Wireless communication link high-fidelity simulation method for high-dynamic Internet of Vehicles

Abstract

The invention provides a wireless communication link high-fidelity simulation method for a high-dynamic Internet of Vehicles, which comprises the following steps of: performing channel modeling by using a channel simulation platform according to an input network layout, a terminal track, a propagation scene and an antenna mode to generate a time-varying channel coefficient; accurate modeling is carried out on key physical phenomena such as multipath propagation, Doppler frequency shift, path gain and propagation time delay in the Internet of Vehicles, and high-fidelity simulation of channel dynamic change characteristics in a high-mobility scene is realized; a transmitter class object, a receiver class object and a channel class object are constructed based on an object-oriented thought, a plurality of processing modules such as QAM modulation, multi-carrier modulation, pilot frequency design, channel estimation and equalization, frequency offset correction and the like are supported, and good transportability and expansibility are achieved. According to the invention, high-fidelity Internet of Vehicles communication system simulation is realized, a communication transceiver signal processing flow is included, and an efficient fidelity verification platform is provided for design and test of an Internet of Vehicles communication protocol and algorithm.

Description

Technical Field

[0001] This invention belongs to the field of vehicle-to-everything (V2X) communication technology. It combines simulation platform construction and link-level simulation method design, specifically involving a simulation method based on ray tracing channel modeling and link-level wireless communication transmission technology, which is applicable to high-dynamic V2X systems in sixth-generation mobile communication systems. Background Technology

[0002] The Internet of Vehicles (IoV), relying on the rapid development of traditional wireless communication network technology, has spurred many emerging intelligent transportation applications while ensuring vehicle safety, expanding application scenarios, bringing many new user experiences, and greatly enhancing road safety. Therefore, the IoV is playing an increasingly important role. Unlike general communication needs in traditional life scenarios, the IoV has its own unique characteristics, with intelligent transportation applications often having higher requirements for communication quality. Firstly, the IoV has the paramount requirement of safety. The rapid development of intelligent driving has greatly improved the IoV's communication support for many application scenarios, such as high-bandwidth high-definition map downloads and media streaming services, high-speed real-time calls and media interaction services, and intelligent driving decision-making services requiring extremely high transmission reliability. Secondly, the IoV inherently has requirements for high mobility and high dynamism. High-speed driving scenarios and complex and ever-changing driving trajectories pose challenges to IoV communication, including strong Doppler shift and frequent cell switching. Therefore, the IoV has stringent communication quality requirements and a need for extensive and comprehensive simulation verification.

[0003] Due to the stringent communication requirements for realizing intelligent functions in the Internet of Vehicles (IoV), numerous high-fidelity simulation methods and strategies are needed. Traditional IoV verification methods are field verification models with high requirements for the site, requiring the construction of real-world verification sites. These verification methods are very costly and limited by uncertainties such as the environment and location, resulting in long lead times and high reproducibility. In contrast, software simulation platforms offer many advantages over field verification, including lower resource and time costs, the ability to comprehensively consider uncertain environmental factors, and reproducibility that ensures the accuracy and consistency of test results.

[0004] Existing vehicle-to-everything (V2X) simulators can be mainly divided into three categories: simulators focusing on traffic flow, simulators focusing on V2X channel modeling, and simulators focusing on physical layer signal processing. Simulators focusing on traffic flow primarily aim to simulate the dynamics of traffic flow, with only minimal consideration for communication. While they can reproduce complex traffic flows with high quality, their communication simulation capabilities are very limited. Simulators focusing on V2X channel modeling, such as QuaDRiGa using ray tracing, are relatively complete in channel modeling and can meticulously characterize V2X communication. However, their function focuses on generating channel parameters and does not include a complete physical layer communication transceiver signal processing simulation process from bitstream to bitstream. Therefore, they cannot effectively combine a high-fidelity channel model with an equally high-fidelity communication link-level simulator, making it difficult to accurately evaluate the performance of different physical layer technologies under real-world channel conditions. Simulators focusing on physical layer signal processing, such as Sionna developed by NVIDIA, an open-source Python library built on TensorFlow, focus on link-level simulation of digital communication systems. It can be seen that existing vehicle network simulators are developing independently, with powerful functions but clear boundaries: traffic flow simulators cannot reflect the physical characteristics of the channel, and therefore cannot reflect the real Doppler shift and frequency-selective fading in the high-dynamic vehicle network environment; although channel modeling simulators can generate these channel effects, they lack an architecture that can accurately map their complex time-varying channel coefficients to the micro-processing links, making it impossible to integrate with specific transceiver signal processing flows, thus making it difficult to assess their final impact on bit-level error performance; while physical layer simulators generally lack mechanisms for accurately modeling and importing these non-stationary channel characteristics driven by high-dynamic scenarios, and their built-in general algorithms, such as channel estimation and equalization, often have insufficient performance or even fail when facing high-dynamic vehicle network channel conditions; at the same time, the design of existing simulation tools is mostly fixed on mature communication technologies, and their analysis framework is usually limited to the traditional time-frequency (TF) domain, making it difficult to support new modulation technologies such as Orthogonal Time-Frequency Space (OTFS) that operate in the delay-Doppler (DD) domain. These limitations in the scenario adaptability and multi-domain analysis capabilities of key algorithms collectively constitute the current difficulties in conducting simulation verification and high-fidelity experiments in highly dynamic vehicle-to-everything (V2X) environments. Summary of the Invention

[0005] Purpose of the invention: The purpose of this invention is to provide a high-fidelity simulation method for wireless communication links in highly dynamic vehicle-to-everything (V2X) networks. Addressing the problem of missing dynamic characteristic modeling in V2X simulations, this method enables accurate channel modeling of highly mobile V2X links. Furthermore, through an object-oriented V2X link-level simulation program, it achieves high-performance communication transceiver signal processing, providing tool support for the design and optimization of future high-precision, high-reliability V2X communication systems.

[0006] Technical solution: To achieve the above-mentioned objectives, the present invention adopts the following technical solution:

[0007] A high-fidelity simulation method for wireless communication links in highly dynamic vehicle-to-everything (V2X) networks includes the following steps:

[0008] Based on the input network layout, terminal trajectory, propagation scenario and antenna mode, a channel simulation platform is used to perform channel modeling and generate time-varying channel coefficients. The time-varying channel coefficients are used to characterize the multipath fading, Doppler frequency shift and delay spread characteristics in a high-dynamic vehicle network communication environment.

