A method for constructing a radio propagation model based on characteristics of an indoor propagation environment

Abstract

The application discloses a radio propagation model construction method based on indoor propagation environment characteristics, which comprises the following steps: firstly, taking a transmitting source position TX as a reference, and partitioning the whole indoor area according to the condition that each receiving area is separated from TX by several walls; secondly, obtaining the path loss distribution of the indoor scene of the radio wave propagation model to be constructed based on simulation, and dividing the path loss simulation values according to the partition condition; finally, fitting the path loss distribution data in each area by using the logarithmic path loss model containing the azimuth angle φ and the propagation distance d proposed by the application. The indoor area to be constructed by the radio wave propagation model is partitioned according to the condition that it is separated from the transmitting source by several walls, the simulation field intensity values of each area are fitted by using the logarithmic path loss model containing the azimuth angle φ and the propagation distance d proposed by the application, the error between the predicted value and the true value can be greatly reduced, and the prediction precision under the complex indoor environment is improved.

Description

Technical Field

[0001] This invention relates to the field of radio propagation coverage prediction, and in particular to a method for constructing a radio propagation model based on indoor propagation environment characteristics. Background Technology

[0002] Wireless communication, as the most fundamental communication technology in the 5G era, primarily transmits data information via radio waves through wireless channels. Therefore, channel modeling to predict radio wave propagation characteristics is fundamental to building wireless communication. However, due to the complex and variable environment of wireless channels, radio waves arrive at the receiving point in different transmission forms, resulting in different received and transmitted signals. Only by accurately predicting the propagation characteristics of wireless signals can reasonable strategies be provided for the design and deployment of wireless networks. Currently, wireless channel modeling methods can be mainly divided into three types: statistical modeling methods based on channel measurements; deterministic modeling methods that analyze and predict using the propagation environment and electromagnetic wave propagation theory; and semi-deterministic modeling methods that combine the advantages of the above two approaches to reduce complexity. The mainstream modeling methods are generally semi-deterministic and deterministic. These methods fit the data to create a model by understanding detailed channel environment information, such as geographical features, building structures, transceiver locations, and material properties. Compared to statistical modeling methods, this method saves a lot of experimental work, achieving prediction of propagation characteristics over a large area solely through the propagation environment.

[0003] However, constructing radio wave propagation models applicable to complex indoor scenarios is quite challenging. Traditionally, indoor radio wave models primarily employ logarithmic distance path loss models. These models reflect the propagation characteristics between transmitting and receiving devices by constructing a one-dimensional function of the logarithmically scaled distance between the transmitter (TX) and receiver (RX). Such models offer good prediction accuracy in unobstructed environments or with simple architectural structures, but become increasingly difficult to predict accurately as the environment becomes more complex. This is because the environment in which the wireless channel operates is complex and variable, leading to different transmission patterns in different areas. The received signal at the RX is often influenced by the indoor architectural environment. For example, the path loss at different RX points at the same distance from a fixed point TX is considered the same, but in reality, due to different indoor architectural structures, the path loss at the same distance can be completely different. Therefore, traditional one-dimensional logarithmic distance path loss models cannot accurately predict the propagation characteristics of radio waves in indoor environments. Therefore, how to construct a radio wave propagation model applicable to complex indoor environments has become a problem worthy of research. Summary of the Invention

[0004] The technical problem to be solved by the present invention is to overcome the defects of the prior art and provide a method for constructing a radio propagation model based on the characteristics of indoor propagation environment.

[0005] To solve the above-mentioned technical problems, the present invention provides the following technical solution:

[0006] This invention provides a method for constructing a radio propagation model based on indoor propagation environment characteristics, specifically including the following steps:

[0007] S1. Divide the indoor area where the radio wave propagation model is to be constructed into zones: Set up a transmission point TX, and divide the area into zones based on how many walls separate each receiving area of ​​the radio wave propagation model from the transmission point. That is, areas separated by the same number of walls are divided into the same area.

[0008] S2. Based on simulation, obtain the path loss distribution of the indoor scene of the radio wave propagation model to be constructed, and divide the simulated path loss values ​​in each partition according to the regional partitioning in step S1.

[0009] S3. For the path loss in each partitioned region, the proposed method including azimuth angle is adopted. By fitting the improved logarithmic path loss model with the propagation distance d, the fitting parameters that minimize the error between the predicted and simulated values ​​in each region are obtained, thus obtaining a radio wave propagation model suitable for each region.

