Displacement monitoring multi-mode transmission solution system

Abstract

The present disclosure relates to a displacement monitoring multi-mode transmission solution system, comprising: a plurality of monitoring stations, a reference station, a monitoring platform; wherein the monitoring station is configured to collect displacement original information of a monitoring target and transmit it to the reference station; the reference station is configured to: for each set of displacement original information of each monitoring station, determine a solution priority matrix according to a timestamp deviation value, a standardized data volume and an integrity factor, assign a solution resource to each set of displacement original information according to the solution priority matrix and perform solution processing to obtain displacement result data, determine a target transmission link from a plurality of transmission links according to a data priority of the displacement result data and a link quality coefficient, and transmit the displacement result data to the monitoring platform through the target transmission link; the monitoring platform is configured to process based on the displacement result data. According to the technical solution of the present disclosure, the transmission adaptability, solution efficiency and data reliability of the displacement monitoring system are improved.

Description

Technical Field

[0001] This disclosure relates to the field of positioning and displacement monitoring technology, and in particular to a multi-mode transmission and calculation system for displacement monitoring. Background Technology

[0002] Displacement monitoring systems have important applications in fields such as geological disaster prevention and safety maintenance of large-scale projects.

[0003] Currently, displacement monitoring systems suffer from poor transmission adaptability, inefficient calculation, and unreasonable resource allocation in practical applications, making it difficult to meet the high requirements of complex monitoring scenarios. Summary of the Invention

[0004] To address the aforementioned technical problems, this disclosure provides a displacement monitoring multi-mode transmission calculation system.

[0005] This disclosure provides a displacement monitoring multi-mode transmission calculation system, including: Multiple monitoring stations, reference stations, and monitoring platforms; The monitoring station is configured to: collect raw displacement information of the monitored target and transmit it to the base station via radio; The reference station is configured to: for each set of original displacement information of each monitoring station, determine the solution priority matrix based on the timestamp deviation value, the standardized data volume, and the integrity factor; wherein, the timestamp deviation value is obtained by standardizing the difference between the current data timestamp and the reference timestamp, the standardized data volume is obtained by the ratio of the current data volume to the maximum single set of data volume, and the integrity factor is calculated by the proportion of effective data fields to the total number of fields; According to the solution priority matrix, solution resources are allocated to each group of original displacement information and solution processing is performed to obtain displacement result data; Based on the accuracy level and timeliness requirements of the displacement data, data priority is determined, and link quality coefficients for various transmission links are calculated. Based on the data priority and the link quality coefficient, a target transmission link is determined from the multiple transmission links so that the displacement results data can be transmitted to the monitoring platform through the target transmission link; The monitoring platform is configured to receive displacement data transmitted by the base station, and to perform data processing, early warning, and issue demand commands based on the displacement data.

[0006] According to the technical solution provided in the embodiments of this disclosure, dynamic matching of transmission strategies is achieved through adaptive optimization and intelligent adaptation of multiple transmission methods, which improves the stability and adaptability of data transmission in complex environments, effectively copes with scenarios such as complex terrain and electromagnetic interference, and ensures real-time and continuous transmission of key data. Furthermore, task priority allocation and parallel processing are achieved through intelligent scheduling of edge computing, which realizes reasonable allocation of computing resources, adapts to edge computing power and reduces cloud dependence, improves the data processing efficiency of multiple monitoring stations, and achieves dual optimization of computing efficiency and accuracy, thereby improving the transmission adaptability, computing efficiency and data reliability of the displacement monitoring system. Attached Figure Description

[0007] The accompanying drawings, which are incorporated in and form a part of this specification, illustrate embodiments consistent with this disclosure and, together with the description, serve to explain the principles of this disclosure.

[0008] To more clearly illustrate the technical solutions in the embodiments of this disclosure or the prior art, the accompanying drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, for those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0009] Figure 1 This is a schematic diagram of a displacement monitoring multi-mode transmission calculation system provided in an embodiment of this disclosure. Detailed Implementation

[0010] To better understand the above-mentioned objectives, features, and advantages of this disclosure, the solutions disclosed herein will be further described below. It should be noted that, unless otherwise specified, the embodiments and features described herein can be combined with each other.

[0011] Numerous specific details are set forth in the following description in order to provide a full understanding of this disclosure, but this disclosure may also be implemented in other ways different from those described herein; obviously, the embodiments in the specification are only some, and not all, of the embodiments of this disclosure.

[0012] Figure 1 This is a schematic diagram of a displacement monitoring multi-mode transmission calculation system provided in an embodiment of this disclosure, as shown below. Figure 1 As shown in the embodiments of this disclosure, the displacement monitoring multi-mode transmission calculation system may include: multiple monitoring stations, a reference station, and a monitoring platform.

[0013] In this embodiment, the monitoring station is configured to: collect raw displacement information of the monitored target and transmit it to the base station via radio. The base station is configured to: receive the raw displacement information transmitted by the monitoring station, perform displacement data calculation through an edge server, and transmit the displacement result data based on multiple transmission links. The monitoring platform is configured to: receive the displacement result data transmitted by the base station, and perform data processing, early warning, and issue demand commands based on the displacement result data.

[0014] The base station includes an edge computing intelligent scheduling unit, which is used to: allocate computing tasks to data transmitted from multiple monitoring stations to achieve parallel computing, and perform lightweight iterative optimization of the computing model. Optionally, the computing process is as follows: for each set of raw displacement information from each monitoring station, a computing priority matrix is ​​determined based on the timestamp deviation, standardized data volume, and integrity factor. Computing resources are allocated to each set of raw displacement information according to the computing priority matrix, and computing is performed to obtain displacement result data.

[0015] As an example, parallel task allocation constructs a solution priority matrix based on the timestamps, data volume, and data integrity of data from each monitoring station, as shown in the following expression:

[0016] In the formula The priority for solving the j-th group of data at the i-th monitoring station (unitless, range [0,1]). For timestamp weights (unitless, range [0,1]), For data volume weights (unitless, range [0,1]), This is the timestamp deviation value of the j-th group of data from the i-th monitoring station (unitless, range [0,1]). The standardized data volume (unitless, range [0,1]) of the j-th group of data from the i-th monitoring station. Let be the integrity factor of the j-th group of data from the i-th monitoring station (unitless, range [0,1]). Adjust the timestamp sensitivity factor (no unit, default 1.2). Adjust the data volume sensitivity factor (unitless, default 0.8). Lightweight model iteration removes redundant parameters from the solution model, retains the core computational links, and ensures that the computing power of edge servers is adapted.

[0017] The priority matrix expression is optimized based on three dimensions: timeliness, data size, and completeness. The timestamp deviation is obtained by standardizing the difference between the current data timestamp and the baseline timestamp. The standardized data volume is obtained by the ratio of the current data volume to the largest single data set volume. The completeness factor is calculated as the proportion of valid data fields to the total number of fields. Weights , satisfy Default value , Adjustment coefficient , Used to adapt to priority sensitivity in different scenarios Enhance timeliness sensitivity To reduce the impact of excessive data growth on priority, priority differentiation is enhanced through non-linear weighting.

[0018] The base station includes a multi-transmission mode intelligent adaptation unit, which is used to: determine the data priority according to the accuracy level and timeliness requirements of the displacement data, calculate the link quality coefficient of multiple transmission links, and then determine the target transmission link from multiple transmission links according to the data priority and the link quality coefficient, so as to transmit the displacement data to the monitoring platform through the target transmission link.

[0019] In one embodiment of this disclosure, the base station is specifically configured to calculate the link quality coefficient of multiple transmission links through the following steps: for each transmission link, determining the actual transmission rate, data packet loss rate and stability factor of the transmission link, and calculating the link quality coefficient of the transmission link based on the actual transmission rate, data packet loss rate and stability factor of the transmission link.

[0020] As an example, the dynamic link quality detection collects the actual transmission rate and data packet loss rate of each transmission link at a preset period, and calculates the link quality coefficient, as shown in the following expression:

[0021] In the formula is the quality coefficient of the kth transmission link (unitless, range [0,1]). This represents the actual transmission rate (in bytes per second) of the k-th transmission link. Let be the data packet loss rate of the k-th transmission link (unitless, range [0,1]). For transmission rate weights (unitless, range [0,1]), The link reliability weight (unitless, range [0,1]). is the stability factor of the k-th transmission link (unitless, range [0,1]).

[0022] The link quality coefficient expression is based on an evaluation model constructed using logarithmic enhanced transmission rate sensitivity and stability verification. The actual transmission rate is calculated by the number of data bytes successfully transmitted per unit time. The packet loss rate is obtained by the ratio of lost data bytes to the total number of transmitted data bytes. The stability factor is calculated by the average transmission success rate over multiple consecutive detection periods, such as three consecutive detection periods. Weights , Default value , ,satisfy ; To avoid This addresses computational anomalies and enhances the performance differentiation of low-speed links.

[0023] Optionally, data priority can be divided into multiple levels based on the accuracy level and timeliness requirements of the solution results. For example, displacement data can be divided into first-level data, second-level data, and third-level data according to data priority. Multiple transmission links include satellite communication links, 4G links, and short message links. The base station is specifically configured to transmit first-level data to the monitoring platform via satellite communication links, second-level data via 4G links, and third-level data via short message links.

