A base station dynamic switching method based on a distributed communication navigation integrated cabinet

Abstract

The application provides a base station dynamic switching method based on a distributed communication navigation integrated cabinet, comprising the following steps: a distributed communication navigation base station network is established based on a plurality of distributed communication navigation integrated cabinets; when a mobile terminal needs to access the network, an access request message is broadcasted and sent; when each cabinet receives the access request message, network quality data and positioning accuracy data are sent to a management platform; the management platform calculates the comprehensive evaluation score of each cabinet according to the network quality data and the positioning accuracy data; and the cabinet with the highest comprehensive evaluation score is selected as the current base station of the mobile terminal. The application uses a distributed high-precision communication navigation integrated cabinet to establish a distributed architecture network which can be expanded quickly, and meets the communication navigation demand in a long-distance mobile scene; based on a network performance and positioning accuracy evaluation model, the cabinet resources in the distributed network are effectively managed and optimized through an algorithm, and high-quality communication navigation services are provided for the mobile terminal.

Description

Technical Field

[0001] This invention relates to the fields of communication and satellite positioning technology, and more specifically, to a method for dynamic switching of base stations based on a distributed communication and navigation integrated cabinet. Background Technology

[0002] Traditional wireless local area network-based communication and navigation systems rely on base stations provided by network operators or fixed base stations built by users to provide network access services. In mountainous areas, forest areas, deserts, and water environments, there are often signal coverage blind spots. Their coverage range is limited and their flexibility is insufficient, making it difficult to adapt to the real-time positioning needs in complex terrains or dynamic environments.

[0003] Using self-built fixed base stations for communication and navigation also presents challenges such as long construction cycles and high costs in scenarios requiring frequent changes in network access locations, including experiments, tests, emergency rescue, and command and control. While using network base stations provided by operators can quickly achieve network access, the speed and stability of data transmission are difficult to guarantee.

[0004] The quality of communication and navigation services is affected by multiple indicators such as service type, transmission distance, network quality, and satellite signal quality. However, there is a lack of comprehensive evaluation models for positioning accuracy and network quality, and the base station handover mechanism lacks environmental adaptability. Summary of the Invention

[0005] This invention addresses the technical problems existing in the prior art by providing a method for dynamic switching of base stations based on a distributed communication and navigation integrated cabinet, thereby enabling dynamic switching of the serving base station for mobile terminals and ensuring the signal service quality of mobile terminals.

[0006] This invention provides a method for dynamic base station handover based on a distributed communication and navigation integrated cabinet, comprising:

[0007] Multiple distributed communication and navigation integrated cabinets are deployed in different areas that require network coverage to form a distributed communication and navigation base station network.

[0008] When a mobile terminal needs to access the network, signal scanning is activated, and an access request message is sent to all cabinets in the network.

[0009] When each rack receives the access request message, it sends its own network quality data and positioning accuracy data to the management platform in real time.

[0010] The management platform calculates a comprehensive evaluation score for each cabinet based on the network quality data and the positioning accuracy data of each cabinet.

[0011] Select the cabinet with the highest comprehensive evaluation score as the current base station for the mobile terminal;

[0012] When the handover conditions are met, another candidate cabinet is selected as the current base station, and the base station handover is performed.

[0013] This invention provides a dynamic base station switching method based on a distributed communication and navigation integrated cabinet. A distributed communication and navigation base station network is constructed using multiple distributed communication and navigation integrated cabinets. When a mobile terminal accesses the network, it broadcasts an access request message. When each cabinet receives the access request message, it sends network quality data and positioning accuracy data to a management platform. The management platform calculates a comprehensive evaluation score for each cabinet based on the network quality data and positioning accuracy data. The cabinet with the highest comprehensive evaluation score is selected as the mobile terminal's current base station. This invention uses a distributed high-precision communication and navigation integrated cabinet to establish a distributed architecture that facilitates rapid network expansion, meeting the communication and navigation needs of long-distance mobile scenarios. Based on a network performance and positioning accuracy evaluation model, an algorithm is used to effectively manage and optimize cabinet resources in the distributed network, providing high-quality communication and navigation services for mobile terminals. Attached Figure Description

[0014] Figure 1 A flowchart of a base station dynamic handover method based on a distributed communication and navigation integrated cabinet is provided in this embodiment of the invention;

[0015] Figure 2 This is a schematic diagram of a base station dynamic handover system based on a distributed communication and navigation integrated cabinet, provided as an embodiment of the present invention. Detailed Implementation

[0016] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention. In addition, the technical features of the various embodiments or individual embodiments provided by the present invention can be arbitrarily combined with each other to form feasible technical solutions. Such combinations are not constrained by the order of steps and / or structural composition patterns, but must be based on the ability of those skilled in the art to implement them. When the combination of technical solutions is contradictory or cannot be implemented, it should be considered that such a combination of technical solutions does not exist and is not within the scope of protection claimed by the present invention.

