Intelligent runway fusion positioning system

Abstract

The present application relates to the technical field of positioning, in particular to a smart runway fusion positioning system, which comprises a terminal sensing module, which synchronously receives signals based on infrared emitters, Bluetooth base stations and GPS reference stations, records optical frequency band amplitude and radio frequency intensity value correction time sequence marks in time, and revises amplitude deviation according to reference scale difference comparison. In the present application, through synchronous collection and amplitude calibration of infrared emission signals, Bluetooth radio frequency signals and GPS reference signals, real-time multi-dimensional continuous positioning of athletes in the runway is realized, the runway position perception is freed from the problem of being easily restricted by single positioning technology, and the problem of a significant decline in positioning accuracy in complex lighting conditions, indoor environments or signal shielding areas is solved. A stable threshold calibration method is established based on cross verification of optical signals and radio frequency power, which effectively resists environmental interference and signal fluctuation, and improves the accuracy and reliability of the positioning result.

Description

Technical Field

[0001] This invention relates to the field of positioning technology, and in particular to a smart track fusion positioning system. Background Technology

[0002] In traditional track and field events and training, the monitoring of athletes' positions mainly relies on human referees or simple photoelectric sensors. Human referees' judgments are subject to subjectivity and error, and it is difficult to accurately record the athlete's position throughout the entire race; while simple photoelectric sensors can only provide limited information at specific detection points and cannot achieve continuous, real-time tracking of athletes' positions on the track.

[0003] With the development of technology, although some systems based on a single positioning technology (such as GPS) have emerged, these systems have limitations in practical applications. For example, the accuracy of GPS positioning drops significantly indoors or in environments with severe signal obstruction; infrared sensing positioning is easily affected by ambient light and obstacles; and Bluetooth positioning has a relatively small coverage area. Therefore, a single positioning technology cannot meet the needs of complex and ever-changing runway environments and high-precision positioning. Summary of the Invention

[0004] The purpose of this invention is to address the shortcomings of existing technologies by proposing a smart track fusion positioning system.

[0005] To achieve the above objectives, the present invention adopts the following technical solution: The intelligent track fusion positioning system includes:

[0006] The terminal sensing module synchronously receives signals from infrared transmitters, Bluetooth base stations, and GPS reference stations. It records optical band amplitude and radio frequency intensity values ​​in a time-division manner to calibrate timing marks. It corrects amplitude deviations based on the difference compared with the reference scale. It retrieves location information identifiers and matches identification codes, organizes multiple observation records to summarize coordinate values, and generates a sensing sequence.

[0007] The data calibration module compares the radio frequency power value and optical amplitude value based on the sensing sequence, extracts the speed threshold and acceleration threshold, retrieves ground marking data around the runway, compares the difference based on adjacent sampling intervals and marks the fluctuation range, and obtains the calibration parameter set by comparing the position coordinates of the infrared transmitter and confirming the synchronization time point.

[0008] The edge processing module filters optical amplitude thresholds and radio frequency energy thresholds based on the calibration parameter group, compares timestamps in segments and uses runway plane coordinate information, compares radio frequency energy fluctuations in groups after each time period and identifies positioning differences, performs orientation judgment on Bluetooth base station spatial index and corrects inconsistent coordinates, and generates fused tracking vectors.

[0009] The interactive communication module retrieves the cloud communication address and edge gateway number based on the fused tracking vector, counts the interference level and determines the available frequency band during occupancy scanning, performs occupancy scanning on multiple Bluetooth base station radio frequency channels and records the effective upload time period, associates runway coordinates with data transmission rate and outputs transmission path markers, and generates a network exchange index.

[0010] Better options also include:

[0011] The motion analysis module acquires athlete speed and acceleration data based on the network exchange index and parses the interval division from the start to the finish line of the track. When parsing the interval, it extracts the start and finish lines, counts the step frequency value and compares it with the Bluetooth base station positioning trajectory. It performs spatiotemporal reconstruction in combination with GPS reference station coordinate information and stores the results in groups to generate a motion feature set.

[0012] The visualization output module filters the coordinate values ​​of the inner and outer boundaries of the track based on the action feature set to generate continuous trajectory nodes. After distinguishing the athlete's speed segments, it superimposes the Bluetooth base station position relationship. When marking, it sets a speed threshold range to map the information of each segment with the coordinates of the Bluetooth base station. It combines the selected nodes and trajectory information and dynamically draws the layered coordinate projection to generate a graphic coordinate distribution.

[0013] Preferably, the terminal sensing module includes:

[0014] The synchronous acquisition submodule receives signals from infrared transmitters, Bluetooth base stations, and GPS reference stations synchronously. It records optical band amplitude and radio frequency strength values ​​in a time-division manner, calibrates timing markers, parses the timestamps of each record, compares amplitude differences, associates record indexes to mark the receiving order, and generates an acquisition record set.

[0015] The difference correction submodule, based on the acquired record set, reads the optical frequency band amplitude one by one according to the reference scale to correct the difference correction magnitude deviation, corrects the difference between the optical frequency band amplitude and the reference scale, determines the correction magnitude of each record and updates the observation index, and generates corrected observation data.

[0016] The coordinate aggregation submodule, based on the corrected observation data, retrieves the location information identifier matching identification code, compares the identification code with the observation index and merges duplicates, summarizes the spatial coordinates of all records and aggregates them into a unified index, and generates a sensing sequence.

[0017] Preferably, the data calibration module includes:

[0018] The amplitude comparison submodule compares the radio frequency power value and the optical amplitude value based on the sensing sequence. After reading the radio frequency power value and the optical amplitude value, it calculates the difference respectively, records the corresponding observation number according to the absolute value of each difference, and generates a comparison result set.

[0019] The threshold extraction submodule extracts velocity and acceleration thresholds based on the comparison result set and retrieves ground marking data around the runway. It divides the range according to the velocity and acceleration values ​​of each record in the comparison result set, summarizes the observation numbers within the same range and marks the location information, and generates a threshold mark set.

[0020] The coordinate comparison submodule, based on the threshold mark set, compares the difference between adjacent sampling intervals and marks the fluctuation range, reads the amplitude change within adjacent observation numbers and determines the fluctuation threshold, confirms the synchronization time point by referring to the position coordinates of the infrared transmitter, and obtains the calibration parameter set.