[0024] In one embodiment of this disclosure, the monitoring station includes a radio transmission adaptive optimization unit, which is used to: detect the radio channel quality in real time before transmitting to the base station via radio, and then perform lightweight compression processing on the collected original displacement information according to the channel quality to adapt to the radio bandwidth.

[0025] Specifically, the system monitors radio channel quality in real time by collecting signal strength and bit error rate parameters from the radio link, generating a channel quality assessment value. Lightweight data compression reduces data volume by preserving key feature components from the original displacement information.

[0026] In one embodiment of this disclosure, the base station further includes an edge data intelligent caching unit, which is used to: perform hierarchical caching of the received data and the calculated displacement result data based on their access frequency, time span, and data correlation factor for each set of original displacement information and / or displacement result data; calculate a heat value based on the data access frequency, time span, and data correlation factor; and perform hierarchical caching of the original displacement information and / or displacement result data based on the heat value.

[0027] As an example, the popularity score is calculated based on the frequency of data access, time span, and data relevance, using the following expression:

[0028] In the formula This represents the popularity value of the m-th data group (unitless, range [0,1]). This represents the number of times the data was accessed (in units of time). Total number of accesses to all data (in times). Current time (in hours). Data caching start time (in hours). For access frequency weight (unitless, range [0,1]), For time decay weights (unitless, range [0,1]), This is the time decay constant (unit: hours). This is the data correlation factor (unitless, range [0,1]).

[0029] The popularity value expression integrates frequency weighting, exponential time decay, and data relevance to construct a popularity evaluation model. Data access frequency is calculated as the ratio of current data access frequency to the total number of accesses for all data. Time span is calculated as the difference between the current time and the data cache start time. The data relevance factor is calculated based on the proportion of reference associations between the current data and cached data. Weighting coefficients are also used. , satisfy Default value , Time decay constant The default value is 24 hours, which controls the rate at which heat decays over a period of time. It achieves a non-linear characteristic where the heat decays faster over time, unlike traditional linear decay.

[0030] The edge data intelligent caching unit is also used to: intelligently mark interrupted transmission data. Optionally, the intelligent marking of interrupted transmission records the interruption location by adding a unique identifier to the data segment whose transmission was interrupted, and locates the data segment based on the identifier when resuming transmission.

[0031] In one embodiment of this disclosure, the monitoring platform includes a displacement data early warning linkage unit. This unit is used to: dynamically adjust the early warning threshold based on displacement data, generate early warning information, and push it back to the base station. Optionally, it performs mean scaling based on the mean and standard deviation of historical displacement data to obtain a mean scaling result, calculates the skewness coefficient based on the third central moment of historical displacement data, and then corrects the mean scaling result based on the skewness coefficient to obtain a dynamic early warning threshold. It then generates early warning information based on the displacement data and the dynamic early warning threshold, and transmits the early warning information to the base station.

[0032] As an example, the warning threshold is dynamically adjusted based on the statistical distribution of historical displacement data, and the expression is as follows:

[0033] In the formula The warning threshold (unit: millimeters) This represents the average of historical displacement data (unit: millimeters). The standard deviation of historical displacement data (in millimeters). The warning coefficient is a unitless coefficient; it is 2.1 for general scenarios and 1.6 for high-risk scenarios. This is the skewness correction factor (unitless, default 0.15). This represents the skewness coefficient of historical displacement data (unitless, range [-3, 3]). The early warning information includes the displacement exceeding the limit value and the time of occurrence, and is pushed to the base station via a multi-transmission intelligent adaptation unit.

[0034] The warning threshold expression is based on a threshold model constructed using mean scaling and skewness correction. The skewness coefficient is calculated using the third central moments of historical displacement data, and the formula is as follows:

[0035] (in the formula) This is the i-th historical displacement data, in millimeters; This represents the total amount of historical displacement data (unit: sets), used to correct threshold bias in non-normally distributed data. Skewness correction coefficient. The default value is 0.15, which balances the impact of skewness on weights. Based on the security level setting of the monitoring scenario, the threshold is dynamically adapted through mean scaling.

[0036] In one embodiment of this disclosure, the monitoring platform further includes a multi-link data unified access unit, which is used to: automatically adapt the format and calibrate the receiving timing of data transmitted by the base station through different transmission methods.

[0037] The data format is automatically adapted and parsed for the data packet structure corresponding to each transmission method, extracting core data fields and converting them into a unified data format. Reception timing calibration, based on the timestamps inherent in the data, corrects timing deviations caused by different transmission links, ensuring that the data is arranged in the order of acquisition time.

[0038] In one embodiment of this disclosure, the monitoring station is further configured to send data volume forecast information to the base station before transmitting the original displacement information. The base station is also configured to pre-allocate computational power based on the data volume forecast information and the computation priority matrix, and to provide feedback on the computation pressure status to the monitoring station. The monitoring station is further configured to adjust the compression ratio of the lightweight compression processing based on the computation pressure status. Thus, a radio transmission computational resource pre-adaptation mechanism is established between the monitoring station and the base station. The monitoring station sends data volume forecast information to the base station before transmitting data, and the base station pre-allocates computational power based on the forecast information and the computation priority matrix expression. The base station provides real-time feedback on the computation pressure status to the monitoring station, and the monitoring station adjusts the compression ratio of the lightweight data compression based on the feedback results to ensure that data transmission matches computational resources.

[0039] In one embodiment of this disclosure, the monitoring platform is further configured to: provide real-time feedback on data reception status to the base station, wherein, when reception is abnormal, the base station is triggered to switch transmission links or retransmit; and issue accuracy requirement instructions to the base station according to actual application needs. The base station is also configured to: adjust the calculation accuracy parameters according to the accuracy requirement instructions. Thus, a multi-transmission data requirement linkage mechanism is established between the base station and the monitoring platform. The monitoring platform provides real-time feedback on data reception status to the base station, and when reception is abnormal, the base station is triggered to switch transmission links or retransmit. The monitoring platform issues accuracy requirement instructions to the base station according to actual application needs, and the base station adjusts the calculation accuracy parameters of the edge calculation intelligent scheduling unit according to the instructions, achieving adaptation between requirements and calculation.

[0040] In one embodiment of this disclosure, the base station performs calculations via an edge server. The base station is further configured to: add a precision level marker to the displacement result data when outputting it; and feed back transmission delay data to the edge server so that the edge server adjusts the output frequency of the displacement result data based on the transmission delay data. Thus, a calculation result transmission strategy adaptation mechanism is established between the edge server of the base station and the multi-transmission mode intelligent adaptation unit. The edge server adds a precision level marker to the displacement result data when outputting it. The multi-transmission mode intelligent adaptation unit matches the corresponding transmission link bandwidth based on the precision level marker and the link quality coefficient formula. The multi-transmission mode intelligent adaptation unit feeds back the transmission delay data to the edge server, and the edge server adjusts the output frequency of the displacement result data based on the delay data, balancing transmission efficiency and data timeliness.

[0041] In one embodiment of this disclosure, the base station is further configured to: optimize the calculation threshold of the calculation process based on the early warning information; and if abnormal data is detected during the calculation process, feed the abnormal information back to the monitoring platform to trigger the data review process of the monitoring platform. Thus, an early warning calculation parameter optimization mechanism is established between the monitoring platform and the base station. The monitoring platform sends early warning information back to the base station, the base station optimizes the calculation threshold of the edge calculation intelligent scheduling unit based on the early warning information, and if abnormal data is detected during the calculation process, the base station feeds the abnormal information back to the monitoring platform, triggering the monitoring platform to review the corresponding data, ensuring data reliability.

[0042] According to the technical solution of this disclosure, through radio transmission adaptive optimization and multi-transmission mode intelligent adaptation mechanism, dynamic matching of transmission strategy with channel quality and data requirements is achieved, improving the stability and adaptability of data transmission in complex environments, effectively avoiding the limitations of a single transmission mode, and ensuring the real-time transmission of key monitoring data. By leveraging the priority allocation and lightweight iterative optimization of the edge computing intelligent scheduling unit, reasonable allocation of computing resources is achieved, improving the computing efficiency of data from multiple monitoring stations. Simultaneously, the optimized design of edge data intelligent caching and breakpoint resumption ensures the rationality of data storage and the continuity of transmission, avoiding data loss or duplicate transmission, and improving the integrity and reliability of monitoring data. Through the collaborative interaction mechanism between modules, end-to-end adaptation optimization of transmission, computing, and early warning is achieved, enabling the system to dynamically adjust working parameters according to the actual monitoring scenario, adapting to monitoring tasks with different environments and requirements, improving the system's versatility and practicality, and providing accurate and continuous monitoring data support for subsequent security risk assessment.

[0043] The following description uses a specific example, Implementation Example 1.

