High-speed rail 5G system switching method and device based on dynamic beam forming, equipment and medium

Abstract

The invention discloses a high-speed rail 5G system switching method, device and equipment based on dynamic beam forming and a medium, and relates to the technical field of high-speed rail communication, and the method comprises the steps: constructing a beam forming signal model; constructing a beam forming signal switching model according to the beam forming signal model; the beam forming signal switching model is used for judging to trigger a switching activation event when the received signal receiving power of the target base station is continuously higher than the signal receiving power of the source base station in the moving process of the train and the difference value exceeds a preset lag threshold; constructing a beam direction optimization problem; the objective of the beam direction optimization problem is to minimize the beam forming signal switching failure probability and the beam forming adjustment overhead; and solving the beam direction optimization problem by adopting deep reinforcement learning to obtain the optimal beam direction of the source base station and the target base station under each time slot in the switching period of the beam forming signal, and realizing high-reliability and low-delay switching control in a high-speed railway scene.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of high-speed rail communication, and particularly relates to a high-speed rail 5G system switching method and device based on dynamic beamforming, equipment and medium. BACKGROUND

[0002] With the rapid development of intelligent railway and 5G communication technology, the fifth generation mobile communication technology-railway dedicated system (5G for Railway, 5G-R) for train-to-ground communication (T2G) has become a key support for new generation high-speed railway information transmission. Compared with the performance limitation of the railway global system for mobile communications (Global System for Mobile Communications-Railway, GSM-R) with only 16 kHz bandwidth and hundreds of milliseconds of delay, 5G-R has the ability of megahertz-level bandwidth, millisecond-level delay and large connection, which can effectively support high-speed rail operation control, vehicle monitoring and passenger service and other businesses. However, in the high-speed railway scene, the train running speed is as high as 350 km / h or even higher, and the high-speed mobility and weak coverage effect at the cell edge are superimposed, so that the wireless channel presents significant time-varying characteristics and multipath fading characteristics, resulting in frequent switching of communication links and fluctuation of signal quality. The signal interruption and delay in the switching process are important bottlenecks restricting the reliability of the 5G-R system.

[0003] The current 5G switching mechanism is mainly based on the A3 event triggering criterion, that is, when the reference signal received power (RSRP) of the target base station exceeds the RSRP of the source base station and lasts for more than a hysteresis threshold for a period of time, the switching is triggered. This method performs well in low-speed scenarios, but in high-speed scenarios, there are two outstanding problems: (1) Hysteresis criterion failure: high-speed motion causes RSRP measurement lag, triggering time deviates from the optimal switching point, and is prone to early or late switching; (2) Fixed beam coverage is insufficient: in high-frequency communication (such as millimeter wave), the beam direction is not adjusted adaptively with the train movement, which will cause the main lobe of the signal to shift and the gain to drop sharply, thereby increasing the switching failure probability and interruption probability.

[0004] To solve the above problems, researchers have proposed various optimization methods. For example, in the article "Adaptive Handover Algorithm for LTE-R System in High-Speed Railway Scenario", an adaptive A3 threshold adjustment is proposed. In the article "Fast Adaptive Handover Using Fuzzy Logic for 5G Communications on High-Speed Trains", a handover strategy based on location prediction is designed. The former dynamically adjusts the threshold to improve flexibility, but lacks fault tolerance. The latter relies on accurate trajectory and channel priori, which is difficult to cope with rapidly changing environments.

[0005] Beamforming technology improves the signal strength at the cell edge by directional radiation, providing a new way to improve the reliability of high-speed scenario communication. In the article "High-Speed based Adaptive Beamforming Handover Scheme in LTE-R", a fixed beam gain scheme is introduced to enhance the edge signal in a specific coverage interval. However, the fixed beam gain method still has the problem of high tracking overhead, as it needs to adjust the beam direction in real time. Especially in high-speed scenarios, how to achieve optimal matching of beam direction in dynamic channels becomes a key factor in determining the performance of handover.

[0006] In summary, how to fully utilize the directional gain and spatial diversity capability of beamforming, while maintaining high data rates and reducing handover failure rates and beam adjustment overhead, has become a key scientific problem for 5G-R systems. SUMMARY

[0007] The purpose of the present application is to provide a high-speed rail 5G system handover method, device, equipment and medium based on dynamic beamforming, to realize high-reliability and low-latency handover control in high-speed railway scenarios.

