Single-step reconfiguration method of six-dimensional movable antenna in internet of vehicles

Abstract

The application relates to a single-step reconfiguration method of a six-dimensional movable antenna in a vehicle Internet, which comprises the following steps: constructing a system model and a link budget; constructing an optimization problem containing a single-step movement constraint based on a predicted cumulative vehicle distribution; generating a longitude and latitude grid discrete position and deducing a movement and time cost lower bound of antenna reconfiguration based on graph theory; utilizing 6DMA direction sparsity, fusing offline environment priori and online historical feedback, constructing a scoring function and selecting an optimal configuration by using a greedy strategy; the application fundamentally eliminates the service interruption risk by limiting the antenna adjustment in a first-order neighborhood, solves the service interruption problem of the 6DMA in a high-mobility vehicle Internet scene caused by the difficulty in obtaining instantaneous channel state information and the large mechanical reconfiguration delay, breaks away from the dependence on instantaneous CSI, significantly reduces mechanical overhead and time delay, improves the system spectrum efficiency, and guarantees the feasibility and robustness in a high-dynamic environment.

Description

Technical Field

[0001] This invention relates to the field of vehicle networking technology, and in particular to a single-step reconfiguration method for a six-dimensional movable antenna in vehicle networking. Background Technology

[0002] With the evolution of sixth-generation mobile communication technology and its deep integration with intelligent transportation systems, vehicle networks are undergoing a critical transformation from information interaction to collaborative perception and control. This process not only requires the network to provide the ultimate connection reliability, but also poses severe challenges to data transmission capacity, latency, and adaptability to dynamic environments. Traditional multi-antenna technology increases array gain by increasing the number of static antennas, but due to its hardware cost, serial number, and inherent rigid spatial layout, it is difficult to meet the sudden demands of rapidly changing users and highly explosive business needs in vehicle networks. Therefore, under limited antenna scale, deeply exploring and utilizing spatial degrees of freedom has become an important issue for improving the spectrum efficiency and reliability of vehicle networks.

[0003] In recent years, mobile antenna technology (MA) has received widespread attention as a new paradigm. Its basic principle is to actively reconstruct the wireless channel by physically adjusting the position or orientation of the antenna, thereby enhancing the target signal and suppressing interference. Early research focused on fluid antenna systems and two-dimensional mobile antenna (2DMA) technology, which tracks small-scale channel variations (CSI) by moving a single antenna element in a limited one-dimensional or two-dimensional space. Although 2DMA technology allows the antenna to move in a two-dimensional plane, giving it a certain degree of spatial freedom, this approach requires extremely high movement frequencies to cope with rapidly changing instantaneous channel conditions. In scenarios where vehicles move at high speeds and the channel changes rapidly, its mechanical wear, control complexity, and power consumption problems are particularly prominent, limiting its practicality.

[0004] To address the aforementioned issues, a novel six-dimensional movable antenna (6DMA) architecture is proposed. 6DMA divides the antenna array unit into multiple independent surfaces, each integrating a small rectangular array, which is connected to the central processing unit via independently controllable telescopic and rotating masts. Flexible cables within the masts are used for power supply and signal transmission. This design allows each antenna surface to not only translate in three-dimensional space but also rotate in three-dimensional space, thus providing high spatial flexibility. By optimizing the position and orientation of these surfaces, 6DMA can effectively focus on dynamically changing user groups. When user distribution exhibits spatial clustering, its performance far surpasses that of traditional fixed antennas and low-dimensional movable antenna technologies.

[0005] Despite the great potential of 6DMA, its actual deployment in high-speed vehicle network scenarios still faces a series of severe challenges: (1) The high-speed movement of vehicles leads to rapid changes in the optimal service antenna configuration, which requires frequent adjustments. However, the response speed of mechanical systems is limited, and multi-step movement across large spaces will introduce significant time delays, which can easily cause communication service interruptions and cannot meet the high reliability and low latency requirements of vehicle network services; (2) The mobility of antennas makes the channel a complex function of position and attitude. Traditional channel estimation methods based on fixed pilots have huge overhead and are difficult to complete in real time. Existing optimization schemes mostly rely on statistical channel information or ideal Monte Carlo sampling, which lacks effective adaptability to instantaneous channel dynamics; (3) Although existing technologies have made preliminary explorations into the discretization modeling, directional sparsity utilization and hierarchical optimization strategies of 6DMA, such as using spherical discretization to generate candidate positions and heuristic search based on channel statistical characteristics, these methods have not made accurate quantitative modeling and theoretical analysis of the mobile energy consumption and reconstruction time cost in the antenna reconstruction process, nor have they proposed a lightweight reconstruction mechanism that can avoid service interruptions. Summary of the Invention

[0006] In view of this, the present invention provides a single-step reconfiguration method for a six-dimensional movable antenna in the Internet of Vehicles (IoV), which solves the problems of frequent antenna reconfiguration, large mechanical overhead and service interruption in existing IoV technologies, and achieves high spectral efficiency, low mechanical overhead and 6DMA real-time reconfiguration that does not depend on instantaneous channel information.

[0007] To achieve the above objectives, the present invention provides a single-step reconfiguration method for a six-dimensional movable antenna in a vehicle-to-everything (V2X) network, comprising the following steps: S1. A system for constructing a six-dimensional movable antenna 6DMA model and a vehicle-to-everything (V2X) communication scenario, including a city scenario with K vehicles and a ground base station BS. The ground base station BS is configured with a six-dimensional movable antenna system consisting of U movable antenna surfaces. The K vehicles are configured with omnidirectional antennas at fixed positions. The position vector, rotation direction, antenna radiation pattern, and antenna motion constraints of the omnidirectional antennas are defined. An uplink communication and rate calculation model from the vehicles to the ground base station BS is established. S2. Construct a periodic single-step reconfiguration strategy based on motion prediction. Based on the predicted cumulative vehicle spatial distribution, construct a 6DMA reconfiguration optimization problem with the goal of maximizing the average sum rate within the prediction window. The position change of the six-dimensional movable antenna follows single-step movement constraints, minimum distance constraints, mutual reflection constraints, and occlusion constraints. The single-step movement constraints restrict each antenna element to only move to the first-order physical neighborhood of its current position during reconfiguration. S3. Construct a set of feasible discrete antenna locations and corresponding spherical discrete location rotation sets that satisfy the minimum distance constraint in the space around the base station, based on the latitude and longitude grid method. Generate feasible discrete six-dimensional movable antenna configurations. Establish a model of movement cost and time cost describing the antenna configuration switching process based on graph theory. Calculate the lower bound of movement cost and time cost using the breadth-first search method BFS and the improved Hungarian algorithm. S4. Obtain the current antenna configuration from the set of discrete spherical positions. Based on the directional sparsity of the 6DMA channel, construct the antenna position rotation-effective gain space mapping offline and the historical rate obtained online. Construct a comprehensive scoring function for each candidate antenna position. Use a greedy strategy to select the optimal position and rotation state in the neighborhood for each antenna element.