[0044] The technical system corresponding to this embodiment is a BeiDou displacement monitoring multi-mode transmission and calculation system, specifically including: a BeiDou monitoring station, used to collect raw displacement information of the monitored target and transmit it via radio; a BeiDou reference station, used to receive data transmitted by the BeiDou monitoring station, complete displacement data calculation through an edge server, and transmit data using 4G, BeiDou short message, or satellite communication methods; and a monitoring platform, used to receive data transmitted by the BeiDou reference station, complete data processing, early warning, and demand command issuance. The BeiDou monitoring station also includes a radio transmission adaptive optimization unit; the BeiDou reference station also includes an edge calculation intelligent scheduling unit, a multi-transmission mode intelligent adaptation unit, and an edge data intelligent caching unit; the monitoring platform also includes a multi-link data unified access unit and a displacement data early warning linkage unit; a radio transmission calculation resource pre-adaptation mechanism is established between the BeiDou monitoring station and the BeiDou reference station; a multi-transmission data demand linkage mechanism is established between the BeiDou reference station and the monitoring platform; a calculation result transmission strategy adaptation mechanism is established between the edge server of the BeiDou reference station and the multi-transmission mode intelligent adaptation unit; and an early warning calculation parameter optimization mechanism is established between the monitoring platform and the BeiDou reference station.

[0045] After the system starts, the initial configuration of each module is completed first, and the initial values ​​of the core parameters are determined: in the edge computing intelligent scheduling unit, the weights Initially set to 0.55. Initially set to 0.45, adjustment coefficient Initially set to 1.2. Initially set to 0.8; in the multi-transmission mode intelligent adaptation unit, link performance weight. Initially set to 0.6. Initially set to 0.4; in the edge data intelligent caching unit, the weight coefficient... Initially set to 0.5. Initially set to 0.5, time decay constant Initially set to 24 hours; in the displacement data early warning linkage unit, the skewness correction coefficient Initially set to 0.15, this is the warning coefficient for general scenarios. Set to 2.1, high-risk scenario warning coefficient Set to 1.6. After initialization, the system enters a continuous running state, and each module works collaboratively according to preset logic.

[0046] The core function of the Beidou monitoring station is achieved through the collaboration of the data acquisition unit and the radio transmission adaptive optimization unit. First, the data acquisition unit collects the raw displacement information of the monitored target at a preset sampling frequency. The sampling frequency can be dynamically adjusted based on the subsequent feedback of the pressure status. The initial sampling frequency is set to 1 time / second. During the acquisition process, the radio transmission adaptive optimization unit simultaneously performs radio channel quality detection. Through its built-in channel detection module, it collects two core parameters of the radio link in real time: signal strength and bit error rate. The signal strength is obtained instantaneously through the RF receiving module and averaged over 100ms. The bit error rate is obtained by calculating the ratio of the number of erroneous bits received per unit time to the total number of bits (calculation logic correct, range [0,1]). Based on the collected signal strength and bit error rate, a channel quality assessment value is generated, expressed as follows:

[0047] In the formula This is a channel quality assessment value (unitless, range [0,1]). The actual signal strength (unit: dBm, typical range [-120, -30]). This is the minimum signal strength corresponding to the system's receiving sensitivity (unit: dBm, default -110 dBm). This represents the ideal maximum signal strength (unit: dBm, default -40 dBm). The bit error rate (unitless, range [0,1]). The range is [0,1], which comprehensively reflects the channel transmission capability.

[0048] Based on the generated channel quality assessment value, the radio transmission adaptive optimization unit performs lightweight compression processing on the original displacement information. The compression algorithm is an improvement on the discrete cosine transform, and the specific process is as follows: First, the original displacement information is divided into fixed-length data blocks according to the time series, and each data block contains N sampling points; a discrete cosine transform is performed on each data block to obtain the frequency domain coefficient matrix; the proportion of coefficients to be retained is determined according to the channel quality assessment value. When, retain the top 30% of high-frequency coefficients, when When the coefficient is between 0.5 and 0.8, the top 20% of high-frequency coefficients are retained. At that time, the top 10% of high-frequency coefficients are retained; the retained coefficients are quantized and encoded to generate compressed raw data. After compression, before transmitting the compressed raw data, the Beidou monitoring station sends a data volume forecast to the Beidou reference station via radio. This information includes the length of the data block to be transmitted, the sampling time range, and the data integrity identifier, providing a basis for the pre-allocation of computing power by the Beidou reference station.

[0049] The BeiDou reference station receives compressed raw data and data volume forecast information transmitted from the BeiDou monitoring station via a radio receiving unit. During reception, data integrity is verified simultaneously by comparing checksums to confirm whether data is lost or erroneous. If an anomaly is found, a retransmission request is sent to the BeiDou monitoring station. After data reception is complete, the edge computing intelligent scheduling unit initiates the computing task allocation process, first calculating the timestamp deviation based on the sampling time range in the data volume forecast information. In the formula This is the timestamp (in seconds) for the collection of the j-th data set at the i-th monitoring station. The system reference timestamp (unit: seconds) for the BeiDou reference station. The preset maximum allowable time deviation threshold (unit: seconds, default value is 30 seconds) ensures The range is normalized to [0,1]. Then, the standardized data volume is calculated. In the formula This represents the number of bytes (in bytes) of the compressed original data for the j-th group at the i-th monitoring station. This function normalizes the data volume to reflect the maximum number of bytes in a single data set supported by the system (unit: bytes, default value is 1024 bytes).

[0050] Meanwhile, the edge computing intelligent scheduling unit calculates the integrity factor. In the formula This represents the number of valid fields (in units) in the j-th group of data from the i-th monitoring station. This represents the total number of fields in the data block (in units of fields). Valid fields are determined by field identifier checks to ensure that no missing or erroneous fields are included in the valid field count. The range is [0,1]. Based on the above parameters, the solution priority is calculated through the solution priority matrix expression. The optimization objectives are the timeliness of the solution task, data size, and completeness. The timestamp deviation reflects the timeliness of the data; a smaller value indicates stronger timeliness. The standardized data volume reflects the complexity of data processing; a larger value indicates more computing power is required. The completeness factor reflects the reliability of the data; a larger value indicates more reliable data. Weights , satisfy By adjusting the weights, the impact of timeliness and data scale is balanced; the adjustment coefficients are adjusted. , Using a non-linear exponential form, Enhance timeliness sensitivity To reduce the impact of excessive data growth on priority, the default value is used. , This makes the priority allocation more in line with the actual solution requirements.

[0051] Based on the solution priority The edge computing intelligent scheduling unit distributes the computing tasks of multiple monitoring stations to multiple computing cores of the edge server, achieving parallel computing. The edge server's computing process is based on the BeiDou relative positioning principle. First, the compressed original data is decompressed and restored to reconstruct the time-series data of the original displacement information. Then, combined with the known precise coordinates of the BeiDou reference station (error ≤ ±0.1mm), carrier phase differential technology is used to calculate the real-time coordinates of the monitoring station (computation accuracy ≤ ±1mm). The displacement is calculated by the difference between coordinates at adjacent time points, yielding the displacement result data. During the computing process, the edge computing intelligent scheduling unit performs lightweight iterative optimization of the computing model, analyzing the contribution of each parameter in the model. The contribution calculation formula is as follows:

[0052] In the formula The contribution of the p-th parameter (unitless, range [0,1]). The calculation deviation (in millimeters) after removing this parameter. Total number of model parameters (unit: number of parameters), removing contribution factors Redundant parameters are removed, and the core computing link is retained. The core computing link includes two key links: coordinate solution and displacement calculation. This ensures that the computing power of the edge server is adapted. The iteration cycle is set to 24 hours. After each iteration, the optimization effect is verified by comparing with the historical best solution result.

[0053] During the calculation process, the multi-transmission mode intelligent adaptation unit of the Beidou reference station simultaneously conducts transmission link quality detection, collecting the actual transmission rates of three transmission links—4G, Beidou short message, and satellite communication—at a preset period of 60 seconds. Data packet loss rate Actual transmission rate The calculation is based on the cumulative number of bytes of data packets successfully transmitted per unit time. In the formula This represents the number of bytes transmitted (in bytes) on the k-th link at second t. The detection cycle (unit: seconds) (seconds). Data packet loss rate In the formula This represents the number of data packets lost (in units) on the k-th link at second t. The number of data packets sent by the k-th link in second t (unit: packets) (calculation logic correct, range [0,1]). Simultaneously, calculate the stability factor. In the formula Let be the transmission success rate (unitless, range [0,1]) of the k-th link in the nth detection period. , The number of data packets successfully received during this period (unit: packets). The range is [0,1], reflecting the stability of the link.

[0054] Based on the above parameters, the link quality coefficient is calculated using the link quality coefficient expression, where a logarithmic function is employed. Perform nonlinear transformation on the transmission rate to avoid The calculation anomaly occurred at that time, while enhancing the low-speed link ( Performance differentiation; Reflects the reliability of the link; a higher value indicates higher reliability; stability factor. Perform a secondary verification of link quality to avoid misjudging a link as instantaneously high-quality. Weighting , Different importance is assigned to transmission rate and reliability, with default values. , Prioritize ensuring transmission speed.

