Electromagnetic field estimation system, electromagnetic field estimation device, electromagnetic field estimation method, and program for electromagnetic field estimation

Abstract

To mitigate a calculation load in analyzing an analysis region including a human body and the like using the FDTD method, in an electromagnetic field estimation system suitable for characteristic evaluation of high frequency band radio communication.SOLUTION: A human body model 18 is disposed in an analysis region 10. The whole area of the analysis region 10 is divided into a plurality of voxels. A voxel group covering a surface of the human body model 18 is defined as surface voxels. A voxel group located inside the surface voxels is defined as non-calculation voxels. Boundaries between the surface voxels and the non-calculation voxels are set to a lossy absorption boundary. To the lossy absorption boundary, a lossy absorption boundary condition for inhibiting electromagnetic field reflection is applied. After the non-calculation voxels are released from a calculation target, electric field components and magnetic field components are successively calculated with respect to the other voxels in the analysis region. Until convergence of the electric field components and the magnetic field components is recognized, the successive calculation is executed repeatedly at predetermined time steps.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The present disclosure relates to an electromagnetic field estimation system, an electromagnetic field estimation device, an electromagnetic field estimation method, and an electromagnetic field estimation program, and in particular to an electromagnetic field estimation system, an electromagnetic field estimation device, an electromagnetic field estimation method, and an electromagnetic field estimation program that are suitable for evaluating the characteristics of high-frequency band wireless communication. [Background technology]

[0002] In microwave and millimeter wave wireless communications, shielding and scattering by human bodies have a large effect on communication waves, so these must be taken into account in the evaluation of radio wave propagation characteristics when designing base stations. Patent Document 1 listed below discloses a device that evaluates radio wave propagation characteristics using the finite-difference time-domain method (FDTD method).

[0003] In the FDTD method, the analysis domain is divided into a large number of blocks (hereafter referred to as "voxels") with dimensions of approximately 1 / 10 or less of the wavelength corresponding to the frequency of the analysis target. The FDTD method is an electromagnetic field analysis technique in which the electric field and magnetic field generated by a wave source within the analysis domain are updated over time for each voxel based on Maxwell's equations. [Prior art documents] [Patent documents]

[0004] [Patent Document 1] Japanese Patent Application Publication No. 2017-11518 Summary of the Invention [Problem to be solved by the invention]

[0005] Compared to other analysis methods such as geometrical optics approximation, the FDTD method has the advantage of being able to accurately reflect the effects of reflection, absorption, scattering, and other factors caused by the electrical properties and shape of the human body in the analysis results. However, the FDTD method imposes a heavy computational load because it sequentially calculates the electric and magnetic fields for all voxels in the three-dimensional space that make up the analysis domain, including voxels corresponding to the interior of the human body. In particular, when evaluating communication waves in the short-wavelength, high-frequency band, the number of voxels that need to be set becomes extremely large, which poses problems such as the required main memory capacity and the required computation time.

[0006] This disclosure has been made to solve the above-mentioned problems, and its primary objective is to provide an electromagnetic field estimation system for reducing the computational load when analyzing an analysis domain including a human body, etc., using the FDTD method. A second object of this disclosure is to provide an electromagnetic field estimation device for reducing the calculation load when analyzing an analysis region including a human body or the like using the FDTD method. A third object of this disclosure is to provide an electromagnetic field estimation method for reducing the calculation load when analyzing an analysis region including a human body or the like using the FDTD method. Furthermore, a fourth object of this disclosure is to provide an electromagnetic field estimation program for reducing the calculation load when analyzing an analysis region including a human body or the like using the FDTD method. [Means for solving the problem]

[0007] In order to achieve the above object, a first aspect is an electromagnetic field estimation system for estimating electromagnetic field components occurring inside an analysis region, comprising: at least one processor; at least one memory storing computer program code; The computer program code and the memory, together with the processor, configure the electromagnetic field estimation system to: Placing an object model having an electric field absorbing property within the analysis region; Dividing the entire analysis region including the object model into a plurality of voxels; setting a group of voxels covering a surface of the object model as surface voxels; setting a group of voxels located inside the surface voxels as non-calculation voxels; setting a boundary between the surface voxel and the non-calculation voxel to a lossy absorbing boundary; applying a lossy absorbing boundary condition to the lossy absorbing boundary so as not to reflect an electric field; releasing the non-calculation voxels from the calculation target and sequentially calculating electric field components and magnetic field components for other voxels in the analysis region; repeating the iterative calculations at predetermined time steps until convergence of the electric field components and the magnetic field components is observed; Preferably, the method is configured to cause the device to perform operations including:

