A data communication method based on a highway electromechanical system

Abstract

The application relates to the technical field of communication, and particularly discloses a data communication method based on a highway electromechanical system, which comprises the following steps: synchronously collecting strain and temperature parameters along an optical fiber, converting the strain and temperature parameters into transmission time delay variation according to predetermined physical characteristics, and generating real-time time delay dynamic data of each communication path through path integration; according to the real-time time delay dynamic data, a time delay variation file containing a trend component and a disturbance margin index is established and continuously updated for each access unit; a reference round-trip time delay of each unit is dynamically corrected according to the file in a communication station, a dynamic protection interval is determined by combining a trend change rate and a disturbance index, and an uplink time slot window with a protection band is re-distributed; through comparison between a periodically measured sample and a predicted value, a conversion relationship between an error and a physical parameter is reversely corrected to a time delay parameter based on historical credibility weighting, and adaptive optimization is realized; and the application can accurately predict and compensate time delay drift caused by dynamic deformation, and ensures reliable transmission of uplink data.

Description

Technical Field

[0001] This invention relates to the field of communication technology, and more specifically to a data communication method based on a highway electromechanical system. Background Technology

[0002] As highway electromechanical systems evolve towards full digitalization and intelligence, broadband access solutions, represented by Passive Optical Network (PON) technology, are widely deployed in data communications for critical services such as monitoring, toll collection, and tunnel management, forming dedicated highway communication networks that support various types of terminal equipment. Especially in critical infrastructure sections such as cross-river and cross-sea bridges and long tunnels, communication optical fibers are often laid along the bridge structure or tunnel walls, forming the physical transmission backbone of the system. In existing technologies, such systems typically employ PON protocols based on Time Division Multiple Access (TDMA) to achieve multi-point access and centralized control. Their standard operating mechanism relies on the Optical Line Terminal (OLT) performing fixed or periodic ranging on each Optical Network Unit (ONU) to acquire and lock the round-trip delay, thereby allocating fixed uplink transmission time slots.

[0003] In existing communication methods for highway electromechanical systems based on Passive Optical Networks (PON), unpredictable drift in communication delays occurs due to random changes in the physical state of the transmission medium in dynamically deformable infrastructure environments such as bridges and tunnels. This leads to technical problems such as uplink time slot synchronization instability, data packet collisions, and failure of critical control commands in Time Division Multiple Access (TDMA) protocols. Existing technologies rely on fixed or periodic ranging to compensate for delays, but cannot perceive and respond in real time to nanosecond-level dynamic changes in delay caused by structural deformation, temperature gradients, etc., resulting in severe degradation of communication quality in scenarios with high reliability requirements. Summary of the Invention

[0004] The purpose of this invention is to provide a data communication method based on a highway electromechanical system to solve the problems mentioned above.

[0005] The objective of this invention can be achieved through the following technical solutions:

[0006] A data communication method based on a highway electromechanical system includes the following steps:

[0007] S1: Synchronously collect real-time physical state parameters of various spatial points along the highway infrastructure, including strain parameters and temperature parameters;

[0008] S2: Based on real-time physical state parameters and according to the predetermined optical fiber physical characteristic conversion relationship, the optical fiber path is segmented and numerical comprehensive analysis is performed independently on each segment to convert the changes in physical state parameters of each segment into corresponding changes in transmission delay. The transmission delay changes calculated for all segments are integrated to generate a real-time data set reflecting the dynamic changes in delay of the entire optical path from the communication center to each remote access unit.

[0009] S3: Based on real-time data sets, establish and continuously update a propagation path latency change profile for each remote access unit;

[0010] S4: The communication center dynamically corrects the pre-stored reference round-trip time delay value based on the current propagation path delay change file of each remote access unit; based on the corrected reference round-trip time delay value, it recalculates and allocates uplink data transmission time slot windows for each remote access unit, and sends the updated time slot allocation scheme to each corresponding unit.

[0011] S5: The communication center periodically selects remote access units to sample the actual round-trip delay. The sampled actual delay value is compared with the delay value predicted based on the current propagation path delay change profile. Based on the error generated by the comparison, the conversion relationship in step S2 is reversed to achieve adaptive optimization of the physical parameter to delay parameter conversion process.

[0012] As a further aspect of the present invention: the process for obtaining the transmission delay change is as follows:

[0013] Based on the conversion relationship of optical fiber physical characteristics, a local mapping relationship library is constructed for each segment. The local mapping relationship library defines the mapping rules from a multi-dimensional vector composed of strain parameter change value and temperature parameter change value to the single transmission delay change value of the corresponding segment.