[0017] See Figure 1 This invention provides a method for dynamic base station handover based on a distributed communication and navigation integrated cabinet, comprising the following steps:

[0018] Step 1: Deploy multiple distributed communication and navigation integrated cabinets in different areas that require network coverage to form a distributed communication and navigation base station network.

[0019] Understandably, the server racks are deployed to predetermined locations in the areas requiring network coverage and powered on. The positioning module uses a single-point positioning algorithm to calculate initial coordinates (meter-level accuracy). If high-precision coordinates (centimeter-level accuracy) are required, a static measurement mode is activated, using a static relative positioning algorithm to calculate centimeter-level precision coordinates, which serve as the initial coordinates for the base station. Server racks broadcast their own coordinates to each other, and the wireless ad hoc network module calculates the optimal communication link based on efficient channel coding and decoding algorithms and a decentralized multi-hop ad hoc network protocol to construct a distributed topology network.

[0020] Step 2: When the mobile terminal accesses the network, signal scanning is activated, and an access request message is sent to all cabinets in the network.

[0021] Understandably, when a mobile terminal accesses a distributed network, it sends an access request frame carrying a unique identifier. The scan period is 100ms, and the request frame is broadcast across the entire network, ensuring a response from all cabinets within the coverage radius. The request frame format is shown in Table 1 below:

[0022] Table 1 Request Frame Format

[0023] Frame header uint8 2 Synchronization header (0xAA 0x55) + Protocol version (0x01) Terminal ID uint64 8 Terminal unique identifier (MAC address or IMEI) Business type enum 1 Used to describe business applications: control commands, voice calls, data transmission, drone navigation, autonomous driving, ship navigation, information dissemination, and others. GNSS observation coordinates double 24 WGS84 coordinates (longitude, latitude, elevation, 8 bytes each) Timestamp uint64 8 (UTC) Microsecond-level timestamp Number of available satellites int 8 The number of satellites currently observable by the mobile terminal (number N). Satellite PRN code list string 8N List of PRN codes for observable satellites Extended fields variable 0-32 Reserved fields

[0024] Step 3: When each cabinet receives the access request message, it sends its own network quality data and positioning accuracy data to the management platform in real time.

[0025] Understandably, when each cabinet in the network receives an access request message from a mobile terminal, it obtains its own network quality data and positioning accuracy data, and preprocesses the network quality data and positioning accuracy data.

[0026] Among them, network quality data mainly includes RSSI signal strength, SINR signal-to-noise ratio and interference level I. Interference level I is calculated by detecting the power of interference signals in non-serving frequency bands through broadband spectrum scanning, and the normalized value is 0 ~ 1.

[0027] The positioning accuracy data mainly includes the known precise coordinates of the cabinet (WGS84 coordinate system, longitude, latitude, and elevation), the GNSS observation coordinates of the cabinet, and the satellite observation status (number of available satellites and PDOP position accuracy factor value).

[0028] The acquired network quality data and positioning accuracy data are preprocessed, including generating differential correction data based on the GNSS observation coordinates and known coordinates of each cabinet.

[0029] Outlier filtering: Outlier filtering is performed on the acquired network quality data and positioning accuracy data. Specifically, the 3σ criterion is used to remove outliers (such as samples with RSSI mutations of ±15dBm) and retain stable data for three consecutive periods.

[0030] Standardized format: Encapsulate rack response data (network quality data and positioning accuracy data) and mobile terminal request data into JSON format.

[0031] Data compression and transmission: The data dimensionality is reduced by half by using a lightweight PCA algorithm before being sent to the integrated management platform, reducing transmission bandwidth usage by 60%.