[0008] To achieve the above purpose, the present application provides the following solutions: In a first aspect, the present application provides a high-speed rail 5G system handover method based on dynamic beamforming, comprising: According to the high-speed railway communication and handover scenario model of the high-speed rail 5G system, a beamforming signal model is constructed. The beamforming signal model is used to determine the signal reception power of the train from the source base station and the target base station during movement. The high-speed rail 5G system supports multiple-input multiple-output and beamforming functions; Based on the beamforming signal model, a beamforming signal switching model is constructed; the beamforming signal switching model is used to determine and trigger a switching activation event when the signal reception power of the target base station received by the train is continuously higher than the signal reception power of the source base station and the difference exceeds a preset hysteresis threshold during the train's movement. After the switching activation event is triggered, a beam direction optimization problem is constructed; the objective of the beam direction optimization problem is to minimize the beamforming signal switching failure probability and the beamforming adjustment overhead. The beam orientation optimization problem is solved by deep reinforcement learning to obtain the optimal beam orientation of the source base station and the target base station in each time slot during the switching period of the beamforming signal.

[0009] Optionally, the beamforming signal switching failure probability is expressed as: ; in, Indicates the probability of beamforming signal switching failure. Indicates the probability of switching activation. For intermediate parameters, , This indicates the probability of interruption when communication services are provided by either the source base station or the target base station; ; when When the probability of interruption is when the communication service is provided by the source base station, it is expressed as: ; Where t represents the current time, and T represents the switching period of the beamforming signal. Let represent each discrete moment in the time window from tT to t. Represents probability operators; Describes the right-tail function of the standard normal distribution. and These respectively represent the source base station and the target base station at Signal reception power at any given time This indicates the preset hysteresis threshold. and These respectively represent the source base station and the target base station at Path loss at time step, This represents the path loss of the source base station at time t. and These respectively represent the source base station and the target base station at Beam gain in the direction of the beam at any given time. This represents the beam gain of the source base station in the beam direction at time t. This indicates the preset communication interruption threshold. This represents the standard deviation of the shadowed fading modeled as a zero-mean Gaussian random variable. This represents the standard deviation of the shadowing fading between the source base station and the train, modeled as a zero-mean Gaussian random variable.

[0010] Optionally, solving the beamdirection optimization problem specifically includes: solving the objective function of the beamdirection optimization problem, wherein the objective function is expressed as: ; in, This represents the beam direction optimization problem, where T represents the switching period of the beamforming signal. Indicates the probability of beamforming signal switching failure. and These represent the time intervals of the source base station and the target base station, respectively. Beam direction angle; Indicates the number of beam direction adjustments; The unit beam adjustment overhead factor; and All are weighted factors. The spacing between adjacent base stations deployed at equal intervals along a railway track. This represents the coverage radius of the base station in omnidirectional mode. and These represent the Euclidean distances between the train and the source base station, and between the train and the target base station, respectively.

[0011] Optionally, deep reinforcement learning is used to solve the beam orientation optimization problem to obtain the optimal beam orientation of the source base station and the target base station in each time slot during the switching period of the beamforming signal, specifically including: The arrival directions of the train to the source base station and the target base station, as well as the path loss between the train and the source base station and the target base station, are used as the state space; the beam directions of the source base station and the target base station are used as the action space; and the negative objective function is used as the reward function. The objective function is the objective function for solving the beam direction optimization problem. Based on the state space, the action space, and the reward function, the objective function of the beam direction optimization problem is solved using the deep deterministic policy gradient algorithm, thereby obtaining the optimal beam directions of the source base station and the target base station in each time slot during the switching period of the beamforming signal.

[0012] Optionally, the signal received power of the source base station is expressed as: ; The signal received power of the target base station is expressed as: ; in, and These represent the signal received power of the source base station and the target base station, respectively. This refers to the transmission power of the base station, which is deployed at equal intervals along the railway track. and These represent the path losses of the source base station and the target base station, respectively. and These represent the beam gains of the source base station and the target base station in the current beam direction, respectively. and These represent the shadow fading of the source base station and the target base station, respectively.

[0013] Optionally, the path loss between the source base station and the target base station is expressed as: ; in, Indicates the center frequencies of the source base station and the target base station. Indicates the average height of the building. and These represent the Euclidean distances between the train and the source base station, and between the train and the target base station, respectively.

[0014] Optionally, the beam gains of the source base station and the target base station in the current beam direction are expressed as: ; ; in, and Let represent the directional vectors of the antenna arrays of the source base station and the target base station, respectively. and These represent the directional weight vectors for beamforming of the source base station and the target base station, respectively. Indicates the direction of arrival between the train and the source base station. Indicates the direction of arrival between the train and the target base station. Indicates the beam direction angle of the source base station. This indicates the beam direction angle of the target base station.