[0014] Extract the changes in the current physical state parameters of each segment and form a corresponding multi-dimensional vector. By querying the local mapping database and performing interpolation calculations, output the changes in transmission delay of the corresponding segment.

[0015] As a further aspect of the present invention: the generation of a real-time data set reflecting the dynamic changes in delay along the entire optical path from the communication central office to each remote access unit specifically includes:

[0016] Based on the physical topology and logical connection relationship of the optical fiber network, identify and determine the continuous segment sequence that the optical signal transmission actually takes from the communication center to each remote access unit;

[0017] According to the order of the continuous segment sequence corresponding to the remote access unit, a dynamic weighted accumulation operation is performed on the transmission delay change of all segments in the continuous segment sequence;

[0018] The dynamically weighted cumulative result calculated by each remote access unit is bound to a unique identifier and a timestamp to construct an independent custom data structure. The custom structures corresponding to all remote access units together constitute a real-time data set.

[0019] As a further aspect of the present invention: S3 specifically includes:

[0020] From the real-time data set, extract the total delay change of the entire optical path corresponding to each remote access unit within a specified time window to form an independent time series of delay change for each remote access unit;

[0021] Multi-scale fluctuation characteristic analysis is performed on the time series of delay change of each remote access unit to identify and separate the trend component caused by long-term structural deformation and the random component caused by instantaneous environmental disturbance.

[0022] The trend components obtained from the analysis are integrated with the existing data in the current propagation path delay change archive to generate the updated archive body; at the same time, the statistical characteristics of the random components are stored in the archive as an auxiliary disturbance margin index, together forming a complete updated propagation path delay change archive.

[0023] As a further aspect of the present invention: the generation of the updated file body specifically includes:

[0024] The continuity and smoothness of the trend component are evaluated, and the confidence level of the change of the trend component relative to the trend component of the previous period recorded in the current archive is calculated.

[0025] Based on the changing confidence level, the fusion weight coefficient between 0 and 1 is dynamically determined;

[0026] By using the fusion weighting coefficient, the data of the current trend component is weighted and synthesized with the trend component data of the previous period recorded in the current archive body. The calculation result covers the trend component records in the original archive body, and an updated archive body is generated.

[0027] As a further aspect of the present invention: the dynamic correction of the pre-stored reference round-trip delay value specifically includes:

[0028] Read the main body of the archive recorded by the remote access unit from the current propagation path latency change archive;

[0029] Extract the trend components of the records and the associated perturbation margin indicators from the main body of the archives;

[0030] The value of the trend component is used as the main reference value for time delay correction, and the value of the disturbance margin index is multiplied by a predefined scaling factor and then superimposed on the main reference value to form the dynamic time delay correction amount.

[0031] The dynamic delay correction adjusts the pre-stored reference round-trip delay value to generate the corrected reference round-trip delay value.

[0032] As a further aspect of the present invention: the recalculation and allocation of uplink data transmission time slot windows for each remote access unit specifically includes:

[0033] Historical data of trend components are extracted from the current propagation path delay change archive to calculate the trend component change rate, and combined with the disturbance margin index to jointly determine the protection interval parameter;

[0034] For remote access units, based on the basic time slot start point determined by the corrected reference round-trip delay value, the time slot window is defined by extending the dynamic protection interval parameter forward and backward by half the duration.

[0035] A logical channel identifier is assigned to each time slot window, and the logical channel identifier is bound to the corresponding time slot window boundary parameters to form a time slot allocation scheme, which is then sent to each corresponding remote access unit.

[0036] As a further aspect of the present invention: S5 specifically includes:

[0037] Prioritize the sampling of the measured round-trip time of remote access units whose disturbance margin index recorded in the propagation path delay change archive exceeds a set threshold;

[0038] Calculate the error between the measured delay value and the delay value predicted based on the current archive, and then distribute the error in reverse to the corresponding segments according to the continuous segment sequence traversed by the optical path of the sampling unit;

[0039] For each affected segment, the mapping rules defined in the local mapping relation library of the segment in step S2 are progressively adjusted according to the error value assigned to the corresponding segment, so as to reduce the prediction delay deviation generated in the future under the same physical state parameter input.

[0040] As a further aspect of the present invention: the progressive adjustment of the mapping rules defined in the segmented local mapping relation library in step S2 specifically includes:

[0041] Based on the error value assigned to the corresponding segment and the current physical state parameters of the corresponding segment, the affected discrete nodes are located in the multi-dimensional space corresponding to the local mapping relationship library.

[0042] Based on the direction and magnitude of the error value, the required output value offset for the discrete node is calculated, and the offset corrects the corresponding node's currently defined transmission delay change output.