[0055] Based on the accuracy level and timeliness requirements of the calculation results, data priority is divided into three levels: Level 1 data consists of displacement data with an accuracy better than ±1mm and a timeliness requirement of ≤5 minutes; Level 2 data consists of displacement data with an accuracy of ±1-3mm and a timeliness requirement of ≤15 minutes; and Level 3 data consists of raw and output data with an accuracy ≥3mm or non-real-time requirements (the classification is reasonable and adaptable to different transmission needs). The multi-transmission mode intelligent adaptation unit adjusts the data priority based on data priority and link quality coefficient. Matching transmission method, when When (satellite communication) ≥ 0.7, primary data will preferentially use satellite communication; when When (4G) ≥ 0.6, secondary data uses 4G; when When the (BeiDou short message) quality coefficient is ≥0.5, the third-level data uses BeiDou short message. If the quality coefficient of a certain link does not meet the threshold requirement, it will automatically switch to the suboptimal link.

[0056] The BeiDou reference station's edge data intelligent caching unit classifies and caches the received compressed raw data and the calculated displacement results. The caching medium adopts a combined architecture of solid-state drives (SSDs) and flash memory arrays. SSDs are used to store frequently accessed data, while flash memory arrays are used to store less frequently accessed data. The cached data popularity classification is calculated using a popularity value expression, where the data access frequency ratio reflects the data usage demand. It is an exponential time decay term. This is the time decay constant, with a default of 24 hours, which enables the data popularity to decay non-linearly over time, with the decay accelerating the longer the time. For data correlation factor, In the formula This represents the number of reference associations (in times) between the m-th data group and other cached data. The maximum number of references (in times) across all data points, reflecting the correlation between data. Weighting coefficient. , satisfy Default value , To balance the impact of access frequency and timeliness.

[0057] Based on heat value The data was divided into high-popularity categories ( Medium heat () Low heat () Level 3: High-intensity data is stored on solid-state drives, medium- and low-intensity data is stored on flash memory arrays, and low-intensity data is automatically cleaned up if it is not accessed for more than 72 hours to release storage resources.

[0058] In response to potential interruptions during transmission, the edge data intelligent caching unit adds a unique identifier to the interrupted data segment. The identifier adopts a combination format of "base station ID-monitoring station ID-data block number-time stamp" to ensure global uniqueness. At the same time, it records the byte offset (unit: bytes) of the interruption position and the number of bytes already transmitted (unit: bytes). When transmission is resumed, the cached untransmitted data segment is queried based on the unique identifier, and transmission continues from the recorded interruption position to avoid data duplication or loss.

[0059] The BeiDou reference station transmits displacement data (or raw data) to the monitoring platform via a matched transmission method. Upon receiving the data, the monitoring platform's multi-link data unified access unit first performs automatic data format adaptation. Different transmission methods correspond to different data packet structures: 4G data packets use JSON format, BeiDou short message data packets use binary format, and satellite communication data packets use XML format. The multi-link data unified access unit has built-in parsers for these three formats. By identifying the format identifier field in the data packet header, it calls the corresponding parser to parse the data packet structure and extract core data fields. These core data fields include the monitoring station ID, collection timestamp, displacement value, accuracy level, and data integrity identifier. The extracted core data fields are then converted into a system-wide unified binary data format to ensure data format consistency.

[0060] After format adaptation is completed, the multi-link data unified access unit performs reception timing calibration. Due to differences in transmission delays across different transmission links (satellite communication delay approximately 0.5-1 second, 4G delay approximately 0.1-0.3 seconds, and BeiDou short message delay approximately 1-3 seconds), data with the same acquisition timestamp may exhibit reception timing discrepancies. Timing calibration is based on the data's built-in acquisition timestamps, constructing a timing calibration model: first, extracting the acquisition timestamps of all received data. (Unit: seconds), sorted in ascending order by timestamp to obtain the theoretical timing sequence; calculate the actual reception time of each data point. (Unit: seconds) and theoretical reception time Difference (unit: seconds) In the formula Estimation of reception time based on adjacent data using linear interpolation; according to The data storage timing is corrected by reordering the data according to the theoretical timing corresponding to the collection timestamp, ensuring the consistency of data timing in subsequent processing stages.

[0061] After time-series calibration is completed, the data is transmitted to the displacement data early warning and linkage unit of the monitoring platform. This unit first performs statistical analysis on the displacement data to construct a historical data sample set. The sample set contains displacement data from the past 30 days, with a sample size of no less than 1000 sets. The mean of the historical displacement data is then calculated based on the sample set. (in the formula) This is the i-th historical data point, in millimeters. (Sample size, unit: groups), standard deviation Unbiased standard deviation is used to improve statistical accuracy. Simultaneously, the skewness coefficient of historical displacement data is calculated. , The range is [-3, 3], used to determine the symmetry of the data distribution. The data shows a right-skewed distribution. The distribution is left-skewed. It approximates a normal distribution.

[0062] Based on the above statistical parameters, the warning threshold is calculated using the warning threshold expression, where the mean is used. Based on, through Implement standard deviation scaling of the threshold to reflect the impact of data dispersion on the threshold; This is a skewness correction term used to correct threshold bias in non-normally distributed data, preventing false alarms or missed alerts due to data skewness. Skewness correction coefficient. The default value is 0.15, which is the weight for balancing skewness; warning coefficient. Based on the security level settings of the monitoring scenario, general scenario High-risk scenarios By adjusting The warning sensitivity is adapted to different scenarios.

[0063] Current displacement data (Unit: mm) and warning threshold (Unit: millimeters) for comparison, when At that time, an early warning message is generated, which includes the monitoring station ID, data collection timestamp, current displacement value, and threshold. Exceeding the standard ( Core fields include (unit: millimeters), precision level, etc.

[0064] The generated early warning information is pushed back to the BeiDou reference station through a multi-transmission mode intelligent adaptation unit. The push link prioritizes the link consistent with the data transmission link. If the quality coefficient of this link is... If the connection fails, the system will automatically switch to another available link. After receiving the early warning information, the BeiDou reference station will simultaneously store it in the edge data intelligent cache unit and send an early warning notification to the corresponding BeiDou monitoring station via radio to guide on-site personnel to pay attention to the status of the monitored target.

[0065] During continuous system operation, the interaction mechanisms between modules operate dynamically according to preset logic to ensure overall system performance optimization. In the pre-adaptation mechanism for radio transmission and computational resources between the BeiDou monitoring station and the BeiDou reference station, the intelligent scheduling unit for edge computation at the BeiDou reference station continuously monitors the CPU utilization and memory usage of the edge servers and calculates the computational pressure value.

[0066] In the formula To calculate the pressure value (unitless, range [0,1]), This represents the current CPU utilization (unitless, range [0,1]). This represents the maximum allowed CPU utilization (no unit, default is 90%). This represents the current memory usage (unitless, range [0,1]). This represents the maximum allowed memory usage (no unit, default is 85%, which is within the safe range for memory usage). The range is [0,1]. When At that time, it sends high-pressure feedback information to the Beidou monitoring station; when When low pressure is detected, a low-pressure feedback message is sent. After receiving the feedback message, the BeiDou monitoring station adjusts the compression ratio of the lightweight data compression: the compression ratio is increased under high pressure (e.g., when the channel quality assessment value is 0.6, the compression ratio is increased from 1:5 to 1:8), and the compression ratio is decreased under low pressure (e.g., when the channel quality assessment value is 0.6, the compression ratio is decreased from 1:5 to 1:3), balancing the data transmission volume and the computational resource load.

[0067] In the multi-transmission data demand linkage mechanism between the BeiDou reference station and the monitoring platform, the real-time statistical data reception success rate of the multi-link data unified access unit of the monitoring platform is calculated using the following formula: In the formula Data reception success rate (unitless, range [0,1]). The number of data packets successfully received (unit: packets). Number of data packets sent to the BeiDou reference station (unit: packets). When receiving abnormal data, the system sends feedback to the BeiDou reference station, including the abnormal link identifier and the lost data packet number. Upon receiving the feedback, if the abnormal link is 4G, the BeiDou reference station switches to satellite communication or BeiDou short message service and retransmits the lost data packet; if the abnormal link is satellite communication or BeiDou short message service, it directly retransmits the lost data packet. Simultaneously, based on actual application needs (such as high-precision data required during engineering acceptance and a balance between accuracy and efficiency during daily monitoring), the monitoring platform issues accuracy requirement instructions to the BeiDou reference station. These instructions include the target accuracy level (e.g., ±0.5mm, ±1mm, ±2mm, with the accuracy level meeting actual requirements). Upon receiving the instructions, the BeiDou reference station adjusts the solution accuracy parameters of the edge solution intelligent scheduling unit. For example, when the target accuracy is ±0.5mm, the number of solution iterations is increased (from 3 to 5) to retain more carrier phase observations; when the target accuracy is ±2mm, the number of solution iterations is reduced (from 3 to 2) to improve solution efficiency.