[0021] Preferably, the edge processing module includes:

[0022] The threshold filtering submodule filters optical amplitude thresholds and radio frequency energy thresholds based on the calibration parameter group, reads the amplitude range marked in the calibration parameter group and limits the upper limit of radio frequency energy, identifies records that are below the lower limit or above the upper limit and summarizes the index to generate a set of filtering thresholds.

[0023] The time period comparison submodule, based on the set of filtering thresholds, compares timestamps in segments and uses runway plane coordinate information to mark outliers based on amplitude fluctuations within consecutive time periods, merges and compares radio frequency energy in the same time period, and generates time period comparison data.

[0024] The coordinate revision submodule performs orientation judgment and corrects inconsistent coordinates on the Bluetooth base station spatial index based on the time period comparison data. It reads the base station coordinates corresponding to the abnormal point and calculates the offset with the reference coordinates, replaces the mismatched coordinates and updates the index synchronously, and generates a fused tracking vector.

[0025] Preferably, the interactive communication module includes:

[0026] The address retrieval submodule retrieves cloud communication addresses and edge gateway numbers based on the fused tracking vectors. It matches the base station indexes in each retrieval vector with the cloud list, reads the corresponding gateway numbers and maps them to a specified address range, and generates communication mapping data.

[0027] The interference assessment submodule, based on the communication mapping data, counts the interference level and determines the available frequency bands during the occupancy scan, tracks the occupancy duration and interference amplitude of the radio frequency channel, records the interference value and the corresponding frequency band index, and generates a set of available channels.

[0028] The path marking submodule, based on the available channel set, associates runway coordinates with data transmission rates and outputs transmission path markings, reads gateway information corresponding to available frequency bands and cross-compares rate values ​​and coordinate information to generate a network switching index.

[0029] Preferably, the motion analysis module includes:

[0030] The data extraction submodule, based on the network exchange index, acquires athlete speed and acceleration data and parses the interval division from the start to the finish line of the track, reads speed and acceleration values ​​and locates the start and finish lines, records the interval index and corresponding speed level, and generates a speed and acceleration set.

[0031] The interval division submodule, based on the velocity and acceleration set, extracts the start and end points when parsing the interval, counts the step frequency value and compares it with the Bluetooth base station positioning trajectory, checks the observation sequence corresponding to the rate level and marks the track difference, and generates step frequency comparison value.

[0032] The spatiotemporal reconstruction submodule, based on the step frequency comparison value, combines GPS base station coordinate information and stores the result grouping, reads the step frequency and coordinate pairing data and arranges the time series, organizes adjacent sampling points under the same series and merges them into a unified interval index, and generates an action feature set.

[0033] Preferably, the visualization output module includes:

[0034] The boundary filtering submodule, based on the action feature set, filters the coordinate values ​​of the inner and outer boundaries of the runway to generate continuous trajectory nodes, reads the coordinate index in the action feature set and removes data that is not within the runway range, generates continuous coordinates according to the node order that meets the range, and generates a coordinate node set.

[0035] The speed segmentation submodule, based on the coordinate node set, distinguishes the relationship between the Bluetooth base station position after the athlete's speed is segmented, reads the coordinate nodes and obtains the corresponding speed values ​​from the action feature set, marks the segments that exceed the threshold and identifies them in the coordinate index, and generates superimposed position information.

[0036] The coordinate mapping submodule, based on the superimposed position information, merges the selected node and trajectory information and dynamically draws the layered coordinate projection, reads the coordinate index corresponding to the superimposed position information and assigns the layered structure to the rendering sequence, organizes the drawing instructions and outputs the positioning graphic, generating the graphic coordinate distribution.

[0037] Compared with the prior art, the advantages and positive effects of the present invention are as follows:

[0038] This invention achieves real-time, multi-dimensional, continuous positioning of athletes within the track by synchronously acquiring and calibrating infrared transmission signals, Bluetooth radio frequency signals, and GPS reference signals. This overcomes the environmental limitations inherent in single-positioning technologies and addresses the significant decrease in positioning accuracy under complex lighting conditions, indoor environments, or areas with signal obstruction. A stable threshold calibration method based on cross-validation of optical signals and radio frequency power effectively resists environmental interference and signal fluctuations, improving the accuracy and reliability of positioning results and ensuring high-fidelity recording of athletes' dynamic trajectories. Through segmented grouping and fusion processing of spatial coordinates and temporal sequences, differences and abnormal fluctuations in spatial coordinate data are corrected in real time, enhancing the coherence and spatiotemporal consistency of location data and providing continuous and detailed motion trajectories. Based on network scanning, frequency band filtering, and interference avoidance strategies, stable and smooth information transmission is ensured, eliminating the risk of communication blockage, improving the efficiency of location tracking data transmission, and further optimizing real-time performance. Combining athlete movement characteristics with track coordinate information enables in-depth analysis and detailed spatiotemporal reconstruction of the movement process, enhancing the support capability of location data for motion state analysis and providing an intuitively traceable and precise representation of the motion trajectory. Attached Figure Description

[0039] Figure 1 This is a system flowchart of the present invention. Detailed Implementation

[0040] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention.

[0041] Please see Figure 1 The present invention provides a technical solution: a smart track fusion positioning system comprising:

[0042] The terminal sensing module synchronously receives signals from infrared transmitters, Bluetooth base stations, and GPS reference stations. It records optical band amplitude and radio frequency intensity values ​​in a time-division manner to calibrate timing marks. It corrects amplitude deviations based on the difference compared with the reference scale. It retrieves location information identifiers and matches identification codes, organizes multiple observation records to summarize coordinate values, and generates a sensing sequence.

[0043] The data calibration module compares the radio frequency power value and optical amplitude value based on the sensing sequence, extracts the speed threshold and acceleration threshold, retrieves the ground marking data around the runway, compares the difference according to the adjacent sampling interval and marks the fluctuation range, and obtains the calibration parameter set by comparing the position coordinates of the infrared transmitter and confirming the synchronization time point.

[0044] The edge processing module filters optical amplitude thresholds and radio frequency energy thresholds based on calibration parameter groups, compares timestamps in segments and uses runway plane coordinate information, compares radio frequency energy fluctuations in groups after each time period and identifies positioning differences, performs orientation judgment on Bluetooth base station spatial index and corrects inconsistent coordinates, and generates fused tracking vectors.

