Cooperative detection online evaluation method and system based on joint driving of space-time performance

Abstract

The present application relates to the technical field of detection evaluation, and proposes a collaborative detection online evaluation method and system based on joint driving of space-time efficiency, which comprises the following steps: performing space-time registration fusion on real-time original signal data of a detection device to obtain synchronous signal data; performing space-domain coherent accumulation on the synchronous signal data to obtain enhanced detection data; analyzing the task correlation characteristics of the enhanced detection data, and performing multi-dimensional efficiency evaluation on the detection device to obtain a detection area coverage index, a tracking accuracy index and an identification accuracy index; performing parameter factor fusion on the detection area coverage index, the tracking accuracy index and the identification accuracy index to obtain a comprehensive efficiency index; performing time sequence trend deduction on the comprehensive efficiency index to obtain an efficiency evolution trajectory; and performing running robustness analysis on the detection device to generate an online efficiency evaluation report; the present application can improve the efficiency of collaborative detection online evaluation.

Description

Technical Field

[0001] This invention relates to the field of detection and evaluation technology, and in particular to a collaborative online detection evaluation method and system based on spatiotemporal performance joint driving. Background Technology

[0002] Collaborative detection systems consist of a network of multiple heterogeneous detection devices (such as radar, photoelectric, and infrared). Performance evaluation of these devices is crucial for system operation optimization and mission assurance. Existing evaluation methods generally suffer from the following interconnected technical challenges: In the data preprocessing stage, they fail to effectively address the spatiotemporal heterogeneity of real-time raw signals from multiple detection devices. Inconsistent signal timestamps and spatial coordinates lead to low data synchronization accuracy, and the lack of effective spatial signal enhancement mechanisms results in insufficient effectiveness of subsequent feature extraction. Regarding the evaluation system, many methods employ a fixed-weight, single-dimensional evaluation model, failing to dynamically adjust indicator weights according to the importance of the detection mission, leading to a disconnect between evaluation results and actual application requirements. In the time dimension, traditional methods typically use the results of indicator calculations within a single timeframe or fixed time window as the final performance conclusion, representing a "snapshot" evaluation. This lacks the ability to explore the evolution of comprehensive performance over time, making it difficult to accurately identify equipment performance fluctuation trends and potential degradation risks.

[0003] Existing methods directly calculate indicators after aligning timestamps and transforming spatial coordinates of multi-source signals during data preprocessing, without considering the impact of the spatial distribution characteristics of the detected signals on subsequent feature extraction. Due to the uneven spatial distribution of various detection devices, the directly fused signal data suffers from spatial energy dispersion, resulting in insufficient effectiveness of extracted coverage, tracking, and identification features. Consequently, the reliability of the evaluation results decreases, creating a dilemma of "data-layer errors propagating to the performance layer." Furthermore, different detection tasks (such as area surveillance focusing on coverage, target close-range reconnaissance focusing on tracking, and attribute discrimination focusing on identification) have significantly different priorities for the three dimensions of coverage, tracking, and identification. Fixed-weight evaluation models cannot dynamically adjust the focus according to task requirements, resulting in a lack of task-layer adaptation. In addition, detection equipment exhibits performance fluctuations and degradation trends due to environmental changes, equipment aging, and target maneuvering during actual operation. Static "snapshot" evaluations cannot identify dynamic risks, creating a time-layer risk blind spot. The three problems mentioned above are not isolated, but constitute a progressive technical dilemma of "error propagation at the data layer → lack of adaptation at the task layer → risk blind spot at the time layer". Existing technologies have not yet proposed a complete solution that can break down the barriers between the data layer, task layer and time layer. Therefore, there is an urgent need for an online evaluation method that can achieve accurate, dynamic and forward-looking collaborative detection system performance. Summary of the Invention

[0004] This invention provides a collaborative detection online evaluation method and system based on spatiotemporal performance joint driving, in order to solve the problems mentioned in the background art.

[0005] To achieve the above objectives, the present invention provides a collaborative online assessment method for detection based on spatiotemporal performance joint driving, comprising: S1. Perform spatiotemporal registration and fusion on the real-time raw signal data of the detection device to obtain the synchronous signal data of the detection device; S2. Perform spatial coherent accumulation on the synchronization signal data to obtain enhanced detection data of the synchronization signal data; S3. Based on the detection task requirements of the detection device, analyze the task association features of the enhanced detection data, and based on the task association features, perform a multi-dimensional performance evaluation of the detection device to obtain the detection area coverage index, tracking accuracy index and identification accuracy index of the detection device. S4. Based on the importance of the detection task requirements, the detection area coverage index, the tracking accuracy index, and the identification accuracy index are fused using parametric factors to obtain the comprehensive performance index of the detection equipment. S5. Perform time-series trend analysis on the comprehensive performance index to obtain the performance evolution trajectory of the detection equipment; S6. Based on the performance evolution trajectory, perform an operational robustness analysis on the detection equipment to generate an online performance evaluation report for the detection equipment.

[0006] In this process, the spatiotemporal registration and fusion in step S1 provides a phase consistency benchmark for the spatial coherent accumulation in step S2; the enhanced detection data in step S2 provides a high signal-to-noise ratio signal basis for the feature extraction in step S3; the detection area coverage index, tracking accuracy index, and identification accuracy index in step S3 provide sub-item performance data for the parametric factor fusion in step S4; the comprehensive performance index in step S4 provides a comparable quantitative sequence for the temporal trend extrapolation in step S5; and the performance evolution trajectory in step S5 provides a dynamic baseline reference for the operational robustness analysis in step S6. Steps S1 to S6 are progressively advanced and together constitute a complete technical closed loop for the online evaluation of the collaborative detection system's performance.

[0007] In a preferred embodiment, the step of performing spatiotemporal registration and fusion on the real-time raw signal data of the detection device to obtain the synchronization signal data of the detection device includes: Collect real-time raw signal data from the detection equipment; The real-time raw signal data is timestamped to obtain the time registration data of the detection device; Based on the position and attitude parameters of the detection device, the time registration data is transformed into spatial coordinates to obtain the spatial registration data of the detection device. The spatial registration data is processed to standardize its format to obtain the synchronization signal data of the detection device.

[0008] In a preferred embodiment, the step of performing spatial coherent accumulation on the synchronization signal data to obtain enhanced detection data of the synchronization signal data includes: The synchronization signal data is spatially distributed and sampled to obtain a spatial sample set of the synchronization signal data. Based on the relative positions of signal sources of the samples in the spatial sample set, phase compensation is performed on the spatial sample set to obtain the phase alignment data of the synchronization signal data. The phase-aligned data is coherently superimposed to obtain the enhanced detection data of the synchronization signal data.

[0009] In a preferred embodiment, the step of parsing the task-related features of the enhanced detection data based on the detection task requirements of the detection device includes: The detection task requirements of the detection device are semantically parsed to obtain the spatial constraints and performance constraints of the detection device. Feature extraction is performed on the enhanced detection data to obtain multidimensional features of the enhanced detection data; Based on the spatial constraints and performance constraints, recursive feature elimination is performed on the multidimensional features to obtain the task-related features of the enhanced detection data.

[0010] In a preferred embodiment, the step of performing a multi-dimensional performance evaluation of the detection device based on the task-related features to obtain the detection area coverage index, tracking accuracy index, and identification accuracy index of the detection device includes: The task-related features are deconstructed and separated to obtain the coverage features, tracking features, and identification features of the enhanced detection data; Based on the coverage characteristics, the airspace coverage rate of the detection device is analyzed to obtain the detection area coverage index of the detection device. Based on the tracking features, the positioning accuracy of the detection device is estimated to obtain the tracking accuracy index of the detection device; Based on the identification features, the attribute discrimination confidence quantification of the detection device is performed to obtain the identification accuracy index of the detection device.

[0011] In a preferred embodiment, the step of fusing parametric factors of the detection area coverage index, the tracking accuracy index, and the identification accuracy index based on the importance of the detection task requirements to obtain the comprehensive performance index of the detection equipment includes: Prioritize the requirements of the detection mission to determine their degree of importance. Based on the task type identifier required by the detection task, the degree of importance is mapped to a multi-dimensional weighting coefficient of the detection area coverage index, the tracking accuracy index, and the identification accuracy index; Based on the multidimensional weighting coefficients, the detection area coverage index, the tracking accuracy index, and the identification accuracy index are weighted and fused to obtain the comprehensive performance index of the detection device.