[0008] Preferably, the six-dimensional movable antenna 6DMA model includes U movable antenna surfaces connected to the central processing unit via independent telescopic robotic arms; The definition of the omnidirectional antenna's position vector, rotation direction, antenna radiation pattern, and antenna motion constraints specifically includes: In the 6DMA model of the six-dimensional movable antenna, the time dimension adopts a discrete time slot mechanism, and the time interval is... The vehicle position remains unchanged within a single time slot, and is only updated at the time slot boundaries based on the sampling rate of the truncated Gaussian distribution; Each of the movable antenna surfaces has the ability to independently adjust its three-dimensional spatial position and three-dimensional rotation attitude. The feasible configuration state space of the movable antenna surfaces is discretized and defined as a set. ,in , These represent the total number of discrete positions and rotation states, respectively. Indicates the first The center coordinates of the candidate locations Represents Euler angles, in any rotational state. The corresponding unit outward normal vector The expression is: Using a binary decision matrix Describe the deployment status of the system, which must meet the following constraints: Capacity constraint: The system activates a total of U antenna surfaces ( ), which needs to satisfy capacity constraints: Mutual non-obstruction constraint: The normal vectors of any two active antenna surfaces must not point to each other's positions; Central processing unit visibility constraint: The radiation direction of all antenna surfaces must be away from the central processing unit; Minimum spatial spacing constraint: The Euclidean distance between any two active antenna surfaces must be greater than a preset safety threshold; Single occupancy constraint: At most one antenna surface can be deployed at each discrete location.

[0009] Preferably, the uplink communication adopts a hybrid near-field-far-field channel model: the propagation between antenna surfaces is regarded as far-field, and the propagation between array elements inside the antenna surface retains near-field phase characteristics; Establishing uplink communication and rate calculation specifically includes: S101. Establish a local coordinate system to describe the direction of arrival; S102. Define antenna gain based on 3GPP radiation direction chart, establish line-of-sight and non-line-of-sight probability path loss based on 3GPP UMi standard and log-normal shadowing fading respectively, and then establish channel model through multipath Rayleigh small-scale fading. S103, Vehicles The channel vector is determined by its position and 6DMA configuration, and is calculated based on the interference-to-noise ratio of each vehicle signal under the matched filter receiver. The expression is: in, Indicates vehicle The signal interference plus noise ratio, This indicates the allocated bandwidth.

[0010] Preferably, the periodic single-step reconfiguration strategy discretizes time into time units of length 1. The antenna configuration is updated every N time slots. vehicle In the time slot The position is recorded as ,vehicle The motion follows a linear kinematic model, and the vehicle's position in future time slots is recursively predicted based on the vehicle's instantaneous average velocity, expressed as: in, This indicates the predicted position of the vehicle at the next moment. This indicates the predicted position of the vehicle at the current moment. Represents the direction vector. This represents the instantaneous average velocity, with the initial condition set to the observed value at the current location, i.e. ; Based on the predicted vehicle positions in future time slots, the average sum of speeds within a vehicle cycle is calculated using the following expression: in, Indicates average and rate. Indicates vehicle The signal interference plus noise ratio, Indicates time vehicle The predicted location, This indicates the antenna configuration.

[0011] Preferably, the single-step movement constraint specifically means that the set of candidate target positions for the u-th antenna in the current period is restricted to the first-order physical neighborhood of the position in the previous period, that is, each antenna element moves at most one step or remains stationary during each reconfiguration, as expressed by: in, Represents the position in the discrete grid The set of physical neighbors.

[0012] Preferably, the discrete location generation method based on latitude and longitude grids in step S3 includes: Given the number of meridians, determine the position of the first circle of latitude that satisfies the minimum distance; Calculate the maximum allowable interval between weft coils to keep all neighboring points from meeting the minimum distance constraint, and calculate the number of intermediate weft coils that can be accommodated based on the total available height, which is obtained by summing the number of positions on each coil. A feasible rotation state is obtained, including the radial vector pointing to the center of the sphere and the normal vectors of the triangular faces formed by adjacent positions.

[0013] Preferably, calculating the lower bound of the mobility cost and time cost includes: S301. Calculate the shortest path length from all initial positions to the target position using a breadth-first search algorithm, and form a distance matrix. S302. Construct a cost function that includes a quadratic penalty term. The expression is: in, This represents the penalty coefficient that prioritizes minimizing the total number of moves. This represents the distance between positions u and v; S303. Solving for the optimal matching using an improved Hungarian algorithm. The expression for the lower bound of the movement cost is obtained as follows: in, This represents the energy consumption per unit distance when a single antenna element is moved to an adjacent discrete location. This represents the matching matrix; a value of 1 indicates that antenna movement has occurred. This represents the distance the antenna moves from position u to position v; The expression for the lower bound of time cost is: in, Indicates the maximum number of steps a single antenna can move. It represents the unit time it takes for a single antenna element to move to an adjacent discrete position.

[0014] Preferably, step S4 specifically includes the following steps: S401. Offline construction of spatial grid - static mapping library for preferred antenna configuration: The ground service area is gridded, and the back position in each grid is removed by geometric hemisphere clipping. A two-level hierarchical sampling strategy is adopted to reduce the search space. The candidate positions and their rotation states are traversed by rotation fine-tuning. The theoretical average spectral efficiency is calculated. The top-ranked configurations are retained for each grid to form an offline preferred candidate set. S402, Online Scoring and Decision-Making: The set of active grids covered by the current period's forecast. For each candidate position, a comprehensive score is constructed, which is then combined with the historical cumulative rate score and the stability reward for positions activated in the previous cycle to obtain the overall score. ; Within the first-order neighborhood of each antenna element, the candidate position with the highest comprehensive score is selected. If neighborhood overlap leads to a conflict, a greedy algorithm is used to prioritize the allocation of antenna-position pairs with higher scores, forming a set of conflict-free activation positions. ; The set of conflict-free activation locations is statistically analyzed. The frequency of each rotation state in the offline static mapping library corresponding to the preferred candidate set of the grid is selected for each active position. If there is no match, the radial direction pointing to the base station is selected as the default rotation.