[0010] As a preferred embodiment of the present invention, step S3 includes the azimuth angle. The improved logarithmic path loss model for the propagation distance d is: Equation 1

[0011]

[0012] Where d is the distance between the field strength receiving point RX and the transmitting point TX; The angle of counterclockwise rotation of the receiving point with the transmitting point as the origin of the polar coordinate system; This indicates that when the distance between RX and TX is d and the angle is... The path loss under the given conditions; d0 represents the reference distance, which is typically taken as 1 meter. This represents the reference path loss value at the reference distance d0; Indicates the path loss index; Indicates shadow decay; X σ Following a Gaussian distribution with a mean of 0 and a standard deviation of σ, σ typically ranges from 3.0 to 14.1 dB, considering environmental factors; parameter X σThe value of is determined as follows: under each partition condition, a value satisfying a Gaussian distribution with a mean of 0 and a standard deviation of 3 is randomly generated, and the average is taken after 10,000 random generation; the value of parameter n is determined by: using X generated each time. δ To determine the value of n, we take the average after 10,000 iterations.

[0013] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0014] This invention proposes a method that includes azimuth angles. The logarithmic path loss model, with propagation distance d as a parameter, divides the indoor research area into several regions based on different wall obstructions. By analyzing the azimuth angle changes between each region and the fixed TX coordinate, as well as the propagation distances between the minimum and maximum receiving points within that region, each region is fitted individually to obtain fitting models for different regions. This model can more accurately predict the radio propagation characteristics under various indoor conditions. The root mean square error between the predicted and actual values ​​at different RX points at the same distance remains consistently low, enabling more precise analysis of complex indoor environments. Attached Figure Description

[0015] The accompanying drawings are provided to further illustrate the invention and form part of the specification. They are used in conjunction with embodiments of the invention to explain the invention and do not constitute a limitation thereof. In the drawings:

[0016] Figure 1 This is a schematic diagram of the indoor area division in Example 1;

[0017] Figure 2 This is a graph showing the relationship between Euclidean distance and path loss for RX in each region of Example 1;

[0018] Figure 3 These are the fitted curves for each region under the TX1 condition in Example 1;

[0019] Figure 4 This is a graph showing the results of fitting data for all regions in Example 1;

[0020] Figure 5 These are the simulated path loss diagram and the model predicted path loss diagram of Example 1;

[0021] Figure 6 These are the RMSE diagrams for each model under the TX1 case in Example 1;

[0022] Figure 7 This is a graph showing the relationship between the Euclidean distance and path loss of RX in each area when the transmitter TX2 is inside the room in Example 2.

[0023] Figure 8This refers to the RMSE of each model in the TX2 case of Example 2. Detailed Implementation

[0024] The preferred embodiments of the present invention will be described below with reference to the accompanying drawings. It should be understood that the preferred embodiments described herein are for illustration and explanation only and are not intended to limit the present invention.

[0025] Example 1

[0026] like Figure 1-6 As shown, this invention provides a method for constructing a radio propagation model based on indoor propagation environment characteristics. Specifically, it involves dividing the indoor area into two scenarios: (a) a scenario where the transmitting source is in a corridor; and (b) a scenario where the transmitting source is inside a room. Figure 1 As shown.

[0027] Taking a point on the corridor as the emission point, labeled TX1; using TX1 as the origin of the polar coordinates, rotating counterclockwise, and dividing the room into six regions based on the number of walls penetrated, as shown in the figure, denoted as P1, P2, P3, P4, P5.1, and P5.2 respectively. In region P1, there is no wall obstruction between RX and TX1; in region P2, there is one wall obstruction between RX and TX1; in region P3, most RX and TX1 are obstructed by two walls; in region P4, most RX and TX1 are obstructed by three walls; in region P5.1, most RX and TX1 are obstructed by three walls, a very small number are obstructed by only two walls, and RX is also affected by multipath propagation; in region P5.2, most RX and TX1 are obstructed by four walls, a very small number are obstructed by three or two walls, and are also affected by multipath propagation. The black square represents the receiving point RX, which is a 0.1m × 0.1m square evenly distributed throughout the simulation area.

[0028] Electromagnetic field simulation software Altair WinProp was used to simulate radio wave propagation. The relative permittivity of air was set to 1, and the permittivity for concrete, wood, and glass was set to 6.8, 2.5, and 3.75, respectively. The simulation yielded the relationship between the distance and path loss between the receiving point RX and the transmitting point TX1 in each region, as shown below. Figure 2 As shown.

