A radio wave propagation channel modeling method for predicting received signal power

Abstract

The present invention relates to a radio wave propagation channel modeling method for predicting received signal power, comprising the following steps: S1, inputting a digital map of a target measurement area; S2, determining an ellipsoidal Fresnel zone, a first spherical zone, and a second spherical zone according to the position of a transceiver; S3, setting a scatterer position and the probability that the scatterer can affect radio wave propagation according to the area determined in step S2, and updating the scatterer distribution according to the movement of the transceiver; S4, simulating radio wave propagation using a hybrid simulation model of ray tracing and propagation graph theory; S5, training a statistical model of line-of-sight propagation path fading under set obstruction according to measurement data, embedding it into a simulation model, and obtaining an optimized prediction model; S6, outputting a predicted reception level and channel characteristic parameters. Compared with the prior art, the present invention simulates radio wave propagation in a line-of-sight path with certain obstructions in an urban scene, and has the advantages of low algorithm complexity and high prediction accuracy.

Description

Technical Field

[0001] The present invention relates to the field of wireless communication technology, and in particular to a radio wave propagation channel modeling method for predicting received signal power. Background Art

[0002] With the increasing development of wireless communication technology, the research on radio wave propagation prediction has received more and more attention, especially the modeling of wireless channels in modern urban environments. The prediction of radio wave propagation in complex urban environments plays an important role in automotive electronics, intelligent transportation, urban Internet of Things, and electromagnetic warfare in national defense and military. In view of the difficulty and high cost of large-scale complex urban measurements, ray tracing algorithms and point cloud-based modeling algorithms have been proposed and applied to urban channel prediction. The random scattering components generated by the propagation graph theory algorithm can effectively predict the overall trend of radio wave propagation. Based on the successful application of propagation graph theory in large-scale scenes such as high-speed rail, if it is applied to urban scenes, it can effectively reduce the complexity of the algorithm and has more advantages than ray tracing in terms of computing time. Summary of the invention

[0003] The purpose of the present invention is to overcome the defects of the above-mentioned prior art and provide a radio wave propagation channel modeling method for predicting the received signal power.

[0004] The purpose of the present invention can be achieved by the following technical solutions:

[0005] A radio wave propagation channel modeling method for predicting received signal power comprises the following steps:

[0006] S1. Input the digital map of the target measurement area;

[0007] S2, according to the transmitter T x With the receiving end R x Position, determine the ellipsoidal Fresnel zone, and then determine the first spherical area and the second spherical area with the transmitting end and the receiving end as the center and the modeled propagation distance as the radius;

[0008] S3, setting the scatterer position and the probability that the scatterer can affect the propagation of radio waves according to the ellipsoidal Fresnel zone, the first spherical zone and the second spherical zone determined in step S2, and updating the scatterer distribution according to the movement of the transceiver;

[0009] S4, according to the scatterer updated in step S3, using a hybrid simulation model of ray tracing and propagation graph theory to simulate radio wave propagation;

[0010] S5, measuring the power level measurement data under the same urban environment, training a statistical model of line-of-sight propagation path fading under set obstruction based on the measured data, optimizing and adjusting the simulation model parameters and area size of step S4, and obtaining an optimized prediction model;

[0011] S6. According to the optimized model, the simulation output predicts the receiving level and channel characteristic parameters.

[0012] Furthermore, the construction of the Fresnel zone in step S2 is based on the transmitting end and the receiving end as the focus, and the straight path |T x R x | and the polyline path with inflection point P |PT x |+|PR x |When the travel difference is nλ / 2, the first, second, ..., nth Fresnel zones are constructed by the first, second, ..., nth ellipsoidal surfaces formed by the inflection point P.

[0013] Furthermore, the radii of the first spherical area and the second spherical area in step S2 are modeled propagation distances, and the modeled propagation distance in an urban area scenario is selected to be in the range of 200m to 300m.

[0014] Furthermore, the specific settings of the random scatterers in the selected area and the probability that the scatterers can affect the propagation of radio waves in step S3 are:

[0015] S301, based on the outer surface of the building, the top surface of the building and the ground without buildings in the selected area, set a uniform scatterer density to generate an original random scatterer s raw ;

[0016] S302, setting the probability that the scatterers can affect the propagation of radio waves, i.e., the lighting probability, in descending order according to the constructed first to nth Fresnel zones;

[0017] S303, according to actual environmental requirements, further dividing the first spherical area and the second spherical area, and according to the different divided areas, the closer the scatterer is to the transmitting end and the receiving end, the higher the probability of the scatterer being lit;

[0018] S304, according to the lighting probability of the scatterers in each area, some scatterers of the original scatterers are selected as effective scatterers s e , used for simulation calculation of propagation graph theory.