[0045] The interactive communication module retrieves cloud communication addresses and edge gateway numbers based on fusion tracking vectors, counts interference levels and determines available frequency bands during occupancy scans, performs occupancy scans on multiple Bluetooth base station radio frequency channels and records valid upload time periods, associates runway coordinates with data transmission rates and outputs transmission path markers, and generates a network exchange index.

[0046] The motion analysis module acquires athlete speed and acceleration data based on network exchange index and parses the interval division from the start to the finish line of the track. When parsing the interval, it extracts the start and finish lines, counts the step frequency value and compares it with the Bluetooth base station positioning trajectory. It combines GPS reference station coordinate information to perform spatiotemporal reconstruction and stores the results in groups to generate a motion feature set.

[0047] The visualization output module generates continuous trajectory nodes by filtering the coordinate values ​​of the inner and outer boundaries of the track based on the action feature set. After distinguishing the athlete's speed segments, it superimposes the Bluetooth base station position relationship. When marking, it sets a speed threshold range to map the information of each segment to the coordinates of the Bluetooth base station. It combines the selected nodes and trajectory information and dynamically draws the layered coordinate projection to generate a graphic coordinate distribution.

[0048] The terminal sensing module includes:

[0049] The synchronous acquisition submodule receives signals from infrared transmitters, Bluetooth base stations, and GPS reference stations synchronously. It records optical band amplitude and radio frequency strength values ​​in a time-division manner, calibrates timing markers, parses the timestamps of each record, compares amplitude differences, associates record indexes to mark the receiving order, and generates an acquisition record set.

[0050] The difference correction submodule, based on the acquired record set, compares the difference correction magnitude deviation with the reference scale, reads the optical frequency band amplitude one by one, corrects the difference between the optical frequency band amplitude and the reference scale, determines the correction magnitude of each record and updates the observation index, and generates corrected observation data;

[0051] The coordinate aggregation submodule, based on the corrected observation data, retrieves the location information identifier matching code, compares the identification code with the observation index and merges duplicates, summarizes the spatial coordinates of all records and aggregates them into a unified index, and generates a sensing sequence.

[0052] Specifically, based on the synchronous reception of signals from the infrared transmitter, Bluetooth base station, and GPS reference station, the amplitude and radio frequency intensity values ​​of the optical band are recorded time-divisionally during each reception. Each record is initially compared according to the amplitude and intensity ranges derived from a large amount of historical observation. For example, the reference range for optical band amplitude is set to 0 to 90, and the reference range for radio frequency intensity is set to -100dBm to -30dBm. If either value exceeds this range, the record is marked as requiring further verification. Subsequently, the records are sequentially labeled according to their timestamps, and each record is compared with a pre-set time deviation threshold T. time Comparison, for example, comparing T time Set to 0.05 seconds and determine through empirical observation, when Δt > T time The corresponding record is considered a time anomaly and classified into a separate sequence. When performing the comparison, the amplitude difference is calculated for the optical frequency band amplitude, for example, defining a threshold T for the amplitude difference ΔA. amp Assume T amp =5 was determined by selecting a step size of 0.5 through multiple sets of control data obtained in previous experiments and adjusting it step by step. When ΔA>T amp The recorded value is considered an amplitude anomaly, requiring manual or automated verification of its source. A threshold T is also established for radio frequency intensity in the same manner. rf For example, using -50dBm as the initial benchmark value and iterating step by step based on the site environment and historical data to obtain -45dBm as the final threshold, if the intensity of a certain record is detected to be lower than -90dBm or higher than -25dBm, it is directly determined that its fluctuation is too large. When recording, it is compared with the surrounding adjacent values ​​and marked. After combining the records of optical amplitude and radio frequency intensity, they are arranged in chronological order. The index of each record is uniformly associated. After the comparison is completed, a valid data queue that meets the defined range is obtained. Finally, the timestamp sorting and index labeling update are combined to generate the collection record set.

[0053] Based on the collected recordings, and according to the deviation correction based on the reference scale comparison, the optical frequency band amplitude of each record must be calculated against the currently known reference scale. The method for setting the reference scale can be determined through confirmed optical comparison tests. For example, the reference scale can be set to 70, and the average value can be recalculated every 10 sampling points. The current reference value is continuously adjusted based on the difference between the measured average value and 70. When the optical frequency band amplitude in the record is A... i The revision amount δ can be set at any time. i =A i -A ref A ref This represents the reference value currently in use. A deviation threshold T is first established during revision. d Assume T d=10 was determined after calculating the mean square error of multiple batches of data. If |δ i |>T d If this record deviates significantly from the benchmark, it is considered to be in a special list, and subsequent records are gradually compared to determine if it is a one-off fluctuation. If |δ| continues to appear... j |>T d In such cases, it is necessary to check whether the initial reference value setting is reasonable or to reacquire the reference scale. Next, quantize the revised values ​​of all records that meet the requirements, and store the revised amplitude and the original record together in the memory structure. After each record is revised, its observation index must be updated, and an incremental step size is established according to the observed deviation range. For example, if the deviation value is kept in the range of -3 to 3, a step size of 0.2 is used for revision; if the deviation value is greater than 3 but not more than 10, a step size of 0.5 is used for revision. After all records are revised, the revision results are stored uniformly. Finally, after all records have been compared and corrected, the corrected observation data is generated.

[0054] Based on corrected observation data, the identification code is matched by retrieving location information. First, an identification code list needs to be stored in the internal data structure. This list can be derived from the distribution of infrared transmitters and Bluetooth base stations installed in the actual site. If each location corresponds to a unique identification code, possible matching identification codes can be found for each observation record. The comparison results are then paired with the observation index. If the same identification code is repeated multiple times in adjacent time periods, these duplicate entries are merged, and only the latest index entry is retained. To ensure the merging process covers all scenarios, a time window W needs to be set. For example, W can be set to 3 seconds and determined through statistical analysis of actual site tests. If the identification code appears multiple times within W seconds and the record numbers are consecutive, they are determined to belong to the same spatial location. After merging, the remaining records are scanned sequentially in chronological order. After the merging operation is completed, all spatial coordinates need to be summarized. The longitude, latitude, and altitude information stored in each record are organized into a unified index format. This can be obtained by referring to the infrared transmitter location lookup table and the Bluetooth base station corresponding location lookup table, thus obtaining an index set with spatial coordinates corresponding to identification codes. Finally, a sensing sequence is generated after merging is complete.