[0015] Compared with the prior art, the beneficial effects of the present invention are: This invention introduces 6DMA technology into a highly dynamic vehicle-to-everything (V2X) scenario for the first time, and establishes a channel model based on multipath characteristics. It proposes a deterministic discrete position and rotation state construction scheme based on latitude and longitude grids, which naturally possesses clear neighbor topology relationships. Based on graph theory, this invention uses breadth-first search (BFS) and an improved Hungarian algorithm to define and theoretically derive the lower bounds of the mobility cost and time cost required for antenna configuration switching. Through periodic optimization based on predicted cumulative distribution (rather than chasing instantaneous fluctuations), the system can capture macroscopically stable user distribution trends, significantly reducing unnecessary mechanical wear and energy consumption, reducing reconfiguration latency, and achieving the best balance between system performance and physical overhead. This invention introduces a single-step reconfiguration mechanism, which strictly limits the adjustment range of each antenna element to the first-order neighborhood of its physical location. Regardless of changes in vehicle distribution, each antenna only needs to move one step (or remain stationary) in each decision cycle. This solves the problem of communication service interruption caused by the arbitrary movement of antennas in the global range and the need for multi-step mechanical operations for cross-regional adjustments in existing technologies. It meets the stringent requirements of high reliability and low latency for vehicle-to-everything (V2X) networks. This invention addresses the problem of drastic changes in user distribution caused by high-speed vehicle movement by proposing an optimization perspective based on predicting cumulative distribution rather than instantaneous distribution. By extending the prediction time window, the cumulative user distribution can capture relatively stable distribution trends in the environment, thereby significantly reducing the antenna reconfiguration frequency and the physical overhead of each reconfiguration, and enhancing system robustness. This invention abandons the traditional instantaneous CSI-based optimization path and adopts offline prior knowledge to pre-construct a static mapping library of "spatial grid-optimized configuration" using the relatively static characteristics of environmental scatterers. It also uses online historical feedback to dynamically correct prior errors using service plastic data from historical periods. The two are integrated to form an instantaneous CSI-free optimization framework. It makes full use of the directional sparsity of 6DMA to establish an offline mapping library and associate the joint position-rotation state with the effective coverage area. This avoids a lot of pilot overhead and transforms the complex real-time channel estimation problem into an efficient scoring problem. It solves the problem of CSI being difficult to obtain and changing extremely rapidly in vehicle-to-everything (V2X) networks, and significantly reduces computational complexity and implementation cost. Attached Figure Description

[0016] Figure 1 This is a schematic diagram of the scenario of the present invention; Figure 2 This is a schematic diagram illustrating the generation of discrete conventional positions according to the present invention; Figure 3 This is a schematic diagram illustrating the generation of the discrete top position in this invention; Figure 4 This is a schematic diagram illustrating the generation of the discrete first latitude coil position according to the present invention; Figure 5 This is a heatmap showing the achievable rate of different grid regions corresponding to the optimal single antenna position in this invention. Figure 6 This is a comparison chart of the average achievable rates of different methods with different transmission powers according to the present invention; Figure 7 This is a comparison chart of the average reachability of different methods with different numbers of users according to the present invention; Figure 8 This is a comparison chart of the average achievable rates for different transmission powers and different prediction intervals according to the present invention. Figure 9 This is a comparison chart of the average reachability rates for different numbers of users and different prediction intervals according to the present invention. Figure 10A comparison of the migration costs required for configuring different methods at different prediction intervals; Figure 11 A comparison of the time cost required to configure different methods for different prediction intervals. Detailed Implementation

[0017] To further illustrate the technical means and effects of the present invention in achieving its intended purpose, the following detailed description of the specific implementation methods, structures, features, and effects of the present invention, in conjunction with the accompanying drawings and preferred embodiments, is provided below.

[0018] This embodiment proposes a single-step reconfiguration method for a six-dimensional movable antenna in vehicle-to-everything (V2X) networks, specifically including the following steps: S1. A system for constructing a six-dimensional movable antenna 6DMA model and a vehicle-to-everything (V2X) communication scenario, including a city scenario with K vehicles and a ground base station BS. The ground base station BS is configured with a six-dimensional movable antenna system consisting of U movable antenna surfaces. The K vehicles are configured with omnidirectional antennas at fixed positions. The position vector, rotation direction, antenna radiation pattern, and antenna motion constraints of the omnidirectional antennas are defined. An uplink communication and rate calculation model from the vehicles to the ground base station BS is established. In the IoV system model, the time dimension is modeled using a discrete time-slot mechanism, with a time-slot interval of [missing information]. Assuming the vehicle's motion possesses quasi-static characteristics, i.e., its position... It remains unchanged within a single time slot, and is only updated at the time slot boundaries according to the sampling rate of the truncated Gaussian distribution; The six-dimensional movable antenna 6DMA model includes U movable antenna surfaces connected to a central processing unit via independent telescopic robotic arms. Each surface has the ability to independently adjust its three-dimensional spatial position and three-dimensional rotational attitude. Due to limitations in mechanical precision and control complexity, the state space of the antenna surfaces is restricted to a discrete domain, defined as a set. ,in , These represent the total number of discrete positions and rotation states, respectively. Indicates the first The center coordinates of the candidate locations Euler angles are represented, describing the angles from the reference pose (where the normal vector points to). The rotation process to transform to the current posture, any rotation state. The corresponding unit outward normal vector The expression is: To describe the deployment state of the system, a binary decision matrix is ​​introduced. ,like This indicates that the person is in office. A rotating state was activated at that location. The system activates a total of U antenna surfaces. The following constraints must be met: Capacity constraint: The system activates a total of U antenna surfaces ( ), which needs to satisfy capacity constraints: Furthermore, physically feasible deployment schemes must simultaneously satisfy the following geometric and hardware constraints: Mutual non-obstruction constraint: To prevent signal blocking or reflection interference between antenna surfaces, the normal vectors of any two active antenna surfaces must not point to each other. and (Right now and ),surface The normal vector must not point to the surface. The location is expressed as: Central Processing Unit (CPU) Visibility Constraint: The radiation direction of all antenna surfaces must be away from the central processing unit located at the origin to avoid obstructed line of sight. The expression is: Minimum spatial spacing constraint: The Euclidean distance between any two active antenna surfaces must be greater than a preset safety threshold. To prevent mechanical collisions, the expression is: Single Occupancy Constraint: At most one antenna surface can be deployed at each discrete location, expressed as: Considering the uplink from vehicle k to the base station, each 6DMA surface integrates an FPA consisting of Q array elements, and the total number of receiver array elements in the system is... Given the large aperture characteristics of 6DMA, we adopt a hybrid near-field-far-field channel model: that is, the propagation between surfaces is regarded as far-field, while the phase difference between array elements inside the surface retains the near-field spherical wave characteristics. The establishment of uplink communication and rate calculation specifically includes: S101. Establish a local coordinate system to describe the direction of arrival, and set... Let K be the coordinates of vehicle k, pointing towards the antenna surface. The unit line-of-sight vector is Local elevation angle With azimuth for: S102. Define antenna gain based on 3GPP radiation direction chart, establish line-of-sight and non-line-of-sight probability path loss based on 3GPP UMi standard and log-normal shadowing fading respectively, and then establish channel model through multipath Rayleigh small-scale fading. The antenna gain conforms to the 3GPP radiation pattern standard, defining the attenuation in the horizontal and vertical directions as follows: and The synthesized linear gain coefficient is: in, Indicates half-power beamwidth. Indicates the sidelobe suppression ratio. Indicates peak gain; The path loss model adopts the 3GPP UMi standard. For path distance... View distance (LoS) probability Given by a piecewise exponential function (parameter set to...) The expression is: The propagation state is determined by Monte Carlo sampling, and the path loss is calculated under the Loss of Sorrow (LoS) state. Only with distance and frequency Related, the expression is: If in a non-line-of-sight (NLoS) state, the standard deviation needs to be superimposed. Log-normal shadowing fading in dB The expression is: The corresponding large-scale channel gain is: At the microscopic level, consider the surface The first Each array element has a global coordinate of . ,in This represents the relative displacement within the array, and the near-field phase shift received by this array element is... ; Assume the channel consists of a multipath set Each path undergoes independent Rayleigh decay. The comprehensive channel coefficients of this array element are then modeled as follows: The effects of array gain, large-scale fading, near-field phase, and small-scale fading were accurately decoupled. To derive the aggregated channel vector, we first define a vector that satisfies... Specific activation unit surface The array response vector is: S103, Vehicles The channel vector is determined by its location and 6DMA configuration; By corresponding to the matrix All The response vectors of the activated elements (in order of position index) and rotation index By vertically assembling (sorted), a vehicle can be constructed. Total channel vector Therefore, the multi-user channel matrix is ​​represented as At this time, the channel matrix This can be viewed as a set of vehicle location distributions and a 6DMA configuration. The nonlinear mapping function is expressed as: Based on a matched filter receiver, the vehicle The expression for the received signal interference plus noise ratio (SINR) is: in, Index indicating interfering vehicles. The regularization factor introduced for numerical stability is represented by the signal interference plus noise ratio of each vehicle. The total uplink sum and rate of the system are expressed as follows: in, Indicates vehicle The signal interference plus noise ratio, This indicates the allocated bandwidth.