[0019] Furthermore, in step S3, a height threshold is set for buildings within a certain range outside the selected area, and random scatterers and a high scatterer lighting probability are set on the building surface above the threshold, and the generated high-rise building scatterers s h , used for simulation calculation of propagation graph theory.

[0020] Furthermore, in step S3, as the position of the transceiver moves, the two spherical areas and the Fresnel area centered on the transceiver are updated, and the scatterers s and s are updated respectively. raw 、s e and h Used for simulation calculations.

[0021] Furthermore, the channel transfer function of the hybrid model of ray tracing and propagation graph theory in step S4 is:

[0022] H(f)=H RT,0:1 (f)+H PG,2:∞ (f)

[0023] Where H(f) is the channel transfer function of the hybrid model of ray tracing and propagation graph theory, H PG,2:∞ (f) is the channel transfer function between the transmitter and the receiver generated by the propagation graph theory, H RT,0:1 (f) is the channel function generated by the 0th to 1st order path of ray tracing.

[0024] Furthermore, the channel transfer function between the transmitter and the receiver generated by the propagation graph theory is:

[0025] H PG,2:∞ (f) = D (f) + R (f) (B (f) + B (f) 2 +…)T(f)

[0026] =D(f)+R(f)(B(f)(IB(f)) -1 )T(f)

[0027] Wherein, D(f) is the channel transfer function of the direct path from the transmitter to the receiver, R(f) represents the channel transfer function from the discretized scatterer to the receiver, T(f) represents the channel transfer function from the transmitter to the discretized scatterer, and B(f) represents the channel transfer function between the discretized scatterers.

[0028] Furthermore, the statistical model of line-of-sight propagation path fading under set obstruction uses the received power value P of the measured data and processes the received power value P using a sliding window to obtain P ′ , calculate the fading value (PP ′ ) and count the probability of occurrence, and calculate the average receiving end sample interval in each interval.

[0029] Furthermore, after the training, a statistical model of line-of-sight propagation path fading under set obstruction is obtained and embedded in the hybrid simulation model of step S4. The first simulation sample is used as a reference to calculate the intervals between other samples and the first simulation sample. When the integer multiple of the receiving end sample interval is satisfied, the sample simulated receiving power P minus the fading value under the corresponding interval (PP ′ ) to simulate the line-of-sight propagation path scenario.

[0030] Compared with the prior art, the present invention has the following beneficial effects:

[0031] 1. The present invention proposes a method for modeling the area based on the Fresnel zone and the area centered on the transceiver for the urban scene and setting the probability of the scatterer to generate the original random scatterer s raw , effective scatterer s e and high-rise building scatterers h , and is used in the simulation calculation of propagation graph theory, which can effectively capture the main contribution of the propagation path. As the position of the transmitter and receiver changes, the scatterer distribution is updated in real time, which can effectively simulate the channel characteristics in time-varying scenarios.

[0032] 2. Based on the propagation graph theory, the present invention introduces a hybrid model of first-order ray tracing, by first calculating the channel transfer function H between the transmitter and the receiver PG,2:∞ (f), and then calculate the channel transfer function H(f) of the hybrid model of ray tracing and propagation graph theory. This ensures that the accuracy of channel simulation based on propagation graph theory is improved under the premise of slightly increasing the computational complexity.

[0033] 3. The present invention trains the statistical model of line-of-sight propagation path fading through measured data, optimizes and adjusts the simulation model parameters and area size, and embeds them into the simulation results, which can effectively simulate the fading characteristics of the receiving level in real propagation and further improve the accuracy of the channel simulation model. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] Figure 1 is a flow chart of the present invention;

[0035] Figure 2 is a schematic diagram of a simulation environment in an embodiment of the present invention;

[0036] Figure 3 Schematic diagram of a method for simulating scene scatterer distribution in an embodiment of the present invention;

[0037] Figure 4 It is a path propagation diagram of a simulation scenario in an embodiment of the present invention;

[0038] Figure 5 A probability distribution diagram of the fading value in the simulation process in an embodiment of the present invention;