[0015] Secondly, this application provides a high-speed rail 5G system switching device based on dynamic beamforming, comprising: The beamforming signal model construction module is used to construct a beamforming signal model based on the high-speed railway communication and handover scenario model of the high-speed rail 5G system. The beamforming signal model is used to determine the signal reception power received by the train from the source base station and the target base station during the train's movement. The high-speed rail 5G system supports multiple input multiple output and beamforming functions. The beamforming signal switching model construction module is used to construct a beamforming signal switching model based on the beamforming signal model. The beamforming signal switching model is used to determine and trigger a switching activation event when the signal reception power of the target base station received by the train is continuously higher than the signal reception power of the source base station and the difference exceeds a preset hysteresis threshold during the train's movement. A beam orientation optimization problem construction module is used to construct a beam orientation optimization problem after a switching activation event is triggered; the objective of the beam orientation optimization problem is to minimize the beamforming signal switching failure probability and the beamforming adjustment overhead. The solution module is used to solve the beam orientation optimization problem using deep reinforcement learning, and obtain the optimal beam orientation of the source base station and the target base station in each time slot during the switching period of the beamforming signal.

[0016] Thirdly, this application provides a computer device, including: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the steps of the high-speed rail 5G system switching method based on dynamic beamforming as described above.

[0017] Fourthly, this application provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the high-speed rail 5G system handover method based on dynamic beamforming as described above.

[0018] Fifthly, this application provides a computer program product, including a computer program that, when executed by a processor, implements the steps of the high-speed rail 5G system handover method based on dynamic beamforming as described above.

[0019] According to the specific embodiments provided in this application, the following technical effects are disclosed: This application provides a handover method, apparatus, equipment, and medium for a high-speed rail 5G system based on dynamic beamforming. Through a beamforming signal handover model, the handover conditions for beamforming signals are clarified, and a beam direction optimization problem is constructed. The optimization objective is to minimize the beamforming signal handover failure probability and beamforming adjustment overhead. The relationship between beam direction and communication performance is quantified. A beam direction optimization mechanism is introduced during the handover decision process, enabling coordinated adjustment and dynamic alignment of the source and target base station beams, thereby improving link signal gain and handover success rate. Specifically, deep reinforcement learning (DRL) is used to solve the beam direction optimization problem, which can significantly reduce beam tracking overhead while ensuring communication continuity, achieving high-reliability, low-latency handover control in high-speed rail scenarios. Attached Figure Description

[0020] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0021] Figure 1 This is a flowchart illustrating a high-speed rail 5G system handover method based on dynamic beamforming, provided as an embodiment of this application.

[0022] Figure 2 This is a schematic diagram of a train switching scenario provided in an embodiment of this application.

[0023] Figure 3 This is a schematic diagram illustrating the dynamic change of beam direction with direction of arrival (AoD) in a dynamic beamforming optimization method provided in an embodiment of this application.

[0024] Figure 4 A comparison chart showing the change in switching failure probability of the dynamic beamforming optimization method provided in an embodiment of this application under different beamforming strategies.

[0025] Figure 5 This is a schematic diagram of the functional modules of a high-speed rail 5G system switching device based on dynamic beamforming, provided in an embodiment of this application.

[0026] Figure 6 This is a schematic diagram of the structure of a computer device provided in an embodiment of this application. Detailed Implementation

[0027] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0028] To make the above-mentioned objectives, features and advantages of this application more apparent and understandable, the application will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0029] In one exemplary embodiment, this application provides a high-speed rail 5G system handover method based on dynamic beamforming, such as... Figure 1 As shown, the high-speed rail 5G system switching method based on dynamic beamforming includes steps 101-104.

[0030] Step 101: Construct a beamforming signal model based on the high-speed railway communication and handover scenario model of the high-speed rail 5G system; the beamforming signal model is used to determine the signal reception power received by the train from the source base station and the target base station during the train's movement; the high-speed rail 5G system supports multiple input multiple output and beamforming functions.

[0031] Step 102: Construct a beamforming signal switching model based on the beamforming signal model; the beamforming signal switching model is used to determine and trigger a switching activation event when the signal reception power of the target base station received by the train is continuously higher than the signal reception power of the source base station and the difference exceeds a preset hysteresis threshold during the train's movement.

[0032] Specifically, the signal reception power of the target base station being continuously higher than that of the source base station means that, within a set time period, the signal reception power of the target base station is continuously higher than that of the source base station.

[0033] Step 103: After triggering the switching activation event, construct the beam direction optimization problem; the objective of the beam direction optimization problem is to minimize the beamforming signal switching failure probability and beamforming adjustment overhead.

