Modeling method and electronic terminal of hybrid channel in subway tunnel

Abstract

The present invention provides a subway tunnel hybrid channel modeling method and an electronic terminal, the method comprising: establishing a three-dimensional model of a subway tunnel; configuring parameters of the three-dimensional model of the subway tunnel; based on a ray tracing method, a transmitter radiates a signal to a receiver; in the process of radiating a signal from the transmitter to the receiver, a radial basis function neural network is constructed, and the radial basis function neural network structure is configured; test data is selected to train the radial basis function neural network, and the difference obtained by subtracting the simulated path loss value from the test path loss is used as the input of the RBF neural network, and the output is the predicted value of the path loss at the receiving point, so as to form the required radial basis function neural network model of the subway tunnel hybrid channel. The present invention improves the prediction accuracy of a single ray tracing simulation; compared with the traditional ray tracing method, the present invention effectively reduces the number of adjustments of simulation parameters, and is more efficient when used for wireless channel characteristic prediction in a variety of complex scenarios.

Description

Technical Field

[0001] The invention belongs to the technical field of software simulation modeling, and in particular relates to a subway tunnel hybrid channel modeling method and an electronic terminal. Background Art

[0002] The existing wireless channel simulation modeling has poor prediction accuracy for wave propagation characteristics in tunnels. Traditional ray tracing simulation modeling lacks scientific methods to adjust simulation parameters such as roughness, number of reflections, and ray spacing. It takes multiple simulation experiments to obtain prediction results that are closer to the actual situation. In addition, once the propagation scenario changes, the original simulation parameters also need to be recalibrated. Summary of the invention

[0003] In view of the above-mentioned shortcomings of the prior art, an object of the present invention is to provide a subway tunnel hybrid channel modeling method and an electronic terminal, which are used to solve the technical problem of poor prediction accuracy of channel simulation modeling in the prior art.

[0004] To achieve the above-mentioned purpose and other related purposes, an embodiment of the present invention provides a subway tunnel hybrid channel modeling method, comprising: establishing a three-dimensional model of a subway tunnel; configuring parameters of the three-dimensional model of the subway tunnel; based on a ray tracing method, a transmitter radiates a signal to a receiver; in the process of radiating the signal from the transmitter to the receiver, a radial basis function neural network is constructed, and the radial basis function neural network structure is configured; test data is selected to train the radial basis function neural network, and the difference obtained by subtracting the simulated path loss value from the test path loss is used as the input of the RBF neural network, and the output is the predicted value of the path loss at the receiving point, so as to form the required radial basis function neural network model of the subway tunnel hybrid channel.

[0005] In one embodiment of the present application, the parameters of the configured three-dimensional model of the subway tunnel include: roughness of the tunnel wall, ray interval, number of ray reflections, and number of ray diffractions.

[0006] In one embodiment of the present application, based on the ray tracing method, the transmitter radiates signals to the receiver, including: the transmitter radiates signals in all directions, and the radiated signals reach the receiver after being transmitted through the spatial channel; the receiver is a plurality of concentric spheres with different radii, and the radii thereof correspond to and match the path length of the rays and the angle between adjacent rays.

[0007] In one embodiment of the present application, the radial basis function neural network structure is configured as follows: the input layer is configured as a vector x, the dimension is m, the number of samples is n, the hidden layer is fully connected to the input layer, and there is no connection within the layer.

[0008] In one embodiment of the present application, the maximum number of neurons in the hidden layer is equal to the number of samples, and the number of neurons in the hidden layer is set to 30-40.

[0009] In one embodiment of the present application, the transmitting and receiving distance, the roughness of the tunnel wall, the ray interval, the number of ray reflections, and the number of ray diffractions are used as the detailed information of the input layer vector x.

[0010] In one embodiment of the present application, the transfer function of the radial basis function neural network is a radial basis function; and the distribution density of the radial basis function is configured to be 1.5.

[0011] In one embodiment of the present application, the basis function of the radial basis function neural network is configured as a Gaussian function.

