Indoor interference source positioning method based on radio wave propagation algorithm and nearest neighbor algorithm

Abstract

The invention belongs to the field of electromagnetic interference source positioning, and provides an indoor interference source positioning method based on a radio wave propagation algorithm and a nearest neighbor algorithm, which comprises the steps of simulation model construction, first interference source distribution and second interference source distribution initialization, receiver setting, first training data set simulation, second training data set simulation and KNN model screening. According to the invention, by combining the ray tracing method, the finite difference time domain method and the KNN model, the interference source in the complex environment is positioned; and by constructing the simulation model, the reconstruction cost of the entity environment is reduced.

Description

Technical Field

[0001] The invention relates to the field of electromagnetic interference source positioning, and in particular to an indoor interference source positioning method based on a radio wave propagation algorithm and a nearest neighbor algorithm. Background Art

[0002] With the continuous improvement of the automation level of production lines in the modern industrialized automobile production process, the number of various frequency-consuming and power-consuming equipment has increased significantly, and there are a large number of potential electromagnetic interference sources in space, and the electromagnetic environment formed by them is becoming increasingly complex. If targeted protection is carried out, the above electromagnetic environment may affect the operation of the equipment and the cooperation between them. In severe cases, it will not only lead to a decline in production quality and production efficiency, but also seriously affect the operation safety of the entire production line. The radio wave propagation algorithm can be divided into two types: high-frequency approximation method and full-wave numerical method. Among them, the high-frequency approximation method equates electromagnetic waves to rays, which is suitable for calculating relatively regular geometric shapes and has the advantage of fast calculation speed. It is usually used to calculate electrically large obstacles; the full-wave numerical method starts from the Maxwell equations and simulates the change process of the electromagnetic field. The calculation accuracy of the scattering field of the geometric shape is related to the density of the space grid. It is suitable for calculating electrically small obstacles and complex structural obstacles. Common high-frequency approximation methods include geometric optics, physical optics, and consistent diffraction theory. Among them, the geometric optics method mainly considers the reflection and refraction propagation of electromagnetic waves in geometric structures. In the radio wave simulation of electrically large objects, its accuracy depends on the modeling accuracy of the geometric structure. Common geometric optics methods are ray tracing methods based on the mirror method or the bouncing ray method, abbreviated as RT. For the simulation of indoor high-frequency radio wave propagation, the use of RT for simulation can greatly improve the computational efficiency. Common full-wave numerical methods include the time-domain finite difference method, abbreviated as FDTD, the finite element method and the moment method. For broadband signal simulation and electrically small obstacle simulation, the use of FDTD can achieve high simulation accuracy while having low computational complexity. Therefore, FDTD is suitable for the simulation of scattered fields of electrically small obstacles in indoor environments.

[0003] Existing positioning methods all use the characteristics of received signals as the basis for positioning estimation, and the obstacle environment is treated as a black box problem without integrating into the real environment. When changing the obstacle environment requires a high cost, electromagnetic simulation needs to be performed in advance to reduce construction costs. In addition, since the indoor radio wave propagation environment is more complex than the outdoor environment, it will produce more reflection and diffraction phenomena than the outdoor environment. At the same time, the impact of obstacles with smaller electrical size on the signal is difficult to estimate using empirical methods. Summary of the invention

[0004] In order to overcome the shortcomings of the prior art, the purpose of the present invention is to provide a method for locating indoor interference sources based on radio wave propagation algorithm and nearest neighbor algorithm, which can realize the positioning of interference sources in complex environments by combining ray tracing method, time-domain finite difference method and KNN model; and reduce the reconstruction cost of the physical environment by constructing a simulation model.

[0005] To achieve the above object, the present invention provides the following solutions:

[0006] A method for locating indoor interference sources based on a radio wave propagation algorithm and a nearest neighbor algorithm, comprising:

[0007] Constructing a simulation model; the simulation model includes: a metal plate obstacle and a metal room;

[0008] Initializing a first interference source distribution and a second interference source distribution according to the simulation model;

[0009] evenly arranging receivers in the metal room;

[0010] Circularly lighting up the interference source virtual devices in the first interference source distribution, and calculating the signal strength value of the signal received by each of the receivers using a ray tracing method and a radio wave propagation algorithm to obtain a first training data set;

[0011] Circularly lighting up the interference source virtual devices in the second interference source distribution, and calculating the signal strength value of the signal received by each receiver by using a finite difference time domain method and a radio wave propagation algorithm to obtain a second training data set;

[0012] The first training data set and the second training data set are respectively input into a preset KNN model for nearest neighbor screening to obtain a high-frequency interference source positioning result and a low-frequency interference source positioning result.