[0012] In a preferred embodiment, the formula for calculating the comprehensive performance index is as follows: ; In the formula, This indicates the comprehensive performance index. This represents the weight coefficient of the coverage dimension. This represents the weight coefficient of the tracking dimension. This represents the weight coefficient of the recognition dimension. This represents the normalized detection area coverage index. This represents the normalized tracking accuracy index. This represents the accuracy index of identification after normalization.

[0013] In a preferred embodiment, the step of performing time-series trend analysis on the comprehensive performance index to obtain the performance evolution trajectory of the detection device includes: The comprehensive performance indicators are sorted by time series to obtain the performance time series sequence of the comprehensive performance indicators; Trend analysis is performed on the performance time series to obtain the trend parameters, periodic parameters, and residual fluctuation parameters of the performance time series. Based on the trend parameter, the periodic parameter, and the residual fluctuation parameter, the macroscopic evolution stage and microscopic state transition point of the performance time series are identified. Based on the macroscopic evolution stage and the microscopic state transition point, the performance time series is reconstructed to obtain the performance evolution trajectory of the detection device.

[0014] In a preferred embodiment, the step of performing an operational robustness analysis on the detection device based on the performance evolution trajectory to generate an online performance evaluation report for the detection device includes: By performing a safety baseline derivation on the performance evolution trajectory, the lower limit of stable operation of the detection equipment is obtained. The time-series feature deconstruction of the performance evolution trajectory yields the performance fluctuation data and performance decay data of the detection device. Based on the performance fluctuation data and the performance degradation data, the degradation risk of the detection equipment is assessed to obtain the near-time domain risk level of the detection equipment; The comprehensive performance index, the stable operation lower limit, and the near-time domain risk level are integrated into an online performance evaluation report for the detection equipment.

[0015] To address the aforementioned problems, this invention also provides a collaborative online detection evaluation system based on spatiotemporal performance joint driving, the system comprising: The spatiotemporal registration and fusion module is used to perform spatiotemporal registration and fusion on the real-time raw signal data of the detection device to obtain the synchronous signal data of the detection device. A spatial coherent accumulation module is used to perform spatial coherent accumulation on the synchronization signal data to obtain enhanced detection data of the synchronization signal data; The performance evaluation module is used to analyze the task association features of the enhanced detection data based on the detection task requirements of the detection device, and to perform multi-dimensional performance evaluation of the detection device based on the task association features, so as to obtain the detection area coverage index, tracking accuracy index and identification accuracy index of the detection device. The dynamic weight fusion module is used to perform parametric factor fusion on the detection area coverage index, the tracking accuracy index and the identification accuracy index based on the importance of the detection task requirements, so as to obtain the comprehensive performance index of the detection equipment. The time-series trend simulation module is used to perform time-series trend simulation on the comprehensive performance index to obtain the performance evolution trajectory of the detection equipment. The robustness analysis module is used to perform operational robustness analysis on the detection equipment based on the performance evolution trajectory, so as to generate an online performance evaluation report of the detection equipment.

[0016] Compared with the prior art, the present invention has the following beneficial effects: 1. This invention achieves temporal synchronization, spatial unification, and spatial energy focusing of multi-source detection data at the signal layer through the cascaded processing of spatiotemporal registration fusion and spatial coherent accumulation, providing high-quality input for performance feature extraction. Unlike existing methods where data preprocessing and performance evaluation are independent, this invention's spatiotemporal registration fusion step provides a phase consistency benchmark for spatial coherent accumulation, while the coherent accumulation step provides a high signal-to-noise ratio signal foundation for feature extraction, forming a causal progression of "preprocessing supporting accumulation, accumulation supporting evaluation," effectively reducing the stepwise propagation of data layer errors to the performance layer. Simultaneously, based on the semantic parsing results required by the detection task, this invention maps the task importance to weight coefficients for coverage, tracking, and identification dimensions through priority analysis, achieving dynamic adaptation of the evaluation index system to the task scenario. Unlike existing fixed-weight evaluation models, the weight coefficients of this invention adaptively adjust with the task type identifier, enabling the comprehensive performance index to prominently reflect the core focus dimensions of the current task, avoiding a disconnect between evaluation results and actual application needs, and effectively improving the rationality and relevance of the comprehensive performance evaluation.

[0017] 2. This invention extracts trend parameters, periodic parameters, and residual fluctuation parameters by extrapolating the time-series trends of comprehensive performance indicators, identifies macro-evolutionary stages and micro-state transition points, and constructs performance evolution trajectories, achieving in-depth mining of the temporal variation patterns of performance. Based on this, operational robustness analysis is performed to derive the lower limit of stable operation and assess the near-time domain risk level, realizing a leap from "snapshot-style" static judgment to "trajectory-style" dynamic prediction in performance evaluation. This enables proactive identification of equipment performance degradation trends and potential risks, providing quantitative decision-making basis for system operation and maintenance and task planning. Ultimately, this invention starts with the acquisition of real-time raw signal data, and through spatiotemporal registration and fusion, spatial coherence accumulation, multi-dimensional performance evaluation, dynamic weight fusion, time-series trend extrapolation, and operational robustness analysis, the six steps are interconnected, forming a progressive closed-loop technical link. There are clear causal dependencies between each step, and none of them can be omitted. Together, they solve the comprehensive technical problems of data layer error propagation, task layer adaptation deficiency, and time layer risk blind spot in the existing technology, effectively improving the accuracy, dynamic adaptability, and foresight of the online evaluation of the collaborative detection system's performance. Attached Figure Description

[0018] Figure 1 This is a flowchart illustrating an embodiment of the online evaluation method for collaborative detection based on spatiotemporal performance joint driving. Figure 2 This is a functional block diagram of a collaborative online detection evaluation system based on spatiotemporal performance joint driving provided in an embodiment of the present invention; The realization of the objective, functional features and advantages of the present invention will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0019] It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.

[0020] This application provides an online evaluation method for collaborative detection based on spatiotemporal performance. The executing entity of this online evaluation method includes, but is not limited to, at least one of the following electronic devices that can be configured to execute the method provided in this application: a server, a terminal, etc. In other words, the online evaluation method for collaborative detection based on spatiotemporal performance can be executed by software or hardware installed on a terminal device or a server device. The server includes, but is not limited to, a single server, a server cluster, a cloud server, or a cloud server cluster. The server can be an independent server or a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, content delivery networks (CDN), and big data and artificial intelligence platforms.

[0021] Reference Figure 1 The diagram shown is a flowchart illustrating an online evaluation method for cooperative detection based on spatiotemporal performance joint driving according to an embodiment of the present invention. In this embodiment, the online evaluation method for cooperative detection based on spatiotemporal performance joint driving includes: S1. Perform spatiotemporal registration and fusion on the real-time raw signal data of the detection device to obtain the synchronous signal data of the detection device; In this embodiment of the invention, the step of performing spatiotemporal registration and fusion on the real-time raw signal data of the detection device to obtain the synchronization signal data of the detection device includes: Collect real-time raw signal data from the detection equipment; The real-time raw signal data is timestamped to obtain the time registration data of the detection device; Based on the position and attitude parameters of the detection device, the time registration data is transformed into spatial coordinates to obtain the spatial registration data of the detection device. The spatial registration data is processed to standardize its format to obtain the synchronization signal data of the detection device.

[0022] During operation, the detection equipment sets a fixed sampling interval of 100 times per second based on its own signal generation frequency. This interval can completely capture the dynamic changes of the signal, ensuring that no signal characteristics are missed. The equipment continuously acquires the raw signal data it generates through its built-in high-speed flash memory storage module. This data includes not only the signal strength, frequency, and generation time, but also key basic information such as the signal phase and amplitude. The acquisition process begins when the equipment starts operating and ends when the data acquisition task is completed. During this period, the signal information obtained from each sampling is completely recorded, and each data point is assigned a unique sampling number to ensure that the signal data at each sampling moment is accurately captured without any time gaps. Ultimately, this forms a real-time raw signal data set of the detection equipment, containing all sampling information and stored in a structured data table format.

[0023] The system time, recorded by a high-precision clock at the microsecond level built into the detection device, is used as the reference time. This reference time is continuously accumulated from the moment the device is started, ensuring the continuity and accuracy of time measurement. The built-in time record field of each real-time raw signal data is read one by one, and the signal generation timestamp is extracted. The timestamps, which may have existed in different formats such as microseconds and seconds, are uniformly converted to a millisecond-level time format calculated from the moment the device is started. By calculating the millisecond difference between each data timestamp and the reference time at the same point in time, this difference is added to the original timestamp, adjusting the time record value of each data point. This ensures that the timestamps of all real-time raw signal data accurately correspond to the specific moment on the reference time axis, ensuring that the unit of measurement for signal data generated at different sampling moments is uniformly milliseconds, and that the time consistency error is controlled within 1 millisecond. Finally, time registration data from the detection device is obtained, each containing an adjusted standard timestamp and basic information about the original signal.