[0055] The data calibration module includes:

[0056] The amplitude comparison submodule compares the radio frequency power value and the optical amplitude value based on the sensing sequence. After reading the radio frequency power value and the optical amplitude value, it calculates the difference respectively, records the corresponding observation number according to the absolute value of each difference, and generates a comparison result set.

[0057] The threshold extraction submodule extracts velocity and acceleration thresholds based on the comparison result set and retrieves ground marking data around the runway. It divides the range according to the velocity and acceleration values ​​of each record in the comparison result set, summarizes the observation numbers within the same range and marks the location information, and generates a threshold mark set.

[0058] The coordinate comparison submodule, based on the threshold mark set, compares the difference between adjacent sampling intervals and marks the fluctuation range, reads the amplitude change within adjacent observation numbers and determines the fluctuation threshold, confirms the synchronization time point by referring to the infrared transmitter position coordinates, and obtains the calibration parameter set.

[0059] Specifically, the comparison between radio frequency power values ​​and optical amplitude values ​​is based on the sensing sequence. During the execution process, the corresponding radio frequency power value P in each record is first read. rf With optical amplitude value A optical The difference Δ is defined as Δ = P rf -A optical And calculate |Δ| and the threshold T preset in advance based on a large amount of historical data. diff Comparisons can be made, for example, by collecting joint measurement results of hundreds of RF power and optical amplitude data, calculating their mean and variance to obtain an initial threshold, and then adjusting it in 0.5-step increments during multiple tests to ultimately lock in T. diff The value is 10, when |Δ|>T diff The record is considered to have an excessively large difference and is marked with the corresponding observation number. The difference values ​​of all records are stored in order and the corresponding numbers are retained. By checking these marked data again, abnormal fluctuations can be identified. If Δ deviates significantly and remains high or low in the next few records, they are merged and summarized into the same abnormal sequence. Otherwise, the record is stored as an independent abnormality. After all records have completed the difference calculation, |Δ| is summarized together with the radio frequency power value and optical amplitude value in the array structure according to the number. The entire record sequence is continuously scanned to confirm whether there are any omissions. Finally, all the extracted difference information is integrated with the corresponding number into a comparison result set.

[0060] Based on the comparison result set, velocity and acceleration thresholds are extracted, and ground marking data around the runway is retrieved. During execution, the mapping relationship between the radio frequency power value and optical amplitude value of each record in the comparison result set, combined with the timestamp and motion position information obtained in the previous step, is used to calculate velocity and acceleration. Velocity can be obtained from the spatial coordinate difference and time interval between adjacent moments. If the time interval between two records is Δt and the position change is Δs, then velocity can be defined as v = Δs / Δt. Acceleration is calculated by subtracting the velocity from two adjacent velocity records and then dividing by the corresponding Δt. In this process, velocity and acceleration thresholds need to be given for specific items. For example, the velocity threshold can be initially selected as 5 m / s through local testing and finely adjusted in increments of 0.5 m / s each time. After more than ten repeated observations, it was finally determined to be 6 m / s. The acceleration threshold can be initially assumed to be 2 m / s. 2 and at 0.2 m / s 2 The step length was determined to be 2.4 m / s based on multiple rounds of testing. 2 If the velocity or acceleration of a certain record exceeds the corresponding threshold, it is marked with an additional highlight in the comparison result set. Then, the ground marking data around the runway is retrieved and the location coordinates of the areas with high velocity and acceleration are compared. In this way, the observation numbers that meet the same range conditions are summarized into the same index list, and the correspondence between records and locations is completed in turn. Finally, the threshold mark set is output.

[0061] Based on the threshold label set, the fluctuation range is marked by comparing the difference between adjacent sampling intervals. During the execution process, the observation numbers contained in the threshold label set are first scanned sequentially, the time difference of each pair of adjacent records is counted, and the amplitude change is analyzed. For example, the amplitude change ΔA = A current -A previous Compare the absolute value of ΔA with the pre-set fluctuation threshold T. wave This threshold can be obtained by referring to multiple sets of infrared emitter amplitude data and radio frequency power data collected in the same runway environment in the early stage. The average value of the fluctuation range of multiple experiments is used as the initial threshold, and then the threshold is repeatedly verified at intervals of 0.2 to finally obtain T. wave =3, when |ΔA|>T wave If the fluctuations are significant, the record will be traced back to its location, and the same fluctuation characteristics will be checked in several adjacent records to see if they are continuous. At the same time, the position coordinates of the infrared transmitter will be used to check whether the observation time points are in the same period. If the fluctuations in the same period exceed a certain frequency, the record segment and its corresponding velocity and acceleration values ​​will be marked separately. After all the markings are completed, they will be summarized in a unified list according to the time sequence and amplitude change. Then, they will be compared according to the runway position and transmitter coordinates to compile the final calibration basis and obtain the calibration parameter set.

[0062] The edge processing module includes:

[0063] The threshold filtering submodule filters optical amplitude thresholds and radio frequency energy thresholds based on the calibration parameter group. It reads the amplitude range marked in the calibration parameter group and limits the upper limit of radio frequency energy. It identifies records that are below the lower limit or above the upper limit and summarizes the index to generate a set of filtering thresholds.

[0064] The time period comparison submodule, based on the set of filtering thresholds, compares timestamps in segments and uses runway plane coordinate information. It marks outliers based on amplitude fluctuations within consecutive time periods, merges and compares radio frequency energy within the same time period, and generates time period comparison data.

[0065] The coordinate revision submodule performs orientation judgment and corrects inconsistent coordinates on the Bluetooth base station spatial index based on time period comparison data. It reads the base station coordinates corresponding to the abnormal point and calculates the offset with the reference coordinates, replaces the mismatched coordinates and updates the index synchronously, and generates a fused tracking vector.