[0013] Preferably, the K value of the KNN model is 4.

[0014] Preferably, the interference frequency in the first interference source distribution is 2.4 GHz; and the interference frequency in the second interference source distribution is 80 MHz.

[0015] Preferably, the spatial step length when acquiring the second training data set is 0.25m.

[0016] Preferably, the distances between adjacent interference source virtual devices in the x-, y- and z-axis directions in the first interference source distribution are 0.5m, 0.5m and 0.5m respectively; the distances between adjacent interference source virtual devices in the x-, y- and z-axis directions in the second interference source distribution are 0.25m, 0.25m and 0.25m respectively.

[0017] The present invention discloses the following technical effects:

[0018] The present invention provides an indoor interference source positioning method based on a radio wave propagation algorithm and a nearest neighbor algorithm. By combining a ray tracing method, a time-domain finite difference method and a KNN model, the problem of positioning small-size interference sources by traditional methods is solved, and the positioning of interference sources in complex environments is realized. By constructing a simulation model, the problem of high cost of traditional methods is solved, and the function of locating interference sources through environmental simulation is realized. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0020] Figure 1 A schematic diagram of the indoor interference source positioning process based on the radio wave propagation algorithm and the nearest neighbor algorithm provided in an embodiment of the present invention;

[0021] Figure 2 A schematic diagram of the mirror image method principle provided by an embodiment of the present invention;

[0022] Figure 3 A schematic diagram of the secondary reflection principle provided by an embodiment of the present invention;

[0023] Figure 4 A three-dimensional schematic diagram of a room and obstacles provided by an embodiment of the present invention;

[0024] Figure 5 A schematic diagram of modeling visualization provided by an embodiment of the present invention;

[0025] Figure 6 A schematic diagram of radio wave propagation visualization provided by an embodiment of the present invention;

[0026] Figure 7 A three-dimensional schematic diagram of a simulation environment provided by an embodiment of the present invention;

[0027] Figure 8 A schematic diagram of the location of the emission source provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0028] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0029] The purpose of the present invention is to provide an indoor interference source positioning method based on radio wave propagation algorithm and nearest neighbor algorithm, which realizes the positioning of interference sources in complex environments by combining ray tracing method, time-domain finite difference method and KNN model; and reduces the reconstruction cost of the physical environment by constructing a simulation model.

[0030] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the present invention is further described in detail below with reference to the accompanying drawings and specific embodiments.

[0031] Figure 1 A schematic diagram of the indoor interference source positioning process based on the radio wave propagation algorithm and the nearest neighbor algorithm provided in an embodiment of the present invention is shown in FIG. Figure 1 As shown, the present invention provides an indoor interference source positioning method based on radio wave propagation algorithm and nearest neighbor algorithm, comprising:

[0032] Step 100: constructing a simulation model; the simulation model includes: a metal plate obstacle and a metal room;

[0033] Step 200: Initializing a first interference source distribution and a second interference source distribution according to the simulation model;

[0034] Step 300: evenly arranging receivers in the metal room;

[0035] Step 400: cyclically lighting up the interference source virtual devices in the first interference source distribution, and calculating the signal strength value of the signal received by each of the receivers using a ray tracing method and a radio wave propagation algorithm to obtain a first training data set;

[0036] Step 500: cyclically lighting up the interference source virtual devices in the second interference source distribution, and calculating the signal strength value of the signal received by each receiver using a finite difference time domain method and a radio wave propagation algorithm to obtain a second training data set;

[0037] Step 600: Input the first training data set and the second training data set into a preset KNN model respectively to perform nearest neighbor screening to obtain a high-frequency interference source positioning result and a low-frequency interference source positioning result.

[0038] Preferably, the K value of the KNN model is 4.

[0039] Specifically, the interference frequency in the first interference source distribution is 2.4 GHz; and the interference frequency in the second interference source distribution is 80 MHz.

[0040] Optionally, the spatial step length when acquiring the second training data set is 0.25m.