[0043] Query the historical reputation adjustment records of discrete nodes;

[0044] Based on historical reputation adjustment records, the offset is weighted and decayed, and the output value of the mapping rule defined for discrete nodes is updated with the decayed offset. At the same time, this adjustment is recorded to update the historical reputation adjustment records.

[0045] The beneficial effects of this invention are:

[0046] (1) This invention accurately calculates the dynamic change in time delay for each communication path by integrating real-time strain and temperature parameters collected by the sensing fiber with a local mapping relation library and path. Based on this, a time delay profile containing trend components and disturbance margins is dynamically constructed for each remote unit. Based on this profile, not only is the reference round-trip time delay corrected in real time, but a dynamic protection interval determined by the trend change rate and disturbance index is also introduced to adaptively widen the uplink time slot. This enables the system to reserve an elastic transmission window for data packets in advance when the bridge experiences slight deformation or vibration, effectively absorbing time delay jitter, thereby ensuring strict synchronization and near-zero collision reliable transmission of uplink data even in harsh physical environments.

[0047] (2) This invention achieves continuous optimization of the conversion relationship by designing a closed-loop feedback and progressive correction mechanism. The system prioritizes the actual sampling of paths with severe delay fluctuations, distributes the error between prediction and measurement inversely to each fiber segment, and adjusts the offset of discrete nodes in the segment mapping relationship library by weighting the reputation. The reputation mechanism can effectively distinguish between random noise and trend deviation, ensuring that the adjustment responds to real changes while maintaining stability. This process enables the mapping relationship between "physical state and transmission delay" to evolve and self-calibrate over time, improving the long-term accuracy of delay prediction and reducing the system's dependence on the initial calibration accuracy and maintenance requirements. Attached Figure Description

[0048] The invention will now be further described with reference to the accompanying drawings.

[0049] Figure 1 This is a flowchart of the method of the present invention. Detailed Implementation

[0050] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0051] Please see Figure 1 As shown, the present invention is a data communication method based on a highway electromechanical system, comprising the following steps:

[0052] S1: Synchronously collect real-time physical state parameters of various spatial points along the highway infrastructure, including strain parameters and temperature parameters;

[0053] S2: Based on real-time physical state parameters and according to the predetermined optical fiber physical characteristic conversion relationship, the optical fiber path is segmented and numerical comprehensive analysis is performed independently on each segment to convert the changes in physical state parameters of each segment into corresponding changes in transmission delay. The transmission delay changes calculated for all segments are integrated to generate a real-time data set reflecting the dynamic changes in delay of the entire optical path from the communication center to each remote access unit.

[0054] S3: Based on real-time data sets, establish and continuously update a propagation path latency change profile for each remote access unit;

[0055] S4: The communication center dynamically corrects the pre-stored reference round-trip time delay value based on the current propagation path delay change file of each remote access unit; based on the corrected reference round-trip time delay value, it recalculates and allocates uplink data transmission time slot windows for each remote access unit, and sends the updated time slot allocation scheme to each corresponding unit.

[0056] S5: The communication center periodically selects remote access units to sample the actual round-trip delay. The sampled actual delay value is compared with the delay value predicted based on the current propagation path delay change profile. Based on the error generated by the comparison, the conversion relationship in step S2 is reversed to achieve adaptive optimization of the physical parameter to delay parameter conversion process.

[0057] In S1, real-time physical state parameters of various spatial points along the highway infrastructure are collected synchronously. These physical state parameters include strain parameters and temperature parameters, specifically including:

[0058] To achieve high-precision sensing of dynamic deformation of the optical fiber transmission medium, this method employs an integrated communication and sensing approach for data acquisition. Specifically, the communication optical fiber deployed within the highway infrastructure is simultaneously used as a distributed sensing medium, with a coherent probe light pulse of a specific wavelength coupled and injected into the fiber at the communication station end. As this probe light pulse propagates along the fiber, a portion of its energy is backscattered due to the inherent Rayleigh scattering effect within the fiber.

[0059] By coherently detecting and demodulating the returned backscattered light signal, we can obtain information on the changes in its optical phase or frequency characteristics over time and space. Since the strain and temperature changes of the optical fiber directly modulate these characteristics of the light propagating inside it, by calculating the correspondence between the changes in these characteristics and the position of the optical fiber, we can simultaneously resolve the real-time physical state parameters at every spatial point along the optical fiber with meter-level spatial resolution.

[0060] Finally, the system converts the continuously calculated changes in optical phase or frequency corresponding to spatial locations into strain and temperature parameters for each spatial point, based on known calibration coefficients for fiber optic elastic and thermo-optic effects, thus forming a spatiotemporally synchronized dataset of physical state parameters covering the entire optical fiber.