[0024] The detection equipment acquires the current installation location coordinates and attitude angle information collected by the built-in GPS positioning module and three-axis gyroscope. The location coordinates include longitude, latitude, and altitude, while the attitude angles include pitch angle, roll angle, and yaw angle. The 2000 National Geodetic Coordinate System, adopted by the state, was selected as the preset unified spatial coordinate system to ensure the universality and accuracy of the spatial coordinates. Based on the longitude, latitude, and altitude values ​​of the detection equipment, the distance offset of the equipment relative to the origin of the 2000 National Geodetic Coordinate System in the X, Y, and Z axes was calculated. Then, based on the pitch angle, roll angle, and yaw angle, the unit vectors of the signal in the X, Y, and Z axes relative to the center point of the equipment in its own local coordinate system were calculated. The direction vectors in the local coordinate system were rotated and converted to the direction vectors in the geodetic coordinate system through coordinate transformation rules. Subsequently, the converted direction vectors were translated and superimposed with the three-dimensional coordinate offsets to obtain the absolute coordinates of the signal in the geodetic coordinate system. Thus, all spatial position information corresponding to all signals in the time registration data was transformed from the local coordinate system of the equipment to the 2000 National Geodetic Coordinate System. The spatial coordinate uniformity error was controlled within 1 meter. Finally, spatial registration data of the detection equipment was obtained, each containing three-dimensional coordinates in the geodetic coordinate system, standard timestamps, and basic information of the original signal.

[0025] To meet the needs of subsequent spatial coherent accumulation processing, a pre-defined and unified data format standard was established to ensure that the data can be directly recognized and retrieved by subsequent processing modules. This standard clearly specifies the field order of signal data as signal strength, signal frequency, timestamp, and spatial coordinates. The spatial coordinate field includes three subfields: X, Y, and Z. All fields are uniformly of decimal number type. Signal strength is retained to 3 decimal places, signal frequency to 2 decimal places, timestamp as an integer, and spatial coordinates to 1 decimal place. The data storage length is 32 bits, including 1 sign bit, 10 integer bits, and the decimal part is allocated the appropriate number of bits according to the requirements of different fields. Each spatial registration data is checked one by one according to the field order, field type, and storage length. Data whose field order does not conform to the preset standard is rearranged. Non-decimal number fields are converted to decimal according to the numerical conversion rules. Data with a storage length of less than 32 bits is padded with zeros at the lower bits. Data with a storage length of more than 32 bits is truncated to 32 bits according to the rounding principle. This ensures that the format of all spatial registration data fully complies with the preset standard and there are no format differences. Finally, the synchronization signal data of the detection equipment is obtained in the form of a standardized data file that can be directly called by the spatial coherent accumulation module.

[0026] The beneficial effects are as follows: Through the above specific implementation steps, the real-time raw signal data of the detection equipment can be consistently aligned in the time dimension based on the same millisecond-level reference time axis, effectively avoiding signal misalignment caused by time differences. In the spatial dimension, it is unified to the 2000 national geodetic coordinate system, ensuring that the signal data generated by detection equipment at different locations are spatially comparable and consistent. In terms of data format, it follows a unified standard specification adapted to the needs of subsequent processing, and can be directly used for subsequent processing without additional conversion, greatly improving processing efficiency. The resulting synchronization signal data completely retains key information such as the strength, frequency, phase, and amplitude of the original signal, and completely eliminates problems such as temporal heterogeneity, spatial heterogeneity, and format heterogeneity. The data quality is significantly improved, providing accurate, unified, and high-quality data support for subsequent spatial coherent accumulation processing. Meanwhile, the entire implementation process clearly defines fixed parameters such as sampling interval, coordinate system type, and data format standards, as well as specific operating steps, adjustment rules, and error control ranges. Different implementers, under the same detection equipment and task requirements, can obtain synchronous signal data with errors within the allowable range by following this process. This effectively ensures the reproducibility of the technical solution and the reliability of data processing, laying a solid data foundation for the accuracy of online performance evaluation of the entire collaborative detection system.

[0027] S2. Perform spatial coherent accumulation on the synchronization signal data to obtain enhanced detection data of the synchronization signal data; In this embodiment of the invention, the step of performing spatial coherent accumulation on the synchronization signal data to obtain enhanced detection data of the synchronization signal data includes: The synchronization signal data is spatially distributed and sampled to obtain a spatial sample set of the synchronization signal data. Based on the relative positions of signal sources of the samples in the spatial sample set, phase compensation is performed on the spatial sample set to obtain the phase alignment data of the synchronization signal data. The phase-aligned data is coherently superimposed to obtain the enhanced detection data of the synchronization signal data.

[0028] The synchronization signal data already possesses a unified format, timestamp, and spatial coordinate information in the geodetic coordinate system. First, the boundary range of the detection area is clearly defined. Based on the preset start and end coordinates of the detection mission, the extreme ranges of the X, Y, and Z axes of the detection airspace are determined to ensure coverage of the entire area required for the detection mission. The detection airspace is divided into several non-overlapping spatial grid units along the X, Y, and Z axes using a fixed three-dimensional grid size of 10 meters × 10 meters × 5 meters. The boundary coordinates of each grid unit are determined progressively by superimposing the grid size onto the starting coordinates. For example, in the X-axis direction, starting from the starting coordinates, the X-axis end coordinate of each grid unit is determined by adding 10 meters to the grid size. This process is repeated to define the X, Y, and Z axis boundary coordinate ranges of each grid unit, and each grid unit is treated as an independent sampling unit. A unique 16-bit identifier code is assigned to each grid unit. The first 8 bits represent the detection area number, and the last 8 bits represent the three-dimensional coordinate offset of the grid within the area, ensuring unique and traceable identification. All synchronization signal data are traversed, and the X, Y, and Z three-dimensional spatial coordinates in the geodetic coordinate system corresponding to each data point are extracted one by one. The coordinates are then compared with the boundary coordinate ranges of all grid cells to determine whether the data coordinates are simultaneously within the X, Y, and Z axis boundary ranges of a certain grid cell. If they are, the data is assigned to that grid cell group. A dual verification mechanism is set up in the assignment determination process: first, preliminary matching is performed through the coordinate range, and then secondary confirmation is performed through the grid identifier code. This ensures that the signal data in each grid cell is completely collected without grid omissions, data duplication, or assignment errors. Finally, a spatial sample set of synchronization signal data is formed, with grid cells as the grouping unit, each group of data associated with a unique grid identifier code, and containing all basic signal information and characteristic parameters of each group.

[0029] The X, Y, and Z three-dimensional spatial coordinates of each sample in the spatial sample set are extracted one by one in the geodetic coordinate system. For each grid cell, the average values ​​of its X-axis boundary coordinates, Y-axis boundary coordinates, and Z-axis boundary coordinates are calculated to obtain the coordinates of the geometric center point of the grid cell. This value is then used as the reference position for that cell, ensuring that the reference position is located in the central region of the grid cell and guaranteeing the uniformity of phase compensation. For each sample, the differences between its X-axis coordinate and the X-axis coordinate of the corresponding grid cell reference position, the differences between its Y-axis coordinate and the Y-axis coordinate of the reference position, and the differences between its Z-axis coordinate and the reference position are calculated. These distance differences in three directions are used to accurately determine the relative position of each sample to the reference position. Based on the fixed value that the speed of electromagnetic wave propagation in vacuum is 3 × 10^8 m / s, the three-dimensional distance differences between the sample and the reference position are vectorized to obtain the straight-line distance from the sample to the reference position. This straight-line distance is then divided by the speed of electromagnetic wave propagation to calculate the phase delay time required for each sample signal to propagate from the signal source to the reference position. Based on the fixed correspondence between phase and time, i.e., the phase change of electromagnetic wave propagation per unit time is 360 degrees / cycle, the phase change per unit time can be calculated by combining the signal frequency. The phase delay time is then converted into a specific phase delay amount. For example, when the signal frequency is 1GHz, every 1 nanosecond delay corresponds to a phase change of 360 degrees. According to the calculated phase delay amount, the signal phase of each sample is adjusted by reverse compensation. The negative number of the phase delay is added to the original phase value of the sample to correct the delayed phase to the level consistent with the phase of the reference signal. After compensation, by comparing the phase values ​​of all samples in the same grid cell, it is ensured that the phase difference is controlled within 0.1 degrees, achieving complete phase alignment of all sample signals with no phase deviation. Finally, phase-aligned data is obtained, in which each data point contains the compensated phase information and is consistent with the phase of the reference signal.