[0041] Preferably, the distances between adjacent interference source virtual devices in the x-, y- and z-axis directions in the first interference source distribution are 0.5m, 0.5m and 0.5m respectively; the distances between adjacent interference source virtual devices in the x-, y- and z-axis directions in the second interference source distribution are 0.25m, 0.25m and 0.25m respectively.

[0042] Preferably, ray tracing technology is a technology used in the field of mobile and personal communications to predict radio wave propagation characteristics. It identifies all possible ray paths and calculates parameters such as amplitude, phase, delay and polarization of each path. Combined with the antenna pattern and system bandwidth, the coherent synthesis result of the receiving point is obtained. Ray tracing technology The ray tracing method is based on the principle of geometric optics and determines the reflection, refraction and shadow effects by simulating the propagation path of the ray. For the diffraction effect of obstacles, the ray tracing method is supplemented and improved by introducing the general diffraction theory. This embodiment uses the mirror method as the basis of the ray tracing method. The mirror method is constructed based on the principles of reflection law, refraction law and analytic geometry, and its core basis is the uniqueness theorem. Within the scope of geometric optics, the reflection line is determined by identifying the mirror point, so that the propagation path of the ray can be determined. Therefore, the mirror theory is applicable to the prediction of radio wave propagation in microcellular areas. In the traditional mirror method, a source point generates a mirror point on the reflection surface, and this mirror point can generate more mirror points on other reflection surfaces. These new mirror points can also continue to generate more mirror points, forming a continuous mirror point generation process. However, in complex environments, it becomes quite difficult to determine which objects act as scatterers to generate mirror points, so this method is usually applicable to scenes that can be simplified to planar polyhedral models.

[0043] Specifically, assuming that the source point S and the polyhedron face are known, the rays reflected by the polyhedron face can be considered as the mirror source point I directly radiating the rays according to the principle of the mirror image method. The position of the mirror source point I and the position of the real source point S are symmetrical about the plane where the polyhedron face is located (refer to Figure 2). By analyzing the radiation characteristics of the real source point S and the electromagnetic properties of the polyhedron, we can obtain relevant data about the radiation field of the mirror source point I. For a given observation point O, the reflection point R is the intersection of the line segment IO and the polyhedron face. In the case of N planar polyhedrons, the number of mirrors is N, and the number of reflected rays reaching the observation point will not exceed N. This is because the size of the polyhedron face is finite, and the reflected rays can only be received by the observation point located in the reflection space of the polyhedron face. When the reflection point R is located on the polyhedron face, the observation point must be located in the reflection space of the polyhedron face. The analysis method of the secondary reflected rays is similar. The source of the secondary reflected ray is the mirror image of the mirror image that has undergone one reflection, that is, the second-order mirror image. For secondary reflection, three conditions must be met: (a) the observation point must be located in the reflection space of the second polyhedron face; (b) the second reflection point must be located on the first polyhedron face; (c) in Figure 3 None of the three paths depicted in (S-R1, R1-R2, R2-O) are obscured by other polyhedral faces in the environment. Multiple reflections can be analyzed in a similar way. For k reflections, the number of mirror images will be N(N-1)k-1. Mirror theory allows determining the exact path of the reflected ray and therefore the phase information of the field associated with the reflected ray. Therefore, the contribution of the multipath transmission components to the received field at the observation point can be obtained by the method of coherent summation.

[0044] Preferably, the ray tracing method has the following advantages: for electrically large-sized objects, the ray tracing method has the advantages of fast calculation speed and high simulation accuracy; the ray tracing method can reduce the requirements of computer resources; the ray tracing method is suitable for high-frequency electromagnetic wave simulation.

[0045] Specifically, this embodiment establishes a metal room with a length of 10m, a width of 8m, and a height of 4m, corresponding to the x-axis, y-axis, and z-axis respectively. A metal plate obstacle with a length of 3m and a height of 2m is placed 5.6m away from the y-axis direction. The coordinates of the center point of the metal plate obstacle are (6.5, 6.2). Figure 4 . This embodiment uses 2.4GHz electromagnetic waves as the interference source. Since the exact location of the interference source cannot be determined, it is necessary to assume that interference sources may appear everywhere in the room. To this end, before RT simulation, it is necessary to traverse the interference sources in a uniformly placed manner, that is, to place an interference source at an equal distance. In this method, an interference source is placed every 0.5m in the x-axis direction, an interference source is placed every 0.5m in the y direction, and an interference source is placed every 0.5m in the z direction. The total number of interference sources is 39×31×7=8463 interference sources.