[0008] A second aspect is an electromagnetic field estimation device for estimating electromagnetic field components occurring inside an analysis region, comprising: a parameter setting unit that places an object model having an electric field absorption characteristic within the analysis region; an analysis region determination unit that divides the entire analysis region including the object model into a plurality of voxels; a human body surface region extraction unit that sets a group of voxels covering the surface of the object model as surface voxels; a non-voxel release unit that sets a group of voxels located inside the surface voxels as non-calculation voxels; a lossy absorbing boundary setting unit that sets the boundary between the surface voxel and the non-calculation voxel as a lossy absorbing boundary; an electric field component absorption condition application unit that applies a lossy absorbing boundary condition to the lossy absorbing boundary so as not to reflect an electric field; an electric field component calculation unit and a magnetic field component calculation unit that release the non-calculation voxels from calculation targets and sequentially calculate electric field components and magnetic field components for other voxels in the analysis region; a convergence determination unit that repeatedly executes the iterative calculation at a predetermined time step until convergence of the electric field component and the magnetic field component is confirmed; It is preferable that the configuration includes:

[0009] A third aspect is an electromagnetic field estimation method for estimating electromagnetic field components occurring inside an analysis region, comprising: For hardware, Placing an object model having an electric field absorbing property within the analysis region; Dividing the entire analysis region including the object model into a plurality of voxels; setting a group of voxels covering a surface of the object model as surface voxels; setting a group of voxels located inside the surface voxels as non-calculation voxels; setting a boundary between the surface voxel and the non-calculation voxel to a lossy absorbing boundary; applying a lossy absorbing boundary condition to the lossy absorbing boundary so as not to reflect an electric field; releasing the non-calculation voxels from the calculation target and sequentially calculating electric field components and magnetic field components for other voxels in the analysis region; repeating the iterative calculations at predetermined time steps until convergence of the electric field components and the magnetic field components is observed; It is desirable to include causing the

[0010] A fourth aspect is an electromagnetic field estimation program for estimating electromagnetic field components occurring inside an analysis region, comprising: At least one processor Placing an object model having an electric field absorbing property within the analysis region; Dividing the entire analysis region including the object model into a plurality of voxels; setting a group of voxels covering a surface of the object model as surface voxels; setting a group of voxels located inside the surface voxels as non-calculation voxels; setting a boundary between the surface voxel and the non-calculation voxel to a lossy absorbing boundary; applying a lossy absorbing boundary condition to the lossy absorbing boundary so as not to reflect an electric field; releasing the non-calculation voxels from the calculation target and sequentially calculating electric field components and magnetic field components for other voxels in the analysis region; repeating the iterative calculations at predetermined time steps until convergence of the electric field components and the magnetic field components is observed; Preferably, the computer readable program includes a program for causing the computer to execute the steps of: [Effects of the Invention]

[0011] According to the first to fourth aspects, it is possible to set non-calculation voxels inside a human body, etc., and to exempt them from calculation targets for electromagnetic field estimation. Therefore, according to these aspects, in a situation where a human body, etc. is included in the analysis domain, it is possible to significantly reduce the calculation load associated with electromagnetic field analysis by the FDTD method without reducing the accuracy of electromagnetic field estimation. [Brief explanation of the drawings]

[0012] [Figure 1] 1 is a schematic diagram of the analysis domain for electromagnetic field estimation using the FDTD method. [Figure 2] FIG. 2 is an enlarged view of the periphery of the human body model included in the analysis space shown in FIG. [Figure 3] 10A and 10B are diagrams for explaining a first technique for setting surface voxels and non-calculation voxels in a human body model. [Figure 4] FIG. 4 is an enlarged view of the head and surrounding areas of the human body model shown in FIG. 3. [Figure 5] 5 is an enlarged view of region V shown in FIG. 4, showing the vicinity of the boundary between surface voxels and non-calculation voxels along the surface of the human body. [Figure 6] FIG. 6 is an enlarged view of the VI region shown in FIG. 5, showing the region where a lossy absorbing boundary condition is set. [Figure 7]This is an example of evaluation of the electric field strength distribution related to reflection and absorption on the surface of the human body when exposed to a 6 GHz plane wave. [Figure 8] FIG. 10 is a diagram for explaining a second technique for setting surface voxels and non-calculation voxels in a human body model. [Figure 9] FIG. 9 is a diagram showing the vicinity of the boundary between non-calculation voxels and surface voxels set by the second method shown in FIG. 8. [Figure 10] FIG. 9 is a diagram showing a region where a lossy absorbing boundary condition is set when the second technique shown in FIG. 8 is used. [Figure 11] 10 is a flowchart illustrating a processing flow when all voxels including a human body model are targets for electromagnetic field estimation. [Figure 12] 1 is a flowchart illustrating a flow of processing used for electromagnetic field estimation in the first embodiment of the present disclosure. [Figure 13] 1 is a diagram functionally illustrating a configuration of an electromagnetic field estimation system according to a first embodiment of the present disclosure. DETAILED DESCRIPTION OF THE INVENTION

[0013] Embodiment 1 [Summary of this disclosure] Fig. 1 is a schematic diagram of a three-dimensional space that serves as an analysis domain 10 for electromagnetic field estimation using the FDTD method. In the analysis domain 10, a three-dimensional coordinate system is set with, for example, one end of the analysis domain 10 as the origin. In the example shown in Fig. 1, an example of the xy plane is indicated by reference numeral 12, an example of the yz plane is indicated by reference numeral 14, and an example of the zx plane is indicated by reference numeral 16.