[0039] Figure 6 is the fading value within the average Rx sample interval of the simulation process in the embodiment of the present invention;

[0040] Figure 7 A comparison diagram of received power curves of the traditional graph theory simulation method and the actual measurement in the embodiment of the present invention;

[0041] Figure 8 It is a comparison diagram of the improved simulation and measured received power curves in the embodiment of the present invention. DETAILED DESCRIPTION

[0042] The present invention is described in detail below in conjunction with the accompanying drawings and specific embodiments. This embodiment is implemented based on the technical solution of the present invention, and provides a detailed implementation method and specific operation process, but the protection scope of the present invention is not limited to the following embodiments.

[0043] like Figure 1 As shown, a radio wave propagation channel modeling method for predicting received signal power includes the following steps:

[0044] S1. Input the digital map of the target measurement area;

[0045] S2, according to the transmitter T x With the receiving end R x Position, determine the ellipsoidal Fresnel zone, and then determine the first spherical area and the second spherical area with the transmitting end and the receiving end as the center and the modeled propagation distance as the radius;

[0046] S3, setting the scatterer position and the probability that the scatterer can affect the propagation of radio waves according to the ellipsoidal Fresnel zone, the first spherical zone and the second spherical zone determined in step S2, and updating the scatterer distribution according to the movement of the transceiver;

[0047] S4, according to the scatterer updated in step S3, using a hybrid simulation model of ray tracing and propagation graph theory to simulate radio wave propagation;

[0048] S5, measuring the power level measurement data under the same urban environment, training a statistical model of line-of-sight propagation path fading under set obstruction based on the measured data, optimizing and adjusting the simulation model parameters and area size of step S4, and obtaining an optimized prediction model;

[0049] S6. According to the optimized model, the simulation output predicts the receiving level and channel characteristic parameters.

[0050] In this embodiment, the digital map of the urban area in step S1 is as follows: Figure 2As shown, the map covers a total of 4km*4km. The asterisk indicates the transmitting end located at the base station, which is about 37m from the ground. The line connected by the dots indicates the movement trajectory of the receiving end, which is the location information collected by the receiving antenna fixed on the roof of the car.

[0051] like Figure 3 As shown, the construction of the Fresnel zone in step S2 takes the transmitting end and the receiving end as the focus, and the straight path |T x R x | and the polyline path with inflection point P |PT x |+|PR x When the stroke difference is nλ / 2, the first, second, ..., nth Fresnel zones are constructed by the first, second, ..., nth ellipsoidal surfaces formed by the inflection point P. In this embodiment, the number of Fresnel zones is n=5.

[0052] The radii of the first spherical area and the second spherical area in step S2 are the modeled propagation distances. The modeled propagation distance in the urban scene is selected to be in the range of 200m to 300m. In this embodiment, the short propagation distance is selected as 250m through control experiments.

[0053] The specific settings of random scatterers and lighting probability in the selected area in step S3 are:

[0054] S301, based on the outer surface of the building, the top surface of the building and the ground without buildings in the selected area, set a uniform scatterer density to generate an original random scatterer s raw ;

[0055] S302 , according to the constructed first to fifth Fresnel zones, the lighting probability of the scatterers is set in descending order to p=[1, 0.9, 0.8, 0.7, 0.6].

[0056] S303, according to the actual environment needs, a smaller radius can be set in the selected first spherical area and the second spherical area for further segmentation, which is segmented into [0-60, 60-120, 120-180, 180-250], and according to the different areas after segmentation, a higher scatterer lighting probability is set according to the scatterer closer to the transmitting end and the receiving end, specifically p = [1, 0.8, 0.4, 0.2];

[0057] S304, according to the lighting probability of the scatterers in each area, some scatterers of the original scatterers are selected as effective scatterers s e , used for simulation calculation of propagation graph theory.

[0058] In step S3, a height threshold is set for buildings outside the selected area, and random scatterers and high scatterer lighting probability are set on the building surfaces above the threshold. The generated high-rise building scatterers s h, used for simulation calculation of propagation graph theory. In this embodiment, the simulation is for receiving power.