[0019] S2. Construct a periodic single-step reconfiguration strategy based on motion prediction. Based on the predicted cumulative vehicle spatial distribution, construct a 6DMA reconfiguration optimization problem with the goal of maximizing the average sum rate within the prediction window. The position change of the six-dimensional movable antenna follows single-step movement constraints, minimum distance constraints, mutual reflection constraints, and occlusion constraints. The single-step movement constraints restrict each antenna element to only move to the first-order physical neighborhood of its current position during reconfiguration. Due to the high mobility of vehicles in the Internet of Vehicles (IoV) scenario, the spatial distribution of users exhibits rapid time-varying characteristics. Although the 6DMA system can adapt to this change by flexibly adjusting the antenna configuration, real-time reconfiguration within each time slot is impractical due to the limitations of the mechanical servo system's response speed and computing resources. Therefore, this embodiment proposes a periodic single-step reconfiguration strategy based on motion prediction, where time is discretized into time slots with a length of [missing information]. ,vehicle In the time slot The position is recorded as Its motion follows a linear kinematic model: in, Indicates instantaneous velocity. It is a unit direction vector that satisfies ; To balance the system's adaptability and implementation complexity, antenna configuration Set to every Updated once per time slot, defining the first... The set of time slots covered by each reconfiguration cycle is The starting time is During this period, the antenna configuration remains unchanged, that is ; Utilizing the spatiotemporal correlation of traffic flow, this embodiment employs a method based on instantaneous average speed. The prediction mechanism is used to estimate future trajectories and predict locations. The recursive calculation is as follows: The initial conditions are set to the observation values ​​at the current location. Based on this predicted trajectory, the first The average sum rate over a period is defined as: While prediction-based reconfiguration can improve long-term average performance, allowing antennas to arbitrarily jump within the global discrete space can lead to significant mechanical displacement delays, especially in discrete grids where cross-region adjustments often require multiple mechanical operations, causing severe service interruptions. Therefore, this embodiment proposes a single-step reconfiguration mechanism: the set of candidate target locations for the u-th antenna in the current period is restricted to the first-order physical neighborhood of its location in the previous period. That is, each antenna element moves at most one step or remains stationary during each reconfiguration. This represents the set of antenna position indices activated in the previous cycle, where Let represent the spatial location index of the u-th antenna element in the global discrete location set. To resolve ambiguities in antenna correspondence and prevent infeasible long-distance interchanges, we distinguish these U antenna elements and track their individual trajectories. For the u-th antenna, its feasible search space in the current period is restricted to its closed neighborhood, expressed as: in, Represents the position in the discrete grid The set of physical neighbors ensures that each antenna can move at most one step, or remain stationary; By introducing specific decision variables for each antenna to ensure that it remains within its local neighborhood, an optimization model is constructed: in, This represents the configuration matrix of the u-th antenna. Constraint (c) ensures that each antenna selects exactly one position-rotation state, and constraint (d) strictly restricts the u-th antenna to be in its local neighborhood. Intra-region movement effectively avoids the risk of logical teleportation between different areas. Constraint (e) prevents collisions, ensuring that no two antennas occupy the same location even if their neighborhoods overlap. Finally, an aggregated configuration is adopted. To assess physical constraints and system capacity.

