T-BOX and S-BOX fused cross-border transportation supervision communication method

Abstract

The invention discloses a T-BOX and S-BOX fused cross-border transportation supervision communication method, which comprises the following steps: S1, multi-source data acquisition and edge intelligent processing: acquiring vehicle working condition data through a vehicle-mounted T-BOX, carrying out real-time analysis by utilizing an edge AI model, adding a transmission priority identifier to the data according to an analysis result, and carrying out multi-source data acquisition and edge intelligent processing; wherein a high-priority identifier is added to the safety abnormal event, and a low-priority identifier is added to the normal working condition data; s2, intelligent link decision and switching based on multi-factor perception: dynamically selecting a communication link based on the priority identifier and pre-stored network state information; and when a high priority identifier is added to the data, or when it is predicted that communication cannot be performed through the ground network within a preset time threshold according to the vehicle position and the network state information, S-BOX is started to establish a satellite communication link for data transmission. According to the invention, continuous, reliable and intelligent transmission of supervision data in an environment with incomplete cross-border network coverage is realized.

Description

Technical Field

[0001] This invention relates to the field of vehicle-to-everything (V2X) communication technology, and in particular to a cross-border transportation supervision and communication method that integrates T-BOX and S-BOX. Background Technology

[0002] Cross-border commercial vehicle transportation is a crucial link in international logistics and supply chains. However, its routes are long and geographically complex, often traversing multiple countries and regions. Therefore, continuous, real-time monitoring of vehicle location, driving status, and cargo safety is a core requirement for ensuring transportation efficiency, safety, and compliance.

[0003] However, current technological methods still have the following limitations: The continuity of communication links is a significant issue. Currently, the transmission of vehicle monitoring data relies heavily on terrestrial mobile communication networks (such as 4G / 5G) along the route. However, in vast border areas, remote mountainous regions, deserts, and transnational trunk lines, terrestrial network coverage has numerous blind spots or areas with weak signals. Once a vehicle enters these areas, the monitoring data link is immediately interrupted, preventing operators from monitoring vehicle status in real time. This not only leads to a lack of transparency in the transportation process but also makes it impossible to promptly warn and intervene in safety risks such as speeding, fatigued driving, and route deviation.

[0004] The communication strategy is rigid, resulting in inefficient resource utilization. While some solutions incorporate satellite communication as a supplementary means, their operation is mostly a simple "master-slave switchover." Expensive satellite links are only passively activated when the ground network completely fails. This "disconnect and reconnect" mechanism leads to the loss of critical data during the switchover interval. More importantly, existing systems lack intelligent judgment regarding the content of data transmission, employing a uniform transmission strategy regardless of data importance (such as routine engine data and emergency braking signals). This results in valuable satellite bandwidth being frequently occupied by large amounts of low-value, delayable general data, while truly priority-requiring safety alerts may be delayed due to channel congestion.

[0005] The credibility and value of data have not been effectively improved. Cross-border transportation involves multiple entities such as customs, insurance, and finance, and has extremely high requirements for the authenticity and tamper-proof nature of data for key business events (such as electronic lock opening and closing, cross-border timing, and cargo anomalies). Existing technologies mainly focus on the "transmission" of data, while neglecting the "trustworthiness" of the data. When used as electronic evidence, data stored in traditional centralized locations is easily questioned in terms of its timing and content, lacking credibility in dispute resolution. Summary of the Invention

[0006] The purpose of this invention is to solve the above-mentioned problems by proposing a cross-border transportation supervision and communication method that integrates T-BOX and S-BOX.

[0007] To achieve the above objectives, the following technical solution was adopted: A cross-border transport regulatory communication method integrating T-BOX and S-BOX includes the following steps: S1: Multi-source data acquisition and edge intelligent processing; The vehicle-mounted T-BOX collects real-time vehicle operating condition data via the vehicle's CAN bus. The edge AI model built into the T-BOX analyzes the real-time vehicle operating condition data in real time and adds a transmission priority identifier to the real-time vehicle operating condition data based on the analysis results.