[0012] In one embodiment of the present application, the remaining test data is also used to verify the radial basis function neural network model of the subway tunnel hybrid channel.

[0013] An embodiment of the present invention further provides an electronic terminal, characterized in that it includes a processor and a memory, wherein the memory stores program instructions; and the processor executes the program instructions to implement the subway tunnel hybrid channel modeling method as described above.

[0014] As described above, the subway tunnel hybrid channel modeling method and electronic terminal of the present invention have the following beneficial effects:

[0015] 1. The present invention constructs a radial basis function (RBF) neural network model of a subway tunnel hybrid channel through a radial basis function (RBF) neural network algorithm, thereby improving the prediction accuracy of a single ray tracing simulation.

[0016] 2. Compared with the traditional ray tracing method, the present invention effectively reduces the number of times the simulation parameters are adjusted and is more efficient when used for predicting wireless channel characteristics in a variety of complex scenarios. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 Shown is a schematic flow chart of the subway tunnel hybrid channel modeling method of the present invention.

[0018] Figure 2 Shown is a schematic diagram of the radial basis function neural network structure in the subway tunnel hybrid channel modeling method of the present invention.

[0019] Figure 3 Shown is a schematic diagram of the comparison of path loss prediction results in the subway tunnel hybrid channel modeling method of the present invention.

[0020] Figure 4 Shown is a principle block diagram of an electronic terminal of the present invention. DETAILED DESCRIPTION

[0021] The following describes the embodiments of the present invention by specific examples, and those skilled in the art can easily understand other advantages and effects of the present invention from the contents disclosed in this specification. The present invention can also be implemented or applied through other different specific embodiments, and the details in this specification can also be modified or changed in various ways based on different viewpoints and applications without departing from the spirit of the present invention. It should be noted that the following embodiments and features in the embodiments can be combined with each other without conflict.

[0022] The purpose of this embodiment is to provide a subway tunnel hybrid channel modeling method and an electronic terminal, which are used to solve the technical problem that the prior art fails to realize monitoring of the timing accuracy of multi-level clocks.

[0023] The subway tunnel hybrid channel modeling method and electronic terminal of this embodiment designs a three-dimensional tunnel model and a train model for ray tracing simulation, and trains the influence of simulation parameters on the difference between measured data and simulation data through the radial basis function (RBF) neural network algorithm, thereby establishing a hybrid channel model based on ray tracing and neural network. Compared with the traditional ray tracing method, the hybrid model effectively reduces the number of simulation parameter adjustments and improves the accuracy of prediction, and can be used for wireless channel characteristic prediction in various scenarios.

[0024] The principles and implementation methods of the subway tunnel hybrid channel modeling method and electronic terminal of this embodiment will be explained in detail below, so that those skilled in the art can understand the subway tunnel hybrid channel modeling method and electronic terminal of the present invention without creative work.

[0025] Example 1

[0026] like Figure 1 As shown, this embodiment provides a subway tunnel hybrid channel modeling method, including:

[0027] Step S100: establishing a three-dimensional model of a subway tunnel;

[0028] Step S200: configuring parameters of the three-dimensional model of the subway tunnel;

[0029] Step S300: Based on the ray tracing method, the transmitter radiates a signal to the receiver;

[0030] Step S400: constructing a radial basis function neural network and configuring the radial basis function neural network structure during the process of the transmitter radiating a signal to the receiver;

[0031] Step S500: Select test data to train the radial basis function neural network, use the difference between the test path loss and the simulated path loss value as the input of the RBF neural network, and output the predicted value of the path loss at the receiving point to form the required radial basis function neural network model of the subway tunnel hybrid channel.

[0032] The following combination Figure 2 and Figure 3 Steps S100 to S500 of the subway tunnel hybrid channel modeling method of this embodiment are described in detail.

[0033] Step S100: Establishing a three-dimensional model of a subway tunnel.

[0034] Step S200: configuring parameters of the three-dimensional model of the subway tunnel.