[0030] Phase-aligned data is processed by grouping data into grid cells, with each group associated with a corresponding grid identifier code to ensure clear and unambiguous grouping. For each grid cell, key feature parameters such as signal strength, frequency, amplitude, phase, and polarization direction are extracted for each data point. The extraction process strictly adheres to a unified format standard to ensure the accuracy and consistency of feature parameter extraction. Superposition operations are performed according to feature parameter type. For signal strength and amplitude parameters, a sequential summation method is used, adding the signal strength values ​​of all samples within the same grid cell to obtain the total signal strength of that grid cell; similarly, the amplitude values ​​of all samples are added sequentially to obtain the total signal amplitude of that grid cell. For signal frequency parameters, the frequency values ​​of all samples within the same grid cell are counted, and the frequency value with the highest frequency is recorded as the signal frequency of that grid cell. If multiple frequency values ​​have the same highest frequency, the arithmetic mean of these frequencies is taken as the final signal frequency. For other characteristic parameters such as phase and polarization direction, the arithmetic mean of the corresponding values ​​of all samples within the same grid cell is calculated. This involves summing the phase values ​​of all samples and then dividing by the total number of samples to obtain the average phase value. The polarization direction is calculated using vector synthesis to obtain the average polarization direction. During the superposition process, the original variation law of the signal characteristics is strictly followed, without introducing any additional interference factors or correction coefficients, ensuring that the superimposed signal truly reflects the actual propagation state and energy distribution of the signal within the grid cell. After superposition, a list of characteristic parameters is established for the enhanced signal data of each grid cell, clearly recording the values ​​of each superimposed characteristic parameter. Finally, enhanced detection data of the synchronization signal data is obtained, corresponding to a set of enhanced signal data for each grid cell, with complete and unmissing characteristic parameters, concentrated signal energy, and optimized quality.

[0031] The beneficial effects are as follows: By clearly defining the boundaries of the detection area and the size of the three-dimensional grid, combined with a unique grid identification code and a dual attribution verification mechanism, the orderly and accurate division of synchronous signal data is achieved, ensuring comprehensive coverage and balanced distribution of signal data within the detection airspace, with no data omissions or attribution errors; by accurately calculating the geometric center point of the grid unit as the reference position, combined with a fixed electromagnetic wave propagation speed and phase-time correspondence, the accurate conversion and reverse compensation of phase delay are achieved, effectively eliminating phase differences between samples at different locations, maintaining a high degree of consistency in the phase dimension, and achieving a phase alignment accuracy within 0.1 degrees; by performing targeted superposition operations according to the type of characteristic parameters, the signal energy within the same spatial unit is integrated, significantly improving signal strength and clarity, increasing the signal-to-noise ratio by more than 30%, and effectively suppressing random noise interference. The enhanced detection data fully preserves the key characteristic parameters of the synchronization signal, significantly improving signal identification and reliability. This provides a precise and high-quality data foundation for subsequent task-related feature analysis and multi-dimensional performance evaluation. At the same time, the entire implementation process clearly defines the grid division standards, reference position calculation methods, compensation logic, superposition rules, and verification mechanisms. The operation standards are unified and executable, and different implementers can obtain consistent processing results by following the process, effectively ensuring the reproducibility of the technical solution and the stability and reliability of data processing.

[0032] S3. Based on the detection task requirements of the detection device, analyze the task association features of the enhanced detection data, and based on the task association features, perform a multi-dimensional performance evaluation of the detection device to obtain the detection area coverage index, tracking accuracy index and identification accuracy index of the detection device. In this embodiment of the invention, the step of parsing the task-related features of the enhanced detection data based on the detection task requirements of the detection device includes: The detection task requirements of the detection device are semantically parsed to obtain the spatial constraints and performance constraints of the detection device. Feature extraction is performed on the enhanced detection data to obtain multidimensional features of the enhanced detection data; Based on the spatial constraints and performance constraints, recursive feature elimination is performed on the multidimensional features to obtain the task-related features of the enhanced detection data.

[0033] Based on the task-related features, a multi-dimensional performance evaluation of the detection device is performed to obtain the detection area coverage index, tracking accuracy index, and identification accuracy index of the detection device, including: The task-related features are deconstructed and separated to obtain the coverage features, tracking features, and identification features of the enhanced detection data; Based on the coverage characteristics, the airspace coverage rate of the detection device is analyzed to obtain the detection area coverage index of the detection device. Based on the tracking features, the positioning accuracy of the detection device is estimated to obtain the tracking accuracy index of the detection device; Based on the identification features, the attribute discrimination confidence quantification of the detection device is performed to obtain the identification accuracy index of the detection device.

[0034] The detection mission requirements of the detection equipment include explicit information such as detection range, target type, and detection accuracy threshold. The mission requirements are semantically processed by parsing sentence by sentence to extract key information describing the spatial range, clarify the latitude and longitude intervals and altitude ranges corresponding to the detection mission, and form the spatial constraints of the detection equipment. At the same time, the requirements for signal response speed, target resolution capability, and data output frequency in the mission are extracted to determine the performance constraints of the detection equipment. This ensures that the spatial constraints and performance constraints directly correspond to the core content of the mission requirements, without any omissions or additional constraints.

[0035] The enhanced detection data contains information on various aspects such as signal strength, coverage area, target location trajectory, and signal attributes for each grid cell. The enhanced detection data of all grid cells is analyzed cell by cell to extract information such as signal coverage area, signal propagation distance, target occurrence frequency, target movement speed, and signal signature for each cell. This information is then classified and organized into feature items of different dimensions. Each feature item corresponds to a specific attribute of the enhanced detection data. All feature items together constitute a multi-dimensional feature set of the enhanced detection data, ensuring that feature extraction is comprehensive and without duplication.

[0036] Based on spatial and performance constraints, each feature in the multidimensional feature set is verified one by one. First, it is determined whether the feature is related to the spatial constraints, that is, whether the feature can reflect the detection situation of the detection device within the specified spatial range. Then, it is determined whether it meets the performance constraints, that is, whether the feature can support the detection device to meet the performance standards required by the mission. Feature items that do not meet either spatial or performance constraints are directly eliminated. Feature items that only meet a single constraint are marked and temporarily retained. Then, the verification process is repeated to gradually eliminate the marked non-core feature items until the remaining feature items all meet both spatial and performance constraints. Finally, the mission-related features of the enhanced detection data are obtained.

[0037] The task-related features are classified and sorted, and deconstructed and separated according to the detection functions described by the features. Features describing the detection area range and coverage density are grouped into one category as coverage features to enhance the detection data; features describing the target positioning accuracy and motion trajectory tracking continuity are grouped into another category as tracking features to enhance the detection data; and features describing target attribute recognition and feature matching degree are grouped into another category as recognition features to enhance the detection data. This ensures that the boundaries between the three categories of features are clear and there are no overlaps or incorrect classifications.

[0038] Based on information such as the number of grids covered and the effective coverage area in the coverage features, the total number of grids within the specified spatial range of the detection mission is first counted. Then, the number of grids actually covered by the signal in the coverage features is calculated. The actual number of covered grids is divided by the total number of grids to obtain the spatial coverage rate of the detection equipment. This coverage rate value is used as the detection area coverage index of the detection equipment. The index value is retained to two decimal places to ensure that the calculation process is directly based on the coverage features and the result truly reflects the coverage of the detection area.

[0039] Based on the target's actual location record and detection location record in the tracking features, the actual coordinates and detection coordinates of all targets in the tracking features are collected. The difference in latitude, longitude and altitude of each set of coordinates is calculated. The absolute values ​​of all differences are summed and divided by the number of coordinate sets to obtain the average positioning deviation. This average positioning deviation is used as the tracking accuracy index of the detection equipment. The unit of the index is consistent with the coordinate unit to ensure that the index can accurately reflect the positioning accuracy level of the detection equipment.

[0040] Based on information such as target attribute matching signals and feature matching records in the identification features, the number of signals in the identification features that are consistent with the target attributes and the total number of signals participating in attribute discrimination are counted. The confidence level of attribute discrimination is obtained by dividing the number of signals with consistent attributes by the total number of signals. This confidence level value is used as the identification accuracy indicator of the detection device. The indicator value is retained to two decimal places to ensure that the indicator can objectively reflect the accuracy of the detection device in identifying the target attributes.