[0014] The analysis domain 10 shown in Fig. 1 includes a human body model 18 and a wave source 20. The human body model 18 can be realized, for example, by a numerical human body model published by the National Institute of Information and Communications Technology (NICT). The human body model 18 represents the shape of the human body (tissues and organs) as an aggregate of minute elements, and each minute block is assigned a number indicating the name of the tissue or organ, such as muscle or fat.

[0015] In electromagnetic field estimation using the FDTD method, the analysis region 10 is divided into a large number of voxels. Then, the electric and magnetic fields generated by the wave source 20 are updated over time for each voxel based on Maxwell's equations.

[0016] When a human body is included in the three-dimensional space in which wireless communication is carried out, the effects of shielding and scattering by the human body have a significant impact on the communication waves. Therefore, when performing electromagnetic field estimation in a three-dimensional space that includes a human body, it is necessary to accurately reflect the effects of the human body in the estimation results.

[0017] Fig. 2 is an enlarged view of the periphery of the human body model 18 shown in Fig. 1. Specifically, Fig. 2 shows how the interior of the human body model 18 is divided into multiple voxels and how electromagnetic field estimation is performed for these voxels. When the human body model 18 is included in the analysis region 10, electromagnetic field estimation calculations are usually performed including the internal region of the human body model 18, as shown in Fig. 2, in order to evaluate the reflection, scattering, and absorption of communication waves by the human body.

[0018] The dimensions of the voxels used in electromagnetic field estimation using the FDTD method are set based on the target frequency of the analysis. Specifically, the dimensions of each voxel are set to be 1 / 10 or less of the target frequency. For example, if the target frequency is 6 GHz, the voxel dimensions must be set to 3 cm or less. In the human body model 18 used in this embodiment, the standard voxel size is set to 2 mm.

[0019] If the voxel dimension is 2 mm, the number of voxels per human body model 18 will be 16 million. If the electric and magnetic fields were calculated sequentially for all of these voxels, a huge amount of main memory would be required, and the calculation time would also be prolonged.

[0020] Therefore, in this embodiment, the voxels located inside the human body model 18 are excluded from the calculation targets by the following method. (1) First, the surface area of the human body model 18 is extracted. (2) Next, the area is replaced with a two-dimensional reflecting / absorbing surface area in which reflection is 1 / 2 in the wave equation for the electric field (Helmholtz equation). (3) Then, the voxels located inside the above region are released as non-computational regions.

[0021] It is known that the ratio of the power density incident on the human body to the power density absorbed by the human body in the high-frequency band is approximately 1 / 2 (see Li K, Sasaki K, Wake K, Onishi T and Watanabe S, 2021b, Quantitative comparison of power densities related to electromagnetic near-field exposures with safety guidelines from 6 to 100 GHz, IEEE Access, 9, 115801-12.). Therefore, if an appropriate reflection coefficient is set near the surface of the human body model18 based on this ratio of 1 / 2, the effects of the human body on reflection, absorption, and scattering can be accurately reflected in the results of electromagnetic field estimation without evaluating the electromagnetic field inside the body.

[0022] [Key points of this disclosure] A specific method for setting non-calculation voxels will be described below with reference to FIGS.

[0023] 3 is a diagram illustrating a first technique used in the first embodiment of the present disclosure to set surface voxels 22 and non-calculation voxels 24 in the human body model 18. In this technique, as shown in FIG. 3, a group of voxels lined up along the surface of the human body model 18 are set as surface voxels 22. Then, all voxels inside the surface voxels 22 are set as non-calculation voxels 24.

[0024] Fig. 4 is an enlarged view of the head area of the human body model shown in Fig. 3. Fig. 5 is an enlarged view of region V shown in Fig. 4, showing the boundary area between surface voxels 22 along the surface of the human body, non-calculation voxels 24 located inside the surface voxels 22, and voxels 26 outside the human body. The mesh lines shown in Figs. 4 and 5 indicate the boundaries of individual voxels set in the analysis region 10. As shown in Fig. 5, in the first technique, a group of voxels arranged in one layer along the surface of the human body model 18 is defined as surface voxels 22.

[0025] FIG. 6 shows a further enlarged view of the VI region shown in FIG. 5. The surface voxel 22 has a dielectric constant (ε r , σ) are given. For the surface voxels 22, the electromagnetic field is estimated based on Maxwell's equations using the electrical constants.

[0026] The boundary between the surface voxels 22 and the non-calculation voxels 24 is set as a lossy absorbing boundary 28. A lossy absorbing boundary condition for discontinuing the calculation domain is set for the lossy absorbing boundary 28. Because the lossy absorbing boundary 28 is the boundary between adjacent voxels arranged in three-dimensional space, it is either the xy plane, yz plane, or zx plane. Hereinafter, when it is necessary to distinguish between them, they will be referred to as the "xy plane absorbing boundary," the "yz plane absorbing boundary," or the "zx plane absorbing boundary," respectively.