[0009] Construct independent and instantiable transmitter, receiver, and channel objects. The transmitter object receives input parameters, including coding type, modulation type, and input bitstream, and generates a transmit signal containing time-domain modulation and pilot signals. The channel object receives the transmit signal and the channel coefficients, applies the channel coefficients to simulate multipath and Doppler effects, and superimposes noise to generate a receive signal. The receiver object receives the receive signal, modulation type, pilot information, and noise variance, performs channel estimation, equalization, and demodulation processing, and generates a decoded bitstream.

[0010] Performance metrics are calculated based on the input bitstream and the decoded bitstream.

[0011] Furthermore, the network layout includes the initial three-dimensional coordinates of the base station and mobile terminal, the antenna array configuration, and the antenna orientation; the terminal trajectory includes a series of three-dimensional coordinate points arranged in chronological order to drive the time-varying characteristics of the channel; the propagation scenario includes line-of-sight propagation or non-line-of-sight propagation in an urban macrocell environment; and the antenna pattern is used to define the three-dimensional far-field radiation pattern matrix of the transmitting and receiving antennas.

[0012] Furthermore, based on the QuaDRiGa platform, time-varying channel coefficients are generated according to the input parameters. The generation process includes:

[0013] Based on the selected propagation scenario, a set of spatially interrelated large-scale fading parameters is generated, including target delay spread and target angle spread;

[0014] Based on the target delay spread and target angle spread, as well as the transceiver positions in the network layout, the initial propagation delay and initial angle are calculated for each multipath component.

[0015] An initial path power is assigned to each multipath component, and the initial path power is adjusted by a frequency-dependent scaling factor based on the multiband effect;

[0016] The path power is adjusted by applying the Rice K-factor, and the delay spread and angle spread are calibrated to make the overall statistical characteristics match the target delay spread and target angle spread.

[0017] Based on the terminal trajectory, the delay, angle, and phase drift of multipath components on a short time slice are simulated to generate Doppler frequency shift;

[0018] The polarization channel coefficients are calculated, taking into account the physical characteristics of the transmitting and receiving antennas themselves as well as the changes in the polarization state of the signal during propagation.

[0019] The distance-dependent path gain, shadow fading, and Rice's K-factor, which is dynamically varying along the trajectory, are applied to the already generated channel coefficients;

[0020] The terminal trajectory is divided into multiple segments, each segment generates channel coefficients independently, and a smooth transition is performed between adjacent segments to generate a continuous channel sequence.

[0021] Furthermore, the channel class object is initialized by loading the channel coefficient tensor generated by QuaDRiGa, and the channel coefficients are updated at the frame level.

[0022] Furthermore, the step of generating the received signal by simulating multipath effects and Doppler effects and superimposing noise on the channel object includes:

[0023] Create an empty receive signal for subsequent accumulation of the effects of each path;

[0024] For each signal path defined in the parameters, the phase rotation factor is calculated based on the Doppler frequency shift and time index of the path. The original transmitted signal is first multiplied by the phase rotation factor, and then a cyclic shift operation is performed. The number of shift points is determined by the time delay parameter of the path. The generated time delay signal is multiplied by the gain coefficient of the path and then accumulated into the total received signal.

[0025] Gaussian white noise is added to the received signal that has been superimposed with all multipath components, according to the specified signal-to-noise ratio;

[0026] Output the simulated received signal.

[0027] Furthermore, the step of generating a transmit signal containing a time-domain modulated signal and a pilot signal by the transmitter object includes:

[0028] Initialize configuration parameters, including word length, code length, and frame length;

[0029] If the encoding function is enabled, the input bitstream is encoded, and the encoding type includes BCH (BoseChaudhuri Hocquenghem) encoding, Turbo encoding, or no encoding.

[0030] Depending on the modulation type, the data enters different modulation paths: If the modulation type is Orthogonal Frequency Division Multiplexing (OFDM), the encoded data is mapped onto the QAM constellation diagram, pilot signals are inserted in the time-frequency domain, an inverse fast Fourier transform is performed, and a parallel-to-serial conversion is performed to convert the frequency domain signal into a time domain signal, and a cyclic prefix is ​​added; if the modulation type is OTFS, the encoded data is mapped onto the QAM constellation diagram, pilot signals are inserted in the time-delay-Doppler domain, a two-dimensional inverse symplectic finite Fourier transform is performed to convert the time-delay-Doppler domain signal into a time-frequency domain symbol, and then a Heisenberg transform is used to obtain the time domain signal;

[0031] Output time-domain signal and pilot information.

[0032] Furthermore, the steps of the receiver object performing channel estimation, equalization, and demodulation processing to generate the decoded bitstream include:

[0033] Depending on the modulation type, the signal enters different processing paths: If the modulation type is OFDM, after removing the cyclic prefix, a Fast Fourier Transform is performed to convert the time-domain signal to the frequency domain. Channel estimation is performed using known pilot information, and channel equalization is performed based on the noise variance and the channel estimation results. If the modulation type is OTFS, a Symplectic Finite Fourier Transform is performed on the received signal to convert the time-domain received signal to the time-delay-Doppler domain. The multipath parameters of the wireless channel are estimated based on known pilot information. Maximum ratio combining is used to weight and combine the path signals, and linear equalization is performed in combination with the noise variance.

[0034] Perform QAM demapping on the equalized signal;

[0035] If channel coding is enabled, the demodulated information is decoded according to the corresponding coding type.

[0036] Output the decoded bitstream as an estimate of the original transmitted information.

[0037] Furthermore, pilot patterns are designed for both OFDM and OTFS modulation. For OFDM, pilots are inserted in the time-frequency domain, while for OTFS, pilots are inserted in the time-delay Doppler domain. For OFDM modulation, three pilot patterns are implemented: block pilot, comb pilot, and dot pilot. The block pilot inserts pilot signals in the time dimension, the comb pilot inserts pilot signals in the frequency dimension, and the dot pilot inserts pilot signals in a sparse dot matrix in the two-dimensional time-frequency plane. For OTFS modulation, there is only one pilot signal in the center of the time-delay Doppler domain, and a zero-fill protection zone is set around the pilot signal.

[0038] A computer system includes a memory, a processor, and a computer program / instructions stored in the memory and executable on the processor. When the computer program / instructions are executed by the processor, they implement the steps of the high-fidelity simulation method for wireless communication links in a highly dynamic vehicle network.