[0034] Step 104: Use deep reinforcement learning to solve the beam direction optimization problem to obtain the optimal beam directions of the source base station and the target base station in each time slot during the switching period of the beamforming signal.

[0035] This application aims to address the issues of rapid signal attenuation and frequent handover failures in high-speed railway scenarios due to high train mobility and weak cell edge coverage. It proposes a handover method for high-speed railway 5G systems based on dynamic beamforming. Specifically, it is a 5G high-speed railway communication method based on beamforming and intelligent learning optimization. By introducing a beam direction optimization mechanism during the handover decision process and combining it with deep reinforcement learning to achieve adaptive dynamic adjustment of the beam direction, it maintains beam alignment and signal enhancement in a rapidly changing channel environment. This improves the handover success rate, reduces handover latency, and reduces beam adjustment overhead while ensuring communication quality, achieving highly reliable 5G-R system connectivity and improving communication reliability and handover performance in high-speed train environments.

[0036] In an exemplary embodiment, the construction of a high-speed railway communication and handover scenario model specifically includes: in a high-speed railway communication scenario, equidistant lines are formed along the railway track. Deploy multiple base stations (gNBs), with the coverage radius of each base station in omnidirectional mode being [missing information]. .like Figure 2 As shown, let the antenna height of the base station be... The height of the train antenna is This forms a three-dimensional geometric model. The train travels at a speed... Running along the orbit, its motion process is discretized, at time... The position is recorded as ,in Indicates the switching cycle. The measurement period is indicated by the distance between the base station and the railway track. The high-speed rail 5G system is based on the 5th Generation New Radio (5G NR) wireless access architecture, supporting Multiple Input Multiple Output (MIMO) and beamforming functions, providing a foundation for subsequent beam optimization.

[0037] In an exemplary embodiment, the construction of the beamforming signal model specifically includes: (a) the base station employing a method containing... A uniform linear array (ULA) of antenna elements for transmitting signals. After beam weighting, the signal is sent to the train to establish a model of the signal received by the train. .

[0038] ; in, For a moment The signal received by the train This represents the directional vector of the antenna array; This represents the directional weight vector for beamforming; Indicates channel noise; Indicates the direction of arrival (DoA) between the train and the base station; Indicates the beam direction angle; This represents the large-scale channel gain, specifically expressed as... ,in This indicates the base station's transmit power. This represents path loss, calculated based on the 5G-R RMa channel model.

[0039] in, , Indicates the antenna spacing. Indicates wavelength; .

[0040] .

[0041] in, Indicates the center frequency of the base station. Indicates the average height of the building. This represents the Euclidean distance between the train and the base station.

[0042] (b) Joint directional vector With weight vector To obtain beamforming gain .

[0043] .

[0044] (c) During the train's movement, signals are received from both the source base station and the target base station, and the received power from each base station is calculated. and .

[0045] The signal received power of the source base station is expressed as: .

[0046] The signal received power of the target base station is expressed as: .

[0047] in, and These represent the signal received power of the source base station and the target base station, respectively. This refers to the transmission power of the base station, which is deployed at equal intervals along the railway track. and These represent the path losses of the source base station and the target base station, respectively. and These represent the beam gains of the source base station and the target base station in the current beam direction, respectively. and The shadow fading of the source and target base stations, respectively, follows a Gaussian distribution and is expressed in the decibel domain as follows: Shadow fading describes the slow, time-varying, large-scale random fluctuations in received signal power caused by obstruction.

[0048] The path loss between the source base station and the target base station is expressed as: ; in, Indicates the center frequencies of the source base station and the target base station. Indicates the average height of the building. and These represent the Euclidean distances between the train and the source base station, and between the train and the target base station, respectively.

[0049] The beam gain of the source base station and the target base station in the current beam direction is expressed as: ; ; in, and Let represent the directional vectors of the antenna arrays of the source base station and the target base station, respectively. and These represent the directional weight vectors for beamforming of the source base station and the target base station, respectively. Indicates the direction of arrival between the train and the source base station. Indicates the direction of arrival between the train and the target base station. Indicates the beam direction angle of the source base station. This indicates the beam direction angle of the target base station.

[0050] In an exemplary embodiment, the construction of the beamforming signal switching model specifically includes: (a) in the high-speed railway communication system, when the signal reception power of the target base station is continuously higher than the signal reception power of the source base station, and the difference exceeds a preset hysteresis threshold. When this occurs, the system determines that a switching activation event has been triggered.

[0051] The trigger condition is expressed as follows: .

[0052] (b) Under the above handover conditions, calculate the handover activation probability. .