[0020] S3. Construct a set of feasible discrete antenna locations and corresponding spherical discrete location rotation sets that satisfy the minimum distance constraint in the space around the base station, based on the latitude and longitude grid method. Generate feasible discrete six-dimensional movable antenna configurations. Establish a model of movement cost and time cost describing the antenna configuration switching process based on graph theory. Calculate the lower bound of movement cost and time cost using the breadth-first search method BFS and the improved Hungarian algorithm. Although the problem in step S2 is formally closed, due to the complexity of the 6DMA channel, accurately obtaining the channel matrix is ​​crucial. It requires a large amount of pilot overhead. In addition, the objective function is highly non-convex and nonlinear, making direct numerical solution in millisecond-level vehicle networking scenarios infeasible. To address this challenge, in step S2 of this embodiment, the beam direction sparsity of 6DMA will be utilized to propose a fast heuristic algorithm that does not require real-time CSI. Discrete location generation methods based on latitude and longitude grids include: Given number of meridians The expression for determining the position of the first loop of latitude that satisfies the minimum distance is: Calculate the maximum allowable spacing between latitude coils to ensure all neighboring points meet the minimum distance constraint. Calculate the number of intermediate latitude coils that can be accommodated based on the total available height, expressed as: in, This indicates the remaining distributable length in the middle. This represents the floor function; the total available positions are obtained by summing the number of positions on each coil, expressed as: Obtain feasible rotation states, including the radial vector pointing towards the center of the sphere. And the normal vectors of the triangular faces formed by adjacent positions; Given the current set of antenna configurations and the configuration set for the next moment Since the antennas are indistinguishable, the configuration will be changed from... Transform to The process can be viewed as a labelless multi-agent path finding (MAPF) problem. Since the number of discrete locations is usually much larger than the number of antenna elements configured, the probability of mutual blocking during movement is low, so the specific movement path is relatively easy to solve. To define the movement cost and time cost of configuration transition, we assume that the unit energy consumption for moving a single antenna element to an adjacent discrete location is... Unit time is The rotation transformation of the antenna element can be completed simultaneously with the position movement, therefore time and energy costs are not separately considered; from the configuration arrive The lower bounds of the movement cost and time cost can be precisely solved using the breadth-first search (BFS) algorithm and the Hungarian algorithm. The calculation of the lower bounds of the movement cost and time cost includes: S301. Calculate the shortest path length from all initial positions to the target position using a breadth-first search algorithm, and form a distance matrix. Any antenna element from its initial position (belong Move to the target location (belong The shortest path can be found through the graph. The BFS algorithm on the given surface determines that the path length (number of steps) is defined as: For all initial and target positions Running BFS yields the distance matrix. ; S302. The total number of moves can be calculated using the Hungarian algorithm. Due to the high connectivity of the graph, there may be multiple matching schemes with the minimum total number of moves. To ensure that the time cost is minimized simultaneously, this embodiment makes a slight modification to the Hungarian algorithm, constructing a cost function that includes a quadratic penalty term. The expression is: in, This represents the penalty coefficient that prioritizes minimizing the total number of moves. This represents the distance the antenna moves from position u to position v; this is achieved by adding a quadratic term. A lexicographical order optimization mechanism is constructed: The primary objective is to minimize the total number of steps taken (energy consumption). The second objective is to minimize the variance of the number of movement steps, thereby reducing the maximum movement distance of any single antenna and minimizing the time cost. This embodiment rigorously proves the rationality of the cost function using the following propositions: Proposition 1 (Total Energy Consumption First): To ensure the algorithm strictly prioritizes the total number of moves Minimize the term rather than the penalty term, and the penalty coefficient must satisfy: in, Represents the diameter of graph G (the maximum distance between any two nodes); The proof is as follows: set up , For two different valid matching schemes, and These are the total number of steps and the penalty for scheme A (similar to scheme B); Assuming that option A has better energy efficiency, that is And since the number of steps is an integer, the minimum difference is ; To ensure that the improved Hungarian algorithm selects scheme A, the following must be satisfied: Consider the worst-case scenario: And the penalty difference is the largest (i.e. Take the theoretical maximum value , Then we have: The theoretical maximum penalty is the maximum distance all U antennas must move. Therefore ; Therefore, set This ensures that the penalty term weight is always less than the cost of a single step, thus maintaining optimal total energy consumption. Proposition 2 (Optimizing time cost through load balancing): For two schemes with the same total number of moves (i.e.) The cost function with a quadratic penalty term (n=2) ensures that a scheme with a more balanced step distribution (smaller maximum number of steps) has a lower total cost; prove: This property originates from the Majorization Theory. Let the movement step vectors of the two schemes be arranged in descending order as follows: and ; If scheme A has an uneven distribution (e.g., a larger maximum number of steps), but the total number of steps is equal, then it is called... Dominate , recorded as ; According to the Hardy-Littlewood-Pólya inequality, for strictly convex functions ,like ,but: In this cost function, the penalty term Its second derivative It is strictly convex; therefore, the solution with a more balanced step distribution (Solution B) has a lower penalty. The improved Hungarian algorithm will prioritize this scheme, thereby suppressing extreme values ​​and reducing the maximum number of single-antenna movement steps. Reduce time cost ; Define matching variables: The optimal match is obtained by solving the following linear assignment problem: S303. Solving for the optimal matching using an improved Hungarian algorithm. ; The corresponding distance matrix after matching is: in, This indicates element-wise multiplication. Indicates the distance of the selected matching pair; The total number of movement steps (lower bound of energy consumption), which is the sum of the number of movement steps of all antennas multiplied by the unit energy consumption, is expressed as follows: in, This represents the energy consumption per unit distance when a single antenna element is moved to an adjacent discrete location. This represents the matching matrix; a value of 1 indicates that antenna movement has occurred. This represents the distance the antenna moves from position u to position v; The maximum number of steps a single antenna can move is expressed as: The expression for the lower bound of time cost is: in, Indicates the maximum number of steps a single antenna can move. This represents the unit time it takes for a single antenna element to move to an adjacent discrete location. Assuming all antennas start moving synchronously, the overall time is determined by the longest path.

[0021] S4. Obtain the current antenna configuration from the set of discrete spherical positions. Based on the directional sparsity of the 6DMA channel, construct the antenna position rotation-effective gain space mapping offline and the historical rate obtained online. Construct a comprehensive scoring function for each candidate antenna position. Use a greedy strategy to select the optimal position and rotation state in the neighborhood for each antenna element. That is, use the data collected offline to determine the initial position of the antenna and dynamically adjust the antenna configuration based on the offline data and historical service data to adapt to the mobility characteristics of vehicle users. Specifically, the following steps are included: S401. Offline construction of spatial grid - static mapping library for preferred antenna configuration: The ground service area is gridded, and the back position in each grid is removed by geometric hemisphere clipping. A two-level hierarchical sampling strategy is adopted to reduce the search space. The candidate positions and their rotation states are traversed by rotation fine-tuning. The theoretical average spectral efficiency is calculated. The top-ranked configurations are retained for each grid to form an offline preferred candidate set. Given the directional sparsity of 6DMA systems in complex scattering environments, users in a specific physical space can typically only obtain high-gain service through a limited combination of "position-angle". Taking advantage of the relatively static nature of macroscopic scatterers such as buildings in urban environments, a static mapping library of "spatial grid - optimal antenna configuration" is pre-constructed as prior knowledge for online decision-making. Ground service area According to the side length Discretization of a uniform grid, the total number of grids is denoted as . For indexes of The grid, whose geometric center coordinates are set as: in, This represents the typical antenna height of the vehicle, and for quantizing the configuration gain, it is shown in each grid. Inner uniform sampling Typical user locations For global discrete configuration sets Each position-rotation tuple in The theoretical average spectral efficiency within this grid is calculated using the following expression: in, Indicates vehicle The signal-to-interference-plus-noise ratio (SIN) is calculated based on static large-scale channel parameters. Theoretically, a complete priori mapping table can be established by traversing all grids and all feasible configurations. However, due to discrete bits... With rotation state The sheer number of combinations makes exhaustively searching the entire space computationally too complex and impractical. This paper proposes a geometric pruning and hierarchical search strategy based on the geometric properties of 6DMA to dynamically constrain the search space. Geometric hemispherical clipping: Utilizing the LoS relationship between the base station (BS) and the target grid, invalid locations with obvious back-facing orientation are eliminated, restricting antenna placement to the hemisphere facing the grid and constructing a subset of effective locations. The expression is: in, Indicates the base station coordinates. Indicates position The reference normal vector at that location ensures that the angle between the antenna position and the grid center does not exceed [the specified value]. Two-level hierarchical sampling: based on Further reduce the size of the candidate pool: Coarse-grained screening: Using the K-means clustering algorithm, in Y representative anchor points are uniformly selected within the area, and their average velocity is calculated based on the default radial downward orientation to quickly assess coverage potential. The anchor points with the highest velocities are then selected. Seed positions.