[0068] In the adaptation mechanism for the transmission strategy of calculation results between the edge server of the Beidou reference station and the intelligent adaptation unit for multiple transmission modes, the edge server labels the displacement result data according to the calculation accuracy: accuracy better than ±0.5mm is labeled as L1, ±0.5-1mm as L2, ±1-2mm as L3, and ≥2mm as L4. After receiving the labeled displacement result data, the intelligent adaptation unit for multiple transmission modes combines it with the link quality coefficient. Matching transmission link bandwidth: L1 level data requires a link bandwidth of ≥1Mbps, with satellite communication or 4G full-bandwidth transmission preferred; L2 level data requires a link bandwidth of ≥500kbps, with 4G or satellite communication medium-bandwidth transmission preferred; L3 level data requires a link bandwidth of ≥100kbps, with 4G or BeiDou short message service optional; L4 level data has no bandwidth limitations, with BeiDou short message service preferred to save communication costs. Simultaneously, the multi-transmission mode intelligent adaptation unit calculates the transmission delay of each link. (Time difference from data packet transmission to receipt confirmation, unit: seconds), and fed back to the edge server; the edge server adjusts the output frequency of the displacement data according to the transmission delay: when When the second is 1 second, the output frequency remains at the initial value (1 time / second); when 1 second ≤ At 1 second, the output frequency drops to 0.5 times / second; when At a certain time, the output frequency drops to 0.2 times / second to avoid data accumulation due to transmission delay.

[0069] In the early warning calculation parameter optimization mechanism between the monitoring platform and the BeiDou reference station, the monitoring platform transmits data such as the exceedance magnitude and frequency of occurrence in the early warning information back to the BeiDou reference station. After receiving the data, the BeiDou reference station optimizes the calculation threshold of the edge calculation intelligent scheduling unit. For example, if a monitoring station issues three consecutive early warnings with exceedance magnitudes ≥2mm, the integrity factor of the monitoring station's data will be adjusted. The calculation threshold is increased (the proportion of valid fields is increased from 80% to 90%) to enhance data reliability verification; if a monitoring station has no early warnings for an extended period (more than 7 days), the integrity factor is reduced. The calculation threshold was reduced from 80% to 70% to improve calculation efficiency. Simultaneously, during the calculation process, the BeiDou reference station detects abnormal data by comparing displacement data from adjacent time points: when the difference between two adjacent displacements is ≥5mm, it is determined to be abnormal data, generating abnormal information including the monitoring station ID, the abnormal data value, and adjacent data values. After the abnormal information is fed back to the monitoring platform, it triggers the platform's verification process. The monitoring platform calls the historical data of the monitoring station and the contemporaneous data of adjacent monitoring stations for cross-verification. If the verification shows a data error, a data correction command is sent to the BeiDou reference station to guide it to recalculate; if the verification shows a real displacement mutation, the warning level is upgraded and the scope of the warning information is expanded.

[0070] During system operation, the working status, parameter configuration, and data transmission records of all modules are stored in real time to the edge data intelligent cache unit of the Beidou reference station and the storage unit of the monitoring platform, forming a full-link data log. The log includes information such as module identifier, timestamp, operation content, and data summary, supporting subsequent data traceability and system maintenance. Simultaneously, the system performs self-diagnosis according to a preset cycle, checking the hardware status, software operation, and link connectivity of each module. If a fault is detected (such as a failure in the monitoring station's acquisition unit, a decrease in the computing power of the reference station's edge server, or a transmission link interruption), a fault alarm is generated, pushed to the monitoring platform, and displays the fault location, fault type, and handling suggestions, ensuring the continuous and stable operation of the system.

[0071] The following description is based on a specific embodiment two.

[0072] The system corresponding to this embodiment can be referred to in Embodiment 1, and will not be repeated here. This embodiment is based on the scenario of mountainous slope displacement monitoring, which is characterized by complex terrain, uneven 4G signal coverage, scattered monitoring points, and high risk of sudden displacement changes. The various modules of the system operate collaboratively according to the scenario adaptation logic.

[0073] During the system deployment phase, parameter adaptation configuration was completed based on the monitoring needs of mountain slopes: In the edge-solving intelligent scheduling unit, the timeliness requirements of slope displacement monitoring were considered, and the weights were adjusted accordingly. Set to 0.6. Set to 0.4, adjust coefficient Set to 1.3. Setting it to 0.7 enhances timeliness sensitivity; in the multi-transmission mode intelligent adaptation unit, the reliability priority of mountain links is increased, and the link performance weight is adjusted. Set to 0.5. Set to 0.5; In the edge data intelligent caching unit, due to the need for backtracking analysis of mountain data, the weighting coefficient is... Set to 0.6. Set to 0.4, time decay constant Set to 12 hours to shorten the time decay period; in the displacement data early warning linkage unit, the slope displacement sudden change risk is high, and the skewness correction coefficient is adjusted. Set to 0.2, warning coefficient The system is uniformly set to 1.6 for high-risk scenarios (parameter configuration adapted to mountainous areas). After deployment, BeiDou monitoring stations are deployed according to slope zones, with 3-5 BeiDou monitoring stations in each zone. The BeiDou reference station is deployed at the highest signal point in the gentle slope area, and the monitoring platform is deployed at the remote monitoring center. The system achieves full-link monitoring through inter-module collaboration.

[0074] The radio transmission adaptive optimization unit of the Beidou monitoring station optimizes channel quality detection and data compression logic to address channel fluctuations caused by terrain obstruction in mountainous areas. The data acquisition unit sets the initial sampling frequency according to the risk level of the slope monitoring zones: 2 times / second for high-risk zones, 1 time / second for medium-risk zones, and 0.5 times / second for low-risk zones (sampling frequency adapted to risk level, with higher frequency sampling in high-risk areas). During data acquisition, the radio transmission adaptive optimization unit employs a dual-antenna receiving architecture (anti-obstruction design, adapted to mountainous terrain), simultaneously acquiring the signal strength of the main and secondary antennas, and using the average of the two as the final signal strength parameter to reduce signal fluctuation interference caused by terrain obstruction. Bit error rate (BER) detection uses a sliding window statistical method with a window length of 500ms (a reasonable window length balancing real-time performance and stability), and calculates the ratio of erroneous bits to total bits within the window in real time to ensure the real-time performance and stability of BER detection. Based on signal strength and BER, a channel quality assessment value is generated through normalization processing. .

[0075] Based on channel quality assessment values The radio transmission adaptive optimization unit uses a hierarchical compression strategy to process the original displacement information: when When, basic discrete cosine transform compression is used, retaining the first 35% of high-frequency coefficients; when At that time, a combined approach of "trend extraction + detail compression" was adopted. First, the trend components of the displacement data were extracted through linear fitting, and then a discrete cosine transform was performed on the detail components, retaining the first 25% of high-frequency coefficients. At this time, a deep compression mode is activated, removing detail components while retaining trend components and key feature point data, increasing the compression ratio to 1:10 (the compression strategy adapts to channel quality to ensure data transmission). After compression processing, the data volume forecast information sent by the BeiDou monitoring station to the BeiDou base station additionally carries the current channel quality assessment value. This enables the BeiDou reference station to pre-allocate computational resources based on channel conditions. Simultaneously, the BeiDou monitoring station receives real-time feedback on the computational pressure from the BeiDou reference station. When the computational pressure value (obtained by normalizing the edge server CPU utilization rate, ranging from [0,1]) ≥ 0.8, the sampling frequency is automatically reduced, decreasing to 1 time / second for high-risk areas and 0.5 times / second for medium- and low-risk areas. When the computational pressure value ≤ 0.3, the initial sampling frequency is restored, achieving dynamic adaptation between the sampling frequency and computational resources (adjusting the sampling frequency to adapt to computing power, balancing monitoring density and computational pressure).

[0076] The radio receiver unit of the BeiDou reference station employs an anti-interference receiving algorithm (adapted to mountainous electromagnetic interference environments) to filter out clutter signals caused by electromagnetic interference and terrain reflections in mountainous areas. It filters valid data packets through signal amplitude threshold and phase consistency checks. After receiving the compressed raw data and data volume prediction information transmitted from the BeiDou monitoring station, it first uses the channel quality assessment value in the prediction information... To determine data reliability, when When the data integrity is not found, a secondary data integrity check is automatically initiated (adapted to low-channel quality scenarios in mountainous areas). By comparing the data block check code with the preset check rules, the data is confirmed to be intact before proceeding to the calculation process. If the check fails, a retransmission request is sent to the BeiDou monitoring station, and the retransmission request carries the recommended compression mode to guide the BeiDou monitoring station to adjust the compression strategy.

[0077] The edge computing intelligent scheduling unit optimizes the task allocation logic for the dispersed deployment of multiple monitoring stations in mountainous areas. First, it calculates the timestamp deviation value. In the formula This is the timestamp (in seconds) for the collection of the j-th data set at the i-th monitoring station. This is the GPS synchronization time for the BeiDou reference station (unit: seconds; higher synchronization accuracy, suitable for mountainous scenarios). The maximum permissible time deviation threshold (unit: seconds) is dynamically adjusted according to the slope risk level, with high-risk zones... Seconds, medium- and low-risk zones (High-risk areas have even higher timeliness requirements), ensuring that data timeliness is given higher priority in high-risk areas. Standardized data volume. In the formula This represents the number of bytes (in bytes) of the compressed original data for the j-th group at the i-th monitoring station. The maximum number of bytes per data group supported by the system (unit: bytes, adapted to 2048 bytes based on the transmission method, adapting to the maximum load of data transmission in mountainous areas), and the maximum load of data transmission in mountainous areas. Integrity factor. In addition to field validation, a data trend consistency check is added. By comparing the displacement trend of the current data block with that of historical data blocks, if the trend deviation exceeds a threshold (default ±5mm), the error is reduced. The value can be determined using the following formula:

[0078] In the formula For integrity factor (unitless, range [0,1]), Number of valid fields (unit: number). Total number of fields (unit: number of fields). The trend consistency coefficient (unitless, range [0,1], takes the value of 1 when the trend is consistent, and the value decreases as the deviation increases) is used to optimize the integrity factor and adapt to the data reliability requirements of mountainous areas.