[0061] In S2, based on real-time physical state parameters and according to predetermined optical fiber physical characteristic conversion relationships, the optical fiber path is segmented, and numerical synthesis analysis is performed independently on each segment to convert the changes in physical state parameters of each segment into corresponding changes in transmission delay. The transmission delay changes calculated for all segments are then integrated to generate a real-time data set reflecting the dynamic changes in delay along the entire optical path from the communication center to each remote access unit. Specifically, this includes:

[0062] First, for each pre-defined fiber segment, a local mapping database is established based on the known fiber elastic-optic coefficient and thermo-optic coefficient. The database is constructed as follows: Beforehand, through laboratory testing or theoretical calculations, the reference values ​​for the transmission delay change of the fiber segment under several discrete, representative strain and temperature combinations are determined. After deployment, each discrete state is recorded as a multi-dimensional vector entry, consisting of a specific set of strain and temperature parameter changes, and associated with a defined reference value for transmission delay change. All these entries together constitute the local mapping database for that segment, which is essentially a discrete lookup table defining the transition from "physical state change" to "delay change".

[0063] During system operation, for each segment at the current moment, the changes in physical state parameters from the initial reference state to the current state are extracted, namely the changes in strain and temperature parameters, forming a real-time multidimensional vector. This real-time multidimensional vector is then compared with each discrete vector entry stored in the local mapping database. A bilinear interpolation method is used for calculation: in the multidimensional space of the mapping database, several nearest-neighbor discrete vector entries surrounding the real-time vector are found. Then, based on the numerical distance ratio between the real-time vector and these neighboring entries, a weighted average of the transmission delay change reference values ​​corresponding to each neighboring entry is calculated, ultimately outputting the accurate transmission delay change of the current segment under the real-time physical state.

[0064] After obtaining the transmission delay changes for all segments, the path integration phase begins. First, based on the physical connection topology of the fiber optic network, a unique optical path from the central office to each remote access unit within the network is pre-stored or dynamically determined. This optical path is formed by sequentially connecting a series of consecutive fiber optic segments, creating a unique continuous segment sequence for that unit. For each remote access unit, the current transmission delay changes for all segments are extracted sequentially from its corresponding continuous segment sequence.

[0065] Next, a dynamic weighted summation operation is performed on the extracted series of transmission delay changes to obtain the total delay change for the entire optical path. Each segment in the sequence is independently assigned a dynamic weight coefficient. This weight coefficient is determined based on two factors: the relative position ordinal number of the segment in the overall path and the historical stability data of the segment over a past period. Historical stability data is obtained by calculating the statistical variance of the segment's historical transmission delay changes. Specifically, segments with later position ordinal numbers and worse historical stability data are assigned smaller dynamic weight coefficients, thus reducing the impact of unstable segments on the overall summation result. The transmission delay change of each segment is multiplied by its corresponding dynamic weight coefficient, and then summed to obtain the total delay change of the remote access unit along the entire optical path.

[0066] Finally, a data record unit is generated for each remote access unit. This data record unit contains at least three fields: the first field is the globally unique identifier of the remote access unit; the second field is the current timestamp; and the third field is the calculated total latency change value. All data record units corresponding to all remote access units are organized and stored in chronological order of their generation, forming a complete, real-time updated data set that reflects the dynamic latency changes of all communication paths across the entire network.

[0067] In S3, based on real-time data sets, a propagation path latency change profile is established and continuously updated for each remote access unit, specifically including:

[0068] First, for each remote access unit, extract all total latency changes within the most recent specified time window (e.g., 30 minutes) from the real-time data set. Arrange these records in chronological order according to their timestamps to form an ordered numerical sequence, i.e., the latency change time series for that unit. Each data point in this time series contains information on the total latency change of the entire optical path corresponding to that moment.

[0069] Subsequently, multi-scale fluctuation characteristic analysis is performed on each obtained time series of time delay changes to distinguish the sources of time delay changes of different natures. Specifically, over a relatively long time scale (e.g., 10 minutes as an analysis interval), the least squares method is used to perform linear fitting on the data within the interval. The slope and intercept of the fitted line together describe the average trend of time delay changes within that time period; this is the trend component, which mainly reflects the slow creep of the bridge structure or diurnal periodic thermal expansion and contraction. Next, the theoretical value of the trend component at each corresponding moment is subtracted from the original time series data point by point to obtain a series of residuals. The standard deviation of these residuals is calculated within a shorter time window (e.g., 1 minute). This standard deviation is defined as the intensity characterization of the random component within that time window, mainly reflecting rapid random disturbances caused by instantaneous wind vibration, vehicle passage, etc.