[0008] Specifically, the edge AI model performs real-time analysis of the vehicle's real-time operating data, including the following steps: S101: Baseline modeling of operating conditions; Based on normal historical vehicle operating condition data, a dynamic baseline model is established to represent the normal operating state of the vehicle. S102: Real-time anomaly detection; The collected real-time vehicle operating condition data is compared with the dynamic baseline model, and the deviation between the real-time vehicle operating condition data and the normal vehicle operation data in the dynamic baseline model is calculated. The obtained deviation is compared with a preset abnormal threshold. If the deviation exceeds the abnormal threshold, it is determined to be abnormal. S103: Event Classification; For detected anomalies, they are classified into different security event types according to their data characteristics, and corresponding transmission priority identifiers are attached to the data based on a preset event type-priority relationship mapping table.

[0009] Among them, identified security anomalies are marked with high priority, while normal operating data are marked with low priority or placed in a cache queue.

[0010] Specifically, after attaching priority identifiers, data scheduling is required. The specific steps are as follows: for data with low priority identifiers, extract its key feature values ​​to generate periodic reports and store the periodic reports in a local cache queue; for security anomaly event data with high priority identifiers, directly place it into a real-time sending queue.

[0011] S2: Intelligent link decision-making and switching based on multi-factor perception; Based on the transmission priority identifier and pre-stored network status information output in step S1, a corresponding communication link is selected for the data.

[0012] Specifically, the pre-stored network status information consists of a high-precision electronic map and a cross-border network protocol adaptation dataset. In the vector data layer of the high-precision electronic map, the geographical boundaries of ground network signal blind spots and weak zones generated based on historical communication data statistics are pre-marked. The cross-border network protocol adaptation dataset contains a list of compatible ground network frequency bands, protocol types, and cross-border handover adaptation parameters for the countries or regions covered by the cross-border transportation route. When a vehicle enters the target country or region through a cross-border node, the communication parameters of the ground network are automatically adjusted based on this dataset.

[0013] Specifically, the satellite communication link established by S-BOX will be activated for data transmission when at least one of the following conditions is met: A: The data is appended with a high-priority identifier; B: Based on the vehicle's current location and the pre-stored network status information, it is predicted that communication via the terrestrial network will be impossible within a preset time threshold.

[0014] Specifically, the prediction steps for condition B include: obtaining the vehicle's current real-time location information; based on the vehicle's real-time location information, querying the pre-stored network status information to see if there are signal blind spots or weak areas on the path ahead, and calculating the estimated arrival time for the vehicle to reach the boundary of the blind spot from its current location; and setting a preset time threshold. If the estimated arrival time is less than or equal to the time threshold, the satellite communication link of S-BOX is activated.

[0015] Specifically, during the selection of a communication link, a first threshold for network strength is preset. For data that is marked with a low priority and is currently transmitted using the terrestrial mobile communication network by default, the signal strength of the terrestrial network is monitored in real time. If the signal strength of the terrestrial network is lower than the first threshold, all or part of the data packets in the low priority data are reallocated to the satellite communication link provided by the S-BOX for transmission.

[0016] Specifically, the first threshold can also be set as a dynamic adjustment value, which can be dynamically calculated and updated based on the success rate of data transmission through the terrestrial network or the cost factor of satellite connection usage within a historical period.

[0017] Specifically, when S-BOX is started, if S-BOX supports multiple satellite communication networks, the optimal network is dynamically selected to establish a connection based on at least one of the following: the current link quality prediction parameters, tariff parameters, and current availability status of each network.

[0018] Specifically, after the S-BOX activation operation is triggered, when the estimated arrival time changes from less than or equal to the time threshold to greater than the sum of the time threshold and the preset hysteresis value, the system will switch back to prioritizing the use of the terrestrial network to avoid frequent switching of the satellite link at the network blind zone boundary.

[0019] Specifically, when both the terrestrial mobile communication network and the satellite communication link are unavailable, short-range direct communication technology is used to send regulatory data packets to one or more neighboring network nodes for multi-hop relay transmission; the neighboring network nodes include at least vehicles that can use the terrestrial mobile communication network and the satellite communication link or roadside units deployed on the roadside and connected to the fixed network.

