A train continuous accurate positioning method and device based on bidirectional optical fiber sensing and time division multiplexing fusion and a storage medium

Abstract

This invention provides a method, apparatus, and storage medium for continuous and accurate train positioning based on bidirectional fiber optic sensing and time-division multiplexing fusion. The method includes: controlling fiber optic sensing units deployed at both ends of the sensing fiber to alternately transmit probe pulses and receive backscatter response signals in different time slots, with a guard interval set between the two measurement time slots to avoid crosstalk; performing event detection on the backscatter response signals at both ends to obtain a candidate position set; converting the second-end candidate positions to the first-end reference coordinate system and performing consistency verification to eliminate false events; weighting and fusing valid candidate position pairs according to quality indicators to obtain a fused positioning result; determining the confidence level based on an uncertainty model, and performing trajectory correlation on the multi-frame fused positioning results to output a continuous train position trajectory with a confidence level label. This invention improves positioning accuracy and uniformity, reduces false alarm rate, and supports degraded operation.

Description

Technical Field

[0001] This invention relates to the field of optical fiber sensing technology for railway transportation, and in particular to a method, device and storage medium for continuous and accurate train positioning based on bidirectional optical fiber sensing and time-division multiplexing fusion. Background Technology

[0002] As railway transportation systems develop towards intelligent and high-density operation, precise train positioning technology has become a crucial foundation for train operation control, monitoring, and maintenance management. Fiber optic sensing technology, particularly phase-sensitive optical time-domain reflectometers and distributed acoustic sensing technology, is considered a potential technological approach for achieving trackside train positioning due to its advantages such as distributed sensing capabilities, no need for power supply along the line, and resistance to electromagnetic interference.

[0003] However, existing fiber optic sensing train positioning schemes generally employ a single-end injection and single-end reception architecture. In this architecture, the probe pulse is injected from one end of the fiber, and the backscattered signal returns along the same path. Due to cumulative fiber attenuation, scattering loss, and detector chain noise, the signal-to-noise ratio of the returned signal decreases with increasing distance, resulting in uneven positioning accuracy along the line—higher accuracy near the injection end and significantly degraded accuracy at the far end. This distance-dependent error is particularly pronounced in long-distance applications, limiting the system's ability to provide uniform positioning accuracy across the entire line. Furthermore, the single-end scheme lacks a bidirectional redundancy verification mechanism. When localized mechanical shocks, environmental vibrations, electromagnetic interference, or sudden changes in Rayleigh scattering noise occur along the line, the single-end system cannot reliably distinguish between real train events and false events, leading to an increased false alarm rate and affecting system reliability.

[0004] Therefore, there is an urgent need for a train positioning solution that can systematically solve the above problems, so as to achieve uniform high-precision positioning across the entire line, effective suppression of false events, quantifiable quality assessment, and continuous and stable trajectory output. Summary of the Invention

[0005] To address the aforementioned issues, this invention proposes a method, device, and storage medium for continuous and precise train positioning based on bidirectional fiber optic sensing and time-division multiplexing fusion. Through collaborative processing of time-division multiplexing dual-end measurement, candidate position extraction, coordinate unification and consistency verification, adaptive fusion positioning, confidence assessment, and continuous trajectory output, continuous and precise train positioning is achieved, improving positioning accuracy and uniformity, reducing false alarm rate, and outputting quantifiable confidence assessment results.