[0039] A computer program product includes a computer program / instructions that, when executed by a processor, implement the steps of a high-fidelity simulation method for wireless communication links in highly dynamic vehicle networks.

[0040] Beneficial Effects: Compared with existing technologies, this invention has the following advantages: This invention employs a high-fidelity simulation method for wireless communication links in highly dynamic vehicular networks, enabling accurate modeling of channel characteristics and improving simulation fidelity in such environments. Simultaneously, the object-oriented programming approach significantly improves system maintainability and scalability. This invention supports multiple coding, modulation, and precoding methods, offering greater simulation flexibility. This invention supports personalized parameter configuration to adapt to different simulation needs, enabling analysis of performance indicators such as bit error rate and channel characteristics, facilitating system evaluation and optimization. This invention achieves high-fidelity simulation of wireless communication links in highly dynamic vehicular networks, providing technical support for research on vehicular network communication technology. Attached Figure Description

[0041] Figure 1 This is a schematic diagram of the channel modeling steps based on QuaDRiGa in an embodiment of the present invention.

[0042] Figure 2 This is a schematic diagram of the NLOS single-hop model in an embodiment of the present invention.

[0043] Figure 3 This is a transceiver modulation framework diagram in an embodiment of the present invention.

[0044] Figure 4 This is a diagram showing the relationship between OFDM and OTFS domain transformations.

[0045] Figure 5This is a design diagram of OFDM pilot pattern in an embodiment of the present invention.

[0046] Figure 6 This is the OTFS pilot pattern design diagram in an embodiment of the present invention.

[0047] Figure 7 This is a graph showing the bit error rate curves of OFDM and OTFS under a Gaussian channel.

[0048] Figure 8 This is a graph showing the bit error rate curves of OFDM and OTFS under frequency-selective channels.

[0049] Figure 9 This is a graph showing the bit error rate curves of OFDM and OTFS under time-selective channels.

[0050] Figure 10 This is a graph showing the bit error rate curves of OFDM and OTFS under the EVA channel. Detailed Implementation

[0051] To make the objectives, technical solutions, and advantages of this invention clearer, the embodiments of this invention are described in detail below with reference to the accompanying drawings. These embodiments are implemented based on the technical solutions of this invention, providing detailed implementation methods and specific operating procedures. It should be understood that the specific examples described herein are merely illustrative of the invention, but the scope of protection of this invention is not limited to the following embodiments.

[0052] To meet the high simulation accuracy requirements of high-dynamic communication scenarios in vehicle-to-everything (V2X) networks, this invention discloses a high-fidelity simulation method for wireless communication links in high-dynamic V2X networks. The method mainly includes: generating time-varying channel coefficients using a channel simulation platform based on the input network layout, terminal trajectory, propagation scenario, and antenna pattern. These time-varying channel coefficients characterize multipath fading, Doppler shift, and delay spread characteristics under high-dynamic V2X communication conditions. The method also involves constructing independent and instantiable transmitter, receiver, and channel objects. The transmitter object receives input parameters including coding type, modulation type, and input bitstream, and generates a transmit signal containing a time-domain modulated signal and pilot signal. The channel object receives the transmit signal and the channel coefficients, applies the channel coefficients to simulate multipath and Doppler effects, and superimposes noise to generate a received signal. The receiver object receives the received signal, modulation type, pilot information, and noise variance, performs channel estimation, equalization, and demodulation processing, and generates a decoded bitstream. Finally, the method calculates performance indicators based on the input bitstream and the decoded bitstream.

[0053] This embodiment employs the QuaDRiGa channel modeling tool based on geometrical statistical properties and designs a time-domain signal transmission method for high-fidelity vehicular network link simulation. It accurately models key physical phenomena in vehicular networks such as multipath propagation, Doppler frequency shift, path gain, and propagation delay, achieving high-fidelity simulation of the dynamic characteristics of channel changes in high-mobility scenarios. Based on object-oriented principles, this embodiment modularly encapsulates the transmitter and receiver functions, supporting multiple processing modules such as OFDM modulation, multi-carrier modulation, pilot design, channel estimation and equalization, and frequency offset correction. It possesses good portability and scalability, is suitable for simulation verification needs of various communication schemes, and can provide an efficient and high-fidelity verification platform for the design and testing of vehicular network communication protocols and algorithms.

[0054] like Figure 1 As shown, this embodiment employs a QuaDRiGa-based channel modeling method, aiming to implement a complete processing flow that accurately applies high-fidelity channel physical parameters to the time-domain transmission signal. This method ensures accurate modeling of key physical phenomena in vehicular networks, such as multipath propagation, Doppler shift, path gain, and propagation delay, thereby achieving high-fidelity simulation of the dynamic characteristics of channel changes in high-mobility scenarios. The input parameters for QuaDRiGa channel modeling are defined as follows:

[0055] Network layout: Defines the initial three-dimensional coordinates of base stations and mobile terminals, i.e., transmitter locations. and receiver position This includes the antenna array configuration (such as the number of antennas, spacing) and antenna orientation. This parameter is used to determine geometrically relevant parameters such as path loss, initial delay, and angle.

[0056] Terminal trajectory: The motion path of the mobile terminal is described by a series of three-dimensional coordinate points arranged in chronological order. As the core input driving the high dynamic time-varying characteristics of the channel, especially the Doppler effect, it is directly applied to steps E and H of channel modeling.

[0057] Propagation Scenario: Select a specific communication environment, such as the LOS and NLOS signal propagation environment under urban macrocell (UMa), and define the character parameter variables as "BERLIN_UMa_LOS" and "BERLIN_UMa_NLOS". The specific statistical model and parameter set called in step A by the pre-selection of the channel model are determined, such as path loss exponent, shadow fading standard deviation, etc.

[0058] Antenna pattern: Defines the specific three-dimensional far-field radiation pattern matrix of the transmitting and receiving antennas. It is used in step F of channel modeling to calculate the antenna gain of the signal at a specific departure angle and arrival angle, and is the key to accurately modeling multiple-input multiple-output channels.

[0059] Based on the above inputs, the overall form of the channel coefficients can be derived from the time-varying impulse response. This indicates that its core is the superposition of multiple propagation paths, represented as:

[0060] (1)

[0061] in For the index of the antenna element, For the number of multipaths, , and For the first The fading coefficient, phase, and propagation delay corresponding to each path. The generation of these channel coefficients follows... Figure 1 Steps A through H shown are explained in detail below:

[0062] A. Calculate relevant large-scale parameters: To establish a statistically consistent and spatially uniform macroscopic channel environment for the simulated link, based on the selected propagation scenario, a set of spatially correlated large-scale fading parameters is generated, and a set of target statistical values, including target delay spread, is output. and target perspective expansion .