[0070] Then, the archive update phase begins. The main body of the archive consists of the aforementioned trend components. During the update, the confidence level of the newly parsed current trend component, i.e., the change confidence level, is first assessed. This is calculated by comparing the main characteristics of the current trend component (e.g., its slope value) with the corresponding characteristics of the previous period's trend component stored in the archive, calculating the relative difference between the two. Simultaneously, the continuity of the data used in the formation of the current trend component and its smoothness over its time interval are examined (e.g., calculating the variance of the data points). Based on the magnitude of the relative difference and the smoothness of the data, a change confidence level value between 0 and 1 is mapped using a predefined correspondence table; the smaller the difference and the smoother the data, the higher the confidence level.

[0071] Based on the calculated confidence level of the change, a fusion weight coefficient between 0 and 1 is dynamically determined. The principle for determining this coefficient is: the higher the confidence level, the closer it is to a preset upper limit (e.g., 0.7), meaning that the new trend component is more trusted; the lower the confidence level, the closer it is to a preset lower limit (e.g., 0.3), meaning that it relies more on the stability of historical trend data in the archive. Finally, a weighted synthesis calculation is performed: the trend component data of the previous period recorded in the archive is multiplied by the difference between "1 and the fusion weight coefficient", and the current trend component data is multiplied by the fusion weight coefficient. The two products are added together, and the sum is the fused new trend component data. This new data is then used to overwrite the old records in the original archive, completing the update of the archive body.

[0072] Simultaneously, the statistical characteristics of the random components obtained from the above analysis within the current time window, specifically their standard deviation values, are stored as an auxiliary disturbance margin indicator, along with the update timestamp, in the remote access unit's archive. This indicator is used to quantify the intensity level of the instantaneous disturbance experienced by the current path. Thus, the updated archive and the latest disturbance margin indicator together constitute the complete and updated propagation path delay variation archive of the remote access unit.

[0073] In S4, firstly, the baseline round-trip time delay is dynamically corrected. For each remote access unit (hereinafter referred to as a unit) that needs scheduling... The communication exchange reads the data specific to that unit from its currently stored propagation path delay variation archive. The main body of the archive. From this main body of the archive, two key parameters are extracted: one is the current value of the trend component, which represents the long-term trend, denoted as... Second, the current value of the disturbance margin index, which characterizes the intensity of instantaneous fluctuations, is denoted as... Dynamic delay correction amount It is calculated by using the trend component as the primary benchmark value and overlaying it with a buffer value derived from the disturbance margin indicator. The specific calculation formula is as follows: ;

[0074] in, This represents the calculated dynamic delay correction amount, and its unit is time unit (such as microsecond). The trend component values ​​extracted from the archives directly reflect the fundamental and gradual time-delay changes in the path. It is a disturbance margin indicator extracted from the archives, reflecting the random fluctuation range of path delay. It is a predefined scaling factor, a positive real number, whose value is typically set empirically between 0.5 and 2.0 based on network tolerance, used to convert the volatility index into a time buffer. The significance of this formula is that the correction not only compensates for definite trend-based time delay changes (…). It also reserves additional margin ( This is to absorb unpredictable random jitter. Subsequently, this dynamic delay correction is applied to a pre-stored baseline round-trip delay value. Adjustments are made to generate a corrected baseline round-trip time delay value. The calculation relationship is as follows: equal Plus .

[0075] Secondly, the uplink time slot window is recalculated and reassigned based on the corrected delay value. This process first determines a dynamic guard interval parameter. For calculation It is necessary to start from the unit Extracting trend components from time delay change archives By taking the historical data from a recent period, and then averaging the values ​​from adjacent periods, the absolute value of the rate of change of the current trend component is calculated and denoted as . This rate of change reflects the speed of the trend change. (Protection interval parameter) It is determined by both the trend change rate and the disturbance margin index, and its calculation formula is as follows: ;

[0076] in, This represents the calculated protection interval parameter, with the same unit as time. It is the absolute value of the rate of change of the calculated trend component. It is the current disturbance margin indicator. and These are two independent preset weighting coefficients, both of which are positive real numbers. The weight used to adjust the influence of the trend change rate on the protection interval is usually between 0.1 and 1.0; The weighting used to adjust the impact of random fluctuations on the protection interval is typically between 0.2 and 1.5. The specific values ​​of the coefficients are calibrated through historical data analysis, with the goal of minimizing the probability of time slot collisions. The formula obtained... The larger the value, the more unstable the transmission path delay of the unit is, and the wider the time slot guard band is required.