[0020] Compared with the prior art, the beneficial effects of the present invention are as follows: This invention deploys an edge AI model on the vehicle terminal side to understand and classify the data content itself, and integrates predictions of the network environment to actively schedule the most suitable communication link. This transforms the communication system from a passive "pipeline" into an active "resource scheduler," thereby resolving the contradiction between business continuity and limited resources in cross-border scenarios.

[0021] This invention establishes a dynamic baseline model of a vehicle's normal operating state and performs real-time comparison and anomaly detection on real-time data. This enables accurate identification of safety anomalies and normal states, and allows for differentiated processing and prioritization of the data. For normal data, feature extraction and caching are performed to ensure rapid response to vehicle safety conditions, while significantly reducing the amount of data requiring real-time transmission. This improves safety monitoring efficiency and significantly reduces the overall cost of satellite remote communication.

[0022] This invention, by integrating data priority and high-precision network map information, can proactively predict communication blind spots and switch to satellite links in advance, avoiding data loss caused by network switching delays or failures. This ensures that the monitoring platform can obtain stable and continuous vehicle operation trajectories and real-time status, effectively improving the transparency and controllability of cross-border transportation.

[0023] This invention, based on the intelligent switching strategy of the "ground-satellite" primary and backup links, also introduces vehicle-road cooperative relays as a backup solution in extreme cases. When all main communication links fail, a minimum data reporting capability can still be maintained through multi-hop relays, ensuring that the regulatory data link will not be completely interrupted in most extreme scenarios, thus improving the reliability of the transmission link. Attached Figure Description

[0024] Figure 1 This is a flowchart illustrating the steps of the cross-border transportation supervision and communication method integrating T-BOX and S-BOX of the present invention. Detailed Implementation

[0025] To make the above-mentioned objectives, features and advantages of this application more apparent and understandable, the specific embodiments of this application will be described in detail below with reference to the accompanying drawings.

[0026] The main implementing entity of this invention is a collaborative system consisting of an in-vehicle T-BOX (telematics processor) and an S-BOX (satellite communication module). The T-BOX has a built-in main control processor (MCU), a power management module, and an edge AI model, responsible for core data processing and decision-making. The S-BOX is the satellite communication execution unit. The T-BOX and S-BOX exchange data via an internal high-speed bus or an in-vehicle network.

[0027] like Figure 1 As shown, a cross-border transportation supervision communication method integrating T-BOX and S-BOX specifically includes the following steps: S1: Multi-source data acquisition and edge intelligent processing; The vehicle-mounted T-BOX collects real-time vehicle operating condition data via the vehicle's CAN bus, including but not limited to GPS / BeiDou positioning information, vehicle speed, engine speed, fuel level, braking signal, and data from the vehicle's vibration sensors. The collected real-time vehicle operating condition data is then transmitted to the edge AI model built into the T-BOX, which analyzes the real-time vehicle operating condition data in real time.

[0028] Specifically, the real-time analysis of vehicle operating data by the edge AI model includes the following steps: S101: Baseline modeling of operating conditions; During the initial operation phase of a vehicle, a dynamic baseline model representing the vehicle's normal operating status is trained, either in the cloud or locally, based on normal historical vehicle operating condition data. This model can be a threshold model based on the statistical range of normal data (such as mean, variance), or a more complex lightweight AI model. This model is deployed in a T-BOX and can be a set of feature vectors containing normal data {N1, N2, ..., N...} n The database is}, where each vector represents a normal operating condition mode.

[0029] S102: Implement anomaly detection; The T-BOX processes the collected real-time vehicle operating data. In this embodiment, taking vibration data as an example, it performs a Fast Fourier Transform to convert the vibration data from the time domain to the frequency domain, and extracts the feature vector V_real corresponding to the frequency band of the dynamic baseline model. Then, it calculates the deviation D between the feature vector V_real and the closest normal vector N_best in the dynamic baseline model. The deviation D can be calculated using the Euclidean distance formula.

