includes information indicating a height of the transmitting antenna and a height of a receiving antenna that receives the radio wave.
3. The radio wave intensity estimation system according to claim 2 , wherein the additional information includes information on a reflected wave that is reflected from the transmitting antenna by at least one of a floor and a ceiling and reaches the receiving antenna.
4. 4. The radio wave intensity estimation system according to claim 1, wherein the additional information includes information indicating a height of an obstruction.
Citation Information
Cited By
Printing method, printing apparatus, and non-transitory computer-readable storage medium storing program
EP4718833A1