[0066] Specifically, based on the calibration parameter set, by reading the marked optical amplitude range and RF energy setting values, the lowest and highest values ​​of the optical amplitude are first selected as reference thresholds from multiple field measurements and comparisons with historical data. For example, the lower limit of the optical amplitude is set to 10, and experiments are conducted with increments of 2 each time. After multiple rounds of verification, it is finally determined to be 15. The upper limit can be initially set to 80 and gradually reduced by 1 until the frequency of fluctuations exceeding the limit is low in interactive observations. Finally, it is fixed at 75. A similar approach is used for RF energy. Multiple sets of RF values ​​are collected first, and a relatively wide range is determined in combination with the environmental noise level, such as -100dBm to -20dBm. m. By deploying infrared transmitters and Bluetooth base stations in different areas to conduct multiple joint detections and record the corresponding radio frequency changes, the range is gradually narrowed and adjusted about 5 times to determine the final upper and lower limits of radio frequency energy as -85dBm and -25dBm, respectively. When filtering, the system reads the actual observed values ​​of the current optical amplitude and radio frequency energy and compares them one by one to see if they are lower than the lower limit of optical amplitude or higher than the upper limit of optical amplitude, and whether they are lower than the lower limit of radio frequency energy or higher than the upper limit of radio frequency energy. If any one of them exceeds the limit, the index number of the observed data is collected and marked as an abnormal group. Finally, all the exceeding numbers are integrated into a filtering result list to generate a filtering threshold set.

[0067] Based on the set of filtering thresholds, the recorded abnormal observation numbers and corresponding timestamps are first sorted and divided into several consecutive time periods. For each time period, the runway plane coordinate information is read and the optical amplitude is dynamically compared. If the fluctuation of multiple consecutive optical amplitude values ​​between adjacent records exceeds the pre-set optical amplitude difference threshold T, the error is considered. opt Then mark this time period as the set of outliers, T optThe average fluctuation range under several test scenarios can be initially set at 5, and then gradually reduced to 3 after multiple tests in different runway environments before being fixed. For these anomalies, radio frequency energy values ​​are extracted within the same time interval for comprehensive analysis, and the fluctuation amplitude is calculated in a way that approximates the optical amplitude. Records with fluctuation amplitudes below a certain lower limit or above a certain upper limit are aggregated and stored, and the overall fluctuation characteristics of the period are judged by comparing the combined optical and radio frequency conditions. If both radio frequency energy and optical amplitude continuously exceed their set threshold range, the period is marked as a higher priority anomaly. After traversing all time intervals and completing the amplitude fluctuation calibration, the calculated fluctuation information of each period is fused with the time index to finally generate period comparison data.

[0068] Based on time-period comparison data, the location coordinates of marked anomalies are compared with the known spatial indexes of Bluetooth base stations. When determining the orientation, the nearest possible base station is first determined based on the base station distribution list around the runway. The offset of the base station coordinates corresponding to the anomaly point from the reference coordinates is calculated. If the calculated offset exceeds the pre-set allowable range, it indicates that the coordinates may be inconsistent with the base station record. The allowable range can be derived from the requirements for base station positioning accuracy in the actual scenario and the measurement error estimation method. For example, a starting allowable radius of 5 meters is used, and multiple batches of data are continuously observed in increments of 1 meter during the trial operation, and finally determined to be 7 meters. If the offset of an anomaly point reaches or exceeds 7 meters, its number is recorded and the original coordinates are replaced with the new position revision value. The replaced information is then updated back into the data structure. After revising all mismatched coordinates, the time series and spatial index are checked uniformly, and they are included in the final fused coordinate sequence. After traversing all records, the results are output to generate a fused tracking vector.

[0069] The interactive communication module includes:

[0070] The address retrieval submodule, based on the fused tracking vector, retrieves the cloud communication address and edge gateway number, matches the base station index in each retrieval vector with the cloud list, reads the corresponding gateway number and maps it to the specified address range, and generates communication mapping data.

[0071] The interference assessment submodule, based on communication mapping data, counts the interference level and determines the available frequency bands during occupancy scanning, tracks the occupancy duration and interference amplitude of the radio frequency channel, records the interference value and the corresponding frequency band index, and generates a set of available channels.

[0072] The path marking submodule, based on the available channel set, associates runway coordinates with data transmission rates and outputs transmission path markings. It reads the gateway information corresponding to the available frequency bands and cross-compares the rate values ​​and coordinate information to generate a network switching index.

[0073] Specifically, to retrieve cloud communication addresses and edge gateway numbers based on fused tracing vectors, a cloud list needs to be maintained internally. This list contains the mapping relationship between all potentially used base station indexes and their corresponding cloud communication addresses and edge gateway numbers. Specific parameters can be obtained during the initial testing phase by conducting on-site inspections of each base station and comparing them with factory data. Then, a final comparison is performed in the deployment environment to form a fixed lookup table. During execution, the base station index in the fused tracing vector is read line by line and matched against the cloud list. When a match is found, the corresponding edge gateway number is extracted and recorded. At this point, it needs to be further mapped to a specified address range, for example, by first... Each edge gateway number is configured with a start and end address value. If the number falls within a certain range, it is classified into the corresponding address range. The specific division of the address range can be determined based on the observation of the network topology and the estimation of the coverage of each gateway. For example, in a small site, numbers 1 to 100 can be mapped to the first address range, and numbers 101 to 200 can be mapped to the second address range. After multiple tests, if it is found that there are too many base stations or frequent conflicts in a certain range, the range can be appropriately reduced or a new address range can be added in the higher number range. After each retrieval and mapping is completed, the obtained gateway numbers and address information are organized in chronological order and base station index. After all pairings are completed, complete communication mapping data can be obtained.

[0074] Interference levels are statistically analyzed and available frequency bands are determined during occupancy scanning based on communication mapping data. First, RF channel information is read from the matched communication mapping data. The occupancy duration of each channel within a specified time window is continuously observed, and the interference amplitude is obtained. The interference amplitude can be calculated by differentially analyzing the power sample values ​​from multiple measurement points. Assuming n power samples p1, p2, ..., p are acquired within a unit of time... n You can first take the average of all samples. Then compare each sample with The absolute value of the difference is accumulated to measure the interference intensity. If the accumulated result exceeds a certain empirical threshold, it is considered a high interference channel. This threshold can be obtained through preliminary experiments. For example, after arranging multiple dynamometer devices in the field to conduct statistics, the initial threshold is determined to be 50dB and then iterated to 45dB. Subsequently, the channels are divided into multiple levels according to the interference intensity level. The same preset standard should also be set for the occupancy time. If a channel is in a high occupancy state for more than 7 seconds in a 10-second continuous monitoring, it is judged to be overloaded. Correspondingly, this channel is also recorded as an interference state in the result. Finally, the interference value and frequency band index are cross-organized and the channels with low interference or higher idle degree are selected. These channels are marked as available when summarizing and then included in the available channel set.