[0059] The channel transfer function of the hybrid model of ray tracing and propagation graph theory in step S4 is:

[0060] H(f)=H RT,0:1 (f)+H PG,2:∞ (f)

[0061] The channel transfer function between the transmitter and the receiver generated by the propagation graph theory is:

[0062] H PG,2:∞ (f) = D (f) + R (f) (B (f) + B (f) 2 +…)T(f)

[0063] =D(f)+R(f)(B(f)(IB(f)) -1 )T(f)

[0064] Wherein, D(f) is the channel transfer function of the direct path from the transmitter to the receiver, R(f) represents the channel transfer function from the discretized scatterer to the receiver, T(f) represents the channel transfer function from the transmitter to the discretized scatterer, and B(f) represents the channel transfer function between the discretized scatterers.

[0065] like Figure 4 This is a path propagation diagram of the present embodiment, in which the hollow circles represent the discretized scattering points of the scatterers in the environment, the five-pointed star points represent the transmitting end in the environment, the star-shaped points represent the receiving end in the environment, the solid straight line represents the direct propagation path in the environment, the dotted line represents the first-order propagation path of scattering in the environment, the strip line represents the first-order propagation path of diffraction in the environment, and the long and short dashed lines represent the first-order propagation path of reflection in the environment.

[0066] The statistical model of line-of-sight propagation path fading in step S5 is that when the receiving end moves in the actual environment, there are a large number of trees on both sides of the road, which may block the propagation path, showing a line-of-sight propagation path model. If tree scatterers are embedded in the simulation, the large increase in the number of scatterers increases the simulation cost, so the model of the statistical characteristics of fading is used instead. Using the received power value P of part of the measured data, the received power value P is processed using a sliding window to obtain P ′ , calculate the fading value (PP ′ ) and count the probability of occurrence, and calculate the average receiving end sample interval in each interval. In this embodiment, based on the simulated theater scene, the first 20,000 measured received power values ​​P are selected to count their power fading characteristics. The number of samples with the fading value bottomed in the range of -25dB to -5dB is counted, and the probability of the fading value occurring is calculated, such as Figure 5is the distribution diagram of the fading value in each interval. Accordingly, the average receiving end sample interval that appears in each interval is calculated, and the distribution is as follows Figure 6 shown.

[0067] Step S5 embeds the fading statistical model into the simulation model of step S4, takes the first simulation sample as the benchmark, calculates the intervals between other samples and the first simulation sample, and when the sample meets the integer multiple of the receiving end sample interval, the sample simulates the receiving power minus the attenuation value under the corresponding interval to simulate the line-of-sight propagation path scenario. In this embodiment, the fading characteristics within the average receiving end sample interval are used to select 800 consecutive measurement sample values ​​other than the training sample in this scenario and embed them into the fading model for verification. Figure 7 and Figure 8 The figures show the comparison between the simulated and measured received power curves by the traditional graph theory simulation method and the comparison between the improved simulated and measured received power curves.

[0068] like Figure 7 As shown, the horizontal axis represents the snapshot index value of the measured received power, the vertical axis represents the received power, the dots represent the actual measured values, and the stars represent the simulation values ​​of the traditional graph theory simulation method. The results show that there is a large difference between the measured values ​​and the simulated values. The calculated root mean square error between the simulation data and the measured data is 17.2606dB.

[0069] like Figure 8 As shown in the figure, the horizontal axis represents the snapshot index value of the measured received power, the vertical axis represents the received power, the dots represent the actual measured values, and the stars represent the improved model simulation values. The results show that the measured values ​​and the simulated values ​​have a high degree of overlap, and the trend of change and the power level are roughly the same. The calculated root mean square error between the simulated data and the measured data is 5.4196dB.

[0070] The preferred specific embodiments of the present invention are described in detail above. It should be understood that a person skilled in the art can make many modifications and changes based on the concept of the present invention without creative work. Therefore, any technical solution that can be obtained by a person skilled in the art through logical analysis, reasoning or limited experiments based on the concept of the present invention on the basis of the prior art should be within the scope of protection determined by the claims.