[0032] Step 4: The management platform calculates the comprehensive evaluation score for each cabinet based on the network quality data and positioning accuracy data of each cabinet.

[0033] Understandably, after receiving network quality data and positioning accuracy data from each cabinet, the management platform evaluates the network quality and positioning accuracy of each cabinet respectively.

[0034] The network quality of each rack is evaluated, focusing on the stability and transmission efficiency of the communication link. The evaluation formula is as follows:

[0035] ;

[0036] in, The network quality score for the server rack is calculated using α, β, and γ, which are dynamic weighting coefficients.

[0037] RSSI (Signal Strength): Characterizes the signal coverage strength, in dBm, ranging from -110 to -40. It is calculated using the uplink probe reference signal sent by the mobile terminal and processed by sliding window mean filtering.

[0038] SINR (Signal-to-Noise Ratio): Measures the effective transmission capability of a signal in an interference environment, measured in dB, ranging from 0 to 30, and calculated from the uplink reference signal sent by the demodulation terminal.

[0039] I (Interference Level): Calculated by detecting interference signal power in non-serving frequency bands through broadband spectrum scanning;

[0040] Normalization range: Each indicator is mapped to [0,1] to avoid differences in dimensions.

[0041] The positioning accuracy of each cabinet is evaluated, focusing on the GNSS positioning accuracy and the validity of differential data. The positioning accuracy evaluation formula is as follows:

[0042] ;

[0043] in, The positioning accuracy of the cabinet is scored, where δ, ε, and λ are dynamic weighting coefficients.

[0044] PDOP (Position Precision Factor): Based on the positioning signal output of the cabinet, it measures the impact of satellite geometric distribution on positioning error. The smaller the PDOP value, the higher the positioning accuracy.

[0045] SSN (Simultaneous Observation Satellites): The current number of satellites observed simultaneously by the mobile terminal and the cabinet, which directly affects the reliability of positioning solutions (threshold: ≥4 satellites for effective positioning); AOD (Differential Age): The timeliness of differential correction data, defined as the delay (in seconds) between the time the differential data was generated and the current time, which directly affects the success rate of RTK fixed solutions.

[0046] Normalization processing: PDOP is mapped to the interval [1,6], SSN is mapped to the interval [4,20], and AOD is mapped to [0,5] seconds.

[0047] A nonlinear weighted fusion algorithm is employed to deeply couple heterogeneous data on network quality and positioning accuracy. Initial values ​​for the weighting coefficients (α, β, γ) are set and dynamically adjusted by the reinforcement learning module of the management platform. The overall score S is calculated by weighting the network quality score and the positioning accuracy score.

[0048] ;

[0049] Where S is the overall evaluation score of the cabinet, and ω1 and ω2 are dynamic weighting coefficients, with initial values ​​set at ω1=0.6 and ω2=0.4 (preferring network quality).

[0050] tanh (hyperbolic tangent function): Compresses the output to (-1,1) and ensures the reasonableness of the score by truncating the positive part (only the positive part [0,1]).

[0051] Step 5: Select the cabinet with the highest comprehensive evaluation score as the current base station for the mobile terminal.

[0052] Understandably, after calculating the comprehensive evaluation score for each cabinet, the management platform selects the cabinet as the current base station for the mobile terminal from all cabinets. Specifically, the management platform sorts all cabinets by their comprehensive evaluation scores S, and selects the cabinet with the highest comprehensive evaluation score and S>=0.7 as the current base station; if the scores S of all cabinets are less than 0.7, the power enhancement mechanism for neighboring cabinets is triggered, and the comprehensive evaluation score of each cabinet is recalculated to reselect the current base station for the mobile terminal.

[0053] The selected cabinet serves as a network base station to provide network services to mobile terminals. At the same time, the integrated management platform uses the selected cabinet as a reference station to perform real-time dynamic carrier phase differential calculation and record the coordinates of the mobile terminals, and publishes the precise location information of the mobile terminals through the network.

[0054] Step 6: When the handover conditions are met, select another candidate cabinet as the current base station and perform base station handover.

[0055] Understandably, the embodiments of the present invention acquire network quality data and positioning accuracy data of each cabinet in real time, and periodically calculate the comprehensive evaluation score of each cabinet, so as to realize the switching of the current base station at any time.

[0056] Specifically, network quality and positioning data are collected every 100ms, and the average value is calculated using a sliding window (window size 5) as the evaluation basis.