[0053] (c) When the train's signal receiving power is lower than the preset interruption threshold In such cases, system communication may be interrupted. Calculate the probability of interruption when communication service is provided by the source base station or the target base station. , .

[0054] (d) Based on the above handover activation probability and communication interruption probability, define the handover failure probability. This represents the combined probability of no handover being triggered and the interruption occurring after a handover being triggered.

[0055] The mathematical model for the beamforming signal switching failure probability is expressed as: .

[0056] in, Indicates the probability of beamforming signal switching failure. Indicates the probability of switching activation. For intermediate parameters, , This indicates the probability of interruption when communication services are provided by either the source base station or the target base station.

[0057] ; when When the probability of interruption is when the communication service is provided by the source base station, it is expressed as: ; Where t represents the current time, and T represents the switching period of the beamforming signal, which is also the trigger time (Time To Trigger, TTT). This means that for each discrete moment in the time window from tT to t, a switch will only be triggered if the condition is met continuously within the TTT time period. Represents probability operators; This represents the Q-function, also known as the right-tail function of the standard normal distribution. and These respectively represent the source base station and the target base station at Signal reception power at any given time This indicates the preset hysteresis threshold. and These respectively represent the source base station and the target base station at Path loss at time step, This represents the path loss of the source base station at time t. and These respectively represent the source base station and the target base station at Beam gain in the direction of the beam at any given time. This represents the beam gain of the source base station in the beam direction at time t. This indicates the preset communication interruption threshold. This represents the standard deviation of the shadowed fading modeled as a zero-mean Gaussian random variable. This represents the standard deviation of the shadowing fading between the source base station and the train, modeled as a zero-mean Gaussian random variable.

[0058] Similarly, when the communication service is provided by the target base station, the corresponding interruption probability can be obtained. .

[0059] In an exemplary embodiment, solving the beamdirection optimization problem, with the objective of minimizing the handover failure probability and beamforming adjustment overhead, specifically includes: solving the objective function of the beamdirection optimization problem, wherein the objective function is expressed as: .

[0060] in, This represents the beam direction optimization problem, where T represents the switching period of the beamforming signal. Indicates the probability of beamforming signal switching failure. and These represent the time intervals of the source base station and the target base station, respectively. Beam direction angle; Indicates the number of beam direction adjustments; The unit beam adjustment overhead factor; and These are all weighting factors used to balance the weighting relationship between switching reliability and overhead control; The spacing between adjacent base stations deployed at equal intervals along a railway track. This represents the coverage radius of the base station in omnidirectional mode. and These represent the Euclidean distances between the train and the source base station, and between the train and the target base station, respectively. This represents a constraint condition that limits the range of values ​​for the beam direction angle, allowing it to be dynamically adjusted within the feasible beam range where the train passes through the overlapping coverage area of ​​the source and target base stations.

[0061] In one exemplary embodiment, this application obtains the optimal beam direction combination under given constraints by solving a beam direction optimization problem. This aims to jointly minimize the probability of handover failure and beam adjustment overhead.

[0062] Specifically, step 104 includes steps 201-202.

[0063] Step 201: Assign the train the arrival directions from the source and target base stations. and and path loss between the train and the source and target base stations. and As a state space, the beam directions of the source base station and the target base station are... and As the action space, the negative objective function is used as the reward function, which is... The objective function is the objective function for solving the beam direction optimization problem.

[0064] Step 202: Based on the state space, the action space, and the reward function, the objective function of the beamdirection optimization problem is solved using the Deep Deterministic Policy Gradient (DDPG) algorithm to obtain the optimal beamdirection of the source and target base stations in each time slot during the switching period of the beamforming signal. More specifically, the Deep Deterministic Policy Gradient algorithm is used to execute the action space... Select Action Through the reward function Calculate reward value and with the state space Through interactive learning and iterative optimization of the beam direction until the reward value is maximized, the objective function of the beam direction optimization problem is obtained after minimization. and the optimal beam direction of the source base station and the target base station in each time slot. .

[0065] Based on the above optimization model, this invention solves the beam direction optimization problem. This yields the optimal beam direction combination under given constraints, thereby minimizing the optimization objective function that jointly considers the handover failure probability and beam adjustment overhead. (Comprehensive optimization objective function) Represented as: .

[0066] Under the premise of satisfying the beam pointing constraint, the optimal beam pointing angles of the source base station and the target base station can be obtained by minimizing the above objective function, which are expressed as follows: .

[0067] in, and These represent the optimal beam direction angles of the source base station and the target base station at time t, respectively.