[0006] This invention proposes a method for continuous and precise train positioning based on bidirectional fiber optic sensing and time-division multiplexing fusion, comprising: time-division multiplexing dual-end measurement: controlling a first-end fiber optic sensing unit and a second-end fiber optic sensing unit deployed at both ends of a sensing fiber optic cable laid along a railway line, alternately transmitting probe pulses into the sensing fiber optic cable and receiving backscattered response signals in different time slots within a measurement cycle, with a guard interval set between the measurement time slots at both ends, the duration of which is not less than the one-way propagation time of the optical signal in the sensing fiber optic cable, to avoid crosstalk caused by the back propagation of the dual-end probe pulses in the sensing fiber optic cable; candidate position extraction: performing event detection processing on the backscattered response signals obtained at both ends to obtain a first-end candidate position set and a second-end candidate position set; coordinate unification and consistency verification: verifying the coordinates of the first-end candidate position set and the second-end candidate position set. The two-end candidate position sets are transformed to a reference coordinate system with the first end as the origin. Consistency verification is performed between the first-end candidate position set and the transformed second-end candidate position set. Candidate position pairs that meet the consistency constraints are determined as valid candidate position pairs, while those that do not meet the consistency constraints are identified as false events and eliminated. Adaptive fusion positioning: Dynamic weights are assigned to the first-end and second-end candidate positions in the valid candidate position pairs according to quality indicators. Weighted fusion is then performed on the valid candidate position pairs to obtain the fused positioning result. Confidence assessment and continuous trajectory output: The confidence level of the fused positioning result is determined based on the positioning uncertainty model. Trajectory association is performed on the fused positioning results of multiple consecutive measurement frames to output a continuous train position trajectory with a confidence level label. This method can reduce distance correlation errors through symmetrical measurement at both ends, avoid crosstalk between ends through time-division multiplexing frame management, eliminate false events through consistency verification, improve positioning accuracy through adaptive fusion, and output quantifiable accuracy boundaries through uncertainty modeling.

[0007] This invention proposes a continuous and precise train positioning device based on bidirectional fiber optic sensing and time-division multiplexing fusion, comprising: a time-division multiplexing dual-end measurement module configured to control a first-end fiber optic sensing unit and a second-end fiber optic sensing unit respectively deployed at both ends of a sensing fiber optic cable laid along a railway line, alternately transmitting probe pulses to the sensing fiber optic cable and receiving backscattered response signals in different time slots within a measurement cycle, wherein a guard interval is set between the measurement time slots at both ends, the duration of which is not less than the one-way propagation time of the optical signal in the sensing fiber optic cable; a candidate position extraction module configured to perform event detection processing on the backscattered response signals acquired at both ends respectively, and obtain a first-end candidate position set and a second-end candidate position set; and a coordinate unification and consistency verification module configured to convert the second-end candidate position set. Using a reference coordinate system with the first end as the origin, consistency verification is performed on the first end candidate position set and the transformed second end candidate position set. Candidate position pairs that meet the consistency constraints are determined as valid candidate position pairs, while candidate positions that do not meet the consistency constraints are judged as false events and eliminated. The adaptive fusion positioning module is configured to assign dynamic weights to the first end candidate positions and the second end candidate positions in the valid candidate position pairs according to quality indicators, and perform weighted fusion on the valid candidate position pairs to obtain the fused positioning result. The confidence assessment and trajectory output module is configured to determine the confidence level of the fused positioning result based on the positioning uncertainty model, and perform trajectory association on the fused positioning results of multiple consecutive measurement frames to output a continuous train position trajectory with a confidence label. It can reduce distance correlation error through dual-end symmetrical measurement, avoid dual-end crosstalk through time-division multiplexing frame management, eliminate false events through consistency verification, improve positioning accuracy through adaptive fusion, and output quantifiable accuracy boundaries through uncertainty modeling.

[0008] An electronic device according to an embodiment of the present invention includes: at least one processor; and a memory coupled to the at least one processor, the memory being used to store instructions that, when executed by the at least one processor, cause the processor to perform the method described above.

[0009] The present invention provides a computer-readable storage medium storing computer instructions, which, when executed, perform the methods described in any of the above embodiments.

[0010] An embodiment of the present invention provides a computer program product, which includes a computer program that, when executed by a processor, implements the method described in any of the above embodiments. Attached Figure Description

[0011] The following figures are intended only to illustrate and explain the invention and do not limit the scope of the invention.

[0012] Figure 1 A flowchart illustrating the continuous and precise train positioning method based on bidirectional fiber optic sensing and time-division multiplexing fusion provided in an embodiment of the present invention.

[0013] Figure 2 A schematic diagram of the structure of a train continuous and precise positioning device based on bidirectional fiber optic sensing and time-division multiplexing fusion provided in an embodiment of the present invention;

[0014] Figure 3 This is a schematic diagram of the structure of an electronic device provided in an embodiment of the present invention.