[0027] The essence of radio wave propagation characteristic evaluation for the purpose of designing base stations in wireless communication is to evaluate the electromagnetic field in the area outside the human body. On the other hand, electromagnetic field components that penetrate into the human body in a real three-dimensional space are absorbed inside the body. Therefore, in this type of radio wave propagation characteristic evaluation, it is necessary that the electromagnetic field components that penetrate into the human body model 18 are absorbed inside the body.

[0028] When the electric field strength inside the human body according to the depth from the skin surface is evaluated using a high-frequency frequency band, the absorption of electromagnetic waves occurs within 1 mm of the human body surface. Therefore, as described above, if the boundary conditions are satisfied and conditions are set that do not re-reflect the electromagnetic field components inside the human body model 18, the influence of the human body on the external electromagnetic field distribution becomes negligible (this varies depending on the voxel size, but is kept to within 1%).

[0029] FIG. 7 shows an example of an evaluation of the electric field strength distribution obtained by the method of this embodiment. Specifically, it shows the electric field strength distribution related to the reflection and absorption of the human body surface when exposed to a 6 GHz band plane wave. The "no human body" distribution shows the electric field strength in the analysis region 10 where the human body model 18 is not placed. On the other hand, the "human body present" distribution shows the electric field strength when the human body model 18 is placed in the analysis region 10. The electric field strength "with human body" is a relative value to the electric field strength "without human body". Both of these show the instantaneous values of the electric field strength in the analysis region 10 as a one-dimensional distribution.

[0030] The evaluation result "with human body" shown in FIG. 7 indicates that the electric field intensity is absorbed near the surface of the human body model 18. According to the first method described above, the number of surface voxels 22 arranged along the surface of the human body model 18 is approximately 900,000, and approximately 15 million voxels inside these voxels can be excluded from the calculation as non-calculation voxels 24. This makes it possible to reduce the main memory capacity required for calculation by approximately 35% and the calculation time by approximately 30%. When dividing the human body model 18 using smaller voxels or when evaluating radio wave propagation characteristics in an environment where multiple human body models 18 exist, the effects of reducing memory capacity and calculation time become even greater.

[0031] [Variation of non-computational voxel setting method] A second method for setting non-calculation voxels will be described below with reference to FIGS.

[0032] 8 is a diagram illustrating a second technique used in the first embodiment of the present disclosure to set surface voxels 22 and non-calculation voxels 24 in the human body model 18. In this technique, as shown in FIG. 8, a group of voxels that can be grouped into a simple shape inside the human body model 18 are set as non-calculation voxels.

[0033] Specifically, two groups of voxels that belong to the interior of the human body model 18 and can be extracted as part of a simple rectangular parallelepiped are set as non-calculation voxels 30-1 and 30-2. Hereinafter, when there is no need to distinguish between the two non-calculation voxels 30-1 and 30-2, they will be collectively referred to as non-calculation voxels 30. Note that here, for convenience, the non-calculation voxels 30 are extracted as two rectangular parallelepipeds, but the number is not limited to this. Three or more simple rectangular parallelepipeds, including a cube, may be set inside the human body model 18, and the voxels belonging to these may be set as non-calculation voxels 30.

[0034] Fig. 9 is an enlarged view of the area around the nose of the human body model 18 shown in Fig. 8. As shown in the figure, in the second method, voxels that do not belong to the simple rectangular parallelepiped made up of non-calculation voxels 30 and that belong to the human body model 18 become surface voxels 22.

[0035] 10 shows an enlarged view of a surface voxel 22 set by the second method. In the second method, the non-calculation voxel 30 is extracted as a simple rectangular parallelepiped, so the lossy absorbing boundary 32 located at the boundary between the non-calculation voxel 30 and the surface voxel 22 is a simple flat surface with no irregularities. On the other hand, unlike the first method, the surface voxel 22 is not limited to a single layer, and in some places has a layered structure of two or more layers.

[0036] Compared to the first method, the second method increases the number of surface voxels 22, resulting in a decrease in the number of non-calculated voxels 30. This increases the computational load associated with sequential calculations for each voxel. On the other hand, because the lossy absorbing boundary 32 is a simple plane, the computational load associated with its extraction and the computational load required to apply the lossy absorbing boundary condition can be reduced.

[0037] [Processing flow for electromagnetic field estimation using the FDTD method] (Processing flow to compare) 11 is a flowchart for explaining the processing flow when all voxels including the human body model 18 are the targets of electromagnetic field estimation. To facilitate understanding of the processing flow used in this embodiment, the processing flow when all voxels are the targets of calculation will first be explained. Note that the processing explained below is assumed to be executed by an electromagnetic field estimation system including a computer system.

[0038] 11, first, a target model for analysis is read (step 100). Specifically, information about the three-dimensional space in which the electromagnetic field estimation is performed, information about the human body model 18 to be placed in the three-dimensional space, the frequency to be analyzed, the setting values of various parameters, the dielectric constant for each type of medium, etc. are read.

[0039] Next, the analysis region 10 is determined (step 102). Here, the three-dimensional data of the entire region within the three-dimensional space designated as the analysis region 10, that is, the entire region including the interior of the human body model 18, is identified. Also, in step 120, the size of the voxels is determined based on the frequency of the analysis target, and absorbing boundary conditions to be applied to the end faces of the analysis region 10 are set.