[0022] Fine-grained expansion: with this Centered on a seed location, an extended search is performed within its first-order physical neighborhood (which still needs to satisfy the hemispherical constraint) to discover the local optimal location; Rotation fine-tuning: For the high-potential positions selected in the two-stage screening, traverse all feasible rotation states. Finally, the optimal position-rotation combination was determined; The three-level search strategy described above fully leverages the directional sparsity of 6DMA, reducing the search space by several orders of magnitude while ensuring the candidate set covers high-performance configurations. Finally, for each grid... Ranked first in spectral efficiency The configuration constitutes the offline preferred candidate set: This prior library accurately characterizes the directional response characteristics under stable scattering environments and macroscopic geometry, providing a high-quality and compact decision-making basis for subsequent online antenna scheduling; Although offline mapping libraries provide optimal solution references for static environments, actual performance may deviate from prior estimates due to instantaneous channel disturbances caused by vehicle movement and prediction errors. To improve the system's robustness in dynamic environments, this embodiment proposes an adaptive configuration scheme that integrates offline priors and online feedback. This scheme not only relies on the currently predicted vehicle distribution but also fully utilizes historical periodic measurement data. By accumulating and learning historical configuration performance, it continuously corrects the channel quality assessment of each candidate location, thereby guiding the... Antenna configuration for each cycle; No. The feasible search space for each antenna element is limited to its closed neighborhood. To determine the optimal configuration, we first consider all candidate locations belonging to the union of these neighborhoods (i.e., Construct a comprehensive scoring function This rating is an offline predicted rating. Historical cumulative measurement score With stability rewards Weighted sum: in, This represents the balance factor that adjusts the weights of offline priors and online historical feedback. S402, Online Scoring and Decision Making, aims to invoke an offline library based on the predicted vehicle distribution: the set of active grids covered by the current period's prediction. For each candidate position, a comprehensive score is constructed, which is then combined with the historical cumulative rate score and the stability reward for positions activated in the previous cycle to obtain the overall score. ; set up For the first The set of active grids covered by the predicted trajectory, where the grids The predicted demand density is The maximum density is According to candidate position In these grid preferred sets Evaluate the hit rate in the game: in, Represents a grid Predicted demand density, Represents a grid Maximum density, Represents the grid-optimized candidate set, Indicates the position in the grid preferred candidate set The number of different rotation states corresponding to the given values. Indicates an indicator function, Indicates the basic reward. This represents the multimodal reward coefficient, which tends to prioritize robust positions that perform well across multiple rotation states; Historical cumulative rate scoring: Utilizing historical data to correct prior biases, the system maintains a continuously growing historical record database. ,in Indicates the first The average total rate of the system measured over a period of time; For any candidate position Its score depends on the average performance across all historical activation cycles. The historical average contribution rate is defined as: denominator statistical position The number of times a location is selected in history, with the numerator being the sum of the corresponding cumulative rates; if the position... If it has never been selected (i.e., a cold start position), then the current global historical average rate will be used. Fill in the gaps to ensure the unbiasedness of the initial estimate; Normalized historical score ,in The highest historical rate Stability Reward: To suppress frequent mechanical switching caused by slight performance differences (i.e., the "ping-pong effect"), an inertial reward is given to the position activated in the current cycle. in, Indicates the stability threshold; Based on the overall score A new configuration is generated by using a strategy that separates position and rotation. The optimal solution is approximated with the lowest computational complexity. Within the first-order neighborhood of each antenna element, the candidate position with the highest comprehensive score is selected. If neighborhood overlap leads to a conflict, a greedy algorithm is used to prioritize the allocation of antenna-position pairs with higher scores, forming a set of conflict-free activation positions. ; A greedy algorithm that considers conflict is used for position selection, while adhering to the movement constraints of a single antenna; specifically, for the previous cycle... Activated at position The One antenna element ( ), traverse all candidate positions within its specific closed neighborhood. Select the candidate position with the highest score in this local subset. As an antenna To resolve conflicts caused by overlapping neighborhoods, the target location is determined by prioritizing the allocation of antenna-location pairs with higher scores, ultimately forming a set of conflict-free active locations. ; In terms of rotational decision-making, the set of conflict-free activation positions is statistically analyzed. For each active position, the frequency of occurrence of each rotation state in the offline static mapping library's corresponding grid optimization candidate set is used. The rotation state with the highest frequency is selected. If no match is found, the radial direction pointing towards the base station is selected as the default rotation for each active position. The current active grid set is counted through the offline candidate library. The number of times each rotational state occurs: Select the rotation direction with the highest cumulative frequency. As the optimal solution; if there is no prior match (i.e. If the system defaults to rotating in the direction of rotation at that position as radial towards the base station, then the rotation direction is selected to make... smallest Finally, Determine the activation and rotation status of this position.