[0030] Where D represents the deviation, V_real is the feature vector of the vehicle's real-time operating data, and N_best is the closest normal vector in the dynamic baseline model. Before calculating the deviation, an anomaly threshold D_threshold needs to be preset. If the deviation value D eventually exceeds this threshold, it is judged as abnormal; otherwise, it is judged as normal. In addition to calculating the deviation using the Euclidean distance formula, other methods such as cosine similarity can also be used for measurement.

[0031] S103: Event Classification; For detected anomalies, T-BOX classifies them into predefined safety event types based on their data characteristics, including but not limited to severe vibration, sudden braking, or fatigued driving. T-BOX then queries a pre-defined event type-priority mapping table to attach corresponding transmission priority identifiers to abnormal and normal data.

[0032] This relationship mapping table is pre-defined based on expert experience, for example: Events involving immediate safety, such as emergency braking and engine fire, are mapped to the highest priority (level 4). Events such as severe vibrations that foreshadow potential failures are mapped to high priority (level 3). GPS data used for real-time trajectory tracking is mapped to medium priority (level 2). Periodic report data such as average fuel consumption are mapped to low priority (level 1). For ease of description and understanding, in this embodiment, the priority levels in the event type-priority mapping table are logically categorized into two main types: high priority and low priority. High priority covers data types requiring immediate transmission assurance, corresponding to levels 4 and 3 in the mapping table; while low priority covers data types that can tolerate delays or be processed in batches, corresponding to levels 2 and 1. The high-priority and low-priority data mentioned later refer to data categories classified according to this logic. The specific four-level classification and their correspondence with high and low priorities described above are only one preferred embodiment. The core of this invention lies in providing differentiated transmission assurance based on the analysis results of the data content. In actual implementation, any priority classification scheme capable of distinguishing the urgency of data can be adopted.

[0033] T-BOX uses this relationship mapping table to attach a transmission priority identifier to the data, which includes a priority level and a timestamp of data generation. Identified security anomalies are assigned a higher priority identifier to highlight their importance, while normal operating data, such as average fuel consumption, are assigned a lower priority identifier to avoid competing for communication links with more critical security anomalies, thus preventing delays in the transmission of security anomaly data. This improves the system's security and reliability.

[0034] Furthermore, after attaching priority identifiers, the data needs to be scheduled. For data with low priority identifiers, T-BOX extracts its key feature values ​​to generate periodic reports and stores the periodic reports in T-BOX's local cache queue to avoid low-priority data occupying the communication link. For security anomaly event data with high priority identifiers, it is directly placed into the real-time transmission queue to ensure the timely transmission of important high-priority data.

[0035] S2: Intelligent link decision-making and switching based on multi-factor perception; Based on the transmission priority identifier and pre-stored network status information, T-BOX dynamically selects the corresponding communication link for data.

[0036] Specifically, the pre-stored network status information includes at least a set of high-precision electronic maps and a cross-border network protocol adaptation dataset. The vector data layer of the high-precision electronic map pre-marks the geographical boundaries of ground network signal blind spots (no signal) and weak areas (signal strength <-100dBm) based on historical communication data. The cross-border network protocol adaptation dataset contains a list of compatible ground network frequency bands, protocol types, and cross-border handover adaptation parameters for the countries or regions covered by the cross-border transportation route. When a vehicle enters the target country or region through a cross-border node, T-BOX will automatically adjust the ground network communication parameters based on this dataset.

[0037] If the data output in step S1 meets at least one of the following conditions, the satellite communication link established by S-BOX will be activated for data transmission: A: For high-priority data in the real-time transmission queue (corresponding to data with priority level 3 or 4 above), T-BOX will directly decide and instruct S-BOX to activate the satellite communication link for transmission to ensure its immediacy and reliability.