[0040] In electromagnetic field estimation using the FDTD method, electric field components must be absorbed without being reflected at the edge faces of the analysis domain 10, that is, at the boundary between the simulation space and the space outside it. For this reason, in determining the analysis domain 10 in step 120, absorbing boundary conditions for absorbing electric field components at the edge faces of the analysis domain 10 are set by a well-known method using CPML (Convolutional Perfectly Matched Layer) or the like.

[0041] Next, the assignment of medium constants is performed (step 104), where the medium constants of the individual components are assigned to their corresponding voxels based on the components in the analysis domain 10 determined in step 102 above.

[0042] Next, the time step is updated (step 106). In electromagnetic field estimation using the FDTD method, calculations are performed sequentially in accordance with the time it takes for the electromagnetic wave to pass through each voxel. Specifically, in step 106, a process is executed to update the calculation loop each time the time Δt obtained by the following calculation formula elapses: Δt=1 / C√(1 / Δx 2 +1 / Δy 2 +1 / Δz 2 ) where C is the velocity and (Δx, Δy, Δz) are the dimensions of the voxel.

[0043] Next, electric field components are calculated sequentially for all voxels included in the analysis region 10 (step 108). The electric field components are calculated using equations generally used in FDTD analysis, specifically, difference equations obtained by expanding Maxwell's equations in the spatial and temporal domains.

[0044] Next, a process is performed to apply an absorbing boundary condition to the electric field component (step 110). Specifically, an absorbing boundary condition is calculated to prevent reflection of electromagnetic waves at the boundary of the analysis domain 10. This condition is calculated based on the electric field component and the magnetic field component in the analysis domain 10. The calculation result in step 108 above is used for the electric field component. For the magnetic field component, a predetermined initial value is used in the first process, and in the second and subsequent processes, the result calculated in the previous cycle in step 112 below is used.

[0045] Next, magnetic field components are calculated sequentially for all voxels included in the analysis region 10 (step 112). The magnetic field components are also calculated by difference equations based on Maxwell's equations, which are generally used in FDTD analysis.

[0046] Next, an absorbing boundary condition for the magnetic field component is applied (step 114). Specifically, based on the electric field component calculated in step 108 and the magnetic field component calculated in step 112, an absorbing boundary condition for the magnetic field component is calculated to prevent the electromagnetic wave from being reflected at the boundary of the analysis region 10.

[0047] Once the above process is complete, it is determined whether the calculation of the electromagnetic field components has converged (step 116). The calculation of the electric field and magnetic field in the analysis region 10 is performed for all voxels at each period. In this step 116, it is determined that the calculation has converged if the changes caused by the time update have reached a steady state. For example, it is determined that the electromagnetic field components have converged if, as a result of the time update, no changes exceeding a threshold value are observed in the electric field strength or magnetic field strength.

[0048] If convergence is not confirmed in step 116, the processing from step 106 onwards is repeated. On the other hand, if convergence of the electromagnetic field components is confirmed, it is determined that the electromagnetic field analysis has ended and this cycle is ended.

[0049] (Processing flow of the first embodiment of the present disclosure) FIG. 12 is a flowchart illustrating a processing flow for electromagnetic field estimation used in the first embodiment of the present disclosure. Hereinafter, in FIG. 12, steps similar to those shown in FIG. 11 above are denoted by the same reference numerals, and their description will be omitted or simplified. Here, an example of setting surface voxels and non-calculation voxels using the first method described above will be described. However, the following description also applies to the case where the second method is used, and which method to adopt can be appropriately determined depending on the requirements for electromagnetic field estimation.

[0050] 12, after the process of step 100 of reading the calculation object model is completed, the process of extracting the surface region of the human body model 18 and the process of setting the lossy absorbing boundary 28 are executed (step 120). Specifically, a group of voxels in one layer covering the surface of the human body model 18 placed in the analysis region 10 are extracted as surface voxels 22. In addition, the boundary surfaces inside these surface voxels 22 are set as the lossy absorbing boundary 28.

[0051] As described above, the lossy absorbing boundary 28 includes an xy-plane absorbing boundary, a yz-plane absorbing boundary, and a zx-plane absorbing boundary. Here, the yz-plane in the analysis domain 10 can be specified by an x-axis index (i = 1...Nx). Similarly, the zx-plane can be specified by a y-axis index (j = 1...Ny), and the xy-plane can be specified by a z-axis index (k = 1...Ny). However, the interval between each index corresponds to the size of a voxel. In this step 120, the lossy absorbing boundary 28 is defined by the above x-axis index, y-axis index, and z-axis index.

[0052] After the above process is completed, the analysis region 10 is determined (step 122). Here, a group of voxels that are part of the human body model 18 and located inside the surface voxels 22 are set as non-calculation voxels 24. In the following process, the non-calculation voxels 24 are released from the calculation targets of the electromagnetic field analysis.