[0063] B. Calculate initial delay and path angle: using the delay spread generated in step A. and angle expansion Calculate the initial propagation delay for each multipath component. The initial path delay is... By considering a spatially correlated uniformly distributed random variable Taking the logarithm, we find that this variable is directly related to the network layout of the transceiver:

[0064] (2)

[0065] Next, for each multipath component Generate a set of initial angles with Laplace distribution and spatial correlation, including four initial angles: azimuth offset angle. Arrival azimuth Deviation from altitude angle and the angle of arrival Meanwhile, the four initial angles in the model are frequency independent, meaning that the mobile terminal sees the same scattering clusters at different frequencies. Therefore, the same initial angle along the same path is the same at different frequencies.

[0066] C. Calculate path power: Assign an initial power to each multipath component, considering multiband effects, where the initial power of each multipath component is... It is calculated as a function of its initial time delay and angle, and by a frequency-dependent scaling factor. Adjustments will be made:

[0067] (3)

[0068] in This is the scaling factor corresponding to the initial delay. , , and These are the scaling factors corresponding to the deviation from azimuth, arrival azimuth, deviation from altitude, and arrival altitude, respectively.

[0069] D. Applying Rice's K-factor, delay spread, and angle spread: The randomly generated multipath parameters are calibrated to ensure their overall statistical characteristics precisely match the large-scale parameters defined in step A. For delay spread, the overall statistical spread of the initial multipath delay is calculated and compared with the target value defined in the macroscopic scene to obtain a unified scaling factor. This factor is then used to adjust the delay of all paths, ensuring the final delay distribution accurately conforms to the preset macroscopic channel characteristics. For angle spread, the method is logically similar: after centering the initial angle, its overall statistical spread is calculated and compared with the macroscopic target value. Finally, scaling is used to adjust the angles of all paths to ensure the final angle distribution is consistent with the macroscopic scene parameters. Specifically, the K-factor is first used... Adjust the power of the first path as follows:

[0070] (4)

[0071] The path power is then normalized for each frequency:

[0072] (5)

[0073] Next, adjust the propagation delay of all paths so that their overall delay spread equals the target value output in step A. The specific method is based on the initial delay of each path. and normalized power Calculate the initial delay spread of the current set. :

[0074] (6)

[0075] in Is all subpaths in frequency The total power at each point. Due to the formula... Scaling of midpath energy The value is frequency-dependent, therefore delay scaling is required to calculate the correct delay spread from the generated path delay and path potential.

[0076] (7)

[0077] Where F is the total number of carrier frequencies. Similarly, adjust the deviation angle and arrival angle of all paths so that their overall angular spread equals the target value output in step A. The specific method involves first normalizing the initial angle and then calculating the average angle weighted by the power of all paths.

[0078] (8)

[0079] Then, subtract this average value from each initial angle to obtain the normalized angle:

[0080] (9)

[0081] Secondly, based on the normalized angle Calculate the initial angle expansion of the current set. :

[0082] (10)

[0083] Its final angle expansion is:

[0084] (11)

[0085] E. Delay, Angle, and Phase Drift on Short Time Slices: Simulating the highly dynamic time-varying characteristics of the channel driven by the terminal trajectory, thereby generating realistic Doppler frequency shifts. First, based on the initial geometry calibrated in step D, the fixed positions of one or more virtual scatterers are determined in three-dimensional space for each multipath component. Then, as the mobile terminal moves along its trajectory, the positions of the scatterers remain unchanged, but the geometric path between the transceiver and the scatterers changes continuously. The channel parameters are updated by calculating this geometric change. Using methods such as... Figure 2 The NLOS single-hop model shown is used for calculation, where and This represents the path vectors of the receiver and transmitter to the scattering body at the initial moment; The path vector of the receiver to the scatterer at the initial moment. The path vector from transmitter to receiver at the initial moment The included angle; and Indicates the new location point The path vectors of the receiver and transmitter to the scatterer are given, where the subscripts represent the r / t-th element of the receiving antenna, and the path...l Sub-diameter m, position point s; and For the receiver and transmitter in new positions The displacement vectors corresponding to their respective initial positions. Indicates a new location The path vector from transmitter to receiver is calculated. The specific steps are as follows: first, based on the initial angle of arrival and path delay, calculate the unique scatterer (LBS)'s three-dimensional position vector in the global coordinate system. When the mobile terminal TX moves along the terminal trajectory to a new location point At this time, the position of the scatterer remains unchanged, and the calculation of the first step from the transmitter's new position is recalculated. The vector from each antenna element to the fixed scatterer And the distance from the scatterer to the receiver at the new position Vector of antenna elements The total length of the new path is then obtained by the sum of the magnitudes of these two vectors:

[0086] (12)

[0087] The new propagation delay is directly determined by the total path length. The physical process of the terminal moving along the trajectory leads to continuous changes in the multipath offset azimuth and arrival azimuth, i.e., changes in the four initial angles; simultaneously, the change in path length between two discrete time points. This will result in a frequency related to the carrier frequency Related phase drift In the frequency domain, this manifests as a Doppler shift.

[0088] F. Calculation of Polarization Channel Coefficients: This further incorporates the physical characteristics of the transmitting and receiving antennas themselves, as well as the changes in the polarization state of the signal during propagation, into the channel coefficients. The initial polarization state of the transmitted signal is represented as a two-dimensional column vector, i.e., the Jones vector. For N scatterers and their corresponding Jones matrix J, the polarization state of the signal is finally obtained. Simultaneously, based on the network layout and the real-time orientation of the mobile terminal and base station defined by the terminal trajectory input, the local coordinate system of the antenna is rotated to align with the simulated global coordinate system. Then, based on the multipath offset azimuth angle and arrival azimuth angle, values ​​are looked up in the rotated antenna radiation pattern matrix to obtain the complex antenna gain for that path in that specific direction. For the transmitter and receiver, these are denoted as follows: .

[0089] G. Application of Path Gain, Shadow Fading, and K-Factor: The distance-dependent path gain, shadow fading, and the Rice K-Factor (K-Factor), which dynamically varies along the trajectory, are ultimately applied to the generated channel coefficients. Here, the path gain PG is logarithmically scaled with respect to the distance d between the base station and the terminal in meters (m).