[0029] For each partition, the proposed path loss model was used to fit the simulation data, taking into account the different regions during the fitting process. Value range and d-value range. For region P1, relative to TX1, its propagation azimuth angle is 176.2°~185.6°, 354.9°~3.4°, and the propagation distance is 0m~9m; in region P2, the propagation azimuth angle is 6.8°~176.2°, 191°~349.7°, and the propagation distance is 1m~11m; in region P3, the propagation azimuth angle is 3.4°~49.79°, 313.09° The azimuth angle is 324.8° to 354.9° in region P4, with a propagation distance of 10m to 12m; in region P5.1, the azimuth angle is 3.4° to 25.9°, with a propagation distance of 12m to 13m; and in region P5.2, the azimuth angle is 332.13° to 354.9°, with a propagation distance of 12m to 13m.

[0030] The fitted curve is as follows Figure 3 As shown, Figure 4 To fit all points, the comparison shows that the model built by fitting each region separately after partitioning can more accurately predict the propagation characteristics within that region.

[0031] The coefficients and formulas obtained from fitting the radio wave propagation model for each region are shown in Table 1. Table 1 shows that the parameter with the greatest difference in fitting the radio wave propagation model for different regions is n. Without partitioning, the overall parameter n ranges from approximately 3.0; after partitioning, the parameters n are approximately 1.5, 2.7, 3.6, and 4.5 in regions P1–P4, respectively; and approximately 4.3 and 5.1 in regions P5.1 and P5.2, respectively. This indicates that without partitioning, the predicted values ​​obtained from the fitting formula cannot accurately predict the actual situation in each region. Since each region is distinguished based on the number of walls between the receiving area and TX1, Table 1 shows that in the case of no obstruction indoors, the parameter n of the path loss propagation model is approximately 1.5. For each additional wall obstructing the path, the value of n increases by 0.9–1. Therefore, based on the partitioned model, the number of walls between the receiving area and TX1 can be directly estimated and analyzed based on the values ​​of n in the model formula.

[0032] Table 1 Radio wave propagation models for different regions under TX1 conditions

[0033]

[0034]

[0035] Based on the propagation model obtained above, the received field strength is predicted for each region within the study area, as shown below. Figure 5 The propagation path loss heatmap shown in (b) is compared with the propagation path loss heatmap obtained from the simulation. Figure 5 Comparing (a), in region P1 without obstruction, the predicted results are almost identical to the actual results, indicating that the prediction is relatively accurate under these conditions. For region P2, the simulated field strength distribution is clearer in the area closer to the source, while the model prediction is calculated by averaging the field strength values ​​within that radius based on the relationship between angle and radius in the radio wave propagation formula. Therefore, it is not as clear as the simulation, but the model prediction still reflects the propagation characteristics under these conditions. Similarly, in regions P3, P4, P5.1, and P5.2, the proposed model can accurately predict the corresponding propagation characteristics. Observing the entire study area using the predicted model clearly shows significant differences in the field strength distribution within each region, which is in line with expectations.

[0036] To verify the reliability of the model, the propagation model was used to predict path loss values ​​at propagation radii of 1m, 3m, 5m, 7m, 9m, 11m, and 13m. The predictions were divided into those under unpartitioned conditions and those under partitioned conditions within each region. The predicted values ​​are the path loss (PL) values ​​calculated by each formula in Table 1 for each selected propagation radius. To obtain more accurate prediction results, each PL value is the average of the formulas calculated 10,000 times. The root mean square error (RMSE) between the model's predicted values ​​and the actual values ​​for each case is shown below. Figure 6 As shown, in the unpartitioned case, the RMSE increases continuously with the propagation distance, while the RMSE of the partitioned propagation model remains in a low range of 0-2dB. At a propagation distance of 12m, the RMSE after partitioning is approximately 11.5dB lower than that in the unpartitioned case, and the predicted value after partitioning is significantly better than that in the unpartitioned case. This indicates that the partitioned model has better predictive ability and is more representative of the radio wave propagation situation in the specified area.

[0037] Example 2

[0038] like Figure 7-8 As shown, this invention provides a method for constructing a radio propagation model based on indoor propagation environment characteristics, specifically: for Figure 1 The situation of the emission source (TX2) shown in (b) inside the room: with TX2 as the origin of the polar coordinates, rotate counterclockwise, and divide each room into six regions, namely P1, P2, P3, P4, P5 and P6, based on the number of walls it passes through.