[0068] As can be seen from the above technical solution of this invention, this invention addresses the signal attenuation and frequent handover failures caused by high mobility and weak cell edge coverage in high-speed railway communication by proposing a handover method for high-speed railway 5G systems based on dynamic beamforming. This method establishes a beamforming-enhanced 5G-R handover model, quantifies the relationship between beam direction and communication performance, and introduces a beam direction optimization mechanism during the handover decision process. This achieves coordinated adjustment and dynamic alignment of the source and target base station beams, thereby improving link signal gain and handover success rate. Based on the constructed optimization model, this invention achieves joint optimization of beam pointing and system performance by minimizing the handover failure probability and beam adjustment overhead. Verification shows that this method can significantly reduce beam tracking overhead while ensuring communication continuity, achieving high-reliability, low-latency handover control in high-speed railway scenarios. The dynamic beamforming method proposed in this invention has a clear structure, low computational complexity, and strong adaptability. It can effectively cope with rapid channel changes and frequent handover requirements in high-speed railway scenarios, possessing good real-time performance and engineering feasibility, providing key technical support for high-reliability communication in 5G-R systems.

[0069] In one exemplary embodiment, the technical effects of the present invention will be further illustrated below in conjunction with simulation experiments.

[0070] Simulation software: Python is used.

[0071] Simulation Scenario: Simulates the handover process of a train traveling at 350 km / h from the source base station to the target base station. The distance between the two base stations is 3 km, and the vertical distance between the train and the track is 80 m. Base station and train antenna parameters: base station antenna height 24.5 m, train antenna height 4.5 m, number of base station antennas 16, antenna spacing 0.075 m, coverage radius of 2 km when the base station uses omnidirectional beamforming. Communication and channel parameters: communication center frequency 2.1 GHz, noise power -90 dB, shadow fading standard deviation 4 dB, average surrounding building height 5 m. Handover and simulation settings: handover hysteresis value 2 dB, hysteresis threshold 80 ms, signal threshold -72 dBm, simulation time slot length 40 ms, base station transmit power 35 dBm.

[0072] The simulation content and results analysis are as follows.

[0073] Simulation 1: Under the above simulation scenario and conditions, the dynamic beamforming optimization method proposed in this invention is simulated, and the results are as follows: Figure 3 As shown. Figure 3 The dynamic changes in beam direction during the switching process are shown in the figure.

[0074] Depend on Figure 3 As can be seen, the arrival directions of the source and target base stations change rapidly at both ends of the coverage area, while the change becomes gradual in the middle region. This is due to the different rates of angle change caused by the difference in distance between the train and the base station. The method of this invention can adjust the beam direction in real time according to the change in arrival direction, ensuring stability and adaptability in high-speed moving environments. Meanwhile, the beam direction is not perfectly aligned with the arrival direction in some areas because this invention introduces a beam adjustment penalty mechanism. When the train approaches the base station and the signal quality is good, this mechanism allows for a moderate decrease in alignment accuracy to reduce frequent adjustments, thereby effectively reducing beam adjustment overhead while ensuring communication quality.

[0075] Simulation 2: Under the above simulation scenarios and conditions, the switching performance of the dynamic beamforming optimization method proposed in this invention was simulated, and the results are as follows: Figure 4 As shown. Figure 4 The dynamic changes in the handover failure probability during the handover process are shown in the figure.

[0076] Depend on Figure 4As can be seen, the handover failure probability of the beamforming-free scheme remains at a high level in most locations, only decreasing when the train approaches the edge of the target base station, resulting in low handover reliability. In contrast, the dynamic beamforming method of this invention can adjust the beam direction in real time according to the train's position, rapidly reducing the handover failure probability to near zero in the range of approximately 1300 m to 2200 m, significantly improving the handover success rate. Furthermore, compared to the ideal precise beam alignment scheme, the performance curves of the dynamic beamforming method almost completely overlap, indicating that this method can achieve near-optimal beam alignment results without prior acquisition of precise angle of arrival information. The overall results demonstrate that the dynamic beamforming method of this invention exhibits good adaptability and robustness in high-speed moving scenarios, significantly reducing the handover failure probability while ensuring signal quality, effectively improving the system's communication stability and handover reliability.

[0077] Based on the same inventive concept, this application also provides a method for implementing the above-mentioned high-speed rail 5G system handover method based on dynamic beamforming. The solution provided by this device is similar to the solution described in the above method. Therefore, the specific limitations of one or more embodiments of the high-speed rail 5G system handover device based on dynamic beamforming provided below can be found in the limitations of the high-speed rail 5G system handover method based on dynamic beamforming described above, and will not be repeated here.