[0090] (13)

[0091] Where A, B, and C are scheme-specific coefficients. Shadowing fading is modeled by a random variable X, and the power obtained in the previous step is scaled by combining PG and shadowing fading. Converted to an amplitude scaling factor that can be directly multiplied by the signal waveform. ,have

[0092] (14)

[0093] in This corresponds to shadow fading. The K-factor, with the additional scaling factor for path power, is:

[0094] (15)

[0095] Where the indices l=1...L are path numbers. It is a formula Calculate the power of the first path. It is the K factor at the beginning of the trajectory, and It is the K-factor of the s-th position of the user's trajectory.

[0096] In practical applications, the amplitude scaling of the signal waveform can be directly multiplied as the scaling factor, and the entire channel impulse response matrix generated beforehand can be multiplied. When the macroscopic power attenuation of the simulated signal caused by the propagation distance and large obstacles, and the K factor changes due to vehicle movement, the non-uniform scaling factor of formula (15) can be multiplied by the channel coefficient of the corresponding path to ensure that the channel model can accurately reflect the line-of-sight conditions of the terminal at the new location.

[0097] H. Transition between time slices: First, the complete terminal trajectory is divided into several segments with largely unchanged large-scale parameters. Within each segment, the channel can be considered generally stationary. Next, for each independent segment, its corresponding propagation scenario parameters are called, and all steps A to G are executed completely, thereby generating many independent channel trajectories. Finally, in order to integrate these independent trajectory segments into a continuous sequence that reflects long-distance, cross-scenario movement in the real world, this embodiment of the invention employs a smooth transition algorithm at the connection points of adjacent segments. By creating a brief overlapping region at the connection point and cross-fading the energy of the preceding and following channel segments, seamless connection between different channel statistical characteristics is achieved.

[0098] The simulation program of this invention adopts an object-oriented programming approach. Since there are many functional modules of the transceiver in the vehicle network communication link, if a procedural programming approach is adopted, it will lead to a messy code structure and high coupling of functional modules, which is not conducive to development and subsequent system maintenance. Therefore, this invention adopts an object-oriented programming approach, which initializes the transmitter class and receiver class as objects separately, and defines their respective functional modules in the class, thereby defining a system-level transceiver.

[0099] like Figure 3 As shown in the transceiver modulation framework diagram, the transmitter includes channel coding starting from the binary bitstream, QAM modulation, OFDM and OTFS mapped modulation, pilot insertion, and cyclic prefix (CP) insertion. The channel module initializes the channel class and updates the channel coefficients at the frame level by loading the channel coefficient tensor generated by QuaDRiGa. The receiver includes carrier frequency offset (CFO) removal for OFDM applications, CP removal, OFDM and OTFS demodulation, channel estimation and equalization, QAM demodulation and channel decoding, and finally calculates the bit error rate based on the received binary bitstream.

[0100] The pilot-based channel estimation implemented in this embodiment of the invention achieves high-accuracy transmission of various multi-carrier modulation modes under poor channel conditions by implementing and customizing multiple pilot styles. Specifically, it adopts multiple pilot styles such as comb pilot, block pilot, and dot pilot, thereby enabling OFDM modulation to still achieve low bit error rate transmission in high-mobility dynamic vehicle network channel environments.

[0101] The channel coding module of the simulator in this embodiment implements Turbo coding and BCH coding. Furthermore, it allows users to select a no-coding mode based on their needs. These coding methods exhibit different performance characteristics under different signal-to-noise ratios (SNRs) and also vary in complexity and hardware cost in actual implementation. Users can freely switch between these methods by defining parameters, allowing them to choose the channel coding scheme according to their requirements. Common 5G standard coding schemes include LDPC, Turbo, Polar, and convolutional codes. LDPC codes demonstrate excellent performance in approximating the Shannon limit and are easy to implement in hardware due to the inherent parallelism of iterative decoding schemes. Turbo codes are a class of high-performance forward error correction codes that utilize an iterative algorithm that trades time for complexity and can approach the Shannon limit to the greatest extent possible, making them well-suited for vehicle-to-everything (V2X) communication. BCH codes have weaker error correction performance compared to Turbo and LDPC, but their simple coding method results in lower hardware implementation costs, making them suitable for low-cost scenarios such as storage systems, V2X, and certain low SNR wireless communication scenarios such as satellite communication. Furthermore, BCH codes allow for adjustment of code length and error correction capabilities. Through encoding and decoding, the transmission accuracy of communication systems in complex channel environments can be improved, transmission reliability enhanced, and bit error rate and frame error rate reduced. Moreover, according to Shannon's theorem, channel capacity is limited by SNR, while channel coding can increase channel capacity without increasing transmission power and channel bandwidth, bringing data transmission efficiency close to the Shannon limit. Finally, the choice of coding rate allows the system to adjust its coding strategy according to system conditions, improving the system's noise immunity; for example, using a high coding rate under good channel conditions and a low coding rate under poor channel conditions.

[0102] The digital modulation module implemented in this embodiment of the invention employs QAM modulation, a combination of simultaneous amplitude and phase modulation. Compared to single modulation, it can transmit more data information under the same bandwidth conditions, thus providing higher bandwidth utilization and better power utilization. Furthermore, the QAM order can be adjusted; as M increases in M-QAM, the amount of information carried by the modulated signal increases. This module allows for free switching between QAM types, allowing users to choose a suitable QAM scheme according to their needs. The simulation system can implement 4QAM / 16QAM / 64QAM / 256QAM, with 2, 4, 6, and 8 bits per symbol, respectively. In the corresponding constellation diagram, a higher modulation order results in denser constellation points and a reduced distance between them, which increases the probability of decoding errors. Therefore, a higher modulation order makes it easier to be misclassified as a different symbol, leading to a higher bit error rate. However, a higher order also means a higher number of bits carried per symbol, i.e., higher-order modulation carries higher information density. Therefore, an appropriate modulation order should be selected for different communication conditions.

[0103] Multicarrier modulation splits data across multiple subcarriers for transmission, thereby reducing the impact of frequency-selective fading, improving spectral efficiency, and reducing inter-symbol interference. Especially in vehicular network communication environments, multicarrier modulation reduces the symbol rate of each subcarrier to mitigate multipath effects, signal fading, and inter-symbol interference, resulting in better system stability. Furthermore, without multicarrier modulation, equalizer computation is high, while in a multicarrier system, channel equalization only requires subcarrier gain adjustment in the frequency domain, reducing computational complexity. The multicarrier modulation methods used in this embodiment of the invention employ OFDM and OTFS methods, which can be selected according to user needs.