[0057] The base station handover condition is met when the current base station's overall evaluation score S < 0.6 or the number of shared satellites between the mobile terminal and the current base station < 4. When the base station handover condition is met, the overall evaluation score of each of the other cabinets is calculated based on the network quality data and positioning accuracy data of each other cabinet. S >= the current base station's score. The cabinet with the largest S value (S>=0.7) is used as the base station after the handover.

[0058] In this embodiment of the invention, the weight coefficients α, β, γ, δ, ε, λ, ω1, and ω2 of network quality assessment, positioning accuracy assessment, and comprehensive assessment are dynamically adjusted based on reinforcement learning.

[0059] The initial values ​​of the weight coefficients are preset based on prior knowledge to suit different application scenarios. A Double Deep Q-Network (DQN) algorithm is employed to construct two networks: a real decision network and a virtual network. The real decision network is responsible for selecting actions based on the current policy, while the virtual network only needs to calculate the target Q-value and periodically synchronize it with the real decision network to calculate the loss and optimize the parameters of the real decision network. By separating the base station's decision-making actions from value assessment, overestimation bias is reduced, and oscillations caused by changes in the weight parameters of the evaluation model are minimized.

[0060] The specific steps for dynamically adjusting weight coefficients based on a dual-depth Q-network include:

[0061] Obtain the current weight configuration and state S based on a real network. t The state S t This represents the overall evaluation score of the current base station in the t-th iteration;

[0062] Select action A to perform. t(Δα, Δβ, Δγ, Δδ, Δε, Δλ, Δω1, Δω2);

[0063] According to the execution action A t Update the current weight configuration to obtain the new weight;

[0064] Calculate the state S for the next iteration based on the new weights. t+1 ;

[0065] Based on the state S of the next iteration t+1 Calculate the reward R t ;

[0066] Stored in the experience pool (S t A t ,R t ,S t+1 );

[0067] According to reward R t Choose the next action A t+1 ;

[0068] The target Q value is calculated based on the virtual network, and the target Q value is fed back to the real network. The loss is calculated and backpropagated to update the parameters of the real network.

[0069] Continue iterating to select the action to be executed until the calculated loss is less than the preset threshold or the number of iterations reaches the maximum, and obtain the optimal action and optimal weight configuration.

[0070] The evaluation model with dynamically adjusted weight coefficients described above is applicable to a variety of typical scenarios, such as:

[0071] (1) High bandwidth scenarios (such as video transmission): Increase the weights of α and ω1 (signal strength and network quality) to reduce handover oscillations caused by signal fluctuations;

[0072] (2) High real-time scenarios (such as UAV control in urban areas): Increase the weights of β and ω2 (signal-to-noise ratio and positioning accuracy) to improve the control signal transmission capability under interference environment while ensuring positioning accuracy.

[0073] See Figure 2 This invention provides a dynamic base station switching system based on a distributed communication and navigation integrated cabinet. The system includes a mobile terminal 20, multiple cabinets 21, and a management platform 22. The multiple distributed communication and navigation integrated cabinets 21 are deployed in a dispersed manner to areas requiring network coverage to form a distributed communication and navigation base station network.

[0074] Mobile terminal 20 is used to activate signal scanning and send access request messages to all cabinets 21 in the network when accessing the network.

[0075] Each cabinet 21 is used to send its own network quality data and positioning accuracy data to the management platform 22 in real time when it receives the access request message;

[0076] The management platform 22 is used to calculate the comprehensive evaluation score of each cabinet based on the network quality data and the positioning accuracy data of each cabinet; select the cabinet with the highest comprehensive evaluation score as the current base station of the mobile terminal; and when the handover conditions are met, select other candidate cabinets as the current base station to perform base station handover.

[0077] Each cabinet 21 includes a data preprocessing module 210, a wireless self-organizing network module 211, a positioning module 212, a timing module 213, and a power supply module 214.

[0078] The data preprocessing module 210 is mainly used to calculate the error between the actual GNSS observation coordinates and the actual coordinates of the cabinet, generate differential correction data for the cabinet, and filter outliers in network quality data and positioning accuracy data.

[0079] The wireless ad hoc network module 211 is used to calculate the optimal communication link based on efficient channel coding and decoding algorithms and a decentralized multi-hop ad hoc network protocol, and to build a distributed topology network.