[0035] In this embodiment, the parameters of the configured three-dimensional model of the subway tunnel include, but are not limited to: roughness of the tunnel wall, ray interval, number of ray reflections, and number of ray diffractions.

[0036] Step S300: Based on the ray tracing method, the transmitter radiates a signal to the receiver.

[0037] When using the ray tracing method, according to the incident and bounce ray algorithm, the signal emission point is regarded as a point source, and the electromagnetic waves emitted by it are regarded as geometric rays. Each emission line is traced, and before it is completely lost, it is processed by transmission, reflection, and diffraction according to the principles of geometric optics when it encounters a plane. The signal receiving point is also regarded as a geometric point, and all signal rays passing through this point are regarded as being received. The electromagnetic field strength is vector-added to predict the propagation characteristics.

[0038] Specifically, in this embodiment, based on the ray tracing method, the transmitter radiates signals to the receiver, including: the transmitter radiates signals in all directions, and the radiated signals reach the receiver after being transmitted through the spatial channel; the receiver is a plurality of concentric spheres with different radii, and the radii thereof correspond to and match the path length of the rays and the angle between adjacent rays.

[0039] Step S400: In the process of the transmitter radiating a signal to the receiver, a radial basis function neural network is constructed and the radial basis function neural network structure is configured.

[0040] Step S500: Select test data to train the radial basis function neural network, use the difference between the test path loss and the simulated path loss value as the input of the RBF neural network, and output the predicted value of the path loss at the receiving point to form the required radial basis function neural network model of the subway tunnel hybrid channel.

[0041] In this embodiment, the radial basis function neural network structure is configured as follows: the input layer is configured as a vector x, the dimension is m, the number of samples is n, the hidden layer is fully connected to the input layer, and there is no connection within the layer.

[0042] In this embodiment, the maximum number of neurons in the hidden layer is equal to the number of samples; the number of neurons in the hidden layer is set to 30-40, for example, the number of neurons in the hidden layer is set to 40.

[0043] In this embodiment, the transmitting and receiving distance, the roughness of the tunnel wall, the ray interval, the number of ray reflections, and the number of ray diffractions are used as the detailed information of the input layer vector x.

[0044] In this embodiment, the transfer function of the radial basis function neural network is a radial basis function; the distribution density of the radial basis function is configured to be 1.5; and the basis function of the radial basis function neural network is configured to be a Gaussian function.

[0045] That is, in this embodiment, when using the radial basis function (RBF) neural network for channel modeling, the parameters of the network model must be set first, including the input layer vector, dimension, number of samples, the hidden layer is fully connected to the input layer, there is no connection within the layer, the maximum number of neurons in the hidden layer is equal to the number of samples, the transfer function is the RBF function, and the basis function is set to a Gaussian function.

[0046] In this embodiment, the remaining test data is also used to verify the radial basis function neural network model of the subway tunnel hybrid channel.

[0047] The method of combining ray tracing and neural network is as follows: the implementation data uses field test data. On the basis of ray tracing, the difference between the measured path loss and the simulated path loss value is used as the input of the RBF neural network, and the output is the predicted value of the path loss at the receiving point. A part of the samples is used to train the network. Through training, the RBF neural network learns the influence of subtle environmental information and simulation parameters on the simulation results into the network, and the remaining samples are used to verify the model.

[0048] This embodiment takes the 28 GHz hybrid channel modeling based on ray tracing and neural network algorithm of a typical subway tunnel as an example to illustrate the subway tunnel hybrid channel modeling method of this embodiment.

[0049] 1) Establish a 3D model of the subway tunnel.

[0050] 2) Set the tunnel roughness to 0.01m, 0.005m, 0.001m, 0.0005m, 0.0003m and 0.0001m respectively. Set the ray interval to 0.05°, 0.1°, 0.2° and 0.5° respectively, the number of reflections to 6, 12, 18 and 24 respectively, and the number of diffractions to 0, 1, 3 and 5 respectively.