[0075] Based on the available channel set, the location of the edge gateway corresponding to these channels is first found in the existing runway coordinate information. Then, the data transmission rate value is read from the system, and the rate value is compared one-to-one with the runway coordinate. If the transmission rate under the same gateway differs greatly between adjacent locations, it may be necessary to check environmental factors or antenna direction. However, in this process, only the record of the rate value is retained, and the path is marked based on this record. A threshold can be set in advance based on the average rate obtained from the test to determine whether it belongs to a high transmission rate area. For example, the threshold can be set to 2Mbps and gradually adjusted in increments of 0.2Mbps during multiple field tests, and finally fixed at 1.8Mbps as a suitable standard. If the speed at a certain point is higher than 1.8Mbps, it is considered a high-speed transmission location. Then, the high transmission rate range is marked on the coordinate system. If there is a part that overlaps with low interference or idle channels, it can be marked as a priority transmission path. When all this information is combined, a mapping relationship between runway coordinates and rate is formed. After cross-confirmation with the edge gateway number corresponding to the available channel set, the transmission path mark can be output. Finally, the integrated path mark and channel information are aggregated into an overall structure to generate a network switching index.

[0076] The motion analysis module includes:

[0077] The data extraction submodule, based on the network exchange index, acquires athlete speed and acceleration data and parses the interval division from the start to the finish line of the track. It reads speed and acceleration values, locates the start and finish lines, records the interval index and corresponding speed level, and generates a speed and acceleration set.

[0078] The interval division submodule, based on the velocity and acceleration set, extracts the start and end points when parsing the interval, counts the step frequency value and compares it with the Bluetooth base station positioning trajectory, checks the observation sequence corresponding to the rate level and marks the difference in track number, and generates step frequency comparison value.

[0079] The spatiotemporal reconstruction submodule, based on step frequency comparison values, combines GPS base station coordinate information and stores the results in groups, reads step frequency and coordinate pairing data and arranges time series, organizes adjacent sampling points under the same series and merges them into a unified interval index, and generates a motion feature set.

[0080] Specifically, based on network exchange indexing, athlete speed and acceleration data are obtained, and the interval division from the start to the finish line of the track is analyzed. First, the basic dimensional information of speed and acceleration needs to be recorded in the deployment environment beforehand. Then, time-series data collected over several sessions are analyzed. When reading speed values, an observation frequency of once per second is used, and the corresponding displacement and time interval are obtained in each observation. The speed is obtained by dividing the displacement difference by the time difference. If several speed data points exceed a threshold set in advance based on historical sports experience, such as 15 m / s, these records are individually marked. Acceleration is obtained by dividing the difference between adjacent speed data by the corresponding time interval. When the acceleration exceeds a certain preset standard, such as 5 m / s... 2 It is necessary to mark the observation number for subsequent comparison, this 5m / s 2 The value can be initially determined by combining the maximum sprint data from previous training projects, and then adjusted through multiple rounds of trials before finalization. Then, the start and finish lines are located based on the track length information, with the start line as the initial position of record number 1. The finish line is mapped to the track length for corresponding identification. For example, when the length is set to 100 meters, the system distinguishes each integer position from 1 meter to 100 meters. When reading speed and acceleration, the system automatically checks the segment number where the observed data is located and further binds the segment number to the measured speed level. For example, if the speed is less than 3 m / s, it is recorded as a slow segment; if it is 3 to 6 m / s, it is recorded as a medium-speed segment; and if it is greater than 6 m / s, it is recorded as a high-speed segment. After multiple comparisons and confirmations, a segment index is generated and the corresponding speed level information is attached. All information is combined to output the speed and acceleration set.

[0081] Based on the velocity-acceleration set, the start and end points are extracted and step frequency values ​​are counted during the parsing interval. First, the velocity and acceleration corresponding to each record obtained earlier need to be arranged in time sequence. Since the velocity-acceleration set already contains segment indices, the observation number of the start and end points of each segment can be directly determined by the segment number during parsing. After marking them in time sequence, the continuous movement duration within each interval can be obtained. Then, the step frequency of each observation record is counted and counted. For example, when the observation frequency is once per second, the system can detect the acceleration fluctuation high points of three or more consecutive steps to confirm the time when the foot touches the ground. The step frequency (PF) is obtained by taking points. If the PF exceeds a certain range, such as 5 steps / second, the interval needs to be additionally marked. The critical value of 5 steps / second can be adjusted after multiple measurements based on the actual training data of professional athletes. For example, starting from 4.5 steps / second, the step length is gradually increased by 0.2 steps / second, and finally 5 steps / second is selected. Then, the track is compared with the Bluetooth base station positioning trajectory to confirm the lane number corresponding to each record and read the speed level. If multiple identical lanes are found in the same segment with different speed distributions, the observation sequence of that segment needs to be checked again to eliminate possible duplicate or missing records. Finally, the differences of each lane are summarized into the PF comparison value.

[0082] The results are grouped based on cadence comparison values ​​and GPS reference station coordinate information. First, the spatial coordinates associated with each cadence observation are read. The accurate location is then corrected using the latitude, longitude, and altitude information provided by the GPS reference station. If the coordinates of an observation deviate significantly from the reference station's comparison result, for example, by more than 5 meters, the record is marked for further investigation. This 5-meter criterion is determined by comparing multiple samples taken from the same athlete by multiple GPS reference stations installed within the test track and taking the median. Subsequently, adjacent sampling points within the same sequence are continuously merged, with intervals not exceeding 1 second and cadence differences not exceeding 1 step / minute. Several observation points per second are grouped together. The thresholds for 1 second and 1 step / second can be gradually revised to form stable parameters based on the numerical distribution and coach requirements during multiple training measurements on the actual field. The merged data is then screened for lane number to ensure that the athlete coordinates under the same time index can correctly correspond to the same lane number. If a record of lane data has obvious conflicts with its neighbors before and after, it is re-calibrated with GPS coordinates as a reference. The time series obtained after grouping all the data can cover the entire track section and retain key observation features. Finally, after these contents and the corresponding step frequency values ​​are synchronously integrated, the motion feature set is output.

[0083] The visualization output module includes:

[0084] The boundary filtering submodule, based on the action feature set, filters the coordinate values ​​of the inner and outer boundaries of the runway to generate continuous trajectory nodes, reads the coordinate index in the action feature set and removes data that is not within the runway range, generates continuous coordinates according to the order of nodes that meet the range, and generates a coordinate node set.