[0041] The beneficial effects are as follows: semantic parsing of the detection task requirements ensures that subsequent feature extraction and performance evaluation closely align with task needs, avoiding interference from irrelevant features; comprehensive extraction of multi-dimensional features and recursive feature elimination ensure the relevance and effectiveness of task-related features; feature deconstruction and separation clarify the core basis for different performance dimensions; the calculation process of the three types of indicators is directly based on the corresponding features, resulting in accurate results with clear physical meaning; the final detection area coverage indicator, tracking accuracy indicator, and identification accuracy indicator can comprehensively and truthfully reflect the actual performance of the detection equipment, providing reliable sub-item data support for subsequent comprehensive performance evaluation; at the same time, the entire implementation process has clear operation steps and clear judgment criteria, ensuring the reproducibility of the technical solution and the consistency of the evaluation results.

[0042] S4. Based on the importance of the detection task requirements, the detection area coverage index, the tracking accuracy index, and the identification accuracy index are fused using parametric factors to obtain the comprehensive performance index of the detection equipment. In this embodiment of the invention, the step of fusing parametric factors of the detection area coverage index, the tracking accuracy index, and the identification accuracy index based on the importance of the detection task requirements to obtain the comprehensive performance index of the detection equipment includes: Prioritize the requirements of the detection mission to determine their degree of importance. Based on the task type identifier required by the detection task, the degree of importance is mapped to a multi-dimensional weighting coefficient of the detection area coverage index, the tracking accuracy index, and the identification accuracy index; Based on the multidimensional weighting coefficients, the detection area coverage index, the tracking accuracy index, and the identification accuracy index are weighted and fused to obtain the comprehensive performance index of the detection device.

[0043] The formula for calculating the comprehensive performance index is as follows: ; In the formula, This indicates the comprehensive performance index. This represents the weight coefficient of the coverage dimension. This represents the weight coefficient of the tracking dimension. This represents the weight coefficient of the recognition dimension. This represents the normalized detection area coverage index. This represents the normalized tracking accuracy index. This represents the accuracy index of identification after normalization.

[0044] The exploration mission requirements include a detailed description of core needs, such as a coverage area requirement specifying "covering the entire area from 110° to 112° East longitude, 30° to 32° North latitude, and 50 to 500 meters above sea level," a target tracking accuracy requirement of "positioning error for moving targets not exceeding 1 meter," and a target attribute recognition requirement of "accuracy in identifying target type and model not less than 95%." The preset priority criteria are based on the mission's execution logic. The first priority is the core objective that the mission must achieve; failure to achieve this objective results in a failed mission, corresponding to an importance value of 0.6. The second priority is the key supporting objective, directly affecting the mission's performance, corresponding to an importance value of 0.3. The third priority is the auxiliary optimization objective, only supplementing and improving the mission's performance without affecting the achievement of the core objective, corresponding to an importance value of 0.1. The text of the detection task requirements is broken down sentence by sentence according to punctuation marks such as periods and semicolons. For each sentence, the core requirements are extracted, focusing on key expressions related to "coverage," "tracking," and "identification" to ensure no core requirements are omitted. The extracted core requirements are then compared with preset priority criteria. If the core requirement is clearly "ensuring no omissions in the detection area," the coverage dimension is directly assigned first priority with an importance score of 0.6. Subsequently, considering the descriptions of tracking and identification in the task requirements, the tracking dimension is assigned second priority with an importance score of 0.3, and the identification dimension is assigned third priority with an importance score of 0.1. If the core requirement is "accurately locking the target's trajectory," the tracking dimension is assigned first priority with an importance score of 0.6, the coverage dimension second priority with an importance score of 0.3, and the identification dimension third priority with an importance score of 0.1. If the core requirement is "accurately identifying target attribute information," the identification dimension is assigned first priority with an importance score of 0.6, the coverage dimension second priority with an importance score of 0.3, and the tracking dimension third priority with an importance score of 0.1. The entire judgment process strictly follows the rule that each dimension corresponds to a unique priority level, ensuring that the sum of the importance values ​​of the coverage dimension, tracking dimension, and identification dimension is always 1, and finally forming the importance level required for the detection task that includes the specific importance values ​​of the three dimensions.

[0045] The task type identifiers required for the detection missions are pre-defined into three categories based on the detection scenario and core objectives: area monitoring identifiers, target tracking identifiers, and attribute recognition identifiers. Area monitoring identifiers are suitable for tasks with the core objective of large-area, blind-spot-free detection coverage, such as city security full-area monitoring and border coverage detection. Target tracking identifiers are suitable for tasks with the core objective of continuously locking onto moving targets and accurately tracking their trajectories, such as vehicle tracking and aircraft trajectory monitoring. Attribute recognition identifiers are suitable for tasks with the core objective of accurately identifying target attribute information, such as weapon and equipment model identification and personnel identity verification. Simultaneously, a weight coefficient mapping rule is preset for each type of identifier. The mapping rule for area monitoring identifiers is that the coverage dimension weight coefficient equals the importance of the coverage dimension, the tracking dimension weight coefficient equals the importance of the tracking dimension, and the recognition dimension weight coefficient equals the importance of the recognition dimension. The mapping rules for target tracking and attribute recognition identifiers are completely consistent with those for area monitoring identifiers, ensuring consistent mapping logic. First, extract the clearly marked task type identifier from the beginning or end of the text requiring the detection task. This identifier is presented in a fixed format of "Task Type: XX". After extraction, confirm that it belongs to one of the three preset identifiers to avoid identifier identification errors. Then, according to the mapping rules corresponding to the identifier, directly use the importance of the coverage dimension obtained in the first step as the weight coefficient of the coverage dimension, the importance of the tracking dimension as the weight coefficient of the tracking dimension, and the importance of the identification dimension as the weight coefficient of the identification dimension. Since the sum of the importance values ​​of the three dimensions is 1, the sum of the three weight coefficient values ​​after mapping remains 1, thus accurately obtaining the weight coefficients of the coverage dimension, tracking dimension, and identification dimension.

[0046] First, obtain the detection area coverage index, tracking accuracy index, and identification accuracy index obtained through the previous multi-dimensional performance evaluation. These three indices are presented in specific numerical form. The theoretical maximum value for each index is preset to 100. Specifically, a theoretical maximum value of 100 for the detection area coverage index means that the detection equipment completely covers the entire area required for the detection mission, without any omissions; a theoretical maximum value of 100 for the tracking accuracy index means that the detection equipment has zero positioning error for the target, accurately locking the target's location; and a theoretical maximum value of 100 for the identification accuracy index means that the detection equipment's judgment of target attributes is completely accurate, achieving a 100% accuracy rate. A fixed numerical conversion method is used for normalization. The actual value of the detection area coverage index is divided by 100 to obtain the normalized detection area coverage index. This index value is strictly controlled between 0 and 1. For example, if the actual value of the detection area coverage index is 85, the normalized value is 0.85. The actual value of the tracking accuracy index is divided by 100 to obtain the normalized tracking accuracy index. The value range is also controlled between 0 and 1. For example, if the actual value of the tracking accuracy index is 92, the normalized value is 0.92. The actual value of the recognition accuracy index is divided by 100 to obtain the normalized recognition accuracy index. The value range is maintained between 0 and 1. For example, if the actual value of the recognition accuracy index is 88, the normalized value is 0.88. The normalized detection area coverage index is multiplied by itself (i.e., the index value is multiplied by itself), and the result is multiplied by the coverage dimension weight coefficient. Simultaneously, the normalized tracking accuracy index is multiplied by itself and then multiplied by the tracking dimension weight coefficient. The normalized recognition accuracy index is also multiplied by itself and then multiplied by the recognition dimension weight coefficient. These three multiplication results are summed to obtain a total value. A specific value is then calculated such that multiplying this specific value by itself equals the total sum value. This specific value is the comprehensive performance index of the detection equipment, reflecting its overall effectiveness.