[0015] The annotations in the attached figures are explained as follows:

[0016] serial number meaning 100 A Continuous and Precise Train Positioning Method Based on Bidirectional Fiber Optic Sensing and Time-Division Multiplexing Fusion S101 Time-division multiplexing dual-ended measurement S102 Candidate position extraction S103 Coordinate unification and consistency verification S104 Adaptive Fusion Positioning S105 Confidence assessment and continuous trajectory output 200 Train Continuous Precise Positioning Device Based on Bidirectional Fiber Optic Sensing and Time Division Multiplexing Fusion 201 Time-division multiplexed dual-ended measurement module 202 Candidate location extraction module 203 Coordinate unification and consistency verification module 204 Adaptive fusion positioning module 205 Confidence assessment and trajectory output module 300 electronic devices 310 processor Detailed Implementation

[0017] To provide a clearer understanding of the technical features, objectives, and effects of the present invention, specific embodiments of the present invention will now be described with reference to the accompanying drawings.

[0018] Many specific details are set forth in the following description in order to provide a full understanding of the invention. However, the invention may also be practiced in other ways different from those described herein, and therefore the invention is not limited to the specific embodiments disclosed below.

[0019] As indicated in this application and claims, unless the context clearly indicates otherwise, the words "a," "an," "an," and / or "the" do not specifically refer to the singular and may also include the plural. Generally speaking, the terms "comprising" and "including" only indicate the inclusion of explicitly identified steps and elements, which do not constitute an exclusive list, and the method or apparatus may also include other steps or elements.

[0020] Before introducing the method flow of this invention, the overall system architecture involved in this invention will first be described. The train continuous precise positioning system involved in this embodiment of the invention includes: a sensing optical fiber continuously laid along the railway line, a first-end optical fiber sensing unit and a second-end optical fiber sensing unit respectively deployed at both ends of the sensing optical fiber, a dual-end time synchronization module, a time-division multiplexing frame management module, and a processing unit. The sensing optical fiber is a single-mode optical fiber and can be deployed near sleepers, beside rails, in ballast layers, cable ducts, or other suitable locations to sense wheel-rail vibration, track structure coupled vibration, or related dynamic disturbances caused by train operation. In one embodiment, the first-end optical fiber sensing unit and the second-end optical fiber sensing unit can be phase-sensitive optical time-domain reflectometers or distributed acoustic sensing devices. The dual-end time synchronization module is used to achieve a unified time reference for the first-end optical fiber sensing unit and the second-end optical fiber sensing unit. In one embodiment, the dual-end time synchronization module uses a precision time protocol to achieve clock synchronization, and can further combine at least one of the following methods to improve synchronization accuracy: hardware timestamp module, temperature-compensated crystal oscillator, synchronous Ethernet, White Rabbit protocol, or fiber optic time synchronization. Preferably, the synchronization error is controlled at the nanosecond level; let the clock deviation between the two ends be... The distance error introduced by this is approximately the speed of light divided by twice the refractive index and then multiplied by the clock deviation. Therefore, controlling the synchronization error to the nanosecond level can keep the distance error caused by the clock within the meter or sub-meter range. The time-division multiplexing frame management module controls the first-end fiber optic sensing unit and the second-end fiber optic sensing unit to alternately transmit probe pulses in different time slots within a measurement cycle, and sets a guard interval between the measurement time slots at both ends. The time-division multiplexing frame management module can be located in either end of the fiber optic sensing unit, in an edge computing node, or in a central control unit. The processing unit performs candidate location extraction, coordinate unification and consistency verification, adaptive fusion positioning, confidence assessment, and continuous trajectory output. The processing unit can be located in an edge computing node or a data center. In one embodiment, the system also includes a temperature acquisition module for acquiring temperature information along the sensing fiber for the processing unit to perform refractive index drift compensation. The temperature acquisition module can employ a distributed temperature sensing module, an additional temperature sensor, an ambient temperature database interface, or other methods to acquire temperature information along the fiber.