[0079] Based on the above parameters, the solution priority is calculated using the solution priority matrix expression. The edge computing intelligent scheduling unit uses a task pool mechanism to allocate computing resources, dividing the edge server's computing cores into a high-priority task pool and a normal task pool. The high-priority task pool accounts for 40%, and the normal task pool accounts for 60% (the task pool allocation is reasonable, ensuring the processing of high-priority data). Solution Priority The tasks (mainly high-risk partition data) are assigned to the high-priority task pool and solved in parallel using dedicated computing cores; The tasks are assigned to the ordinary task pool and share the computing core using a time-slice round-robin approach. During the solution process, based on the nonlinear characteristics of mountain slope displacement, the solution model is optimized: a terrain elevation correction factor is introduced based on carrier phase differential technology, with the formula: In the formula Elevation correction value (unit: millimeters). This is a terrain correction factor (unitless, set according to slope gradient; for every 10° increase in slope, ...). Increase by 0.1 (default 1.0). Let be the slope of the i-th monitoring station (unit: degrees, range [0,90]). The impact of terrain undulation on positioning accuracy is compensated by a correction factor (correction logic adapts to mountainous terrain to improve calculation accuracy), so that the accuracy of displacement data is improved to within ±0.8mm.

[0080] The lightweight iterative optimization of the solution model is adjusted according to the slope monitoring cycle. The iteration cycle is set to 12 hours for high-risk zones and 24 hours for medium- and low-risk zones (the iteration cycle adapts to the risk level, with more frequent optimization in high-risk areas). During the iteration process, the contribution of model parameters is calculated by comparing the deviation between the solution results and the manual monitoring calibration data. The contribution calculation formula is as follows:

[0081] In the formula The contribution of the p-th parameter (unitless, range [0,1]). The calculation deviation (in millimeters) after removing this parameter. Total number of model parameters (unit: number of parameters), removing contribution factors Redundant parameters (with stricter thresholds to meet the high-precision requirements of mountainous areas) are removed, while core parameters and high-contribution parameters are retained to ensure that the model maintains high-precision solution performance under the computing power constraints of edge servers.

[0082] The BeiDou reference station's multi-transmission mode intelligent adaptation unit optimizes link detection and handover logic to address the uneven 4G signal coverage in mountainous areas. The link quality detection cycle is shortened to 30 seconds (a shorter detection cycle adapts to the rapid link fluctuations in mountainous areas), improving link status response speed and collecting actual transmission rates from 4G, BeiDou short message, and satellite communication links. Data packet loss rate At this time, add link connectivity verification by sending test data packets to confirm link connectivity and avoid invalid link evaluations. Actual transmission rate The calculation uses a weighted average method, and the formula is as follows:

[0083] In the formula This represents the actual transmission rate (unit: bytes / second). Time weighting (unitless, more recent data has higher weight), ), This represents the number of bytes transmitted (in bytes) by the k-th transmission link in the t-th second. The detection cycle (unit: seconds) Enhanced sensitivity to real-time transmission rates (weighted average to adapt to link rate fluctuations in mountainous areas). Data packet loss rate. In addition to counting the number of lost data packets, the system also counts packet delay time. A delay exceeding 30 seconds is considered packet loss (this delay threshold is tailored to the high latency characteristics of mountainous links), ensuring... It can reflect the real-time transmission capability of the link.

[0084] Stability factor The calculation uses a moving average over 5 detection periods, and the formula is as follows: In the formula is the stability factor (unitless, range [0,1]). The transmission success rate of the k-th link in the nth detection period (unitless, range [0,1]) is used to reduce the impact of instantaneous fluctuations in mountainous links on stability assessment (the moving average window is larger, smoothing fluctuations). Based on the above parameters, the link quality coefficient is calculated through the link quality coefficient expression. Data priority is optimized in combination with the monitoring needs of mountainous slopes: Level 1 data is displacement data of high-risk zones (accuracy better than ±0.8mm, timeliness ≤3 minutes), Level 2 data is displacement data of medium-risk zones (accuracy ±0.8-1.5mm, timeliness ≤10 minutes), and Level 3 data is displacement data of low-risk zones and raw data (timeliness ≤30 minutes). Thus, the priority division adapts to the timeliness requirements of mountainous monitoring.

[0085] The multi-transmission mode intelligent adaptation unit constructs a link switching decision matrix to adapt to the multi-link switching needs in mountainous areas. When the quality coefficient of the current transmission link... When the data threshold is lower than the specified threshold (Level 1 data threshold 0.7, Level 2 data threshold 0.6, Level 3 data threshold 0.5), the system automatically switches to a suboptimal link. For example, if the current transmission mode is a 4G link, when... When (4G) < 0.6, if If (satellite communication) ≥ 0.6, then switch to satellite communication; if If the (satellite communication) value is less than 0.6, the system switches to BeiDou short message service. Simultaneously, to address the bandwidth limitations of BeiDou short message service (adapting to its narrow bandwidth characteristics), secondary compression is performed on the three-level data. Huffman coding is used to further compress the already compressed original data, increasing the compression ratio to 1:15, ensuring that the data can be transmitted completely via BeiDou short message service.

[0086] The BeiDou reference station's edge data intelligent caching unit addresses the frequent data transmission interruptions in mountainous areas by optimizing caching and breakpoint resumption logic. The caching architecture employs a "local cache + off-site backup" model (the backup mechanism adapts to the risk of transmission interruptions in mountainous areas). The local cache stores nearly 72 hours of high-frequency data, while the off-site backup stores nearly 96 hours of full data. The off-site backup is periodically synchronized to the monitoring platform's backup storage node via satellite communication. Cache data popularity is graded using a popularity value expression and a data correlation factor. Considering the spatial correlation of slope monitoring, if the data from the m-th group shows a coordinated displacement trend with the data from adjacent monitoring stations (deviation ≤ 2mm, consistent with spatial correlation of slope displacement), then The value is increased by 0.2 to enhance the caching priority of spatially correlated data and optimize the spatial characteristics of mountain monitoring through correlation optimization.

[0087] The breakpoint resume function has been optimized to a "multi-segment marking + incremental transmission" mode to adapt to the frequent transmission interruptions in mountainous areas. When transmission is interrupted, in addition to adding a unique identifier, the untransmitted data segment is divided into multiple sub-data blocks, each with a size of 512 bytes (adapting to the BeiDou short message bandwidth), and a sub-identifier is added to each. When transmission resumes, the sub-identifier list is transmitted first, the monitoring platform reports the identifiers of the received sub-data blocks, and the BeiDou reference station only transmits the unreceived sub-data blocks, reducing bandwidth consumption in mountainous areas. Simultaneously, the edge data intelligent caching unit records the link status information for each transmission, including the link quality coefficient. Transmission delay (unit: seconds), packet loss rate (unitless, range [0,1]). When a link experiences three consecutive transmission interruptions, it is marked as an unstable link, and subsequent data transmissions will preferentially avoid this link until its quality coefficient is within five consecutive detection cycles. ≥ Threshold, adapt to unstable links in mountainous areas through avoidance mechanisms.

[0088] The monitoring platform's multi-link data unified access unit optimizes data reception and processing logic to address the characteristics of mixed multi-link transmission in mountainous areas. After receiving data from the BeiDou reference station, it first distinguishes the data source link based on the link identifier field. A dedicated decompression process is then performed on the compressed data transmitted via BeiDou short messages. The original data structure is restored through Huffman decoding and inverse discrete cosine transform (the decompression process is adapted to the secondary compressed data). During the automatic data format adaptation process, additional checks are added for special fields specific to mountainous areas. These fields include slope correction values, monitoring station altitude, and channel quality assessment values, to adapt to the mountainous scenario and ensure data integrity and accuracy in mountainous environments.

[0089] To address the issue of large latency fluctuations in mountainous links, a dynamic interpolation method is used for receiver timing calibration (the interpolation method adapts to latency fluctuations): First, the average transmission delay of each link is calculated. In the formula Average transmission delay (unit: seconds). The transmission delay (in seconds) for the m-th data packet on the k-th link. This represents the number of data packets recently transmitted on this link (unit: packets, default 10 groups); the collection timestamp is corrected based on the average transmission delay, and the correction formula is as follows: In the formula This is the corrected data collection timestamp (unit: seconds). The original acquisition timestamp (unit: seconds) is used; all corrected acquisition timestamps are arranged in ascending order to achieve time series calibration, ensuring that data from different links and monitoring stations are uniformly sorted according to the actual acquisition time series, providing an accurate time series basis for subsequent early warning analysis.