[0104] OFDM is a multi-carrier transmission technology that uses orthogonal subcarriers to transmit data in parallel. The waveforms of different subcarriers are orthogonal at the center frequency, thereby avoiding interference, improving spectrum utilization, and effectively suppressing inter-symbol interference caused by time spread in the wireless channel. The OFDM technology used in this embodiment of the invention also inserts a cyclic prefix, which transforms the channel from linear convolution to cyclic convolution, thereby reducing the complexity of equalization. OTFS, on the other hand, is a two-dimensional modulation scheme computed in the time-delay Doppler domain, which can adapt well to time- and frequency-selective fading channels, significantly improving the system's transmission performance in fast time-varying channels. Figure 4 The relationship between OFDM and OTFS domain transformations is shown in the diagram. It is the channel response in the DD domain. It's Doppler shift.

[0105] In OTFS modulation technology, the received signal and transmit signals The relationship between them is as follows:

[0106] (16)

[0107] in It is the channel spread function, also known as the time-varying impulse response. relative to The Fourier transform of the transmitted signal to the received signal can be interpreted as the Heisenberg transform, derived from the function... Parameterization is performed. OTFS modulation can be interpreted as a cascade of two two-dimensional transforms. At the transmitter, the information symbols in the time-delay-Doppler domain are first... The finite Fourier transform (FFT) maps the tactic region to the TF domain. Since the actual implementation has a finite block length, this mapping also involves a window function and periodization, denoted as period . Equation (17):

[0108] (17)

[0109] in , , The receiving end uses The Symptotic Finite Fourier Transform (SFFT) is... Perform the reverse process:

[0110] (18)

[0111] The combination of a window function and the inverse SFFT is collectively referred to as the OTFS transform, which converts a time-frequency signal into a time-domain signal using the Heisenberg transform. :

[0112] (19)

[0113] in , It is the time-frequency window function of the transmitting end. Equation (20) is applied to... The Heisenberg operator, with parameters as The basis functions of the transmitter and receiver satisfy the biorthogonality condition:

[0114] (20)

[0115] Therefore, the received signal can be understood as a cascade of two Heisenberg operators acting on basis functions, one describing the modulation process and the other describing channel effects. The two operators are derived from the function... and A parameterized concatenation of Heisenberg operators is still a single Heisenberg operator; therefore, its parametric function is the convolution of the original two functions.

[0116] (twenty one)

[0117] The received signal can be obtained from this:

[0118] (twenty two)

[0119] in, It is the impulse response of the combined transformation:

[0120] (twenty three)

[0121] The receiver performs a cascaded Wigner transform and OTFS transform. That is, the received signal is first filtered with the receiving basis function, which is represented in the time-delay-Doppler domain as:

[0122] (twenty four)

[0123] exist , After sampling, the following was obtained:

[0124] (25)

[0125] Therefore, an end-to-end channel can be described as:

[0126] (26)

[0127] Then the receiving end... After sampling, windowing, and periodization, SFFT is performed to obtain the estimated sequence of information symbols. This is achieved through two-dimensional periodic convolution:

[0128] (27)

[0129] in It is obtained by multiplying the channel response by a window function using an SFFT, then performing a circular convolution followed by sampling.

[0130] (28)

[0131] Finally, the final information symbol estimate is obtained through a two-dimensional equalization algorithm. It is worth noting that the channel gain for each received symbol during transmission is... Diversity gain is obtained simultaneously from both the time delay domain and the Doppler domain.

[0132] The receiver class in this embodiment of the invention implements channel estimation and channel equalization functions. The pilot-based channel estimation module is implemented by inserting pilot symbols into the time-frequency resource block at the transmitting end to track the channel impulse response at the current time-frequency position. Then, at the receiving end, interpolation estimation is performed based on the pilot channel response to calculate the channel estimation matrix for the entire channel plane. .

[0133] This invention provides pilot patterns for both OFDM and OTFS modulation. For OFDM, this invention implements three pilot patterns: block pilot, comb pilot, and dot pilot, as shown below. Figure 5 The OFDM pilot pattern design diagram shows three pilot patterns. The first is block pilot, which is characterized by inserting column by column of pilot signals in the time domain, that is, inserting pilots in the symbol direction. The second is comb pilot, which is characterized by inserting row by row of pilot signals in the frequency domain, that is, inserting pilots in the subcarrier direction. The third is dot pilot, which is characterized by inserting dot-shaped pilot signals in both the time domain and the frequency domain.

[0134] OFDM channel estimation is based on pilot signals inserted at the transmitter. First, the channel estimate is calculated at the pilot location, and then two-dimensional interpolation is used to extend it to the entire time-frequency plane. For example, equation (29) first uses the pilot signal from the receiver. and the pilot signal at the transmitting end Calculate the channel response at the pilot point ,in It is a subcarrier index. It is a symbol position index

[0135] (29)

[0136] Then, an interpolation network is constructed according to formula (30), where It is the number of sub-carriers. The number of symbols:

[0137] (30)

[0138] Finally, one-dimensional interpolation is performed on the block and comb pilots, and two-dimensional interpolation is performed on the point pilots, as shown in formula (31), where Used for unidirectional interpolation Used for two-dimensional interpolation.

[0139] (31)

[0140] OFDM channel equalization uses the estimation matrix obtained through channel estimation to evaluate the received signal. Perform channel estimation matrix The inverse operation is used to obtain the estimated signal, as shown in formula (32), thereby compensating for the influence of channel interference on the channel and reducing inter-symbol interference.

[0141] (32)

[0142] like Figure 6 As shown in the OTFS pilot pattern design diagram, compared with OFDM, the OTFS pilot pattern only has a symbol in the center of the time delay Doppler domain. However, it needs to set a zero-padding (ZP) protection zone around the pilot signal to reduce the interference of data symbols. The OTFS symbol first performs channel estimation based on the pilot information, and then uses the maximum ratio combining (MRC) method to equalize the channel. First, it extracts the pilot signal in the pilot area to calculate the average power. This average power is used for subsequent judgment. That is, if the signal amplitude at a certain position exceeds this average power, it is considered that the position may contain channel information. Then, the received signal is traversed. When its power is higher than the threshold, the possible channel path at that position is extracted and converted to the time domain to generate a time-frequency channel. The time-frequency channel information is used for MRC detection and equalization. Among them, the MRC detection principle is to perform weighted combining of the path of each channel according to the conjugate of the channel gain. Formula (33) calculates the optimal combining weight of MRC. It is the channel gain, normalization factor. Keeping the total power constant, the combined received signal is then calculated using formula (34). ,in It is the first The signal received along the path.