[0021] To avoid crosstalk caused by the back propagation of dual-end probe pulses in the sensing fiber, this invention employs a time-division multiplexing frame structure. A preferred measurement cycle includes the following time slots: a synchronization time slot, used for frame start alignment, time verification, or exchange of synchronization control information between the two ends; this time slot is necessary during the initial communication establishment and can also be set in each measurement cycle; a first-end measurement time slot, in which the first-end fiber sensing unit emits probe pulses and completes the corresponding backscatter reception; a guard interval, used to ensure that the second-end fiber sensing unit enters the measurement time slot after the echo associated with the first-end probe pulse has attenuated; a second-end measurement time slot, in which the second-end fiber sensing unit emits probe pulses and completes the corresponding backscatter reception; and a data processing time slot, used for candidate result exchange, fusion processing, or reporting control information, which can be optionally set according to data transmission needs. A complete measurement cycle can be expressed as the sum of the durations of each time slot. Double guard intervals or more complex alternating dual-end measurement sequences can also be set. Let the length of the sensing fiber be... The refractive index is The speed of light is Then the protection interval should meet the following requirements: The first and second end measurement time slots should meet the requirements for complete echo reception, that is:

[0022] By meeting the above conditions, crosstalk caused by the back propagation of double-ended probe pulses in the optical fiber can be effectively prevented.

[0023] See Figure 2 This invention provides a train continuous and precise positioning device 200 based on bidirectional fiber optic sensing and time-division multiplexing fusion. The device includes: a time-division multiplexing dual-end measurement module 201, a candidate position extraction module 202, a coordinate unification and consistency verification module 203, an adaptive fusion positioning module 204, and a confidence assessment and trajectory output module 205. The functions of each module correspond one-to-one with the following method steps.

[0024] See Figure 1 This invention provides a method 100 for continuous and precise train positioning based on bidirectional fiber optic sensing and time-division multiplexing fusion, the method including steps S101 to S105.

[0025] In step S101, time-division multiplexing dual-end measurement is performed: First-end fiber optic sensing units and second-end fiber optic sensing units, respectively deployed at both ends of the sensing fiber optic cable laid along the railway line, are controlled to alternately transmit probe pulses into the sensing fiber optic cable and receive backscattered response signals in different time slots within a measurement cycle. A guard interval is set between the measurement time slots at both ends, the duration of which is not less than the one-way propagation time of the optical signal in the sensing fiber optic cable, to avoid crosstalk caused by the back propagation of the dual-end probe pulses in the sensing fiber optic cable. In one embodiment, since train events are visible from both ends simultaneously, the candidate positions obtained at both ends satisfy a geometric complementary relationship. When one end is farther from the train, resulting in a decrease in the signal-to-noise ratio, the other end, being relatively closer, usually has better signal quality, thus providing complementary information for subsequent fusion. This step corresponds to the time-division multiplexing dual-end measurement module 201 in the device.

[0026] In step S102, candidate position extraction involves performing event detection processing on the backscatter response signals acquired at both ends to obtain a first set of candidate positions and a second set of candidate positions. In one embodiment, the event detection processing includes preprocessing, background suppression, differential detection, threshold-based or statistical event identification, and candidate peak extraction on the optical time-domain reflectometer traces acquired at both ends. This step corresponds to the candidate position extraction module 202 in the device.

[0027] In step S103, coordinate unification and consistency verification: Since the candidate position at the second end has its origin at the second end, it needs to be transformed to a reference coordinate system with its origin at the first end. Let the equivalent measurement length of the sensing fiber be L, then the transformation result of the candidate position at the second end is:

[0028] in, The second candidate position is defined with the second end as the origin. After the transformation, the consistency verification is performed on the first candidate position set and the transformed second candidate position set. Candidate positions in the first candidate position set are paired with candidate positions in the transformed second candidate position set. When the distance difference between two paired candidate positions does not exceed the consistency tolerance, the candidate position pair is determined to correspond to the same train event and is identified as a valid candidate position pair. Candidate positions that do not meet the consistency constraints are judged as false events and are eliminated. In one embodiment, the consistency tolerance can be dynamically set according to spatial resolution, noise level, speed estimation, or historical trajectory stability. Candidate positions that fail the consistency verification can be directly discarded, temporarily stored for confirmation in subsequent frames, or marked as low confidence. This step corresponds to the coordinate unification and consistency verification module 203 in the device.