[0038] B: The T-BOX presets a time threshold T_threshold. In this embodiment, the time threshold T_threshold is 30 seconds. This time threshold is sufficient time reserved for the S-BOX to start up, register on the network, and establish the link. The T-BOX acquires the vehicle's position p, instantaneous speed v, and heading angle θ in real time. Based on the current vehicle position p and heading angle θ, it queries the high-precision electronic map to see if there is a boundary point p_b of a ground network blind spot or weak zone on the path ahead. Then, it calculates the shortest spatial distance d from the current vehicle position p to the boundary point p_b of the blind spot or weak zone, and combines it with the instantaneous speed v to calculate the estimated arrival time T. The calculation formula is: T = d / v. If the estimated arrival time T ≤ T_threshold, the T-BOX actively wakes up the S-BOX, sends a pre-connection command to the S-BOX, establishes a satellite communication link in advance, and switches all subsequent data to the satellite communication link to achieve seamless connection and avoid data loss. If the estimated arrival time T > T_threshold, the current link is maintained and monitoring continues.

[0039] Furthermore, to avoid frequent link oscillations caused by signal fluctuations at the blind zone boundary, this invention introduces a hysteresis handover mechanism. The conditions for T-BOX to trigger a switchback to the ground network are more stringent. The specific steps are as follows: a hysteresis value H is preset, which is set to 10 seconds in this embodiment; only when the estimated arrival time T changes from being less than or equal to the time threshold T_threshold to being greater than the sum of the time threshold T_threshold and the hysteresis value H, i.e., T > T_threshold + H, will T-BOX instruct to switch back to preferential use of the ground network. This means that the vehicle must stably move a greater distance from the blind zone boundary (corresponding to an additional H seconds in time) before the system switches back to the ground network, ensuring that the link state only changes after a substantial and continuous improvement in network conditions.

[0040] In addition, during the selection of communication links, a first threshold for network strength is preset. In this example, the first threshold is set to -100dBm. For data that is marked with a low priority and is transmitted using the terrestrial mobile communication network according to the default routing policy, T-BOX will monitor the signal strength of the terrestrial network in real time. The default routing policy refers to the rule set when T-BOX is initialized: except for data marked as high priority (level ≥ 3) or data that triggers predictive handover, other data will attempt to use the terrestrial network for transmission by default.

[0041] The signal strength of the terrestrial network is compared with a first threshold. If the signal strength of the terrestrial network is lower than the first threshold and this state continues for more than a preset stable time window of 5 seconds, in order to prevent data transmission errors or repeated retransmissions, T-BOX will reallocate all or part of the low-priority data in the buffer queue to the satellite link to ensure basic communication success rate.

[0042] Specifically, to avoid the satellite link being instantly saturated and to achieve a balance between cost and reliability, this invention designs different strategies for the aforementioned low-priority data, defining the scope of this data. In practical implementation, T-BOX can retrieve data from the cache queue according to at least one of the following strategies: Waiting time priority strategy: Prioritize the allocation of data packets with the longest waiting time in the buffer queue to the satellite link to ensure maximum data timeliness; Priority segmentation strategy: Data is divided according to priority level, and data with higher priority level will be allocated to satellite links first; Proportional allocation strategy: Based on the available bandwidth and cache queue size of the current satellite link, data is extracted and redistributed to the satellite link according to a dynamically calculated ratio. In this embodiment, the ratio is available bandwidth / total data volume in the queue.

[0043] If the signal strength is higher than or equal to the first threshold, or if the low signal strength state does not last long enough, the original transmission strategy is maintained, and the data continues to be transmitted through the terrestrial network or remains in the buffer queue.

[0044] Furthermore, in addition to being set to a fixed value as described above, the first threshold can also be set to a dynamic threshold. This dynamic threshold can be adjusted in reverse based on the historical success rate statistics of terrestrial network transmission, or adjusted collaboratively based on the cost ratio factor of the satellite-terrestrial link. The function representing the dynamic adjustment logic of the first threshold is as follows: Among them, S_rate is the statistical value of the success rate of data packets transmitted through the terrestrial network recently. The lower the success rate, the higher the first threshold can be adjusted. C_ratio is the ratio factor of the unit data cost between the satellite link and the terrestrial link. The higher the satellite cost, the lower the C_ratio can be adjusted. TH_base is a basic threshold.