[0047] The beneficial effects are as follows: By pre-setting clear priority judgment criteria, the basis for judging core requirements at different levels and their corresponding fixed values ​​are clarified. This provides a unified and executable operational standard for analyzing the importance of detection task requirements, completely avoiding biases caused by subjective judgments and ensuring that different implementers can obtain completely consistent importance results when facing the same detection task requirements. Based on pre-set task type identifiers and corresponding mapping rules, a direct and accurate conversion from importance level to weight coefficient is achieved. The weight allocation is highly consistent with the core requirements of the detection task, ensuring that the task focus is prominently reflected in the indicator fusion. The normalization process uses a fixed division by 100 conversion method, unifying the measurement standards of the three performance indicators and completely eliminating fusion interference that may be caused by differences in the original magnitude of the indicators, making indicators of different dimensions comparable and fusionable. The weighted fusion calculation process, through clearly defined numerical computation steps, comprehensively integrates the performance contributions of various dimensions. It fully covers the core performance dimensions of the detection equipment—coverage, tracking, and identification—and, through weighting coefficients, fully reflects the impact of the importance of the detection mission on overall performance. The resulting comprehensive performance index comprehensively, objectively, and accurately reflects the overall performance level of the detection equipment in the current mission scenario. The entire implementation process clearly defines the judgment criteria, mapping rules, calculation steps, and numerical ranges. The operational standards are unified and repeatable, ensuring consistent results for different implementers following the process. This effectively guarantees the reproducibility of the technical solution and the reliability and accuracy of the evaluation results.

[0048] S5. Perform time-series trend analysis on the comprehensive performance index to obtain the performance evolution trajectory of the detection equipment; In this embodiment of the invention, the step of performing time-series trend analysis on the comprehensive performance index to obtain the performance evolution trajectory of the detection device includes: The comprehensive performance indicators are sorted by time series to obtain the performance time series sequence of the comprehensive performance indicators; Trend analysis is performed on the performance time series to obtain the trend parameters, periodic parameters, and residual fluctuation parameters of the performance time series. Based on the trend parameter, the periodic parameter, and the residual fluctuation parameter, the macroscopic evolution stage and microscopic state transition point of the performance time series are identified. Based on the macroscopic evolution stage and the microscopic state transition point, the performance time series is reconstructed to obtain the performance evolution trajectory of the detection device.

[0049] Each comprehensive performance indicator is automatically associated with a unique millisecond-level timestamp upon generation. This timestamp starts at the moment the detection device is activated and accumulates continuously in milliseconds, maintaining complete consistency with the time base in the previous data processing workflow, ensuring the continuity and uniformity of the time dimension throughout the evaluation process. All generated comprehensive performance indicators are comprehensively collected, a temporary dataset is established, and the timestamp and performance values ​​corresponding to each indicator are extracted one by one. A dual verification process is performed on the extracted data. The first verification checks for duplicate timestamps; if multiple indicators correspond to the same timestamp, the indicator data with the earlier generation time is retained, and the remaining duplicate data is removed. The second verification compares the continuous range of timestamps and, for performance indicators corresponding to missing time points, performs linear interpolation based on the values ​​of two adjacent valid indicators to supplement them, ensuring that no indicator at any time point is missed. The comprehensive performance indicators are sorted and arranged in ascending order of timestamp values, i.e., from earliest to latest. After sorting, the correspondence between each timestamp and the performance indicator is checked again to ensure that one timestamp corresponds to only one comprehensive performance indicator and there is no one-to-many or many-to-one correspondence. Finally, a performance time series is formed, in which comprehensive performance indicators are arranged in chronological order and each data point contains a specific millisecond-level timestamp and the corresponding performance value.

[0050] A point-by-point traversal analysis is performed on the performance time series to obtain trend parameters. The traversal begins from the second data point in the series, calculating the difference between the current comprehensive performance indicator value and the value of the preceding adjacent indicator. The difference is rounded to four decimal places. The difference results for all adjacent data points are recorded sequentially, forming a complete difference set. A statistical sample size threshold of 10 is set. Percentage determination is only performed when the number of data points in the difference set reaches 10 or more. The number of positive, negative, and zero values ​​in the difference set is counted, and the percentage of each type of difference relative to the total number of differences in the set is calculated. If the percentage of positive values ​​exceeds 50%, the trend parameter is considered to be upward; if the percentage of negative values ​​exceeds 50%, the trend parameter is considered to be downward; if the percentages of zero, positive, and negative values ​​all do not exceed 50%, the trend parameter is considered to be stable. This ensures that trend determination is based on a sufficient sample size and avoids misjudgments caused by random factors. When analyzing periodic parameters, a fixed time window of 1 hour (3,600,000 milliseconds) is set. Starting from the beginning timestamp of the performance time series, data is extracted segment by segment according to the time window length, forming several non-overlapping time segments. Each data segment contains all performance indicator data within a complete 1-hour period. The changes in performance indicators within each data segment are analyzed to extract the change pattern. The change pattern is described as a combination of rising, stable, and falling. For example, if the indicator in a data segment rises continuously for the first 30 minutes, remains stable for the middle 15 minutes, and gradually declines for the last 15 minutes, then the change pattern of that data segment is rising, stable, and falling. The change patterns of all data segments are compared pairwise, with the comparison criterion being that the difference in the duration of each stage within the same pattern does not exceed 5%. Repeating change patterns are identified, and the time interval between two consecutive occurrences of the repeating pattern is calculated, i.e., the difference between the beginning timestamp of the previous pattern and the beginning timestamp of the next pattern. This difference is the period duration corresponding to the periodic parameter. When calculating the residual fluctuation parameter, firstly, select the corresponding fitting method based on the determined trend parameter. If the trend is upward or downward, use linear fitting; if the trend is stationary, use horizontal fitting. Fit a trend line for the performance time series, with each time point in the trend line corresponding to a fitted value. Then, based on the period duration corresponding to the periodic parameter, use a sine curve fitting method to fit the periodic fluctuation curve of the series. The amplitude of the periodic fluctuation curve is determined based on the fluctuation range of the indicator values ​​in the series. Subtract the fitted value corresponding to the trend line at each time point from the actual value of the comprehensive performance indicator in the performance time series, and then subtract the fitted value corresponding to the periodic fluctuation curve at that time point. The result is the residual fluctuation parameter at that time point. The residual fluctuation parameters of all time points are arranged in chronological order, collectively constituting the residual fluctuation characteristics of the series.

[0051] Macro-evolutionary stages are defined based on trend parameters and duration. The time nodes where trend parameters change are identified from the performance time series. These nodes serve as dividing points to segment different time periods. The duration of the trend parameter within each time period is calculated, i.e., the difference between the end and start timestamps of the time period, converted to hours. If the trend parameter shows an upward trend and lasts for 1 hour or more within a time period, that time period is classified as an upward evolutionary stage; if the trend parameter shows a stable trend and lasts for 1 hour or more, it is classified as a stable evolutionary stage; if the trend parameter shows a downward trend and lasts for 1 hour or more, it is classified as a downward evolutionary stage. Time periods with a duration of less than 1 hour are merged into adjacent evolutionary stages with the same trend parameter. Each evolutionary stage clearly records its start and end timestamps, converted to specific date and time formats, ensuring clear and traceable stage divisions. A preset threshold of ±0.05 is set for the residual fluctuation parameter. This threshold is determined based on the numerical range of the comprehensive performance index from 0 to 1, and has been verified through extensive experiments to effectively distinguish between regular fluctuations and irregular abrupt changes. The residual fluctuation parameter at each time point is checked one by one. When the residual fluctuation parameter at a certain time point is greater than 0.05 or less than -0.05, that is, when the absolute value exceeds 0.05, that time point is identified as a microstate transition point. Detailed information for each transition point is recorded, including the millisecond-level timestamp of the transition point, the corresponding comprehensive performance index value, the specific value of the residual fluctuation parameter, and the fluctuation amplitude. The fluctuation amplitude is the result of subtracting 0.05 from the absolute value of the residual fluctuation parameter, ensuring that the identification of transition points is accurate and complete, and that the relevant information is fully traceable.

[0052] Based on the defined macro-evolutionary stages, the situation trajectory is reconstructed. For each stage, all comprehensive performance indicators within that stage are sequentially connected in ascending order of timestamps using a smooth curve method. The curve avoids abrupt changes at each data point, forming independent performance change curves for each stage. Each data point on the curve clearly corresponds to a specific millisecond-level time and a comprehensive performance indicator value. On the performance change curves of each stage, all micro-state transition points are accurately marked using solid red circles as special markers. The diameter of these markers is set to twice that of regular data points to ensure clear distinction. Next to each special marker, the fluctuation range of that transition point is noted in text form. The font size of the text matches the font size of the curve coordinate axis labels, and the position is 10 pixels to the right of the marker to avoid obscuring the curve and other data points. The performance change curves of each macro-evolutionary stage are seamlessly connected in chronological order, ensuring continuous timestamps and smooth curve transitions without breaks or abrupt turns. Simultaneously, all annotation information and related text descriptions for micro-state transition points are fully preserved. Ultimately, a two-dimensional trajectory curve is formed with millisecond-level time as the horizontal axis and comprehensive performance index value as the vertical axis. This curve is continuous and complete, and clearly shows the division of macroscopic evolution stages and the distribution of microscopic state transition points, which is the performance evolution trajectory of the detection equipment.