[0078] like Figure 5 As shown, a high-speed rail 5G system switching device based on dynamic beamforming includes: The beamforming signal model construction module is used to construct a beamforming signal model based on the high-speed railway communication and handover scenario model of the high-speed rail 5G system. The beamforming signal model is used to determine the signal reception power received by the train from the source base station and the target base station during the train's movement. The high-speed rail 5G system supports multiple input multiple output and beamforming functions.

[0079] The beamforming signal switching model construction module is used to construct a beamforming signal switching model based on the beamforming signal model. The beamforming signal switching model is used to determine and trigger a switching activation event when the signal reception power of the target base station received by the train is continuously higher than the signal reception power of the source base station and the difference exceeds a preset hysteresis threshold during the train's movement.

[0080] A beam orientation optimization problem construction module is used to construct a beam orientation optimization problem after a switching activation event is triggered; the objective of the beam orientation optimization problem is to minimize the beamforming signal switching failure probability and beamforming adjustment overhead.

[0081] The solution module is used to solve the beam orientation optimization problem using deep reinforcement learning, and obtain the optimal beam orientation of the source base station and the target base station in each time slot during the switching period of the beamforming signal.

[0082] In one exemplary embodiment, a computer device is provided, which may be a server or a terminal, and its internal structure diagram may be as follows. Figure 6 As shown, the computer device includes a processor, memory, input / output (I / O) interfaces, and a communication interface. The processor, memory, and I / O interfaces are connected via a system bus, and the communication interface is also connected to the system bus via the I / O interfaces. The processor provides computational and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system, computer programs, and a database. The internal memory provides the environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The database stores handover data for the high-speed rail 5G system based on dynamic beamforming. The I / O interfaces are used for information exchange between the processor and external devices. The communication interface is used for communication with external terminals via a network connection. When the computer program is executed by the processor, it implements the high-speed rail 5G system handover method based on dynamic beamforming.

[0083] Those skilled in the art will understand that Figure 6 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0084] In one exemplary embodiment, a computer device is also provided, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps in the above-described method embodiments.

[0085] In one exemplary embodiment, a computer-readable storage medium is provided storing a computer program that, when executed by a processor, implements the steps in the above-described method embodiments.

[0086] In one exemplary embodiment, a computer program product is provided, including a computer program that, when executed by a processor, implements the steps in the above-described method embodiments.

[0087] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of the relevant data must comply with relevant regulations.

[0088] Those skilled in the art will understand that all or part of the processes in the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium. When executed, the computer program can include the processes of the embodiments described above. Any references to memory, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM).

[0089] The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, etc., and are not limited to these.

[0090] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0091] This document uses specific examples to illustrate the principles and implementation methods of this application. The descriptions of the above embodiments are only for the purpose of helping to understand the methods and core ideas of this application. Furthermore, those skilled in the art will recognize that, based on the ideas of this application, there will be changes in the specific implementation methods and application scope. Therefore, the content of this specification should not be construed as a limitation of this application.