[0051] 3) In the operation of the transmission bouncing ray method, the transmitter radiates signals in all directions, and the signals reach the receiver after being transmitted through the spatial channel. The receiver is a number of concentric spheres with different radii, and its radius depends on the path length of the ray and the angle between adjacent rays.

[0052] 4) Based on ray tracing, an RBF neural network is constructed. The input layer of the RBF neural network is vector x, the dimension is m, the number of samples is n, the hidden layer is fully connected to the input layer, there is no connection within the layer, the maximum number of neurons in the hidden layer is equal to the number of samples, and the number of neurons in the hidden layer is generally set to 40, such as Figure 2 shown.

[0053] 5) Set the parameters of network modeling, using the sending and receiving distance, the roughness of the tunnel wall, the ray interval, the number of ray reflections, and the number of ray diffraction as the detailed information of the input layer vector x.

[0054] 6) The neural network transfer function is RBF. The basis function is set to Gaussian function. To avoid overfitting, the RBF function distribution density is set to 1.5.

[0055] 7) Use the test data of the 28GHz frequency band to train the network, and use the difference between the test path loss and the simulated path loss value as the input of the RBF neural network, and the output is the predicted value of the path loss at the receiving point. Through training, the RBF neural network learns the influence of subtle environmental information and simulation parameters on the simulation results into the network.

[0056] 8) Finally, the remaining test data is used to verify the modeling.

[0057] 9) The path loss prediction experiment of the 28 GHz frequency band was carried out using the hybrid channel modeling constructed above, and compared with the ray tracing method and measured data, such as Figure 3 It can be found that the path loss value predicted by hybrid channel modeling based on ray tracing and RBF neural network algorithm is closer to the measured value.

[0058] Example 2

[0059] like Figure 4 As shown, this embodiment further provides an electronic terminal 10 , which includes a processor 101 and a memory 102 .

[0060] The memory 102 is connected to the processor 101 through a system bus and communicates with each other. The memory 102 is used to store computer programs. The processor 101 is coupled to the memory 1002. The processor 101 is used to run the computer program so that the electronic terminal 10 executes the subway tunnel hybrid channel modeling method described in Example 1. Example 1 has already described the subway tunnel hybrid channel modeling method in detail, which will not be repeated here.

[0061] The subway tunnel hybrid channel modeling method can be applied to various types of electronic terminals 10 .

[0062] In actual implementation, the electronic terminal 10100 is, for example, an electronic terminal 10100 installed with an Android operating system or an iOS operating system, or an operating system such as Palm OS, Symbian, or Black Berry OS, Windows Phone, etc.

[0063] In an exemplary embodiment, the electronic terminal 10 can be implemented by one or more application-specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), field programmable gate arrays (FPGAs), controllers, microcontrollers, microprocessors, cameras or other electronic components to execute the above-mentioned subway tunnel hybrid channel modeling method.

[0064] It should be noted that the system bus mentioned above can be a Peripheral Component Interconnect (PCI) bus or an Extended Industry Standard Architecture (EISA) bus. The system bus can be divided into an address bus, a data bus, a control bus, etc. For ease of representation, Figure 4 The term "communication interface" is represented by only one thick line, but it does not mean that there is only one bus or one type of bus. The communication interface is used to implement communication between the database access device and other devices (such as clients, read-write libraries, and read-only libraries). The memory may include random access memory (RAM) and may also include non-volatile memory, such as at least one disk storage.

[0065] The above-mentioned processor 101 can be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; it can also be a digital signal processor (DSP), an application specific integrated circuit (ASIC), a field programmable gate array (FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components.

[0066] Those skilled in the art can understand that all or part of the steps of implementing the above-mentioned method embodiments can be completed by hardware related to the computer program. The aforementioned computer program can be stored in a computer-readable storage medium. When the program is executed, the steps of the above-mentioned method embodiments are executed; and the aforementioned storage medium includes: ROM, RAM, magnetic disk or optical disk and other media that can store program codes.