[0053] The beneficial effects include ensuring the accuracy, completeness, and consistency of the performance time series by using millisecond-level timestamps with a unified benchmark and employing a dual verification and interpolation completion mechanism for time series sorting, thus laying a solid data foundation for subsequent trend analysis. Trend pattern analysis comprehensively extracts three types of parameters—trend, periodicity, and residual fluctuation—through explicit difference calculations, percentage statistics, fixed time window divisions, change pattern comparisons, and targeted fitting methods. This fully reflects the long-term trends, cyclical patterns, and random fluctuations of performance changes, avoiding the limitations of single-dimensional analysis. The division of macro-evolutionary stages is based on clear trend types and a 1-hour duration standard, while the identification of micro-state transition points is based on a preset threshold of ±0.05 and complete information records, making the hierarchical structure of performance changes and key abrupt change nodes clearly identifiable and enhancing the analytical value of the trajectory. The situation trajectory reconstruction, through smooth curve connections, special marking, and seamless integration, forms an intuitive and comprehensive performance evolution trajectory, clearly and accurately presenting the dynamic changes in the performance of the detection equipment over time, providing precise and detailed analytical basis for subsequent operational robustness analysis. The entire implementation process clearly defines the time baseline, verification rules, calculation steps, judgment criteria, and presentation format. The operation standards are unified and repeatable. Different implementers can obtain a consistent performance evolution trajectory by following this process, which effectively ensures the reproducibility of the technical solution and the reliability and accuracy of the trajectory results.

[0054] S6. Based on the performance evolution trajectory, perform an operational robustness analysis on the detection equipment to generate an online performance evaluation report for the detection equipment.

[0055] In this embodiment of the invention, the step of performing operational robustness analysis on the detection device based on the performance evolution trajectory to generate an online performance evaluation report for the detection device includes: By performing a safety baseline derivation on the performance evolution trajectory, the lower limit of stable operation of the detection equipment is obtained. The time-series feature deconstruction of the performance evolution trajectory yields the performance fluctuation data and performance decay data of the detection device. Based on the performance fluctuation data and the performance degradation data, the degradation risk of the detection equipment is assessed to obtain the near-time domain risk level of the detection equipment; The comprehensive performance index, the stable operation lower limit, and the near-time domain risk level are integrated into an online performance evaluation report for the detection equipment.

[0056] All segments marked as stable evolution phases within the past 72 hours of the performance evolution trajectory are selected. All comprehensive performance index values ​​within these segments are extracted, and the extracted values ​​are sorted. The top 5% of excessively high values ​​and the bottom 5% of excessively low values ​​are removed. The minimum value among the remaining 90% of values ​​is used as the core reference value. Combined with the minimum performance guarantee value specified in the design and operation standards of the detection equipment, if the core reference value is higher than the design minimum guarantee value, then the core reference value is used as the lower limit of stable operation of the detection equipment; if the core reference value is lower than the design minimum guarantee value, then the design minimum guarantee value is used as the lower limit of stable operation, ensuring that the lower limit of stable operation both conforms to the actual operation of the equipment and meets the design requirements.

[0057] Using a fixed 1-hour time window, the entire performance evolution trajectory is divided into several continuous and non-overlapping time segments, each corresponding to performance change data within a window. The comprehensive performance index values ​​within each time window are analyzed, and the difference between the maximum and minimum values ​​within the window is calculated. This difference represents the performance fluctuation data for that window. The fluctuation data from all windows are arranged chronologically to form a complete performance fluctuation dataset. Simultaneously, the average value of the comprehensive performance index within each time window is calculated. The difference between the current window's average value and the previous window's average value is calculated. If the difference is negative, its absolute value is recorded as a single attenuation. When three or more consecutive negative differences occur, these attenuation values ​​are accumulated to obtain the cumulative attenuation. All single attenuation values ​​and the cumulative attenuation together constitute the performance attenuation data of the detection equipment.

[0058] The preset performance fluctuation threshold is 0.1, and the performance decay threshold is 0.05. The near-term time range is set to the next 24 hours. The maximum fluctuation value in the performance fluctuation data is compared with the fluctuation threshold, and the cumulative decay in the performance decay data is compared with the decay threshold. The duration of decay is then considered for comprehensive judgment. If the maximum fluctuation value exceeds 0.1 and the cumulative decay exceeds 0.05 and lasts for 2 hours or more, the near-term risk level is determined to be high risk. If the maximum fluctuation value is between 0.05 and 0.1 and the cumulative decay is between 0.03 and 0.05 and lasts for 1 hour or more, it is determined to be medium risk. If the maximum fluctuation value is below 0.05 and the cumulative decay is below 0.03 or there is no continuous decay, it is determined to be low risk, ensuring that the risk level determination has clear numerical basis and time standard.

[0059] The system collects the current values ​​of comprehensive performance indicators, the specific values ​​of stable operating limits, the near-term risk level, and the performance fluctuation and attenuation data used to determine the risk. These are then integrated into an online performance assessment report with a fixed structure. The first part of the report clearly presents the current values ​​of the comprehensive performance indicators. The second part explains the process and specific values ​​for determining the stable operating limits. The third part details the basis for determining the near-term risk level and the key values ​​of fluctuation and attenuation data. The fourth part provides targeted operational recommendations based on the risk level: high-risk recommendations include immediate shutdown and inspection; medium-risk recommendations include shortening the monitoring cycle to once every 30 minutes; and low-risk recommendations include maintaining the regular monitoring frequency. This ensures the report is complete, logically clear, and has practical guiding significance.

[0060] The beneficial effects are as follows: by selecting historical stable phase data and combining it with design standards to derive the lower limit of stable operation, this ensures that the lower limit not only matches the actual operating state of the equipment but also complies with design specifications, providing a reliable benchmark for robustness analysis; the time series feature deconstruction adopts a fixed time window division method, which makes the extraction of performance fluctuation data and degradation data have a unified standard, and the data results are objective and comparable; the degradation risk assessment is based on clear numerical thresholds and time conditions, avoiding the subjectivity of risk level determination and ensuring the accuracy and reliability of the assessment results; the online performance evaluation report integrates core indicators, judgment criteria, and targeted suggestions, providing comprehensive and practical information, and providing detailed and reliable reference for the operation, maintenance, and risk prevention of detection equipment; the entire implementation process has clear operational standards and clear judgment logic, ensuring the reproducibility of the technical solution and the accuracy of the analysis results.

[0061] like Figure 2 The diagram shown is a functional block diagram of a collaborative online detection evaluation system based on spatiotemporal performance joint driving provided by an embodiment of the present invention.

[0062] The online evaluation system 100 for collaborative detection based on spatiotemporal performance joint drive described in this invention can be installed in an electronic device. Depending on the functions implemented, the online evaluation system 100 may include a spatiotemporal registration and fusion module 101, a spatial coherence accumulation module 102, a performance evaluation module 103, a dynamic weight fusion module 104, a time-series trend extrapolation module 105, and an operational robustness analysis module 106. The module described in this invention can also be called a unit, which refers to a series of computer program segments that can be executed by the processor of an electronic device and can perform a fixed function, stored in the memory of the electronic device.

[0063] In this embodiment, the functions of each module / unit are as follows: The spatiotemporal registration and fusion module 101 is used to perform spatiotemporal registration and fusion on the real-time raw signal data of the detection device to obtain the synchronous signal data of the detection device. The spatial coherent accumulation module 102 is used to perform spatial coherent accumulation on the synchronization signal data to obtain enhanced detection data of the synchronization signal data. The performance evaluation module 103 is used to analyze the task association features of the enhanced detection data based on the detection task requirements of the detection device, and to perform multi-dimensional performance evaluation of the detection device based on the task association features, so as to obtain the detection area coverage index, tracking accuracy index and identification accuracy index of the detection device. The dynamic weight fusion module 104 is used to perform parametric factor fusion on the detection area coverage index, the tracking accuracy index and the identification accuracy index based on the importance of the detection task requirements, so as to obtain the comprehensive performance index of the detection equipment. The time-series trend simulation module 105 is used to perform time-series trend simulation on the comprehensive performance index to obtain the performance evolution trajectory of the detection equipment. The operational robustness analysis module 106 is used to perform operational robustness analysis on the detection equipment based on the performance evolution trajectory, so as to generate an online performance evaluation report of the detection equipment.