[0023] This embodiment was tested according to the provided method: Figure 5 This demonstrates the mapping relationship between the user grid and antenna configuration based on directional sparsity. Figure 5 Each square in the diagram represents a physical location in the scene, and the shade of color indicates the average rate that a single antenna surface can achieve at that location. Users located directly below the base station have relatively low rates due to the large angle of incidence and low antenna gain. The ring area around the base station achieves the highest rates due to low path loss and favorable angle of incidence. Conversely, users at the cell edge perform the worst due to severe path loss and lack of line-of-sight links. The curve showing the change in the total system rate as a function of transmit power for 30 vehicles in the scenario is as follows: Figure 6 As shown, the 6DMA scheme significantly outperforms the traditional fixed-position antenna, indicating that the high spatial flexibility of 6DMA can effectively adapt to user distribution and capture large-scale channel changes. Although the 6DMA scheme with circular trajectory and only rotating discrete position performs better than FPA, its gain is limited by the small spatial degrees of freedom. It is worth noting that the optimization method based on single-step movement achieves a higher rate than the full reconfiguration method. This is because the single-step method has a smaller action space and is more likely to find an approximate optimal solution, while the full reconfiguration method has a large action space, and heuristic methods often cannot fully tap its potential. The trend of the total system rate with the number of users under a fixed transmit power is as follows: Figure 7 As shown, the total rate increases rapidly with the number of vehicles, but eventually saturates due to increased interference between users. The 6DMA framework provided in this embodiment always performs best among the comparative schemes. This is mainly because the distribution of vehicles in the road environment is usually relatively dispersed. The 6DMA system can use its spatial flexibility to dynamically focus antenna resources on relatively concentrated areas, while the physically constrained baseline scheme is difficult to provide such targeted services. With a fixed number of 30 vehicle users, different transmit powers and antenna update intervals The impact on system performance is as follows: Figure 8 As shown, the single-step movement scheme provided in this embodiment is comprehensively superior to the heuristic global reconfiguration scheme; in particular, when The achievable rate is lower than and This is because in The time-based optimization relies on the instantaneous sparse distribution of a small number of vehicles, resulting in a low-density mesh map, making it difficult for the algorithm to distinguish superior antenna positions; while and When the input is the cumulative predicted distribution for a future time period, this aggregation provides a more robust user hotspot statistical heatmap. Even with slight prediction errors, it still provides a better basis for decision-making. This further verifies the feasibility of antenna configuration based on the predicted distribution, which not only reduces the reconfiguration frequency but also improves performance. With a fixed transmit power of 23dBm, the number of users in different vehicles and the antenna update interval The impact on system performance is as follows: Figure 9 As shown; the experiment used 10 random seeds, with 50 simulations per seed, and the global mean was taken; the results show that, although The value has some impact on system performance, but the magnitude is small, and it is larger when the number of users is small. Slightly better As user density increases, smaller reconfiguration intervals begin to show a slight advantage because they can better track rapidly changing micro-dynamics; however, from a practical implementation perspective, larger intervals are still preferable. It can significantly reduce physical overhead with minimal performance loss. The single-step movement method is always superior to the full reconfiguration scheme, demonstrating excellent local adaptability. Figure 10 This demonstrates the theoretical lower bound of the average movement cost required for each position update using both the single-step movement method and the full reconfiguration method; based on previous modeling, the cost of moving the antenna to a nearby location is one unit; when At lower speeds, antenna configuration adjustments are based on the instantaneous actual distribution of vehicle users. Due to the high mobility of vehicles, the grid where users are located may change significantly during the adjacent decision-making cycle, requiring substantial adjustments to antenna positions, resulting in high relocation costs and frequent reconfiguration. As the number of users increases, the cumulative user distribution exhibits a stable macroscopic pattern, with minimal changes in distribution near the decision-making time, resulting in a corresponding reduction in antenna movement amplitude. The average movement cost of the antenna configuration method based on single-step movement is in the single digits, meaning that adapting to the new distribution requires only moving a small number of antennas, providing strong theoretical support for the practical deployment of 6DMA; Figure 11 This demonstrates the theoretical lower bound of the time cost required for each antenna reconfiguration using two methods; the time cost is defined as the number of steps required to move the antenna with the longest path during reconfiguration; similar to the movement cost, the time cost increases with... The increase shows a significant downward trend; the minimum time cost of a single reconfiguration should be 1, but the single-step moving method shows a value less than 1, indicating that the antenna position did not change during some reconfiguration processes. This reflects that the cumulative distribution of predicted vehicle users has limited changes, and the macro distribution remains approximately static within the prediction window; by capturing long-term patterns rather than chasing instantaneous fluctuations, the system avoids unnecessary mechanical adjustments and demonstrates strong feasibility.

[0024] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention in any way. Although the present invention has been disclosed above with reference to preferred embodiments, it is not intended to limit the present invention. Any person skilled in the art can make some modifications or alterations to the above-disclosed technical content to create equivalent embodiments without departing from the scope of the present invention. Any simple modifications, equivalent changes and alterations made to the above embodiments based on the technical essence of the present invention without departing from the scope of the present invention shall still fall within the scope of the present invention.

Claims

1. A single-step reconfiguration method for a six-dimensional movable antenna in a vehicle-to-everything (V2X) network, characterized in that, Includes the following steps: S1. A system for constructing a six-dimensional movable antenna 6DMA model and a vehicle-to-everything (V2X) communication scenario, including a city scenario with K vehicles and a ground base station BS. The ground base station BS is configured with a six-dimensional movable antenna system consisting of U movable antenna surfaces. The K vehicles are configured with omnidirectional antennas at fixed positions. The position vector, rotation direction, antenna radiation pattern, and antenna motion constraints of the omnidirectional antennas are defined. An uplink communication and rate calculation model from the vehicles to the ground base station BS is established. S2. Construct a periodic single-step reconfiguration strategy based on motion prediction. Based on the predicted cumulative vehicle spatial distribution, construct a 6DMA reconfiguration optimization problem with the goal of maximizing the average sum rate within the prediction window. The position change of the six-dimensional movable antenna follows single-step movement constraints, minimum distance constraints, mutual reflection constraints, and occlusion constraints. The single-step movement constraints restrict each antenna element to only move to the first-order physical neighborhood of its current position during reconfiguration. S3. Construct a set of feasible discrete antenna locations and corresponding spherical discrete location rotation sets that satisfy the minimum distance constraint in the space around the base station, based on the latitude and longitude grid method. Generate feasible discrete six-dimensional movable antenna configurations. Establish a model of movement cost and time cost describing the antenna configuration switching process based on graph theory. Calculate the lower bound of movement cost and time cost using the breadth-first search method BFS and the improved Hungarian algorithm. S4. Obtain the current antenna configuration from the set of discrete spherical positions. Based on the directional sparsity of the 6DMA channel, construct the antenna position rotation-effective gain space mapping offline and the historical rate obtained online. Construct a comprehensive scoring function for each candidate antenna position. Use a greedy strategy to select the optimal position and rotation state in the neighborhood for each antenna element.