Claims

1. A method for handover of high-speed-rail 5G system based on dynamic beamforming, characterized in that, The high-speed rail 5G system handover method based on dynamic beamforming comprises: According to the high-speed rail 5G system high-speed railway communication and handover scene model, a beamforming signal model is constructed; the beamforming signal model is used to determine the signal receiving power of the train in the moving process from the source base station and the target base station respectively; the high-speed rail 5G system supports multiple input multiple output and beamforming function; According to the beamforming signal model, a beamforming signal handover model is constructed; the beamforming signal handover model is used to determine the trigger handover activation event when the received signal receiving power of the target base station is continuously higher than that of the source base station, and the difference exceeds the preset hysteresis threshold during the train moving process; After triggering the handover activation event, a beam direction optimization problem is constructed; the target of the beam direction optimization problem is to minimize the beamforming signal handover failure probability and the beamforming adjustment overhead; The beam direction optimization problem is solved by using deep reinforcement learning to obtain the optimal beam direction of the source base station and the target base station in each time slot within the handover period of the beamforming signal. 2.The method of claim 1, wherein, The beamforming signal handover failure probability is expressed as: ； wherein denotes the beamforming signal switching failure probability, denotes the switching activation probability, is an intermediate parameter, , denotes the interruption probability when the communication service is provided by the source base station or the target base station; ； When is the outage probability when the communication service is provided by the source base station. ； Where t represents the current time, and T represents the switching period of the beamforming signal. Let represent each discrete moment in the time window from tT to t. Represents probability operators; Describes the right-tail function of the standard normal distribution. and These respectively represent the source base station and the target base station at Signal reception power at any given time This indicates the preset hysteresis threshold. and These respectively represent the source base station and the target base station at Path loss at time step, This represents the path loss of the source base station at time t. and These respectively represent the source base station and the target base station at Beam gain in the direction of the beam at any given time. This represents the beam gain of the source base station in the beam direction at time t. This indicates the preset communication interruption threshold. This represents the standard deviation of the shadowed fading modeled as a zero-mean Gaussian random variable. This represents the standard deviation of the shadowing fading between the source base station and the train, modeled as a zero-mean Gaussian random variable. 3.The high-speed rail 5G system handover method based on dynamic beamforming according to claim 1, wherein, Solving the beam direction optimization problem specifically includes solving the objective function of the beam direction optimization problem, which is expressed as: ； wherein, denotes the beam direction optimization problem, T denotes the switching period of the beamforming signal, denotes the beamforming signal switching failure probability, denotes the beam direction angle of the source base station at time denotes the beam direction angle of the target base station at time ; denotes the number of beam direction adjustments; is the unit beam adjustment overhead coefficient; and are weighting factors, is the distance between adjacent base stations among the base stations deployed along the railway track at equal intervals, is the coverage radius of the base station in the omnidirectional mode; and denote the Euclidean distance between the train and the source base station, and the train and the target base station, respectively. 4.The method of claim 1, wherein, The beam direction optimization problem is solved by using deep reinforcement learning to obtain the optimal beam direction of the source base station and the target base station in each time slot within the handover period of the beamforming signal, specifically including: The arrival direction of the train, the source base station and the target base station, and the path loss between the train, the source base station and the target base station are taken as the state space, the beam direction of the source base station and the target base station is taken as the action space, and the negative target function is taken as the reward function; the target function is the objective function of the beam direction optimization problem; According to the state space, the action space and the reward function, the deep deterministic policy gradient algorithm is used to solve the objective function of the beam direction optimization problem to obtain the optimal beam direction of the source base station and the target base station in each time slot within the handover period of the beamforming signal. 5.The method of claim 1, wherein, The signal receiving power of the source base station is expressed as: ； The signal receiving power of the target base station is expressed as: ； wherein, and respectively represent the signal reception power of the source base station and the target base station, is the transmission power of the base station, and the base station is a base station deployed at equal intervals along a railway track or the like, and respectively represent the path loss of the source base station and the target base station, and respectively represent the beam gain of the source base station and the target base station in the current beam direction, and respectively represent the shadow fading of the source base station and the target base station. 6.The method of claim 5, wherein, The path loss of the source base station and the target base station is expressed as: ； wherein, denotes the center frequency of the source base station and the target base station, denotes the average height of the building, and denote the Euclidean distance between the train and the source base station, and the train and the target base station, respectively. 7.The high-speed rail 5G system handover method based on dynamic beamforming according to claim 5, wherein, The beam gain of the source base station and the target base station under the current beam direction is expressed as: ； ； wherein, and and and and denotes the direction of arrival between the train and the source base station, denotes the direction of arrival between the train and the target base station, denotes the beam direction angle of the source base station, denotes the beam direction angle of the target base station. 8.A device for handover of high-speed rail 5G system based on dynamic beamforming, characterized in that, The high-speed rail 5G system handover device based on dynamic beamforming comprises: A beamforming signal model construction module is configured to construct a beamforming signal model according to the high-speed rail 5G system high-speed railway communication and handover scene model; the beamforming signal model is used to determine the signal receiving power of the train in the moving process from the source base station and the target base station respectively; the high-speed rail 5G system supports multiple input multiple output and beamforming function; The beamforming signal switching model construction module is configured to construct a beamforming signal switching model according to the beamforming signal model; the beamforming signal switching model is configured to determine a switching activation event when the signal receiving power of the target base station is continuously higher than the signal receiving power of the source base station, and the difference exceeds a preset hysteresis threshold during the movement of the train. The beam direction optimization problem construction module is configured to construct a beam direction optimization problem after the switching activation event is triggered; the beam direction optimization problem aims to minimize the beamforming adjustment overhead and the beamforming signal switching failure probability. The solving module is configured to solve the beam direction optimization problem by using deep reinforcement learning to obtain the optimal beam direction of the source base station and the target base station in each time slot within the switching period of the beamforming signal.

9. A computer device comprising: A memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that the processor executes the computer program to implement the high-speed rail 5G system switching method based on dynamic beamforming according to any one of claims 1-7.

10. A computer-readable storage medium having stored thereon a computer program, characterized in that, The computer program is executed by the processor to implement the high-speed rail 5G system switching method based on dynamic beamforming according to any one of claims 1-7.

Images