[0045] In this embodiment, the following simplified formula can be used for periodic calculation: Where TH is the first threshold, TH_base is the base threshold, S_rate is the most recent terrestrial network transmission success rate statistic, C_ratio is the cost ratio factor between satellite and terrestrial communication, and α and β are weighting coefficients.

[0046] Furthermore, when initiating the S-BOX to establish a satellite communication connection, if the vehicle-mounted S-BOX supports connections to multiple satellite networks, the T-BOX will dynamically select the optimal network to establish a connection based on the real-time link quality, cost parameters, and availability status of each satellite network. In this embodiment, the T-BOX will make a dynamic selection based on a multi-dimensional scoring model: Wherein, Score_i is the score of the i-th satellite network; LinkQuality_i is the estimated quality link of the corresponding satellite network, which can be represented by the signal-to-noise ratio; Cost_i is the satellite network cost; Availability_i is the current availability status of the satellite network, which is 1 if available and 0 if unavailable; w1, w2, and w3 are preset weights. T-BOX selects the network with the highest overall score and issues an instruction to S-BOX to connect to that specific network.

[0047] Furthermore, if the T-BOX detects that neither the ground nor satellite links can interact within a continuous 60-second detection period, it determines that both the current ground mobile communication network and satellite communication link are unavailable. At this point, a backup plan is activated: short-range direct communication technology is used. In this example, a vehicle-mounted V2X interface is selected to send the surveillance data packets to a nearby vehicle or roadside unit with the best signal quality or the closest direction of movement to the network access point as a relay node. The surveillance data packets are then transmitted via multi-hop relay according to a latency-tolerant network protocol. If any ground mobile communication network or satellite communication link becomes available again within the continuous detection period, the relay transmission attempt is immediately stopped, and the system switches back to using the available link.

[0048] When the backup plan is activated, all data to be sent (including high-priority data in the real-time transmission queue and low-priority data in the cache queue) will be uniformly sent into a single intermediate transmission queue. This relay transmission queue is strictly sorted according to the data's priority level from highest to lowest. This ensures that, in extreme cases, the most critical security anomaly data still has the highest chance of transmission, preventing high-priority data from being blocked by a large amount of previously cached low-priority data.

[0049] The above description is merely a preferred embodiment of this application and is not intended to limit the scope of this application. Those skilled in the art may find other optimizations and additional functions in this application. Any modifications, equivalent substitutions, or improvements made within the spirit and principles of this application should be included within the scope of protection of this application.