[0053] Subsequently, medium constants are assigned in step 104, the time step is updated in step 106, and then the electric field components are calculated (step 124). Here, electric field calculations are sequentially performed for all voxels in the analysis region 10 except for the non-calculated voxels 24. This calculation is performed using difference equations based on Maxwell's equations, using electrical constants and the like appropriate for each medium, as in step 108 shown in FIG. 11. For example, for the surface voxel 22, the calculation is performed using electrical constants appropriate for human skin.

[0054] Next, an absorption condition for the electric field component is applied (step 126). In this embodiment, in step 126, the following two absorption conditions are applied to the analysis region 10. (1) Absorbing boundary conditions to prevent the electric field from being reflected at the edge of the analysis region 10 (2) Lossy absorbing boundary conditions for absorbing the electric field at the lossy absorbing boundary 28 set in step 120

[0055] The absorbing boundary condition (1) above is calculated in the same way as in step 110 shown in Fig. 11. On the other hand, the lossy absorbing boundary condition (2) above is calculated as follows from the wave equation (Helmholtz equation) for the electric field:

[0056] Below is the general formula of Helmholtz's wave equation, which describes the behavior of an electric field.

[0057]

number

[0058] The condition that must be satisfied by the two-dimensional absorbing surface that constitutes the lossy absorbing boundary 28 defined in this embodiment is that the electric field component reflected by the absorbing surface is 0. For example, when the yz plane with an x-axis index of 0 (hereinafter expressed as "x=0") is the lossy absorbing boundary 28, it is sufficient that the plane waves Ey and Ez propagated from the region where x>0 satisfy the following equations.

[0059]

number

[0060] Here, Vp is the phase velocity, which can be expressed as follows using variables Sx, Sy, and Sz:

[0061]

number

[0062] Below, we will explain the formulation of the electric field E to be applied to the lossy absorbing boundary 28, using the yz plane as an example. Here, we will show a formulation assuming that the yz plane is the absorbing surface, which absorbs radio waves propagating in the -x direction at the position x = 0. In this case, the condition applied to the electric field component Ey propagating in the -x direction is as follows. Furthermore, by changing the subscript y to z in the conditional equation shown below, we can obtain the condition applied to the electric field component Ez propagating in the -x direction.

[0063]

number

[0064] In the above conditional expressions, Δt is the discrete time (time step) explained in step 106. The superscripts n, n+1, etc. indicate time updates in the FDTD analysis. For example, n+1 represents the value at a time 1×Δt ahead of n. In addition, (0,j,k), (1,j,k), etc. in the conditional expressions are the (x,y,z) coordinates indicating the corresponding position. Furthermore, Δx is the dimension of the voxel in the x direction.

[0065] When the lossy absorbing boundary 28 exists on the yz plane where x = 1, the conditional expression can be obtained by sliding the x coordinate, for example, by replacing (0,j,k) in the expression with (1,j,k). Furthermore, for the lossy absorbing boundary 28 for the electric field component propagating in the y or z direction, a similar formulation can be obtained by replacing Δx in the above expression with Δy or Δz and then substituting the coordinate position of the absorbing surface.

[0066] In step 126, the above conditional equation is applied to the lossy absorbing boundary 28 set in step 120. This realizes the electric field absorption characteristics on the surface of the human body model 18. As a result, it becomes possible to accurately evaluate the radio wave propagation characteristics within the analysis region 10 while freeing the non-calculation voxels 24 from being targets of electromagnetic field calculation.

[0067] After the above process is completed, the magnetic field components are calculated (step 128). Here, as with the electric field components, the non-calculated voxels 24 are excluded from the calculation. Because the electric field components are absorbed by the surface of the human body model 18, the magnetic field components can be evaluated accurately in the analysis region 10 even if the non-calculated voxels 24 are excluded from the calculation.

[0068] Thereafter, in this routine, an absorbing boundary condition is applied to the magnetic field component in step 114, as in the case shown in Fig. 11. Furthermore, if convergence of the electromagnetic field component is confirmed in step 116, the electromagnetic field analysis is terminated.

[0069] [Configuration of the First Embodiment] 13 is a functional diagram illustrating a configuration of an electromagnetic field estimation system 40 according to the first embodiment of the present disclosure. The electromagnetic field estimation system 40 according to the present embodiment can be used to evaluate radio wave propagation characteristics required for station placement design in wireless communication. More specifically, the electromagnetic field estimation system 40 can be used to evaluate what kind of radio wave propagation characteristics a communication wave used in wireless communication exhibits in an analysis region 10.

[0070] Specifically, the electromagnetic field estimation system 40 shown in Fig. 13 can be realized by a computer system or by a combination of a computer system and dedicated hardware. The computer system includes a processor, a memory, and an input / output interface. The memory stores an electromagnetic field estimation program. The processor executes the program to achieve desired functions. The electromagnetic field estimation program may be recorded on an optical disc such as a CD (Compact Disc) or a DVD (Digital Versatile Disc) and distributed, or may be distributed by electronic download.

[0071] The functions provided by the components shown in Figure 13 may be implemented in circuitry or processing circuitry, including general-purpose processors, application-specific processors, integrated circuits, ASICs (Application Specific Integrated Circuits), a CPU (a Central Processing Unit), conventional circuits, and / or combinations thereof, programmed to provide the functions. A processor includes transistors and other circuits and is considered to be circuitry or processing circuitry. A processor may also be a programmed processor that executes a program stored in a memory.