[0143] (33)

[0144] (34)

[0145] In this embodiment of the invention, a CFO removal module is implemented using CP before OFDM demodulation. OFDM is easily affected by CFO, which is caused by Doppler frequency shift. This destroys the orthogonality between subcarriers, resulting in a serious performance degradation of the multicarrier system. CFO will cause each OFDM symbol to generate a certain degree of phase shift, thus presenting a linear arrangement on the constellation diagram. CFO is estimated and compensated by adding a cyclic prefix to the signal. In this embodiment of the invention, the structural characteristics of CP and symbol tail repetition are used to perform autocorrelation operation on the received signal to estimate the carrier frequency shift and complete the frequency offset compensation. It should be noted that the main function of CP is to transform the channel convolution into a cyclic convolution, thereby suppressing inter-symbol interference caused by multipath propagation; while the inter-carrier interference caused by Doppler or frequency shift destroying the orthogonality of subcarriers needs to be eliminated by CFO estimation and equalization. As in formula (35), the first OFDM symbols After the first After obtaining the strip diameter .

[0146] (35)

[0147] in, This represents the number of points in the Inverse Fast Fourier Transform (IFFT). It is Gaussian noise. For the length of CP, This represents the total number of sampling points. The sampling interval is defined. First, the influence of noise is ignored. At that time, due to Therefore, we can obtain:

[0148] (36)

[0149] Comparing the two equations, we can find that... and Same amplitude, phase difference Fixed as Therefore, we can finally obtain an estimate of the frequency offset. :

[0150] (37)

[0151] The embodiments of the present invention use the estimated value calculated by equation (37) to correct the carrier frequency offset of OFDM and suppress inter-symbol interference caused by multipath propagation and subcarrier orthogonality in the CFO removal module.

[0152] The object-oriented programming design of this invention mainly consists of three independent object classes: transmitter, receiver, and channel. The algorithm pseudocode is as follows:

[0153] Table 1 Transmitter Classes

[0154]

[0155] Table 2 Channel Classes

[0156]

[0157] Table 3 Receiver Types

[0158]

[0159] like Figure 7 The figure shows the bit error rate (BER) curves for OFDM and OTFS under a Gaussian channel. Both OFDM and OTFS have a frame structure of 32 subcarriers and 64 symbols, with CP and ZP lengths both set to 8. OFDM uses symbol pilots, while OTFS uses embedded pilots. The BER curves show that under simple channel conditions with only noise, both OFDM and OTFS exhibit good stability under low signal-to-noise ratio (SNR) conditions, achieving BER rates above 10 dB. level.

[0160] like Figure 8 The figure shows the bit error rate curves of OFDM and OTFS under frequency-selective channels. Both OFDM and OTFS have a frame structure of 32 subcarriers and 64 symbols, with CP and ZP lengths both set to 8. OFDM uses symbol pilots, while OTFS uses embedded pilots. Frequency-selective channels are characterized by multipath delay but no Doppler. Within a signal-to-noise ratio range of 4-14 dB, both OTFS and OFDM exhibit good robustness to frequency selectivity.

[0161] like Figure 9 The figure shows the bit error rate curves of OFDM and OTFS under a time-selective channel. Both OFDM and OTFS have a frame structure of 32 subcarriers and 64 symbols, with CP and ZP lengths both set to 8. OFDM uses symbol pilots, while OTFS uses embedded pilots. The time-selective channel is configured with only a single-path 800Hz Doppler shift and no multipath delay. In a time-selective channel, OFDM can compensate for the Doppler shift offset through CFO, but it remains sensitive to time selectivity. Both OFDM and OTFS exhibit low bit error rates.

[0162] like Figure 10The figure shows the bit error rate curves of OFDM and OTFS under the EVA channel. Both OFDM and OTFS have a frame structure of 32 subcarriers and 64 symbols, with CP and ZP lengths both set to 8. OFDM uses symbol pilots, while OTFS uses embedded pilots. The EVA channel is configured with a maximum vehicle speed of 120 km / h, making it a dual-select channel for high-speed movement with both time and frequency selectivity. In this configuration, OTFS effectively captures channel diversity in the DD domain, outperforming OFDM in the dual-select channel and achieving more robust and reliable communication.

[0163] Through the above simulation experiments, it can be seen that the embodiments of the present invention propose a high-fidelity simulation method for wireless communication links in highly dynamic vehicle networks. This method can address the problems of missing dynamic characteristic modeling and poor program design scalability in vehicle network simulation. It can achieve accurate channel modeling of highly mobile vehicle network links and realize high-performance communication transceiver signal processing flow through an object-oriented vehicle network link-level simulation program.

[0164] This invention also discloses a computer system, including a memory, a processor, and a computer program / instructions stored in the memory and executable on the processor. When the computer program / instructions are executed by the processor, they implement the steps of the high-fidelity simulation method for wireless communication links in a highly dynamic vehicle network.

[0165] This invention also discloses a computer program product, including a computer program / instruction, which, when executed by a processor, implements the steps of the high-fidelity simulation method for wireless communication links in a highly dynamic vehicle network.

[0166] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A high-fidelity simulation method for wireless communication links in highly dynamic vehicle-to-everything (V2X) networks, characterized in that, Includes the following steps: Based on the input network layout, terminal trajectory, propagation scenario and antenna mode, a channel simulation platform is used to perform channel modeling and generate time-varying channel coefficients. The time-varying channel coefficients are used to characterize the multipath fading, Doppler frequency shift and delay spread characteristics in a high-dynamic vehicle network communication environment. Construct independent and instantiable transmitter, receiver, and channel objects; the transmitter object is used to receive input parameters including coding type, modulation type, and input bit stream, and generate a transmit signal containing time-domain modulated signal and pilot signal; The channel class object is used to receive the transmitted signal and the channel coefficients, and to apply the channel coefficients to simulate multipath effect and Doppler effect and add noise to generate the received signal. The receiver class object is used to receive the received signal and modulation type, pilot information and noise variance, perform channel estimation, equalization and demodulation processing, and generate decoded bit stream; Performance metrics are calculated based on the input bitstream and the decoded bitstream.