[0077] Subsequently, for the unit Determine the actual uplink data transmission time slot window. Based on the Time Division Multiple Access (TDMA) protocol rules, use the modified reference round-trip time value. A theoretical starting point for a basic time slot can be determined. Using this point as a reference, extend forward (towards earlier points in the timeline). The duration is 2 / 3, extending backwards (towards a later point in the timeline). The duration is increased by 2 / 3, thus forming a total length that is increased from the standard time slot. The duration of the extended time slot window. The starting boundary of this window. equal minus / 2, the termination boundary is based on the standard data packet transmission duration and / 2 extends backward to determine.

[0078] Finally, the time slot allocation scheme is generated and distributed. The communication center assigns a unique logical channel identifier for each thus calculated, widened time slot window within the current scheduling period. Subsequently, a data structure containing the following information is created: the target remote access unit. The unique identifier, the assigned logical channel identifier, and the calculated actual starting boundary of the time slot window. And window length. This data structure constitutes a unit... The communication center distributes these independent time slot allocation schemes to each corresponding remote access unit in the network via downlink broadcast or directed signaling. Upon receiving their assigned scheme, each unit strictly adjusts its uplink data transmission timing according to the specified time slot window boundary parameters, thereby achieving synchronized and collision-free uplink data transmission in a dynamically changing transmission delay environment. This entire process is repeated periodically to ensure that the time slot allocation always matches the latest path delay characteristics.

[0079] In S5, the communication central office periodically selects remote access units to sample the measured round-trip time delay. The sampled measured time delay values ​​are compared with the time delay values ​​predicted based on the current propagation path time delay change profile. Based on the error generated by the comparison, the conversion relationship in step S2 is reversed to achieve adaptive optimization of the physical parameter to time delay parameter conversion process. Specifically, this includes:

[0080] First, the communication center periodically initiates measured round-trip delay sampling. Instead of sampling all remote access units, a priority strategy is employed to optimize signaling overhead and focus on the most likely inaccurate paths. Specifically, the communication center traverses the propagation path delay variation archives of all units, examining the recorded "disturbance margin index." In this method, this index is defined as the standard deviation of its random component, and its value reflects the severity of path delay fluctuations. A selection threshold is set, typically dynamically determined based on the statistical distribution of historical network-wide disturbance margin index values ​​(e.g., taking the 75th percentile of the index values ​​for all units). Remote access units whose current disturbance margin index exceeds this set threshold are preferentially selected as sampling targets. This strategy is based on the judgment that paths with severe delay fluctuations may have a deviated mapping relationship between their physical state and transmission delay due to extreme environmental changes, or may be inherently unstable, thus requiring more calibration through actual measurements.

[0081] After sampling the measured round-trip time delay for a selected unit, the measured value is obtained. Simultaneously, based on the current propagation path delay change profile of the unit (including the trend component and disturbance margin index in the profile body), a "predicted delay value" is calculated using the same logic as in step S4 for correcting the baseline round-trip time delay value. The absolute difference between the measured delay value and the predicted delay value is calculated; this is the total error of this sampling. Next, the error needs to be distributed in reverse. Since the total error originates from the entire optical path of the unit, it needs to be reasonably decomposed into various fiber segments along the optical path. The specific distribution is based on the contribution ratio of each segment to the total delay change. The communication center obtains the "transmission delay change" calculated independently for each segment in step S2, based on the continuous segment sequence of the unit's optical path. The absolute value of the current transmission delay change for each segment is divided by the sum of the absolute values ​​of the current transmission delay changes for all segments along the entire optical path, resulting in a weighting coefficient between 0 and 1. Then, the calculated total error is multiplied by the weight coefficient corresponding to each segment, and the result is used as the error responsibility amount allocated to that segment. If the total error is positive (the measured error is greater than the predicted error), the error responsibility amount allocated to each segment is also positive, indicating that the current mapping relationship of that segment may underestimate the time delay change; the opposite is also true.

[0082] Finally, for each segment assigned a non-zero error responsibility, the relevant mapping rules in its "local mapping relation library" are progressively adjusted. The adjustment process is meticulous and conservative to avoid overfitting due to single-sample noise. First, based on the segment's current physical state parameters (i.e., the strain and temperature changes that triggered the error), one or more "discrete nodes" closest to it are located in the multidimensional discrete space of its local mapping relation library. These nodes store preset output values ​​for transmission delay changes. Next, based on the error responsibility assigned to this segment, a preliminary "output value offset" is calculated. For example, if the error responsibility is positive, the offset is positive, indicating that the output value stored in that node needs to be adjusted upwards.