[0080] The positioning module 212 receives satellite signals and uses a single-point positioning algorithm to calculate known coordinates (accuracy at the meter level). If high-precision coordinates (accuracy at the centimeter level) are required, the static measurement mode is activated, and the static relative positioning algorithm is used to calculate centimeter-accuracy coordinates as the initial coordinates of the cabinet.

[0081] The time synchronization module 213 is used to send high-precision time signals to the mobile terminal 20 and the management platform 22.

[0082] Power module 214 is used to supply power to the other modules.

[0083] Mobile terminal 20 and cabinet 21 achieve positioning services through GNSS satellite signals, and can be located through GNSS satellite signals.

[0084] The present invention provides a method and system for dynamic base station handover based on a distributed communication and navigation integrated cabinet, which has the following beneficial effects:

[0085] (1) This invention uses a distributed high-precision communication and navigation integrated cabinet to establish a distributed architecture that supports dynamic elastic expansion and meets the communication and navigation needs in long-distance mobile scenarios. Based on a comprehensive evaluation of network performance and positioning accuracy, a performance optimization mechanism for dynamic switching of base stations is implemented through an algorithm, which effectively manages and optimizes cabinet resources in the distributed network and provides high-quality communication and navigation services for mobile terminals.

[0086] (2) The present invention designs a mechanism that receives network quality data and location information in real time, performs preliminary processing and formatting on the data, removes noise and abnormal data, and processes and encapsulates the preprocessed data according to a unified data format before feeding it back to the management platform.

[0087] (3) The present invention designs a multimodal comprehensive evaluation algorithm that deeply couples network performance and positioning accuracy, and simultaneously considers the network quality and positioning accuracy of each cabinet to allocate access base stations for mobile terminals.

[0088] (4) The present invention designs a decision-making and dynamic adjustment mechanism that continuously collects and evaluates preprocessed data during the change of mobile terminal location, and always allocates the cabinet with the best evaluation result to the mobile terminal as the base station, thereby realizing the dynamic switching of the base station allocation for the mobile terminal.

[0089] (5) The present invention designs a reinforcement learning model that optimizes and adjusts the weight coefficients of each indicator as the number of mobile terminals and the network environment in the distributed communication and navigation system changes, so as to adapt to the ever-changing business needs and environmental conditions and ensure that the system can always provide high-quality communication and navigation services.

[0090] It should be noted that the descriptions of each embodiment in the above embodiments have different focuses. For parts that are not described in detail in a certain embodiment, please refer to the relevant descriptions in other embodiments.

[0091] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0092] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded computer, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations. Figure 1 One or more processes and / or boxes Figure 1A device that provides the functions specified in one or more boxes.

[0093] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0094] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0095] Although preferred embodiments of the invention have been described, those skilled in the art, upon learning the basic inventive concept, can make other changes and modifications to these embodiments. Therefore, the appended claims are intended to be interpreted as including both the preferred embodiments and all changes and modifications falling within the scope of the invention.

[0096] Obviously, those skilled in the art can make various modifications and variations to this invention without departing from its spirit and scope. Therefore, if these modifications and variations fall within the scope of the claims of this invention and their equivalents, this invention also intends to include these modifications and variations.