[0067] Example 3

[0068] This embodiment provides a computer-readable storage medium on which a computer program is stored, and when the computer program is executed by a processor, the subway tunnel hybrid channel modeling method is implemented. The subway tunnel hybrid channel modeling method has been described in detail above, and will not be repeated here.

[0069] Those skilled in the art can understand that all or part of the steps of implementing the above-mentioned method embodiments can be completed by hardware related to the computer program. The aforementioned computer program can be stored in a computer-readable storage medium. When the program is executed, the steps of the above-mentioned method embodiments are executed; and the aforementioned storage medium includes: ROM, RAM, magnetic disk or optical disk and other media that can store program codes.

[0070] In summary, the present invention realizes monitoring of the timing accuracy of multi-level clocks: the accuracy between the output time of the monitoring system host and the output time of the secondary time server (or ETS server), the accuracy between the output time of the monitoring system station host and the output time of its system host, and the accuracy between the output time of the monitoring station slave and the output time of the station host. Therefore, the present invention effectively overcomes the shortcomings of the prior art and has a high industrial utilization value.

[0071] The above embodiments are merely illustrative of the principles and effects of the present invention, and are not intended to limit the present invention. Anyone familiar with the art may modify or alter the above embodiments without departing from the spirit and scope of the present invention. Therefore, all equivalent modifications or alterations made by a person of ordinary skill in the art without departing from the spirit and technical concept disclosed by the present invention shall still be covered by the claims of the present invention.

Claims

1. A subway tunnel hybrid channel modeling method, characterized by: include: Build a 3D model of a subway tunnel; Configure the parameters of the 3D model of the subway tunnel; The parameters of the configured three-dimensional model of the subway tunnel include: roughness of the tunnel wall, ray interval, number of ray reflections, and number of ray diffraction; Based on the ray tracing method, the transmitter radiates the signal to the receiver; In the process of the transmitter radiating a signal to the receiver, a radial basis function neural network is constructed, and the radial basis function neural network structure is configured; Test data is selected to train the radial basis function neural network, and the difference between the test path loss and the simulated path loss value is used as the input of the RBF neural network, and the output is the predicted value of the path loss at the receiving point, so as to form the required radial basis function neural network model of the subway tunnel hybrid channel.

2. The subway tunnel hybrid channel modeling method according to claim 1, characterized in that: Based on the ray tracing method, the transmitter radiates signals to the receiver, including: the transmitter radiates signals in all directions, and the radiated signals reach the receiver after being transmitted through the spatial channel; the receiver is a plurality of concentric spheres with different radii, and the radii thereof correspond to and match the path length of the rays and the angle between adjacent rays.

3. The subway tunnel hybrid channel modeling method according to claim 1, characterized in that: The radial basis function neural network structure is configured as follows: the input layer is configured as a vector x, the dimension is m, the number of samples is n, the hidden layer is fully connected to the input layer, and there is no connection within the layer.

4. The subway tunnel hybrid channel modeling method according to claim 3 is characterized by: The maximum number of neurons in the hidden layer is equal to the number of samples, and the number of neurons in the hidden layer is set to 30-40.

5. The subway tunnel hybrid channel modeling method according to claim 3 is characterized by: The detailed information of the input layer vector x is taken as the sending and receiving distance, the roughness of the tunnel wall, the ray interval, the number of ray reflections, and the number of ray diffraction.

6. The subway tunnel hybrid channel modeling method according to claim 1, characterized in that: The transfer function of the radial basis function neural network is a radial basis function; the distribution density of the radial basis function is configured to be 1.

5.

7. The subway tunnel hybrid channel modeling method according to claim 6 is characterized by: The basis function of the radial basis function neural network is configured as a Gaussian function.

8. The subway tunnel hybrid channel modeling method according to claim 1 is characterized by: The method also includes using the remaining test data to verify the radial basis function neural network model of the subway tunnel hybrid channel.

9. An electronic terminal, characterized in that: It includes a processor and a memory, wherein the memory stores program instructions; the processor runs the program instructions to implement the subway tunnel mixed channel modeling method as claimed in any one of claims 1 to 8.

Images