[0072] As used herein, a circuit, "part," "unit," or "means" refers to hardware that is programmed to implement a described function or that implements that function. The hardware may be any hardware disclosed herein or any hardware that is programmed to implement a described function or that is known to execute a program. If the hardware is a processor, which is considered a type of circuitry, the circuit, "part," "unit," or "means" refers to a combination of hardware and software used to configure the hardware and / or processor.

[0073] The main functional components of the electromagnetic field estimation system 40 of this embodiment are realized by the multiple elements shown in Fig. 13. The electric field component calculation unit 42 shown in Fig. 13 executes the processing of step 124 shown in Fig. 12. Specifically, here, the electric field components are sequentially calculated for all voxels in the analysis domain 10 except for the non-calculation voxels 24 for each time step.

[0074] The electric field component absorbing boundary condition calculation unit 44 performs part of the processing of step 126 shown in Fig. 12. Specifically, here, an absorbing boundary condition to be applied to the electric field component in order to prevent the electric field from being reflected at the edge (boundary) of the analysis region 10 is calculated.

[0075] The electric field component lossy absorbing condition calculation unit 46 performs the remaining part of the processing of step 126 shown in Fig. 12. Specifically, here, the condition for the electric field to be absorbed on the surface of the human body model 18, that is, the lossy absorbing boundary condition for the electric field to be absorbed at the lossy absorbing boundary 28, is calculated.

[0076] The magnetic field component calculation unit 48 executes the process of step 128 shown in Fig. 12. Specifically, here, the magnetic field components are sequentially calculated for all voxels in the analysis region 10 except for the non-calculation voxels 24 for each time step.

[0077] The magnetic field component absorbing boundary condition calculation unit 50 executes the processing of step 114 shown in Fig. 12. Specifically, here, the absorbing boundary condition to be applied to the magnetic field component in order to prevent the electric field from being reflected at the edge (boundary) of the analysis region 10 is calculated.

[0078] The analysis space information memory 52 stores various information necessary for proceeding with electromagnetic field estimation in the analysis domain 10. For example, information such as the size of the analysis domain 10, its internal structure, electrical constants according to the physical properties of each structure, the frequency to be analyzed, and the number and positions of the human body models 18 is stored.

[0079] The voxel model storage memory 54 stores information about the voxel model that constitutes the analysis region 10 determined in step 122 shown in Fig. 12. Specifically, information about the voxel groups that constitute the surface voxels 22 or non-calculation voxels 24 of the human body model 18 is stored.

[0080] The parameter setting unit 56 is realized by, for example, an input interface provided in a computer system. The parameter setting unit 56 performs the function of setting parameters for specifying the calculation object model to be read in step 100 shown in FIG.

[0081] The array initialization unit 58 initializes the array of voxels in the analysis domain 10. The array initialization unit 58 is realized by, for example, an input interface provided in a computer system. The initialization by the array initialization unit 58 initializes, for example, information on the dielectric constant set for each voxel depending on the type of medium.

[0082] The array data reading unit 60 reads the array data necessary for generating the analytical region 10 determined in step 122 shown in Fig. 12. The data read by the array data reading unit 60 is stored in the voxel model storage memory 54.

[0083] 13 are capable of communicating with each other. These components can be divided into multiple housings, and assembled to form an electromagnetic field estimation system 40. However, these components may also be assembled into a single housing to form an electromagnetic field estimation device.

[0084] [Effects of the First Embodiment] As described above, the electromagnetic field estimation according to this embodiment allows non-calculation voxels 24 to be set inside the human body model 18, thereby eliminating these voxels from being subject to calculation. Therefore, the electromagnetic field estimation according to this embodiment significantly reduces the calculation load associated with the evaluation of the analysis region 10, and significantly reduces the main memory capacity and calculation time required for the calculation.

[0085] [Modification of the first embodiment] In the first embodiment described above, non-calculation voxels are set inside the human body model 18, but the setting region is not limited to the inside of the human body model 18. If an object that absorbs the electric field near its surface, similar to the human body model 18, exists inside the analysis region 10, non-calculation voxels may be set inside the object. [Explanation of symbols]

[0086] 10 Analysis area 12 xy plane 14 yz plane 16 zx plane 18 Human Body Models 20 Wave Source 22 surface voxels 24, 30, 30-1, 30-2 Non-computational voxels 28, 32 Lossy absorbing boundary 40 Electromagnetic Field Estimation System 42 Electric field component calculation section 44 Electric field component absorbing boundary condition calculation section 46 Electric field component lossy absorption condition calculation section 48 Magnetic field component calculation section 50 Magnetic field component absorbing boundary condition calculation section