[0083] However, the offset is not directly applied to the nodes. Each discrete node maintains a "historical adjustment reputation record," a dynamically changing value with an initial value of 1. The update rule is as follows: after each adjustment to the node, if its prediction performance improves (error decreases) in a subsequent period (e.g., in the next sampling based on similar physical state parameters), its reputation record value increases by a fixed small step (e.g., 0.1), but not exceeding an upper limit (e.g., 2.0); if its performance deteriorates (error increases), its reputation record value decreases by a fixed step (e.g., 0.2), but not below a lower limit (e.g., 0.5). During the current adjustment, the current reputation record value of the target node is queried. A decay factor is obtained by dividing 1 by the reputation record value. The calculated initial output value offset is multiplied by this decay factor to obtain the "decayed offset." A higher reputation record value results in a smaller attenuation factor (less than 1), amplifying the adjustment range and indicating that the node's historical adjustments are effective and highly reliable. Conversely, a lower reputation record value results in a larger attenuation factor (greater than 1), suppressing the adjustment range and indicating that the node's historical adjustments are unstable and require caution. Finally, this attenuated offset is used to modify the output value of the transmission delay change stored in the target discrete node. After the update, the details of this adjustment are immediately recorded (including the node value before adjustment, offset, and physical state parameters that triggered the adjustment) so that the node's reputation record can be updated in subsequent evaluations. In this way, the transformation relationship can undergo slow, targeted, and stable self-optimization based on actual network performance.

[0084] The working principle of this invention is as follows: Real-time acquisition of strain and temperature physical parameters along the transmission line is achieved through an integrated optical fiber for communication and sensing. Using a pre-built local mapping database and bilinear interpolation, the changes in physical parameters of each segment are converted into changes in transmission delay. These changes are then integrated according to the topology path to obtain a dynamic delay data set for each remote access unit. Furthermore, a path delay change profile containing trend components and disturbance margin indices is established and continuously updated through multi-scale fluctuation analysis. The communication center dynamically corrects the baseline round-trip delay values ​​of each unit based on this profile and, combined with the calculated dynamic guard interval, reallocates uplink data transmission time slot windows with guard bands to absorb delay jitter. Finally, through periodic comparison of measured and predicted values, errors are back-allocated, and the local mapping database is progressively and self-learningly corrected based on historical adjustment credibility. This achieves continuous adaptive optimization of the physical parameter to delay parameter conversion process, ensuring accurate synchronization and reliable transmission in dynamic deformation environments.

[0085] The foregoing has provided a detailed description of one embodiment of the present invention, but this description is merely a preferred embodiment and should not be construed as limiting the scope of the invention. All equivalent variations and modifications made within the scope of the claims of this invention should still fall within the patent coverage of this invention.

Claims

1. A data communication method based on a highway electromechanical system, characterized in that, Includes the following steps: S1: Synchronously collect real-time physical state parameters of various spatial points along the highway infrastructure, including strain parameters and temperature parameters; S2: Based on real-time physical state parameters and according to the predetermined optical fiber physical characteristic conversion relationship, the optical fiber path is segmented and numerical comprehensive analysis is performed independently on each segment to convert the changes in physical state parameters of each segment into corresponding changes in transmission delay. The transmission delay changes calculated for all segments are integrated to generate a real-time data set reflecting the dynamic changes in delay of the entire optical path from the communication center to each remote access unit. S3: Based on real-time data sets, establish and continuously update a propagation path latency change profile for each remote access unit; S4: The communication center dynamically corrects the pre-stored baseline round-trip delay value based on the current propagation path delay change data of each remote access unit. Based on the corrected baseline round-trip time delay (RTD) value, the uplink data transmission time slot window is recalculated and allocated for each remote access unit, and the updated time slot allocation scheme is distributed to the corresponding units. The dynamic correction of the pre-stored baseline RTD value specifically includes: Read the main body of the archive recorded by the remote access unit from the current propagation path latency change archive; Extract the trend components of the records and the associated perturbation margin indicators from the main body of the archives; The value of the trend component is used as the main reference value for time delay correction, and the value of the disturbance margin index is multiplied by a predefined scaling factor and then superimposed on the main reference value to form the dynamic time delay correction amount. The dynamic delay correction adjusts the pre-stored reference round-trip delay value to generate the corrected reference round-trip delay value; The recalculation and allocation of uplink data transmission time slot windows for each remote access unit specifically includes: Historical data of trend components are extracted from the current propagation path delay change archive to calculate the trend component change rate, and combined with the disturbance margin index to jointly determine the protection interval parameter; For remote access units, based on the basic time slot start point determined by the corrected reference round-trip delay value, the time slot window is defined by extending the dynamic protection interval parameter forward and backward by half the duration. A logical channel identifier is assigned to each time slot window, and the logical channel identifier is bound to the corresponding time slot window boundary parameters to form a time slot allocation scheme, which is then sent to each corresponding remote access unit. S5: The communication center periodically selects remote access units to sample the actual round-trip delay. The sampled actual delay value is compared with the delay value predicted based on the current propagation path delay change profile. Based on the error generated by the comparison, the conversion relationship in step S2 is reversed to achieve adaptive optimization of the physical parameter to delay parameter conversion process.