Claims

1. A method for dynamic base station handover based on a distributed communication and navigation integrated cabinet, characterized in that, include: Multiple distributed communication and navigation integrated cabinets are deployed in different areas that require network coverage to form a distributed communication and navigation base station network. When a mobile terminal needs to access the network, signal scanning is activated, and an access request message is sent to all cabinets in the network. When each rack receives the access request message, it sends its own network quality data and positioning accuracy data to the management platform in real time. The management platform calculates a comprehensive evaluation score for each cabinet based on the network quality data and the positioning accuracy data of each cabinet. Select the cabinet with the highest comprehensive evaluation score as the current base station for the mobile terminal; When the handover conditions are met, another candidate cabinet is selected as the current base station, and the base station handover is performed; When each rack receives the access request message, it sends its own network quality data and positioning accuracy data to the management platform in real time, including: When each cabinet receives the access request message, it acquires its own network quality data and positioning accuracy data. The network quality data includes the cabinet's RSSI signal strength, SINR signal-to-noise ratio, and interference level I. The positioning accuracy data includes the cabinet's known precise coordinates, the cabinet's GNSS observation coordinates, and the cabinet's satellite observation status information. The cabinet's satellite observation status information includes the number of available satellites and the PDOP position accuracy factor. Differential correction data for the computer cabinet based on the GNSS observation coordinates and known precise coordinates of the cabinet; The system preprocesses its network quality data, positioning accuracy data, and differential correction data and then sends them to the management platform. The management platform calculates a comprehensive evaluation score for each server rack based on the network quality data and positioning accuracy data for each rack, including: Calculate the network quality score for each rack based on the network quality data for each rack. Based on the positioning accuracy data of each cabinet, calculate the positioning accuracy score for each cabinet; Calculate the overall evaluation score for each cabinet based on the network quality score and the positioning accuracy score for each cabinet; The step of calculating the positioning accuracy score for each rack based on the positioning accuracy data for each rack includes: ； in, For positioning accuracy scoring, δ, ε, and λ are dynamic weighting coefficients; PDOP is the position accuracy factor of the server rack; SSN is the number of satellites viewed simultaneously by the mobile terminal and the cabinet; AOD is the delay between the differential correction data generation time of the cabinet and the current time; PDOP, SSN, and AOD are normalized values.

2. The method according to claim 1, characterized in that, The calculation of a network quality score for each server rack based on the network quality data for each rack includes: ； in, For network quality scoring, α, β, and γ are dynamic weighting coefficients. The RSSI value ranges from -110 to -40 dBm, and the SINR value ranges from 0 to 30 dB. .

3. The method according to claim 1, characterized in that, The comprehensive evaluation score for each server rack is calculated based on its network quality score and positioning accuracy score, including: ； Where ω1 and ω2 are dynamic weighting coefficients, Rate the network quality. The positioning accuracy is scored, and tanh is the hyperbolic tangent function.

4. The method according to claim 1, characterized in that, The selection of the cabinet with the highest comprehensive evaluation score as the current base station for the mobile terminal includes: The cabinet with the highest comprehensive evaluation score and a comprehensive evaluation score ≥ 0.7 is selected as the current base station for the mobile terminal; If the overall evaluation score of all cabinets is less than 0.7, then the power of each cabinet will be increased, the overall score of each cabinet will be recalculated, and the current base station will be redefined.

5. The method according to claim 1, characterized in that, When the handover conditions are met, selecting another candidate cabinet as the current base station and performing base station handover includes: Regularly collect network quality data and positioning accuracy data of the current base station, and calculate the comprehensive evaluation score of the current base station; The base station handover conditions are met when the overall evaluation score of the current base station is less than 0.6 or the number of co-viewed satellites between the mobile terminal and the current base station is less than 4. When the base station handover conditions are met, the comprehensive evaluation score of each of the other cabinets is calculated based on the network quality data and positioning accuracy data of each other cabinet. In other cabinets, the overall evaluation score will be greater than or equal to the overall evaluation score of the current base station. The cabinet with the highest overall evaluation score (≥0.7) and the highest overall evaluation score will be used as the base station after the handover.

6. The method according to claim 3, characterized in that, The weight coefficients α, β, γ, δ, ε, λ, ω1, and ω2 are dynamically adjusted based on a dual-depth Q-network, which includes a real network and a virtual network. The dynamic adjustment of weight coefficients α, β, γ, δ, ε, λ, ω1, ω2 based on a dual-depth Q-network includes: Obtain the current weight configuration and state S based on a real network. t The state S t This represents the overall evaluation score of the current base station in the t-th iteration; Select actionA t (Δα、Δβ、Δγ、Δδ、Δε、Δλ、Δω1、Δω2); According to the execution action A t Update the current weight configuration to obtain the new weight; Calculate the state S for the next iteration based on the new weights. t+1 ; Based on the state S of the next iteration t+1 Calculate the reward R t ; Stored in the experience pool (S t A t ,R t ,S t+1 ); According to reward R t Choose the next action A t+1 ; The target Q value is calculated based on the virtual network, and the target Q value is fed back to the real network. The loss is calculated and backpropagated to update the parameters of the real network. Continue iterating to select the action to be executed until the calculated loss is less than the preset threshold or the number of iterations reaches the maximum, and obtain the optimal action and optimal weight configuration.

Images