[0029] In step S104, adaptive fusion localization: For valid candidate location pairs that pass the consistency verification, a weighted fusion method is used to obtain the final localization result.

[0030] and The weights are dynamic and can be determined based on at least one quality metric, including but not limited to: real-time signal-to-noise ratio (SNR), peak width, candidate peak shape quality, historical trajectory continuity, residual stability, and channel health status. In one embodiment, when only the real-time SNR of the two ends is used as the basis for weight allocation, the weight of the first end is the SNR of the first end divided by the sum of the SNRs of the two ends, and the weight of the second end is the SNR of the second end divided by the sum of the SNRs of the two ends. When the quality of the two ends is similar, the fusion degenerates to an arithmetic mean. After fusion, a two-end residual metric can also be determined, which is the distance difference between the first-end candidate position and the second-end candidate position in the candidate position pair; a half-residual can also be defined, which is half of the two-end residual, to characterize the internal consistency of the fusion positioning result. This step corresponds to the adaptive fusion positioning module 204 in the device.

[0031] In step S105, confidence assessment and continuous trajectory output: the confidence level of the fused positioning result is determined based on the positioning uncertainty model. The positioning uncertainty model established in this invention includes the following uncertainty factors: spatial resolution constraint uncertainty. Uncertainty of refractive index and environmental drift error Uncertainty of time synchronization error between two ends Candidate peak extraction algorithm error uncertainty Uncertainty of coordinate unification and length calibration error Two-end fusion residual uncertainty Standard uncertainty can be expressed as the combination of various uncertainty factors, for example:

[0032] Through the aforementioned uncertainty modeling, this invention provides a basis for system accuracy evaluation, quality threshold setting, and application adaptation. Confidence levels can include: high confidence, medium confidence, low confidence, and invalid results. When the fused positioning result meets the quality threshold condition, a valid positioning result with a confidence interval is output; otherwise, the result is marked as low confidence or invalid, and at least one of the following operations can be triggered: re-acquisition, trajectory holding, degraded operation output, or issuing a quality alarm to the upper-level system. The fused positioning results of multiple consecutive measurement frames are correlated to form a time-continuous train position trajectory, which is then output. The output continuous train position trajectory carries a confidence label, providing high-quality position input for upper-level train control, operation monitoring, or maintenance management systems. This step corresponds to the confidence evaluation and trajectory output module 205 in the device.

[0033] In one embodiment, when either the first-end fiber optic sensing unit or the second-end fiber optic sensing unit fails, experiences synchronization anomalies, or has insufficient quality, the system can switch to single-end positioning mode, outputting a degraded positioning result and the corresponding degraded confidence level. This degraded operation mode ensures that the system can still provide positioning services even under conditions of partial device malfunction, improving system availability.

[0034] See Figure 3 This invention also provides an electronic device 300, including: at least one processor 310 and a memory 320 coupled to the at least one processor. The memory 320 is used to store instructions, which, when executed by the at least one processor 310, cause the processor to perform the method as described in any of the above embodiments. In one embodiment, the electronic device may be an edge computing node, an industrial control computer, a server, or other device with computing capabilities. The processor may be a central processing unit, a graphics processing unit, a field-programmable gate array, or a digital signal processor. The memory may be random access memory, read-only memory, flash memory, or disk storage.

[0035] This invention also provides a computer-readable storage medium storing computer instructions thereon, which, when executed, perform the method according to any one of the above embodiments. The computer-readable storage medium can be a non-transitory computer-readable storage medium, such as a read-only memory, random access memory, flash memory, magnetic disk, optical disk, etc.

[0036] This invention also provides a computer program product, including a computer program that, when executed by a processor, implements the method described in any of the above embodiments.