[0090] After time-series calibration, the data is transmitted to the displacement data early warning linkage unit of the monitoring platform. This unit optimizes the early warning logic and threshold adjustment mechanism to address the high risk of sudden displacement changes in mountainous slopes. First, a historical data sample set is constructed, containing slope displacement data for the past 60 days (a longer sample period to adapt to the slow-changing characteristics of displacement in mountainous areas). Subsample sets are then established for high, medium, and low-risk zones. The high-risk zone subsample set has at least 1500 samples, and the medium- and low-risk zone subsample sets have at least 1000 samples (more sufficient sample size to ensure statistical reliability). The mean is calculated based on each subsample set. Standard deviation skewness coefficient The calculation method is the same as that in Example 1.

[0091] The warning threshold is calculated using the warning threshold expression, where the warning coefficient is... The skewness correction factor is uniformly set to 1.6 for high-risk scenarios. The value is set to 0.2 to enhance early warning sensitivity and adapt to the high-risk characteristics of mountainous areas. Simultaneously, a displacement change rate early warning index is introduced to calculate the difference between the current displacement value and the displacement value at the previous moment. In the formula The displacement rate (unit: mm / s). This is the current displacement value (unit: millimeters). Set a threshold for the rate of change of the displacement value (in millimeters) from the previous moment. (Based on dynamic adjustment of early warning thresholds, adapted for mutation detection), when At that time, regardless of whether the current displacement value exceeds All of these generate early warning information for sudden changes, avoiding early warning delays caused by sudden slope displacement (the sudden change early warning is adapted to the risk of sudden displacement changes in mountainous areas).

[0092] The generated warning information includes risk zone identifiers and mutation magnitudes. Fields such as slope gradient correction values ​​and monitoring station location coordinates are adapted to mountainous scenarios and pushed back to the BeiDou reference station via a multi-transmission intelligent adaptation unit. After receiving the data, the BeiDou reference station, in addition to storing and notifying on-site personnel, sends a high-frequency sampling command to all BeiDou monitoring stations in the corresponding monitoring zone, temporarily doubling the sampling frequency of the monitoring stations in that zone (from 2 times / second to 4 times / second in high-risk zones, and from 1 time / second to 2 times / second in medium-risk zones) for 30 minutes (high-frequency sampling adapts to sudden change monitoring), strengthening the monitoring density of sudden displacements and obtaining more detailed data on displacement change processes.

[0093] During continuous system operation, the interaction mechanisms between modules are dynamically adjusted according to the adaptation logic for mountainous scenarios, ensuring the system's adaptability to complex environments. In the pre-adaptation mechanism for radio transmission and computational resources between the BeiDou monitoring station and the BeiDou reference station, the computational pressure value of the BeiDou reference station is... The calculation introduces the priority weight of the solution task, and the formula is:

[0094] In the formula To calculate the pressure value (unitless, range [0,1]), This represents the current CPU utilization (unitless, range [0,1]). This represents the maximum allowed CPU utilization (no unit, default is 90%). This represents the current memory usage (unitless, range [0,1]). This represents the maximum allowed memory usage (no unit, default 85%). For the current high priority task ( Total number (unit: pieces) This sets the maximum capacity of the high-priority task pool (unit: tasks, default 20), increases the weight of high-priority tasks on the computational load, and optimizes the weighting to suit the high-priority data requirements in mountainous areas. At that time, high-pressure feedback is sent to the BeiDou monitoring stations in the high-risk zone. In addition to adjusting the compression ratio, the sampling frequency of the monitoring stations in that zone is temporarily reduced to 50% of the initial value to alleviate the calculation pressure. At this time, the sampling frequency is restored and the compression ratio is reduced to improve data accuracy.

[0095] In the multi-transmission data demand linkage mechanism between the BeiDou reference station and the monitoring platform, the reception success rate of the monitoring platform is... Statistics are compiled separately for each link, when a certain link... (When the threshold for mountainous scenarios is lower than that for regular scenarios, adapting to the low success rate of mountain links) a link optimization suggestion is sent to the BeiDou reference station, including the historical quality coefficient change trend and packet loss period distribution of the link; after receiving it, the BeiDou reference station adjusts the detection cycle of the link to 20 seconds (the detection cycle is shorter, adapting to the rapid changes of mountain links), and tracks the link status in real time. If there are 10 consecutive cycles... If the link is not configured as a backup link, other links will be used to transmit data first. Simultaneously, the monitoring platform issues dynamic accuracy requirement commands based on the phased needs of slope monitoring (e.g., high-frequency, high-precision data is needed during the rainy season, and efficiency needs to be balanced during the dry season). The target accuracy is set at ±0.5mm during the rainy season and ±1.2mm during the dry season to adapt to seasonal characteristics. After receiving the data, the BeiDou reference station adjusts the terrain elevation correction factor of the edge calculation intelligent scheduling unit. The higher the precision requirement, The smaller the adjustment step size (±0.05, the more the adjustment step size is adapted to the accuracy requirements), the more likely it is to ensure that the correction accuracy matches the target accuracy.

[0096] In the adaptation mechanism for the transmission strategy of calculation results between the edge server of the Beidou reference station and the intelligent adaptation unit for multiple transmission modes, the displacement data output by the edge server, in addition to being labeled with the accuracy level, is also labeled with the data urgency level: high-risk partition data is labeled "urgent," medium-risk partition data is labeled "normal," and low-risk partition data is labeled "ordinary" to adapt to priority transmission in mountainous areas. The intelligent adaptation unit for multiple transmission modes matches transmission resources based on accuracy level and urgency level: "urgent" + L1 level data occupies 60% of the link bandwidth and is transmitted with priority; "normal" + L2 level data occupies 30% of the bandwidth; "ordinary" + L3 / L4 level data occupies 10% of the bandwidth (bandwidth allocation adapts to the limited bandwidth in mountainous areas), avoiding waste of limited bandwidth resources in mountainous areas. At the same time, the transmission delay feedback is adjusted to differentiated processing according to the data urgency level: the transmission delay threshold for "urgent" data is set to 1 second, and the link is switched immediately if the threshold is exceeded; the threshold for "normal" data is set to 3 seconds; and the threshold for "ordinary" data is set to 5 seconds (the delay threshold is adapted to the urgency level to ensure the transmission of high-urgency data), ensuring the timeliness of transmission of high-urgency data.

[0097] In the early warning calculation parameter optimization mechanism between the monitoring platform and the Beidou reference station, the early warning information returned by the monitoring platform is supplemented with slope environmental parameters (such as rainfall and soil moisture, which are additionally collected and transmitted by the monitoring station, and the environmental parameters are adapted to the risk factors of mountain slopes). The Beidou reference station combines the environmental parameters to optimize the calculation threshold: when the rainfall is ≥50mm / day (unit: millimeters / day, the rainstorm threshold is reasonable) and the soil moisture is ≥80% (unit not specified, range [0,1], the high humidity threshold is reasonable), the calculation priority weight of the corresponding zone monitoring station data will be adjusted. Increased to 0.7, The threshold was lowered to 0.3 to further enhance timeliness; at the same time, the warning threshold was lowered. Reduced by 10% to improve early warning sensitivity and address displacement risks in harsh environments (threshold adjustment adapted to harsh environments). When a BeiDou reference station detects abnormal data, in addition to reporting the anomaly, it simultaneously sends the station's recent channel quality assessment value. Combined with the solution parameters; during the verification by the monitoring platform, in conjunction with Determine whether the data anomaly is caused by channel interference. If so, data retransmission and recalculation will be triggered first; if Then, by combining data from adjacent monitoring stations with environmental parameters, it can be determined whether it is a true displacement change, ensuring the accuracy of the verification results and adapting to the channel interference and environmental impact in mountainous areas.

[0098] After system deployment, maintenance procedures are executed according to the cyclical requirements of mountain slope monitoring: During data transmission off-peak periods without affecting monitoring, module self-checks are performed to check the battery level of the monitoring station, sensor status, edge server computing power and storage capacity of the base station, and the connectivity of the transmission link; if the battery level of the monitoring station is detected to be below 20% (the threshold is adapted to the long-term power supply needs of mountainous areas to avoid frequent maintenance), a low battery alarm is triggered, and the alarm information is pushed to the monitoring platform, while the remaining battery power and estimated battery life are recorded; if the sensor sampling value is detected to exceed the physical range (such as displacement value exceeding ±50mm, which is in line with the conventional range for slope displacement monitoring), the sensor is marked as faulty, its data acquisition is suspended, and a sensor replacement suggestion is sent to maintenance personnel to ensure the validity of the collected data.