[0046] Furthermore, this method uses 4 receivers placed at different positions to receive signals from different interference sources. The receiver coordinates are (4.4, 1.5), (4.6, 1.5), (6.4, 1.5), and (6.6, 1.5). When the program is executed, each interference source is turned on in a loop, and the signal strength value of the arriving signal is received by the 4 receivers. After the loop is completed, the training data set can be generated.

[0047] Specifically, in the process of generating the test data set, 50 coordinate positions are selected by random numbers, and the signal strength values ​​of the arriving signals received by 4 receivers. The training data set and the test data set are input into the KNN algorithm, and K=4 is selected, and the average positioning error of the interference source is 3.72m. The KNN algorithm used in this embodiment can achieve a positioning calculation time of seconds in an ordinary personal computer (only milliseconds are required when only KNN is executed). Unlike the positioning calculation time, as the number of obstacles increases and the number of interference source settings increases, the calculation time of the training set increases, and usually requires hours of calculation time. This is because the calculation time of RT is proportional to the number of obstacles and the number of transmission sources. However, since the calculation speed of RT is independent of the wavelength, it is suitable for high-frequency interference source simulation.

[0048] Furthermore, the method designs a visualization interface for radio wave propagation from transmitter to receiver. Figure 5 A visualization diagram of the program modeling of this embodiment is shown, where the blue "X" is a transmitter and the yellow "X" is a receiver. Figure 6 A diagram showing a visualization of radio wave propagation from a transmitter to a receiver.

[0049] Specifically, the finite-difference time-domain method is a technique widely used in the numerical analysis of electromagnetic fields. Based on the principle of differential calculation, this method directly starts from the Maxwell equations that describe electromagnetic phenomena and converts the continuous differential form of these equations into discrete differential forms. By sampling the electromagnetic field within a specific spatial volume and time span, FDTD is able to simulate physical processes such as the propagation, scattering, reflection and refraction of electromagnetic waves. The key elements of the FDTD method include the Yee grid structure, the division of the calculation area, the application of absorbing boundary conditions, and the transformation from near field to far field.

[0050] Furthermore, in the FDTD method, the design of the Yee cell allows the components of the electric and magnetic fields to be sampled alternately in space and time. Each electric field component is surrounded by four magnetic field components, and vice versa. This unique sampling pattern allows the Maxwell curl equations containing time variables to be converted into a set of difference equations, and then the electromagnetic field problem in space is solved by stepping forward on the time axis.

[0051] Specifically, in the numerical calculation of the simulated electromagnetic field, the new value of the electric field and magnetic field at each grid point depends not only on the value of the point in the previous time step, but also on the value of the field component of the other field at the surrounding adjacent points half a time step ago. By establishing the difference equations of these field components, we can let the time step perform alternating calculations of the electric field and magnetic field for the grid points of the entire simulation area one by one. After a sufficient number of time steps, we can get the required time domain numerical solution.

[0052] Preferably, FDTD has the following advantages: it can accurately calculate the structure of objects with electrical dimensions less than 3 times the wavelength, and the simulation accuracy of obstacles with small electrical dimensions and complex structures is better than that of the ray tracing method; it is more suitable for the simulation of electromagnetic waves with lower frequencies in indoor environments; because it belongs to the time domain method, it is more efficient for the simulation of broadband signals; it can simulate electronic devices such as antennas and circuits. In this embodiment, we use FDTD to simulate the scattered field of obstacles with small electrical dimensions, and when the electromagnetic wave frequency is relatively low (such as electromagnetic waves below 300MHz), the complex scattered field simulation at the corners of the indoor environment.

[0053] Specifically, this embodiment establishes a metal room with a length of 10m, a width of 8m, and a height of 4m, corresponding to the x-axis, y-axis, and z-axis respectively. A metal square obstacle with a length of 1m, a width of 1m, and a height of 1m is placed at the center of the floor. The coordinates of the center point of the square obstacle are (5.5, 4.5, 0.5). The three-dimensional schematic diagram of the room and the obstacle is as follows Figure 7 shown.