2. The high-fidelity simulation method for wireless communication links in highly dynamic vehicle networks according to claim 1, characterized in that: The network layout includes the initial three-dimensional coordinates of base stations and mobile terminals, antenna array configuration, and antenna orientation; the terminal trajectory includes a series of three-dimensional coordinate points arranged in chronological order to drive the time-varying characteristics of the channel; the propagation scenario includes line-of-sight propagation or non-line-of-sight propagation in an urban macrocell environment; the antenna pattern is used to define the three-dimensional far-field radiation pattern matrix of the transmitting and receiving antennas.

3. The high-fidelity simulation method for wireless communication links in highly dynamic vehicle networks according to claim 1, characterized in that: Based on the QuaDRiGa platform, time-varying channel coefficients are generated according to input parameters. The generation process includes: Based on the selected propagation scenario, a set of spatially interrelated large-scale fading parameters is generated, including target delay spread and target angle spread; Based on the target delay spread and target angle spread, as well as the transceiver positions in the network layout, the initial propagation delay and initial angle are calculated for each multipath component. An initial path power is assigned to each multipath component, and the initial path power is adjusted by a frequency-dependent scaling factor based on the multiband effect; The path power is adjusted by applying the Rice K-factor, and the delay spread and angle spread are calibrated to make the overall statistical characteristics match the target delay spread and target angle spread. Based on the terminal trajectory, the delay, angle, and phase drift of multipath components on a short time slice are simulated to generate Doppler frequency shift; The polarization channel coefficients are calculated, taking into account the physical characteristics of the transmitting and receiving antennas themselves as well as the changes in the polarization state of the signal during propagation. The distance-dependent path gain, shadow fading, and Rice's K-factor, which is dynamically varying along the trajectory, are applied to the already generated channel coefficients. The terminal trajectory is divided into multiple segments, each segment generates channel coefficients independently, and a smooth transition is performed between adjacent segments to generate a continuous channel sequence.

4. The high-fidelity simulation method for wireless communication links in highly dynamic vehicle networks according to claim 1, characterized in that: The channel class object is initialized by loading the channel coefficient tensor generated by QuaDRiGa, and the channel coefficients are updated at the frame level.

5. The high-fidelity simulation method for wireless communication links in highly dynamic vehicle networks according to claim 1, characterized in that: The steps for generating a received signal by simulating multipath and Doppler effects and superimposing noise using the channel object include: Create an empty receive signal for subsequent accumulation of the effects of each path; For each signal path defined in the parameters, the phase rotation factor is calculated based on the Doppler frequency shift and time index of the path. The original transmitted signal is first multiplied by the phase rotation factor, and then a cyclic shift operation is performed. The number of shift points is determined by the time delay parameter of the path. The generated time delay signal is multiplied by the gain coefficient of the path and then accumulated into the total received signal. Gaussian white noise is added to the received signal that has been superimposed with all multipath components, according to the specified signal-to-noise ratio; Output the simulated received signal.

6. The high-fidelity simulation method for wireless communication links in highly dynamic vehicle networks according to claim 1, characterized in that: The steps for the transmitter object to generate a transmit signal containing a time-domain modulated signal and a pilot signal include: Initialize configuration parameters, including word length, code length, and frame length; If the encoding function is enabled, the input bitstream is encoded, and the encoding type includes BCH encoding, Turbo encoding, or no encoding. Depending on the modulation type, the data enters different modulation paths: If the modulation type is OFDM, the encoded data is mapped onto the QAM constellation diagram, a pilot signal is inserted in the time-frequency domain, an inverse fast Fourier transform is performed, and a parallel-to-serial conversion is performed to convert the frequency domain signal into a time domain signal, and a cyclic prefix is ​​added; if the modulation type is OTFS, the encoded data is mapped onto the QAM constellation diagram, a pilot signal is inserted in the time-delay-Doppler domain, a two-dimensional inverse symplectic finite Fourier transform is performed to convert the time-delay-Doppler domain signal into a time-frequency domain symbol, and then a Heisenberg transform is used to obtain the time domain signal. Output time-domain signal and pilot information.

7. The high-fidelity simulation method for wireless communication links in highly dynamic vehicle networks according to claim 1, characterized in that: The steps for the receiver object to perform channel estimation, equalization, and demodulation processing to generate the decoded bitstream include: Depending on the modulation type, the signal enters different processing paths: If the modulation type is OFDM, after removing the cyclic prefix, a Fast Fourier Transform is performed to convert the time-domain signal to the frequency domain. Channel estimation is performed using known pilot information, and channel equalization is performed based on the noise variance and the channel estimation results. If the modulation type is OTFS, a Symplectic Finite Fourier Transform is performed on the received signal to convert the time-domain received signal to the time-delay-Doppler domain. The multipath parameters of the wireless channel are estimated based on known pilot information, and the path signals are weighted and combined using maximum ratio combining, and linear equalization is performed in combination with the noise variance. Perform QAM demapping on the equalized signal; If channel coding is enabled, the demodulated information is decoded according to the corresponding coding type. Output the decoded bitstream as an estimate of the original transmitted information.

8. The high-fidelity simulation method for wireless communication links in highly dynamic vehicle networks according to claim 1, characterized in that: Pilot patterns were designed for both OFDM and OTFS modulation. Pilots were inserted in the time-frequency domain for OFDM and in the time-delay Doppler domain for OTFS. For OFDM modulation, three pilot styles are implemented: block pilot, comb pilot, and dot pilot. The block pilot inserts pilot signals in the time dimension, the comb pilot inserts pilot signals in the frequency dimension, and the dot pilot inserts pilot signals in a sparse dot matrix in a two-dimensional time-frequency domain plane. For OTFS modulation, there is only a pilot signal in the center of the time-delay Doppler domain, and a zero-fill protection zone is set around the pilot signal.

9. A computer system comprising a memory, a processor, and computer programs / instructions stored in the memory and executable on the processor, characterized in that, When the computer program / instructions are executed by the processor, they implement the steps of the high-fidelity simulation method for wireless communication links in a highly dynamic vehicle network according to any one of claims 1-8.

10. A computer program product comprising a computer program / instructions, characterized in that, When the computer program / instructions are executed by the processor, they implement the steps of the high-fidelity simulation method for wireless communication links in a highly dynamic vehicle network according to any one of claims 1-8.

Images