[0099] Data calibration is performed according to a preset cycle, for example, once a week. Displacement data from 3-5 typical monitoring points are collected using high-precision manual monitoring equipment (such as a total station with an accuracy of ±0.3mm, suitable for high-precision calibration of mountain slopes) as calibration benchmark values. (Unit: mm). The manually calibrated values ​​are compared with the displacement data calculated by the system concurrently. (Unit: mm) Compare and calculate calibration deviation. ;when When (adapting deviation thresholds to meet high-precision monitoring requirements), adjust the terrain elevation correction factor of the edge calculation intelligent scheduling unit. Adjust the step size to ±0.05 until... Simultaneously, based on the statistical results of multiple sets of calibration deviations, the warning threshold of the displacement data early warning linkage unit is corrected. If the average deviation is positive, then Adjust upwards by 1.2 times the mean deviation. If the mean deviation is negative, then... The deviation is adjusted down by 0.8 times the absolute value of the mean deviation to ensure that the system's calculation accuracy is consistent with the manual calibration results, thus guaranteeing the accuracy of subsequent early warning judgments.

[0100] Link optimization is performed at preset intervals, such as one month. Based on historical transmission link data from the past month, the core performance indicators of each link are calculated: average link quality coefficient. ( (Total number of tests in the month, unit: times) and average transmission delay ( (Total number of data packets transmitted in the month, unit: packets) and average data loss rate The link performance weights of the multi-transmission mode intelligent adaptation unit are adjusted based on statistical results. , If satellite communication For three consecutive months, the signal strength was 0.1 or higher than that of the 4G link. If the bandwidth is lower than 0.05 for 4G links, then... Adjusted to 0.45 Adjust to 0.55 to increase the link reliability weight; if the 4G link... The latency remained below that of satellite communication for more than 0.3 seconds, and Then maintain , The initial value, balancing transmission rate and reliability; if the BeiDou short message... If the value is higher than 0.1, the link detection frequency will be increased to 15 seconds / time to track the link status in real time and avoid data loss due to link instability.

[0101] In addition, the antennas of the Beidou monitoring station are calibrated according to a preset cycle, such as once a quarter. The phase center deviation of the antenna is measured by professional equipment. If the deviation exceeds ±2mm (the phase center deviation threshold is adapted to the Beidou positioning accuracy requirements), the antenna installation angle is adjusted to ensure the stability of the antenna receiving Beidou satellite signals. Every six months, a health check is performed on the edge server storage device of the Beidou reference station. The storage detection tool scans for bad sectors on solid-state drives and flash memory arrays, cleans up low-hot data that has not been accessed for more than 90 days (adapting to the characteristics of limited data storage resources in mountainous areas), and backs up the core calculation model and historical early warning data to avoid data loss or model damage due to hardware failure, and to ensure the long-term stable operation of the system.

[0102] In summary, the system of this disclosure focuses on technologies related to multi-mode data transmission, edge intelligent computation, and dynamic early warning in complex environments. It is applicable to safety monitoring scenarios for structures such as slopes, bridges, and buildings, aiming to optimize the transmission adaptability, computation efficiency, and data reliability of the BeiDou displacement monitoring system. Through radio transmission adaptive optimization and intelligent adaptation of multiple transmission modes, it dynamically matches channel quality and data requirements, avoiding the limitations of single transmission methods. This effectively addresses scenarios such as complex terrain and electromagnetic interference, ensuring real-time and continuous transmission of critical data, and significantly improving transmission adaptability and stability. Intelligent scheduling for edge computation enables task priority allocation and parallel processing. Lightweight model iteration adapts to edge computing power, reducing cloud dependence and improving data processing efficiency across multiple monitoring stations, achieving dual optimization of computation efficiency and accuracy. A heat-based hierarchical caching mechanism optimizes storage resource allocation, employing "multi-segment marking + incremental transmission." The breakpoint resume design avoids data loss or duplicate transmission, providing complete and reliable monitoring data for subsequent analysis and ensuring data integrity and reasonable storage. The full-link dynamic collaboration mechanism allows the system to adjust parameters according to the monitoring scenario (such as mountain slopes and conventional buildings), adapting to different risk levels, terrain environments, and accuracy requirements, improving system practicality and enhancing scenario adaptability and versatility. By combining dynamic early warning thresholds with skewness correction and displacement change rate monitoring, the system can quickly identify and issue graded early warnings for abnormal displacements, and optimize calculation parameters through reverse linkage, reducing the probability of misjudging safety risks, providing strong support for structural safety, and achieving accurate and timely early warning response.

[0103] It should be noted that, in this document, relational terms such as "first" and "second" are used merely to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0104] The above description is merely a specific embodiment of this disclosure, enabling those skilled in the art to understand or implement it. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of this disclosure. Therefore, this disclosure is not to be limited to the embodiments described herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A displacement monitoring multi-mode transmission calculation system, characterized in that, The system includes: Multiple monitoring stations, reference stations, and monitoring platforms; The monitoring station is configured to: collect raw displacement information of the monitored target and transmit it to the base station via radio; The reference station is configured to: for each set of original displacement information of each monitoring station, determine the solution priority matrix based on the timestamp deviation value, the standardized data volume, and the integrity factor; wherein, the timestamp deviation value is obtained by standardizing the difference between the current data timestamp and the reference timestamp, the standardized data volume is obtained by the ratio of the current data volume to the maximum single set of data volume, and the integrity factor is calculated by the proportion of effective data fields to the total number of fields; According to the solution priority matrix, solution resources are allocated to each group of original displacement information and solution processing is performed to obtain displacement result data; Based on the accuracy level and timeliness requirements of the displacement data, data priority is determined, and link quality coefficients for various transmission links are calculated. Based on the data priority and the link quality coefficient, a target transmission link is determined from the multiple transmission links so that the displacement results data can be transmitted to the monitoring platform through the target transmission link; The monitoring platform is configured to receive displacement data transmitted by the base station, and to perform data processing, early warning, and issue demand commands based on the displacement data.

2. The system as described in claim 1, characterized in that, The monitoring station is specifically configured as follows: The radio channel quality is detected in real time before being transmitted to the reference station via radio. The acquired raw displacement information is lightweight and compressed based on the channel quality to adapt to the radio station bandwidth.

3. The system as described in claim 1, characterized in that, The base station is specifically configured to calculate the link quality coefficients of various transmission links through the following steps: For each transmission link, the actual transmission rate and data packet loss rate of the transmission link are obtained; wherein, the actual transmission rate is calculated by the number of data bytes successfully transmitted per unit time, and the data packet loss rate is obtained by the ratio of the number of lost data bytes to the total number of transmitted data bytes; The stability factor is calculated based on the average transmission success rate over multiple consecutive detection cycles. The link quality coefficient of the transmission link is calculated based on the actual transmission rate, data packet loss rate, and stability factor of the transmission link.

4. The system as described in claim 3, characterized in that, The displacement data is divided into primary, secondary, and tertiary data according to data priority. The various transmission links include satellite communication links, 4G links, and short message links. The base station is specifically configured as follows: The primary data is transmitted to the monitoring platform via the satellite communication link. The secondary data is transmitted to the monitoring platform via the 4G link; The third-level data is transmitted to the monitoring platform via the short message link.

5. The system as described in claim 1, characterized in that, The base station is also configured as follows: For the original displacement information and / or displacement result data, the data access frequency and time span are obtained; wherein, the data access frequency is obtained based on the ratio of the current data access count to the total number of data access counts, and the time span is obtained based on the difference between the current time and the data cache start time; Calculate the data relevance factor based on the percentage of references between current data and cached data. The popularity value is calculated based on the frequency of data access, time span, and data relevance factor. The original displacement information and / or displacement result data are cached according to the heat value.

6. The system as described in claim 1, characterized in that, The monitoring platform is specifically configured as follows: The mean scaling process is performed based on the mean and standard deviation of historical displacement data to obtain the mean scaling result. Calculate the skewness coefficient based on the third-order central moment of historical displacement data; The mean scaling result is corrected based on the skewness coefficient to obtain the dynamic early warning threshold; Early warning information is generated based on the displacement data and the dynamic early warning threshold, and the early warning information is transmitted to the base station.

7. The system as described in claim 2, characterized in that, The monitoring station is also configured to send a data volume forecast to the base station before transmitting the original displacement information; The base station is also configured to: pre-allocate computing power according to the data volume forecast information and the computing priority matrix, and report the computing pressure status to the monitoring station; The monitoring station is also configured to adjust the compression ratio of the lightweight compression process based on the calculated pressure status.

8. The system as described in claim 1, characterized in that, The monitoring platform is also configured to: The system provides real-time feedback on the data reception status to the base station; in the event of a reception anomaly, the base station is triggered to switch the transmission link or retransmit. Accuracy requirement instructions are issued to the reference station according to actual application needs; The base station is also configured to adjust the solution accuracy parameters according to the accuracy requirement command.

9. The system as described in claim 1, characterized in that, The base station performs calculations via an edge server, and the base station is further configured as follows: When the edge server outputs displacement result data, it adds a precision level mark to the displacement result data. The transmission delay data is fed back to the edge server so that the edge server can adjust the output frequency of the displacement result data according to the transmission delay data.

10. The system as described in claim 6, characterized in that, The base station is also configured as follows: The solution threshold for the solution process is optimized based on the warning information. If abnormal data is detected during the calculation process, the abnormal information will be fed back to the monitoring platform to trigger the data review process of the monitoring platform.

Images