2. The single-step reconfiguration method for a six-dimensional movable antenna in a vehicle-to-everything (V2X) network according to claim 1, characterized in that, The six-dimensional movable antenna 6DMA model includes U movable antenna surfaces connected to the central processing unit via independent telescopic robotic arms; The definition of the omnidirectional antenna's position vector, rotation direction, antenna radiation pattern, and antenna motion constraints specifically includes: In the 6DMA model of the six-dimensional movable antenna, the time dimension adopts a discrete time slot mechanism, and the time interval is... The vehicle position remains unchanged within a single time slot, and is only updated at the time slot boundaries based on the sampling rate of the truncated Gaussian distribution; Each of the movable antenna surfaces has the ability to independently adjust its three-dimensional spatial position and three-dimensional rotation attitude. The feasible configuration state space of the movable antenna surfaces is discretized and defined as a set. ,in , These represent the total number of discrete positions and rotation states, respectively. Indicates the first The center coordinates of the candidate locations Represents Euler angles, in any rotational state. The corresponding unit outward normal vector The expression is: Using a binary decision matrix Describe the deployment status of the system, which must meet the following constraints: Capacity constraint: The system activates a total of U antenna surfaces ( ), which needs to satisfy capacity constraints: Mutual non-obstruction constraint: The normal vectors of any two active antenna surfaces must not point to each other's positions; Central processing unit visibility constraint: The radiation direction of all antenna surfaces must be away from the central processing unit; Minimum spatial spacing constraint: The Euclidean distance between any two active antenna surfaces must be greater than a preset safety threshold; Single occupancy constraint: At most one antenna surface can be deployed at each discrete location.

3. The single-step reconfiguration method for a six-dimensional movable antenna in a vehicle-to-everything (V2X) network according to claim 2, characterized in that, The uplink communication adopts a hybrid near-field-far-field channel model: the propagation between antenna surfaces is regarded as the far field, while the propagation between array elements inside the antenna surface retains the near-field phase characteristics. Establishing uplink communication and rate calculation specifically includes: S101. Establish a local coordinate system to describe the direction of arrival; S102. Define antenna gain based on 3GPP radiation direction chart, establish line-of-sight and non-line-of-sight probability path loss based on 3GPP UMi standard and log-normal shadowing fading respectively, and then establish channel model through multipath Rayleigh small-scale fading. S103, Vehicles The channel vector is determined by its position and 6DMA configuration, and is calculated based on the interference-to-noise ratio of each vehicle signal under the matched filter receiver. The expression is: in, Indicates vehicle The signal interference plus noise ratio, This indicates the allocated bandwidth.

4. The single-step reconfiguration method for a six-dimensional movable antenna in a vehicle-to-everything (V2X) network according to claim 1, characterized in that, The periodic single-step reconfiguration strategy discretizes time into units of length . The antenna configuration is updated every N time slots. vehicle In the time slot The position is recorded as ,vehicle The motion follows a linear kinematic model, and the vehicle's position in future time slots is recursively predicted based on the vehicle's instantaneous average velocity, expressed as: in, This indicates the predicted position of the vehicle at the next moment. This indicates the predicted position of the vehicle at the current moment. Represents the direction vector. This represents the instantaneous average velocity, with the initial condition set to the observed value at the current location, i.e. ; Based on the predicted vehicle positions in future time slots, the average sum of speeds within a vehicle cycle is calculated using the following expression: in, Indicates average and rate. Indicates vehicle The signal interference plus noise ratio, Indicates time vehicle The predicted location, This indicates the antenna configuration.

5. A single-step reconfiguration method for a six-dimensional movable antenna in a vehicle-to-everything (V2X) network according to claim 4, characterized in that, The single-step movement constraint specifically means that the set of candidate target positions for the u-th antenna in the current period is restricted to the first-order physical neighborhood of its position in the previous period. That is, each antenna element moves at most one step or remains stationary during each reconfiguration, as expressed in the following expression: in, Represents the position in the discrete grid The set of physical neighbors.

6. A single-step reconfiguration method for a six-dimensional movable antenna in a vehicle-to-everything (V2X) network according to claim 1, characterized in that, The discrete location generation method based on latitude and longitude grids in step S3 includes: Given the number of meridians, determine the position of the first circle of latitude that satisfies the minimum distance; Calculate the maximum allowable interval between weft coils to keep all neighboring points from meeting the minimum distance constraint, and calculate the number of intermediate weft coils that can be accommodated based on the total available height, which is obtained by summing the number of positions on each coil. A feasible rotation state is obtained, including the radial vector pointing to the center of the sphere and the normal vectors of the triangular faces formed by adjacent positions.

7. A single-step reconfiguration method for a six-dimensional movable antenna in a vehicle-to-everything (V2X) network according to claim 6, characterized in that, Calculating the lower bounds of the mobility cost and time cost includes: S301. Calculate the shortest path length from all initial positions to the target position using a breadth-first search algorithm, and form a distance matrix. S302. Construct a cost function that includes a quadratic penalty term. The expression is: in, This represents the penalty coefficient that prioritizes minimizing the total number of moves. This represents the distance the antenna travels from position u to position v. S303. Solving for the optimal matching using an improved Hungarian algorithm. The expression for the lower bound of the movement cost is obtained as follows: in, This represents the energy consumption per unit distance when a single antenna element is moved to an adjacent discrete location. This represents the matching matrix; a value of 1 indicates that antenna movement has occurred. This represents the distance the antenna moves from position u to position v; The expression for the lower bound of time cost is: in, Indicates the maximum number of steps a single antenna can move. It represents the unit time it takes for a single antenna element to move to an adjacent discrete position.

8. A single-step reconfiguration method for a six-dimensional movable antenna in a vehicle-to-everything (V2X) network according to claim 1, characterized in that, Step S4 specifically includes the following steps: S401. Offline construction of spatial grid - static mapping library for preferred antenna configuration: The ground service area is gridded, and the back position in each grid is removed by geometric hemisphere clipping. A two-level hierarchical sampling strategy is adopted to reduce the search space. The candidate positions and their rotation states are traversed by rotation fine-tuning. The theoretical average spectral efficiency is calculated. The top-ranked configurations are retained for each grid to form an offline preferred candidate set. S402, Online Scoring and Decision-Making: The set of active grids covered by the current period's forecast. For each candidate position, a comprehensive score is constructed, which is then combined with the historical cumulative rate score and the stability reward for positions activated in the previous cycle to obtain the overall score. ; Within the first-order neighborhood of each antenna element, the candidate position with the highest comprehensive score is selected. If neighborhood overlap leads to a conflict, a greedy algorithm is used to prioritize the allocation of antenna-position pairs with higher scores, forming a set of conflict-free activation positions. ; The set of conflict-free activation locations is statistically analyzed. The frequency of each rotation state in the offline static mapping library corresponding to the preferred candidate set of the grid is selected for each active position. If there is no match, the radial direction pointing to the base station is selected as the default rotation.

Images