[0064] In the several embodiments provided by this invention, it should be understood that the disclosed methods and systems can be implemented in other ways. For example, the system embodiments described above are merely illustrative; for instance, the division of modules is only a logical functional division, and other division methods may be used in actual implementation.

[0065] The modules described as separate components may or may not be physically separate. The components shown as modules may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs.

[0066] Furthermore, the functional modules in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or in the form of hardware plus software functional modules.

[0067] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the present invention can be implemented in other specific forms without departing from the spirit or essential characteristics of the present invention.

[0068] This application embodiment can acquire and process relevant data based on artificial intelligence technology. Artificial intelligence is the theory, method, technology, and application system that uses digital computers or machines controlled by digital computers to simulate, extend, and expand human intelligence, perceive the environment, acquire knowledge, and use that knowledge to obtain optimal results.

[0069] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention.

Claims

1. A collaborative online assessment method for detection based on spatiotemporal performance joint driving force, characterized in that, The method includes: S1. Perform spatiotemporal registration and fusion on the real-time raw signal data of the detection device to obtain the synchronous signal data of the detection device; S2. Perform spatial coherent accumulation on the synchronization signal data to obtain enhanced detection data of the synchronization signal data; S3. Based on the detection task requirements of the detection device, analyze the task association features of the enhanced detection data, and based on the task association features, perform a multi-dimensional performance evaluation of the detection device to obtain the detection area coverage index, tracking accuracy index and identification accuracy index of the detection device. S4. Based on the importance of the detection task requirements, the detection area coverage index, the tracking accuracy index, and the identification accuracy index are fused using parametric factors to obtain the comprehensive performance index of the detection equipment. S5. Perform time-series trend analysis on the comprehensive performance index to obtain the performance evolution trajectory of the detection equipment; S6. Based on the performance evolution trajectory, perform an operational robustness analysis on the detection equipment to generate an online performance evaluation report for the detection equipment.

2. The online evaluation method for collaborative detection based on spatiotemporal performance joint driving as described in claim 1, characterized in that, The process of performing spatiotemporal registration and fusion on the real-time raw signal data of the detection device to obtain the synchronization signal data of the detection device includes: Collect real-time raw signal data from the detection equipment; The real-time raw signal data is timestamped to obtain the time registration data of the detection device; Based on the position and attitude parameters of the detection device, the time registration data is transformed into spatial coordinates to obtain the spatial registration data of the detection device. The spatial registration data is processed to standardize its format to obtain the synchronization signal data of the detection device.

3. The online evaluation method for collaborative detection based on spatiotemporal performance joint driving as described in claim 1, characterized in that, The step of performing spatial coherent accumulation on the synchronization signal data to obtain enhanced detection data of the synchronization signal data includes: The synchronization signal data is spatially distributed and sampled to obtain a spatial sample set of the synchronization signal data. Based on the relative positions of signal sources of the samples in the spatial sample set, phase compensation is performed on the spatial sample set to obtain the phase alignment data of the synchronization signal data. The phase-aligned data is coherently superimposed to obtain the enhanced detection data of the synchronization signal data.

4. The online evaluation method for collaborative detection based on spatiotemporal performance joint driving as described in claim 1, characterized in that, The task association features of the enhanced detection data, based on the detection task requirements of the detection device, include: The detection task requirements of the detection device are semantically parsed to obtain the spatial constraints and performance constraints of the detection device. Feature extraction is performed on the enhanced detection data to obtain multidimensional features of the enhanced detection data; Based on the spatial constraints and performance constraints, recursive feature elimination is performed on the multidimensional features to obtain the task-related features of the enhanced detection data.

5. The online evaluation method for collaborative detection based on spatiotemporal performance joint driving as described in claim 1, characterized in that, Based on the task-related features, a multi-dimensional performance evaluation of the detection device is performed to obtain the detection area coverage index, tracking accuracy index, and identification accuracy index of the detection device, including: The task-related features are deconstructed and separated to obtain the coverage features, tracking features, and identification features of the enhanced detection data; Based on the coverage characteristics, the airspace coverage rate of the detection device is analyzed to obtain the detection area coverage index of the detection device. Based on the tracking features, the positioning accuracy of the detection device is estimated to obtain the tracking accuracy index of the detection device; Based on the identification features, the attribute discrimination confidence quantification of the detection device is performed to obtain the identification accuracy index of the detection device.

6. The online evaluation method for collaborative detection based on spatiotemporal performance joint driving as described in claim 1, characterized in that, Based on the importance of the detection task requirements, the detection area coverage index, the tracking accuracy index, and the identification accuracy index are fused using parametric factors to obtain the comprehensive performance index of the detection equipment, including: Prioritize the requirements of the detection mission to determine their degree of importance. Based on the task type identifier required by the detection task, the degree of importance is mapped to a multi-dimensional weighting coefficient of the detection area coverage index, the tracking accuracy index, and the identification accuracy index; Based on the multidimensional weighting coefficients, the detection area coverage index, the tracking accuracy index, and the identification accuracy index are weighted and fused to obtain the comprehensive performance index of the detection device.

7. The online evaluation method for collaborative detection based on spatiotemporal performance joint driving as described in claim 6, characterized in that, The formula for calculating the comprehensive performance index is as follows: ； In the formula, This indicates the comprehensive performance index. This represents the weight coefficient of the coverage dimension. This represents the weight coefficient of the tracking dimension. This represents the weight coefficient of the recognition dimension. This represents the normalized detection area coverage index. This represents the normalized tracking accuracy index. This represents the accuracy index of identification after normalization.

8. The online evaluation method for collaborative detection based on spatiotemporal performance joint driving as described in claim 1, characterized in that, The step of performing time-series trend analysis on the comprehensive performance index to obtain the performance evolution trajectory of the detection equipment includes: The comprehensive performance indicators are sorted by time series to obtain the performance time series sequence of the comprehensive performance indicators; Trend analysis is performed on the performance time series to obtain the trend parameters, periodic parameters, and residual fluctuation parameters of the performance time series. Based on the trend parameter, the periodic parameter, and the residual fluctuation parameter, the macroscopic evolution stage and microscopic state transition point of the performance time series are identified. Based on the macroscopic evolution stage and the microscopic state transition point, the performance time series is reconstructed to obtain the performance evolution trajectory of the detection device.

9. The online evaluation method for collaborative detection based on spatiotemporal performance joint driving as described in claim 1, characterized in that, The step of performing an operational robustness analysis on the detection device based on the performance evolution trajectory to generate an online performance evaluation report for the detection device includes: By performing a safety baseline derivation on the performance evolution trajectory, the lower limit of stable operation of the detection equipment is obtained. The time-series feature deconstruction of the performance evolution trajectory yields the performance fluctuation data and performance decay data of the detection device. Based on the performance fluctuation data and the performance degradation data, the degradation risk of the detection equipment is assessed to obtain the near-time domain risk level of the detection equipment; The comprehensive performance index, the stable operation lower limit, and the near-time domain risk level are integrated into an online performance evaluation report for the detection equipment.

10. A collaborative online assessment system for detection based on spatiotemporal performance, characterized in that, The system for implementing the online evaluation method for collaborative detection based on spatiotemporal performance joint driving as described in claim 1 includes: The spatiotemporal registration and fusion module is used to perform spatiotemporal registration and fusion on the real-time raw signal data of the detection device to obtain the synchronous signal data of the detection device. A spatial coherent accumulation module is used to perform spatial coherent accumulation on the synchronization signal data to obtain enhanced detection data of the synchronization signal data; The performance evaluation module is used to analyze the task association features of the enhanced detection data based on the detection task requirements of the detection device, and to perform multi-dimensional performance evaluation of the detection device based on the task association features, so as to obtain the detection area coverage index, tracking accuracy index and identification accuracy index of the detection device. The dynamic weight fusion module is used to perform parametric factor fusion on the detection area coverage index, the tracking accuracy index and the identification accuracy index based on the importance of the detection task requirements, so as to obtain the comprehensive performance index of the detection equipment. The time-series trend simulation module is used to perform time-series trend simulation on the comprehensive performance index to obtain the performance evolution trajectory of the detection equipment. The robustness analysis module is used to perform operational robustness analysis on the detection equipment based on the performance evolution trajectory, so as to generate an online performance evaluation report of the detection equipment.

Images