2. The data communication method based on a highway electromechanical system according to claim 1, characterized in that, The process for obtaining the transmission delay change is as follows: Based on the conversion relationship of optical fiber physical characteristics, a local mapping relationship library is constructed for each segment. The local mapping relationship library defines the mapping rules from a multi-dimensional vector composed of strain parameter change value and temperature parameter change value to the single transmission delay change value of the corresponding segment. Extract the changes in the current physical state parameters of each segment and form a corresponding multi-dimensional vector. By querying the local mapping database and performing interpolation calculations, output the changes in transmission delay of the corresponding segment.

3. The data communication method based on a highway electromechanical system according to claim 1, characterized in that, The generation of the real-time data set reflecting the dynamic changes in delay along the entire optical path from the communication center to each remote access unit specifically includes: Based on the physical topology and logical connection relationship of the optical fiber network, identify and determine the continuous segment sequence that the optical signal transmission actually takes from the communication center to each remote access unit; According to the order of the continuous segment sequence corresponding to the remote access unit, a dynamic weighted accumulation operation is performed on the transmission delay change of all segments in the continuous segment sequence; The dynamically weighted cumulative result calculated by each remote access unit is bound to a unique identifier and a timestamp to construct an independent custom data structure. The custom structures corresponding to all remote access units together constitute a real-time data set.

4. The data communication method based on a highway electromechanical system according to claim 1, characterized in that, S3 specifically includes: From the real-time data set, extract the total delay change of the entire optical path corresponding to each remote access unit within a specified time window to form an independent time series of delay change for each remote access unit; Multi-scale fluctuation characteristic analysis is performed on the time series of delay change of each remote access unit to identify and separate the trend component caused by long-term structural deformation and the random component caused by instantaneous environmental disturbance. The trend components obtained from the analysis are integrated with the existing data in the current propagation path delay change archive to generate the updated archive body; at the same time, the statistical characteristics of the random components are stored in the archive as an auxiliary disturbance margin index, together forming a complete updated propagation path delay change archive.

5. A data communication method based on a highway electromechanical system according to claim 4, characterized in that, The generated updated file body specifically includes: The continuity and smoothness of the trend component are evaluated, and the confidence level of the change of the trend component relative to the trend component of the previous period recorded in the current archive is calculated. Based on the changing confidence level, the fusion weight coefficient between 0 and 1 is dynamically determined; By using the fusion weighting coefficient, the data of the current trend component is weighted and synthesized with the trend component data of the previous period recorded in the current archive body. The calculation result covers the trend component records in the original archive body, and an updated archive body is generated.

6. The data communication method based on a highway electromechanical system according to claim 1, characterized in that, S5 specifically includes: Prioritize the sampling of the measured round-trip time of remote access units whose disturbance margin index recorded in the propagation path delay change archive exceeds a set threshold; Calculate the error between the measured delay value and the delay value predicted based on the current archive, and then distribute the error in reverse to the corresponding segments according to the continuous segment sequence traversed by the optical path of the sampling unit; For each affected segment, the mapping rules defined in the local mapping relation library of the segment in step S2 are progressively adjusted according to the error value assigned to the corresponding segment, so as to reduce the prediction delay deviation generated in the future under the same physical state parameter input.

7. A data communication method based on a highway electromechanical system according to claim 6, characterized in that, The progressive adjustment of the mapping rules defined in the segmented local mapping relation library in step S2 specifically includes: Based on the error value assigned to the corresponding segment and the current physical state parameters of the corresponding segment, the affected discrete nodes are located in the multi-dimensional space corresponding to the local mapping relationship library. Based on the direction and magnitude of the error value, the required output value offset for the discrete node is calculated, and the offset corrects the corresponding node's currently defined transmission delay change output. Query the historical reputation adjustment records of discrete nodes; Based on historical reputation adjustment records, the offset is weighted and decayed, and the output value of the mapping rule defined for discrete nodes is updated with the decayed offset. At the same time, this adjustment is recorded to update the historical reputation adjustment records.

Images