Claims

1. A radio wave propagation channel modeling method for predicting received signal power, characterized in that: The following steps are involved: S1. Input the digital map of the target measurement area; S2, according to the transmitter T x With the receiving end R x Position, determine the ellipsoidal Fresnel zone, and then determine the first spherical area and the second spherical area with the transmitting end and the receiving end as the center and the modeled propagation distance as the radius; S3, setting the scatterer position and the probability that the scatterer can affect the propagation of radio waves according to the ellipsoidal Fresnel zone, the first spherical zone and the second spherical zone determined in step S2, and updating the scatterer distribution according to the movement of the transceiver; S4, according to the scatterer updated in step S3, using a hybrid simulation model of ray tracing and propagation graph theory to simulate radio wave propagation; S5, measuring the power level measurement data under the same urban environment, training a statistical model of line-of-sight propagation path fading under set obstruction based on the measured data, optimizing and adjusting the simulation model parameters and area size of step S4, and obtaining an optimized prediction model; The statistical model of line-of-sight propagation path fading under set obstruction uses the received power value P of the measured data, processes the received power value P with a sliding window, obtains P′, calculates the fading value (PP′), and statistically calculates the probability of occurrence, and calculates the average receiving end sample interval that occurs in each interval; S6. According to the optimized model, simulation output predicts the receiving level and channel characteristic parameters; The construction of the Fresnel zone in step S2 is based on the transmitting end T x With the receiving end R x is the focus, straight line path |T x R x | and the polyline path with inflection point P |PT x |+|PR x |When the travel difference is nλ / 2, the first, second, ..., nth Fresnel zones are constructed by the first, second, ..., nth ellipsoidal surfaces formed by the inflection point P; The specific settings of the random scatterers in the selected area and the probability that the scatterers can affect the propagation of radio waves in step S3 are: S301, based on the outer surface of the building, the top surface of the building and the ground without buildings in the selected area, set a uniform scatterer density to generate an original random scatterer s raw ; S302, setting the probability that the scatterers can affect the propagation of radio waves, that is, the lighting probability, in descending order according to the constructed first to nth Fresnel zones; S303, according to actual environmental requirements, further dividing the first spherical area and the second spherical area, and according to the different divided areas, the closer the scatterer is to the transmitting end and the receiving end, the higher the probability of the scatterer being lit; S304, according to the lighting probability of the scatterers in each area, some scatterers of the original scatterers are selected as effective scatterers s e , used for simulation calculation of propagation graph theory; The step S3 specifically sets a height threshold for buildings within a set range outside the selected area, sets random scatterers and a high scatterer lighting probability on the building surface above the threshold, and generates a high-rise building scatterer s h , used for simulation calculation of propagation graph theory.

2. A radio wave propagation channel modeling method for predicting received signal power according to claim 1, characterized in that: The radii of the first spherical area and the second spherical area in step S2 are the modeled propagation distances, and the modeled propagation distance in the urban scene is selected to be in the range of 200m to 300m.

3. A radio wave propagation channel modeling method for predicting received signal power according to claim 1, characterized in that: In step S3, as the position of the transceiver moves, the two spherical areas and the Fresnel area centered on the transceiver are updated, and the scatterers s and s are updated respectively. raw 、s e and h , used for simulation calculations.

4. A radio wave propagation channel modeling method for predicting received signal power according to claim 1, characterized in that: The channel transfer function of the hybrid model of ray tracing and propagation graph theory in step S4 is: H(f)=H RT，0：1 (f)+H PG，2：∞ (f) Where H(f) is the channel transfer function of the hybrid model of ray tracing and propagation graph theory, H PG，2：∞ (f) is the channel transfer function between the transmitter and the receiver generated by the propagation graph theory, H RT，0：1 (f) is the channel function generated by the 0th to 1st order path of ray tracing.

5. A radio wave propagation channel modeling method for predicting received signal power according to claim 4, characterized in that: The channel transfer function between the transmitter and the receiver generated by the propagation graph theory is: H PG，2：∞ (f)=D(f)+R(f)[B(f)+B(f) 2 +…]T(f) =D(f)+R(f){B(f)[I-B(f)] -1 }T(f) Wherein, D(f) is the channel transfer function of the direct path from the transmitter to the receiver, R(f) represents the channel transfer function from the discretized scatterer to the receiver, T(f) represents the channel transfer function from the transmitter to the discretized scatterer, and B(f) represents the channel transfer function between the discretized scatterers.

6. A radio wave propagation channel modeling method for predicting received signal power according to claim 1, characterized in that: After the training, a statistical model of line-of-sight propagation path fading under set obstruction is obtained and embedded in the hybrid simulation model of step S4. The first simulation sample is used as a benchmark to calculate the intervals between other samples and the first simulation sample. When an integer multiple of the receiving end sample interval is met, the sample simulates the received power P minus the fading value (PP′) under the corresponding interval to simulate the line-of-sight propagation path scenario.

Images