Claims

1. A T-BOX and S-BOX fusion cross-border transportation supervision communication method, characterized in that: Comprising the following steps: S1: Multi-source data acquisition and edge intelligent processing; The vehicle-mounted T-BOX collects vehicle real-time working condition data through the vehicle CAN bus, and the edge AI model built-in the T-BOX analyzes the vehicle real-time working condition data in real time, and based on the analysis result, the vehicle real-time working condition data is attached with a transmission priority identifier; Wherein, the identified safety abnormal event is attached with a high priority identifier, and the normal working condition data is attached with a low priority identifier; S2: Intelligent link decision and switching based on multi-factor perception; Based on the transmission priority identifier output by step S1 and the pre-stored network state information, the corresponding communication link is selected for the data; Wherein, when the data at least meets one of the following conditions, the satellite communication link established by the S-BOX is started for data transmission: A: The data is attached with a high priority identifier; B: According to the vehicle current position and the pre-stored network state information, it is predicted that the communication through the ground network will be impossible within a preset time threshold. 2.The T-BOX and S-BOX fusion cross-border transportation supervision communication method of claim 1, wherein: In step S1, the edge AI model analyzes the vehicle real-time working condition data in real time, specifically including the following steps: S101: Working condition baseline modeling; Based on normal historical vehicle working condition data, a dynamic baseline model for representing the normal running state of the vehicle is established; S102: Real-time anomaly detection; The collected vehicle real-time working condition data is compared with the dynamic baseline model, the deviation degree between the vehicle real-time working condition data and the vehicle normal running data in the dynamic baseline model is calculated, and the obtained deviation degree is compared with a preset abnormal threshold value. If the deviation degree exceeds the abnormal threshold value, it is determined to be abnormal; S103: Event classification; For the detected abnormality, it is classified into different safety event types according to its data characteristics, and based on a preset event type-priority relationship mapping table, the data is attached with a corresponding transmission priority identifier. 3.The T-BOX and S-BOX fusion cross-border transportation supervision communication method of claim 1, wherein: In step S1, after attaching the priority identifier, data scheduling is also performed, specifically: for the data attached with a low priority identifier, its key feature value is extracted to generate a periodic report, and the periodic report is stored in a local cache queue; for the safety abnormal event data attached with a high priority identifier, it is directly placed in a real-time sending queue. 4.The T-BOX and S-BOX fusion cross-border transportation supervision communication method of claim 1, wherein: In step S2, the pre-stored network state information is a high-precision electronic map and a cross-border network protocol adaptation data set; in the vector data layer of the high-precision electronic map, the geographical boundaries of the ground network signal blind area and weak area based on the historical communication data statistics are pre-labeled; the cross-border network protocol adaptation data set contains the ground network frequency band compatibility list, protocol type and cross-border switching adaptation parameters of the cross-border transportation route covering countries or regions, when the vehicle passes through the cross-border node and enters the target country or region, the communication parameters of the ground network are automatically adjusted based on the data set. 5.The T-BOX and S-BOX fusion cross-border transportation supervision communication method of claim 1, wherein: In step S2, the prediction step of condition B specifically includes: obtaining the real-time position information of the vehicle; based on the real-time position information of the vehicle, querying whether there is a signal blind area or weak area in the front path in the pre-stored network state information, and calculating the estimated arrival time of the vehicle from the current position to the boundary of the blind area; a time threshold is preset, if the estimated arrival time is less than or equal to the time threshold, the satellite communication link of S-BOX is triggered to be enabled. 6.The T-BOX and S-BOX fusion cross-border transportation supervision communication method of claim 5, wherein: After triggering the operation of enabling S-BOX, when the estimated arrival time changes from less than or equal to the time threshold to greater than the sum of the time threshold and a preset hysteresis value, the ground network is switched back to be preferentially used to avoid frequent switching of the satellite link at the boundary of the network blind area.

7. The T-BOX and S-BOX fusion cross-border transportation supervision communication method of claim 1, wherein: In step S2, during the process of selecting the communication link, a first threshold of network strength is first preset, and for the data currently using the ground mobile communication network for transmission by default and being attached with a low priority identifier, the signal strength of the ground network at this time is monitored in real time; If the signal strength of the ground network at this time is lower than the first threshold, the low-priority data is redistributed to the satellite communication link provided by the S-BOX for transmission; If the signal strength is higher than or equal to the first threshold, the original transmission strategy is maintained, and the data continues to be transmitted through the ground network or remains in the cache queue. 8.The T-BOX and S-BOX fusion cross-border transportation supervision communication method of claim 7, wherein: The first threshold can also be set as a dynamic threshold, which can be dynamically calculated and updated according to the success rate of data transmission through the ground network or the use cost factor of the satellite link in a historical period. 9.The T-BOX and S-BOX fusion cross-border transportation supervision communication method of claim 1, wherein: In step S2, when the S-BOX is started, if the S-BOX supports multiple satellite communication networks, at least one of the current link quality estimation parameters, the cost parameters and the current availability state of each network is used to dynamically select the optimal network to establish a connection. 10.The T-BOX and S-BOX fusion cross-border transportation supervision communication method of claim 1, wherein: When the ground mobile communication network and the satellite communication link are both unavailable, the supervision data packet is sent to one or more adjacent network nodes for multi-hop relay transmission by using short-distance direct communication technology; the adjacent network nodes at least include a vehicle that can use the ground mobile communication network and the satellite communication link or a road side unit deployed on the roadside and connected to the fixed network.

Images