Claims

1. An electromagnetic field estimation system for estimating electromagnetic field components occurring within an analysis region, comprising: at least one processor; at least one memory storing computer program code; The computer program code and the memory, together with the processor, configure the electromagnetic field estimation system to: Placing an object model having an electric field absorbing property within the analysis region; Dividing the entire analysis region including the object model into a plurality of voxels; setting a group of voxels covering a surface of the object model as surface voxels; setting a group of voxels located inside the surface voxels as non-calculation voxels; setting a boundary between the surface voxel and the non-calculation voxel to a lossy absorbing boundary; applying a lossy absorbing boundary condition to the lossy absorbing boundary so as not to reflect an electric field; releasing the non-calculation voxels from the calculation target and sequentially calculating electric field components and magnetic field components for other voxels in the analysis region; repeating the iterative calculations at predetermined time steps until convergence of the electric field components and the magnetic field components is observed; 1. An electromagnetic field estimation system configured to perform operations including:

2. The electromagnetic field estimation system according to claim 1 , wherein the surface voxels are set by a group of voxels in one layer covering the surface of the object model.

3. the non-calculation voxels are set by a group of voxels included in one or more rectangular parallelepipeds that fit inside the object model; The electromagnetic field estimation system according to claim 1 , wherein the surface voxels are set by a group of voxels that belong to the object model and do not belong to the non-calculation voxels.

4. The electromagnetic field estimation system according to claim 1 , wherein the object model is a human body model.

5. the surface constituting the lossy absorbing boundary is any one of a yz-plane absorbing boundary formed on a yz-plane in three-dimensional coordinates set for the analysis domain, a zx-plane absorbing boundary formed on a zx-plane, and an xy-plane absorbing boundary formed on an xy-plane, the lossy absorbing boundary conditions include a yz-plane condition applied to the yz-plane absorbing boundary, a zx-plane condition applied to the zx-plane absorbing boundary, and an xy-plane condition applied to the xy-plane absorbing boundary, The yz-plane condition includes a condition for not reflecting the y-direction electric field component Ey (x direction) propagating in the x direction, and a condition for not reflecting the z-direction electric field component Ez (x direction) propagating in the x direction, The zx plane condition includes a condition for not reflecting the z-direction electric field component Ez (y direction) propagating in the y direction, and a condition for not reflecting the x-direction electric field component Ex (y direction) propagating in the y direction, 2. The electromagnetic field estimation system according to claim 1, wherein the xy plane conditions include a condition for not reflecting an x-direction electric field component Ex (z direction) propagating in the z direction, and a condition for not reflecting a y-direction electric field component Ey (z direction) propagating in the z direction.

6. An electromagnetic field estimation device for estimating electromagnetic field components occurring within an analysis region, comprising: a parameter setting unit that places an object model having an electric field absorption characteristic within the analysis region; an analysis region determination unit that divides the entire analysis region including the object model into a plurality of voxels; a human body surface region extraction unit that sets a group of voxels covering the surface of the object model as surface voxels; a non-voxel release unit that sets a group of voxels located inside the surface voxels as non-calculation voxels; a lossy absorbing boundary setting unit that sets the boundary between the surface voxel and the non-calculation voxel as a lossy absorbing boundary; an electric field component absorption condition application unit that applies a lossy absorbing boundary condition to the lossy absorbing boundary so as not to reflect an electric field; an electric field component calculation unit and a magnetic field component calculation unit that release the non-calculation voxels from calculation targets and sequentially calculate electric field components and magnetic field components for other voxels in the analysis region; a convergence determination unit that repeatedly executes the iterative calculation at a predetermined time step until convergence of the electric field component and the magnetic field component is confirmed; An electromagnetic field estimation device configured to include:

7. An electromagnetic field estimation method for estimating electromagnetic field components occurring inside an analysis domain, comprising: For hardware, Placing an object model having an electric field absorbing property within the analysis region; Dividing the entire analysis region including the object model into a plurality of voxels; setting a group of voxels covering a surface of the object model as surface voxels; setting a group of voxels located inside the surface voxels as non-calculation voxels; setting a boundary between the surface voxel and the non-calculation voxel to a lossy absorbing boundary; applying a lossy absorbing boundary condition to the lossy absorbing boundary so as not to reflect an electric field; releasing the non-calculation voxels from the calculation target and sequentially calculating electric field components and magnetic field components for other voxels in the analysis region; repeating the iterative calculations at predetermined time steps until convergence of the electric field components and the magnetic field components is observed; The electromagnetic field estimation method includes:

8. An electromagnetic field estimation program for estimating electromagnetic field components occurring within an analysis region, At least one processor Placing an object model having an electric field absorbing property within the analysis region; Dividing the entire analysis region including the object model into a plurality of voxels; setting a group of voxels covering a surface of the object model as surface voxels; setting a group of voxels located inside the surface voxels as non-calculation voxels; setting a boundary between the surface voxel and the non-calculation voxel to a lossy absorbing boundary; applying a lossy absorbing boundary condition to the lossy absorbing boundary so as not to reflect an electric field; releasing the non-calculation voxels from the calculation target and sequentially calculating electric field components and magnetic field components for other voxels in the analysis region; repeating the iterative calculations at predetermined time steps until convergence of the electric field components and the magnetic field components is observed; A computer-readable program for estimating an electromagnetic field, comprising:

Images