[0037] Some aspects of the methods and apparatus of this invention can be executed entirely by hardware, entirely by software (including firmware, resident software, microcode, etc.), or by a combination of hardware and software. The aforementioned hardware or software can be referred to as a "data block," "module," "engine," "unit," "component," or "system." The processor can be one or more application-specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), field-programmable gate arrays (FPGAs), processors, controllers, microcontrollers, microprocessors, or combinations thereof. Computer-readable media may include, but are not limited to, magnetic storage devices (e.g., hard disks, floppy disks, magnetic stripes), optical discs (e.g., compact discs (CDs), digital multifunction discs (DVDs)), smart cards, and flash memory devices (e.g., electrically erasable programmable read-only memory (EPROMs), cards, sticks, key drivers).

[0038] It should be understood that the preceding operations do not necessarily need to be performed precisely in sequence. Instead, the various steps can be processed in reverse order or simultaneously. At the same time, other operations may be added to these processes, or one or more operations may be removed from them.

[0039] It should be understood that although this specification is described according to various embodiments, not every embodiment contains only one independent technical solution. This way of describing the specification is only for clarity. Those skilled in the art should regard the specification as a whole. The technical solutions in each embodiment can also be appropriately combined to form other implementation methods that can be understood by those skilled in the art.

[0040] The above description is merely an illustrative embodiment of the present invention and is not intended to limit the scope of the invention. Any equivalent changes, modifications, and combinations made by those skilled in the art without departing from the concept and principles of the present invention should fall within the scope of protection of the present invention.

Claims

1. A method for continuous and precise train positioning based on bidirectional fiber optic sensing and time-division multiplexing fusion (100), characterized in that, The method includes: Time-division multiplexing dual-end measurement (S101): Controlling the first-end fiber optic sensing unit and the second-end fiber optic sensing unit deployed at both ends of the sensing fiber optic cable laid along the railway line to alternately transmit probe pulses to the sensing fiber optic cable and receive backscatter response signals in different time slots within a measurement cycle; wherein, a guard interval is set between the measurement time slots at both ends, and the duration of the guard interval is not less than the one-way propagation time of the optical signal in the sensing fiber optic cable, so as to avoid crosstalk caused by the back propagation of the dual-end probe pulses in the sensing fiber optic cable. Candidate position extraction (S102): Perform event detection processing on the backscatter response signals obtained from both ends respectively to obtain the first end candidate position set and the second end candidate position set; Coordinate unification and consistency verification (S103): Transform the second end candidate position set to a reference coordinate system with the first end as the origin, perform consistency verification on the first end candidate position set and the transformed second end candidate position set, determine the candidate position pairs that meet the consistency constraint conditions as valid candidate position pairs, and determine the candidate positions that do not meet the consistency constraint conditions as false events and remove them; Adaptive fusion localization (S104): Dynamic weights are assigned to the first and second candidate positions in the effective candidate position pairs according to the quality index, and the effective candidate position pairs are weighted and fused to obtain the fusion localization result; Confidence assessment and continuous trajectory output (S105): Based on the positioning uncertainty model, determine the confidence level of the fused positioning result, perform trajectory association on the fused positioning results of multiple continuous measurement frames, and output a continuous train position trajectory with confidence label.

2. The method according to claim 1, characterized in that, The duration of the protection interval is not less than the product of the refractive index of the sensing fiber and the length of the sensing fiber divided by the speed of light; the duration of the measurement time slots at both ends is not less than twice the product of the refractive index of the sensing fiber and the length of the sensing fiber divided by the speed of light.

3. The method according to claim 1, characterized in that, The measurement cycle also includes a synchronization time slot and a data processing time slot; the synchronization time slot is used for frame start alignment and dual-end time verification, and the data processing time slot is used for candidate result exchange or fusion processing.

4. The method according to claim 1, characterized in that, The coordinate unification includes: transforming the candidate position of the second end, with the second end as the origin, to a reference coordinate system with the first end as the origin by subtracting the candidate position of the second end from the equivalent measurement length of the sensing fiber.