[0039] Electromagnetic field simulation software Altair WinProp was used to simulate radio wave propagation. The relative permittivity of air was set to 1, and the permittivity of concrete, wood, and glass was set to 6.8, 2.5, and 3.75, respectively. The simulation yielded the relationship between the distance and path loss between the receiving point RX and the transmitting point TX2 in each region, as shown below. Figure 7 As shown.

[0040] For region P1, the propagation azimuth relative to TX2 is 180°–360°, and the propagation distance is 1m–7m; for region P2, the propagation azimuth is 180°–360°, and the propagation distance is 5m–11m; for region P3, the propagation azimuth is 202.4°–360°, and the propagation distance is 7m–8m; for region P4, the propagation azimuth is 210.5°–240.9° and 299.6°–345°, and the propagation distance is 11m–18m; for region P5, the propagation azimuth is 317.5°–335.5°, and the propagation distance is 15m–18m; for region P6, the propagation azimuth is 323.9°–339.3°, and the propagation distance is 15m–18m. The propagation models obtained by fitting each region based on formula (1) are shown in Table 2.

[0041] Table 2 Radio wave propagation models for each region under TX2 conditions

[0042]

[0043] To verify the reliability of the model, the propagation model was used to predict path loss values ​​at propagation radii of 1m, 3m, 5m, 7m, 9m, 11m, 13m, 15m, 17m, 19m, and 21m. The prediction results were divided into predicted values ​​under unpartitioned conditions and predicted values ​​with RMSE in each partitioned region. Figure 8 As shown in the figure, it can be seen that compared with the overall fitting model, the predicted values ​​of the partitioned model are always kept within a relatively small RMSE, indicating that the prediction accuracy is higher after partitioning.

[0044] Therefore, the radio wave propagation model based on the partitioning concept proposed in this invention has generalizability. After partitioning and adopting the proposed model including azimuth angles... After fitting the improved logarithmic path loss model with the propagation distance d, the error between the predicted value and the true value can be greatly reduced, and the prediction accuracy of radio waves in complex indoor environments can be improved. Compared with the traditional radio wave propagation model that only considers the propagation distance, the propagation model obtained by this method can more clearly analyze the indoor building conditions and radio wave propagation.

[0045] Finally, it should be noted that the above descriptions are merely preferred embodiments of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing embodiments or make equivalent substitutions for some of the technical features. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method of constructing a radio propagation model based on characteristics of an indoor propagation environment, characterized by, Specifically comprising the following steps: S1, partitioning the indoor area to be constructed with the radio wave propagation model: setting a transmitting point TX, and partitioning the receiving area to be constructed with the radio wave propagation model from the transmitting point on the condition that several walls are separated, that is, the areas separated by the same number of walls are divided into the same area; S2, obtaining the path loss distribution of the indoor scene to be constructed with the radio wave propagation model based on simulation, and dividing the path loss simulation values in each partition according to the area partitioning condition in step S1; S3, the path loss in each division area is fitted with the improved log path loss model containing azimuth angle and propagation distance d, the fitting parameters in each area that minimize the error between the predicted value and the simulation value are obtained, and thus the radio wave propagation model suitable for each area is obtained; In step S3, the improved log path loss model including the azimuth angle and the propagation distance d is: Equation 1 wherein d is the distance between the field intensity receiving point RX and the transmitting point TX; is the angle of rotation in the counterclockwise direction corresponding to the receiving point with the transmitting point as the polar coordinate origin; PL(d, ) represents the path loss when the distance between RX and TX is d and the angle is ; d0 represents the reference distance; PL0(d0, ) represents the reference path loss value when the reference distance d0; ; n represents the path loss index; represents the shadow fading; follows a Gaussian distribution with a mean of 0 and a standard deviation of .

2. The method of claim 1, wherein the method further comprises: determining a characteristic of the indoor propagation environment; and selecting the radio propagation model based on the determined characteristic of the indoor propagation environment. Parameter The value mode is: under each partition condition, a randomly generated value satisfying the Gaussian distribution with mean 0 and standard deviation 3, and the average is taken after 10000 times of random generation; the value mode of parameter n is: the value of n this time is determined by X δ generated each time, so the average is taken after 10000 times.

Images