[0085] The speed segmentation submodule, based on the coordinate node set, distinguishes the athlete's speed segmentation and superimposes the Bluetooth base station position relationship, reads the coordinate nodes and obtains the corresponding speed values ​​from the action feature set, marks the segments that exceed the threshold and marks them in the coordinate index, and generates superimposed position information.

[0086] The coordinate mapping submodule, based on the superimposed position information, merges the selected node and trajectory information and dynamically draws the layered coordinate projection, reads the coordinate index corresponding to the superimposed position information and assigns the layered structure to the rendering sequence, organizes the drawing instructions and outputs the positioning graphic, generating the graphic coordinate distribution.

[0087] Specifically, to generate continuous trajectory nodes by filtering the coordinate values ​​of the inner and outer boundaries of the runway based on the motion feature set, the boundary definition of the runway first needs to be obtained. Multiple positioning stakes or calibration points can be deployed on the site, and their coordinate information can be compiled into a boundary list. Each boundary point can include latitude, longitude, and relative site coordinates. The effective range of the runway is represented by connecting segments. Then, each observation record and its corresponding coordinate index in the motion feature set is read, and these indices are compared with the runway boundary list. If the shortest distance between the coordinates of a record and the boundary is less than or greater than a certain empirical threshold, it can be determined whether the record is outside the runway. This threshold can be determined by repeatedly measuring the actual width of the runway and the safe zone on-site, using a baseline value. For example, the threshold can be initially set to 2 meters and increased or decreased in 0.5-meter increments. Multiple verifications are performed, comparing each record with the baseline value to confirm its range. If the vertical distance between the record coordinates and the runway boundary exceeds 2 meters, it is marked as outside the runway range; if the distance is less than 2 meters, it is marked as a valid point within the runway. After completing the inside-outside determination of all observation records, the data within the runway are sorted by time or index, and a trajectory sequence is generated based on the continuity between coordinate points. This continuity can be determined by comparing the distance and time difference between adjacent points. When both distance and time meet the reasonable range of motion change, they are strung together into a continuous trajectory. If jump points or outliers occur in the middle, the coordinates need to be checked again to see if they exceed the preset threshold range before correction. Finally, after removing data outside the runway range, continuous coordinates are generated by the node order, and the coordinate node set is output.

[0088] After segmenting athletes' speeds based on a set of coordinate nodes and overlaying the location relationships of Bluetooth base stations, the process begins by reading coordinate and time information from the node set one by one and searching for corresponding records in the previously calculated speed data. Matching speed and coordinate yields a complete spatiotemporal distribution. Based on this, speed thresholds are set to identify different speed segments. For example, based on data from actual athlete tests, the initial threshold can be set at 5 m / s, with multiple rounds of field practice using a step size of 0.5 m / s. This is eventually adjusted to 6 m / s as the speed segment boundary. When the speed value corresponding to a coordinate node exceeds 6 m / s, it is marked as a high-speed segment; otherwise, it is classified into other levels. If a segment remains at a high speed for an extended period, it will be flagged as such. The system additionally marks the differences between fast and normal segments to clarify the distinction. Meanwhile, the location relationships of Bluetooth base stations can be read using a base station distribution list. During deployment, engineers can measure and record the coordinates of each base station. The distance between the node's latitude and longitude or planar coordinates and the base station coordinates is calculated. If the distance falls within a set reference range, it indicates that the athlete is within the base station's coverage area. For example, the reference range can be set at 30 meters and increased to 35 meters after multiple measurements to accommodate a wider coverage area. When the distance between a node's coordinates and a Bluetooth base station is less than 35 meters, the node is associated with that base station. Once the speed segments of all nodes are combined with the base station coverage relationships, overlay location information can be generated.

[0089] Based on the aggregation of selected node and trajectory information using overlaid location information, and the dynamic rendering of layered coordinate projections, the overlaid location information is first processed into layers. Nodes within the coverage area of ​​the Bluetooth base station are marked as one layer, nodes outside the base station coverage area but exceeding the speed threshold are marked as another layer, and the remaining nodes that do not exceed the threshold and are not within the base station coverage area are assigned to a third layer. After layering, a rendering sequence is formed according to the index value of each node, and the coordinates are drawn sequentially using either planar projection or three-dimensional coordinate methods. If a more detailed display of the trajectory is required, different colors or line styles for different layers can be set in the system. When checking the rendering sequence, the node order should be consistent with the original time order. If some node numbers are found to be discontinuous, it can be backtracked to check whether there are any instances of crossing sections or loss. Once all node rendering data is ready, the drawing instructions are arranged according to the coordinates and layering relationships for real-time drawing. Finally, after completing all rendering steps, the positioning graphic is output, and the corresponding coordinate sequence is summarized and stored to generate a graphic coordinate distribution.

Claims

1. A smart track fusion positioning system, characterized in that, The system includes: The terminal sensing module synchronously receives signals from infrared transmitters, Bluetooth base stations, and GPS reference stations. It records optical band amplitude and radio frequency intensity values ​​in a time-division manner to calibrate timing marks. It corrects amplitude deviations based on the difference compared with the reference scale. It retrieves location information identifiers and matches identification codes, organizes multiple observation records to summarize coordinate values, and generates a sensing sequence. The data calibration module compares the radio frequency power value and optical amplitude value based on the sensing sequence, extracts the speed threshold and acceleration threshold, retrieves ground marking data around the runway, compares the difference based on adjacent sampling intervals and marks the fluctuation range, and obtains the calibration parameter set by comparing the position coordinates of the infrared transmitter and confirming the synchronization time point. The edge processing module filters optical amplitude thresholds and radio frequency energy thresholds based on the calibration parameter group, compares timestamps in segments and uses runway plane coordinate information, compares radio frequency energy fluctuations in groups after each time period and identifies positioning differences, performs orientation judgment on Bluetooth base station spatial index and corrects inconsistent coordinates, and generates fused tracking vectors. The interactive communication module retrieves the cloud communication address and edge gateway number based on the fused tracking vector, counts the interference level and determines the available frequency band during occupancy scanning, performs occupancy scanning on multiple Bluetooth base station radio frequency channels and records the effective upload time period, associates runway coordinates with data transmission rate and outputs transmission path markers, and generates a network exchange index.