[0054] Furthermore, this embodiment uses 80MHz electromagnetic waves as the interference source and uses the Taflove program for simulation. Assume that the interference source is located in the corner of the room, such as Figure 8 The x-coordinate range is 0-2m, the y-direction range is 0-2m, and the z-direction range is 0-2m. An interference source is placed every 0.25m in the x, y, and z-axis directions. The total number of interference sources is 8×8×8=512.

[0055] Optionally, in the FDTD parameter setting, the spatial step of the three coordinate axes is set to 0.25 m to meet the condition that the spatial step is less than 1 / 12 wavelength, the time step meets the Courant condition, and the total iteration time meets the time required for the electromagnetic wave to pass through the longest reflection distance three times.

[0056] Preferably, relative to the 80MHz interference source, the corners of the room and the metal blocks can be regarded as obstacles with smaller electrical size, which can be used to verify the accuracy of FDTD. This method uses 4 receivers placed at different positions to receive signals from different interference sources, and the receiver coordinates are (3, 3.2), (3, 6.2), (6, 3.2), and (6, 6.2). When the program is executed, each interference source is cycled and the signal strength value of the arriving signal is received by the 4 receivers. After the cycle is completed, a training data set can be generated.

[0057] Specifically, in the process of generating the test data set, the test set interference source selects the position when the x, y, and z coordinates are 0.4, 0.8, 1.2, and 1.6 respectively as the position of the test interference source to ensure that the position of the test set interference source does not overlap near the position of the training set interference source. When the program is executed, each interference source is illuminated in a loop, and the four receivers receive the signal strength value of the arriving signal. The training data set and the test data set are input into the KNN algorithm, and K=4 is selected, and the average positioning error of the interference source is 0.93m. It can be seen that FDTD can obtain good simulation accuracy when calculating low-frequency signals.

[0058] The beneficial effects of the present invention are as follows:

[0059] The present invention realizes the positioning of interference sources in complex environments by combining ray tracing method, time-domain finite difference method and KNN model; and reduces the reconstruction cost of the physical environment by constructing a simulation model.

[0060] The various embodiments in this specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the various embodiments can be referenced to each other.

[0061] The principles and implementation methods of the present invention are described in this article using specific examples. The description of the above embodiments is only used to help understand the method and core idea of ​​the present invention. At the same time, for those skilled in the art, according to the idea of ​​the present invention, there will be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as limiting the present invention.

Claims

1. A method for locating indoor interference sources based on radio wave propagation algorithm and nearest neighbor algorithm, characterized in that: include: Build simulation models; The simulation model includes: a metal plate obstacle and a metal room; Initializing a first interference source distribution and a second interference source distribution according to the simulation model; evenly arranging receivers in the metal room; Circularly lighting up the interference source virtual devices in the first interference source distribution, and calculating the signal strength value of the signal received by each of the receivers using a ray tracing method and a radio wave propagation algorithm to obtain a first training data set; Circularly lighting up the interference source virtual devices in the second interference source distribution, and calculating the signal strength value of the signal received by each receiver by using a finite difference time domain method and a radio wave propagation algorithm to obtain a second training data set; The first training data set and the second training data set are respectively input into a preset KNN model for nearest neighbor screening to obtain a high-frequency interference source positioning result and a low-frequency interference source positioning result.

2. According to claim 1, a method for locating indoor interference sources based on radio wave propagation algorithm and nearest neighbor algorithm is characterized in that: The K value of the KNN model is 4.

3. The indoor interference source positioning method based on radio wave propagation algorithm and nearest neighbor algorithm according to claim 1 is characterized in that: The interference frequency in the first interference source distribution is 2.4 GHz; and the interference frequency in the second interference source distribution is 80 MHz.

4. The indoor interference source positioning method based on radio wave propagation algorithm and nearest neighbor algorithm according to claim 1 is characterized in that: The spatial step length when the second training data set is acquired is 0.25m.

5. The indoor interference source positioning method based on radio wave propagation algorithm and nearest neighbor algorithm according to claim 1 is characterized in that: In the first interference source distribution, the distances between adjacent interference source virtual devices in the x, y, and z axis directions are 0.5m, 0.5m, and 0.5m, respectively; in the second interference source distribution, the distances between adjacent interference source virtual devices in the x, y, and z axis directions are 0.25m, 0.25m, and 0.25m, respectively.

Images