5. The method according to claim 1, characterized in that, The consistency verification includes: pairing candidate positions in the first candidate position set with candidate positions in the transformed second candidate position set; when the distance difference between the two paired candidate positions does not exceed the consistency tolerance, it is determined that the candidate positions correspond to the same train event; the consistency tolerance is dynamically set according to spatial resolution, noise level or historical trajectory stability.

6. The method according to claim 1, characterized in that, The quality indicators include at least one of the following: real-time signal-to-noise ratio, peak width, candidate peak shape quality, historical trajectory continuity, residual stability, and channel health status.

7. The method according to claim 1, characterized in that, The positioning uncertainty model synthesizes at least two uncertainty factors from spatial resolution limitation, refractive index and environmental drift error, dual-end time synchronization error, candidate peak extraction algorithm error, coordinate unification and length calibration error, and dual-end fusion residual to obtain the standard uncertainty.

8. The method according to claim 1, characterized in that, The confidence levels include high confidence, medium confidence, low confidence, and invalid results; when the fused positioning result does not meet the quality threshold conditions, at least one of the following operations is triggered: re-acquisition, trajectory holding, degraded operation output, or sending a quality alarm to the upper-level system.

9. The method according to claim 1, characterized in that, The method further includes a degradation operation step: when one end of the first-end fiber optic sensing unit or the second-end fiber optic sensing unit fails, has an abnormal synchronization, or has insufficient quality, it switches to single-end positioning mode and outputs the degradation positioning result and the corresponding degradation confidence level.

10. The method according to claim 1, characterized in that, The first-end fiber optic sensing unit and the second-end fiber optic sensing unit use a precision time protocol to achieve clock synchronization, and combine at least one of the following methods to improve synchronization accuracy: hardware timestamp, temperature-compensated crystal oscillator, synchronous Ethernet, White Rabbit protocol or fiber optic time synchronization.

11. The method according to claim 1, characterized in that, After the weighted fusion, a two-end residual index is determined, which is the distance difference between the first-end candidate position and the second-end candidate position in the candidate position pair.

12. A train continuous and precise positioning device (200) based on bidirectional fiber optic sensing and time-division multiplexing fusion, characterized in that, The device includes: The time-division multiplexing dual-ended measurement module (201) is configured to control a first-end fiber optic sensing unit and a second-end fiber optic sensing unit, respectively deployed at both ends of a sensing fiber optic cable laid along a railway line, to alternately transmit probe pulses to the sensing fiber optic cable and receive backscatter response signals in different time slots within a measurement cycle; wherein a guard interval is set between the measurement time slots at both ends, the duration of which is not less than the one-way propagation time of the optical signal in the sensing fiber optic cable, so as to avoid crosstalk caused by the back propagation of the dual-ended probe pulses in the sensing fiber optic cable. The candidate position extraction module (202) is configured to perform event detection processing on the backscatter response signals obtained from both ends respectively, and obtain a first end candidate position set and a second end candidate position set; The coordinate unification and consistency verification module (203) is configured to transform the second end candidate position set to a reference coordinate system with the first end as the origin, perform consistency verification on the first end candidate position set and the transformed second end candidate position set, determine the candidate position pairs that meet the consistency constraint conditions as valid candidate position pairs, and determine the candidate positions that do not meet the consistency constraint conditions as false events and remove them; The adaptive fusion localization module (204) is configured to assign dynamic weights to the first end candidate position and the second end candidate position in the effective candidate position pair according to the quality index, and to perform weighted fusion on the effective candidate position pair to obtain the fusion localization result. The confidence assessment and trajectory output module (205) is configured to determine the confidence level of the fused positioning result based on the positioning uncertainty model, and to perform trajectory association on the fused positioning result of multiple consecutive measurement frames, and output a continuous train position trajectory with confidence labels.

13. An electronic device (300), comprising: At least one processor (310); as well as A memory (320) coupled to the at least one processor, the memory being used to store instructions that, when executed by the at least one processor, cause the processor to perform the method as described in any one of claims 1-11.

14. A computer-readable storage medium having stored thereon computer instructions that, when executed, perform the method according to any one of claims 1-11.

15. A computer program product, characterized in that, Includes a computer program, which, when executed by a processor, performs the method according to any one of claims 1-11.

Images