2. The intelligent track fusion positioning system according to claim 1, characterized in that, Also includes: The motion analysis module acquires athlete speed and acceleration data based on the network exchange index and parses the interval division from the start to the finish line of the track. When parsing the interval, it extracts the start and finish lines, counts the step frequency value and compares it with the Bluetooth base station positioning trajectory. It performs spatiotemporal reconstruction in combination with GPS reference station coordinate information and stores the results in groups to generate a motion feature set. The visualization output module filters the coordinate values ​​of the inner and outer boundaries of the track based on the action feature set to generate continuous trajectory nodes. After distinguishing the athlete's speed segments, it superimposes the Bluetooth base station position relationship. When marking, it sets a speed threshold range to map the information of each segment with the coordinates of the Bluetooth base station. It combines the selected nodes and trajectory information and dynamically draws the layered coordinate projection to generate a graphic coordinate distribution.

3. The intelligent track fusion positioning system according to claim 1, characterized in that, The terminal sensing module includes: The synchronous acquisition submodule receives signals from infrared transmitters, Bluetooth base stations, and GPS reference stations synchronously. It records optical band amplitude and radio frequency strength values ​​in a time-division manner, calibrates timing markers, parses the timestamps of each record, compares amplitude differences, associates record indexes to mark the receiving order, and generates an acquisition record set. The difference correction submodule, based on the acquired record set, reads the optical frequency band amplitude one by one according to the reference scale to correct the difference correction magnitude deviation, corrects the difference between the optical frequency band amplitude and the reference scale, determines the correction magnitude of each record and updates the observation index, and generates corrected observation data. The coordinate aggregation submodule, based on the corrected observation data, retrieves the location information identifier matching identification code, compares the identification code with the observation index and merges duplicates, summarizes the spatial coordinates of all records and aggregates them into a unified index, and generates a sensing sequence.

4. The intelligent track fusion positioning system according to claim 1, characterized in that, The data calibration module includes: The amplitude comparison submodule compares the radio frequency power value and the optical amplitude value based on the sensing sequence. After reading the radio frequency power value and the optical amplitude value, it calculates the difference respectively, records the corresponding observation number according to the absolute value of each difference, and generates a comparison result set. The threshold extraction submodule extracts velocity and acceleration thresholds based on the comparison result set and retrieves ground marking data around the runway. It divides the range according to the velocity and acceleration values ​​of each record in the comparison result set, summarizes the observation numbers within the same range and marks the location information, and generates a threshold mark set. The coordinate comparison submodule, based on the threshold mark set, compares the difference between adjacent sampling intervals and marks the fluctuation range, reads the amplitude change within adjacent observation numbers and determines the fluctuation threshold, confirms the synchronization time point by referring to the position coordinates of the infrared transmitter, and obtains the calibration parameter set.

5. The intelligent track fusion positioning system according to claim 1, characterized in that, The edge processing module includes: The threshold filtering submodule filters optical amplitude thresholds and radio frequency energy thresholds based on the calibration parameter group, reads the amplitude range marked in the calibration parameter group and limits the upper limit of radio frequency energy, identifies records that are below the lower limit or above the upper limit and summarizes the index to generate a set of filtering thresholds. The time period comparison submodule, based on the set of filtering thresholds, compares timestamps in segments and uses runway plane coordinate information to mark outliers based on amplitude fluctuations within consecutive time periods, merges and compares radio frequency energy in the same time period, and generates time period comparison data. The coordinate revision submodule performs orientation judgment and corrects inconsistent coordinates on the Bluetooth base station spatial index based on the time period comparison data. It reads the base station coordinates corresponding to the abnormal point and calculates the offset with the reference coordinates, replaces the mismatched coordinates and updates the index synchronously, and generates a fused tracking vector.

6. The intelligent track fusion positioning system according to claim 1, characterized in that, The interactive communication module includes: The address retrieval submodule retrieves cloud communication addresses and edge gateway numbers based on the fused tracking vectors. It matches the base station indexes in each retrieval vector with the cloud list, reads the corresponding gateway numbers and maps them to a specified address range, and generates communication mapping data. The interference assessment submodule, based on the communication mapping data, counts the interference level and determines the available frequency bands during the occupancy scan, tracks the occupancy duration and interference amplitude of the radio frequency channel, records the interference value and the corresponding frequency band index, and generates a set of available channels. The path marking submodule, based on the available channel set, associates runway coordinates with data transmission rates and outputs transmission path markings, reads gateway information corresponding to available frequency bands and cross-compares rate values ​​and coordinate information to generate a network switching index.

7. The intelligent track fusion positioning system according to claim 2, characterized in that, The motion analysis module includes: The data extraction submodule, based on the network exchange index, acquires athlete speed and acceleration data and parses the interval division from the start to the finish line of the track, reads speed and acceleration values ​​and locates the start and finish lines, records the interval index and corresponding speed level, and generates a speed and acceleration set. The interval division submodule, based on the velocity and acceleration set, extracts the start and end points when parsing the interval, counts the step frequency value and compares it with the Bluetooth base station positioning trajectory, checks the observation sequence corresponding to the rate level and marks the track difference, and generates step frequency comparison value. The spatiotemporal reconstruction submodule, based on the step frequency comparison value, combines GPS base station coordinate information and stores the result grouping, reads the step frequency and coordinate pairing data and arranges the time series, organizes adjacent sampling points under the same series and merges them into a unified interval index, and generates an action feature set.

8. The intelligent track fusion positioning system according to claim 2, characterized in that, The visualization output module includes: The boundary filtering submodule, based on the action feature set, filters the coordinate values ​​of the inner and outer boundaries of the runway to generate continuous trajectory nodes, reads the coordinate index in the action feature set and removes data that is not within the runway range, generates continuous coordinates according to the node order that meets the range, and generates a coordinate node set. The speed segmentation submodule, based on the coordinate node set, distinguishes the relationship between the Bluetooth base station position after the athlete's speed is segmented, reads the coordinate nodes and obtains the corresponding speed values ​​from the action feature set, marks the segments that exceed the threshold and identifies them in the coordinate index, and generates superimposed position information. The coordinate mapping submodule, based on the superimposed position information, merges the selected node and trajectory information and dynamically draws the layered coordinate projection, reads the coordinate index corresponding to the superimposed position information and assigns the layered structure to the rendering sequence, organizes the drawing instructions and outputs the positioning graphic, generating the graphic coordinate distribution.

Images