Service module coordination method and system
By combining dynamic electronic fences and real-time movement trajectories in a multi-dimensional evaluation method, the problem of incomplete data credibility assessment in existing technologies has been solved, enabling more accurate judgment of data authenticity and anomaly identification, and improving the comprehensiveness and accuracy of data evaluation.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- SHANGHAI TIANLONG DIGITAL TECH CO LTD
- Filing Date
- 2026-06-08
- Publication Date
- 2026-07-03
AI Technical Summary
Existing technologies lack a comprehensive assessment of the location behavior and data credibility of mobile devices in field operation data collection scenarios, resulting in insufficient accuracy in judging the authenticity of data.
By introducing dynamic electronic fences and judging the cumulative time ratio of real-time movement trajectories, and combining time consistency and data content authenticity analysis, the credibility of operation data is evaluated from multiple dimensions, and data interaction and adaptive adjustment are carried out through the service module collaboration mechanism.
It improves the comprehensiveness of data credibility assessment, identifies abnormal situations such as data forgery, post-entry supplementation, or off-site data collection, reduces misjudgments caused by positioning errors and environmental changes, and enhances the accuracy of data assessment and the ability to identify systemic anomalies.
Smart Images
Figure CN122340481A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of data processing technology, and in particular to a service module collaboration method and system. Background Technology
[0002] In field data collection scenarios, mobile devices typically collect on-site data (such as images and location information) and upload it to a cloud system for processing and analysis. Existing data processing systems generally include a data receiving module, an auditing module, and an analysis module. The data receiving module records the timestamp of the uploaded data; the auditing module checks for data completeness and missing information; and the analysis module assesses quality based on the data content itself (e.g., the interval between the upload and collection times). In some solutions, the system performs a preliminary screening of the data's authenticity based on preset geographical area divisions and fixed task assignments.
[0003] For example, patent document CN117197770B discloses an IoT-based inspection process data monitoring system and method. It obtains the matching degree between the real-time location information of personnel and robots and the preset route, sets data requirements, and judges whether the collected data meets the corresponding requirements in order to eliminate invalid data.
[0004] However, the existing solutions primarily focus on the matching degree between location and preset routes, as well as indicators such as the timestamp and completeness of the data itself. The correlation between the location behavior of mobile devices during actual operations (e.g., whether the device is within the work area corresponding to its task) and data credibility has not yet been incorporated into the evaluation system. This results in a relatively singular evaluation dimension and insufficient accuracy when judging the authenticity of data (e.g., whether there is supplementary or falsified data). Therefore, how to combine device location behavior information to more comprehensively evaluate the credibility of collected data has become a key technical issue of concern in this field. Summary of the Invention
[0005] To address the aforementioned technical problems, the technical solution adopted by this invention is as follows: According to a first aspect of the present invention, a service module collaboration method is provided, the method comprising the following steps: The first service module obtains the job data uploaded by the mobile device, the time information of the job data, the real-time movement trajectory of the mobile device, and the task information of the current task.
[0006] The second service module determines the dynamic electronic fence area based on the task information.
[0007] The third service module evaluates the credibility of the operation data based on at least the first, second, and third dimensions, and generates corresponding judgment results. The first dimension is used to determine whether the time information meets the preset time conditions, the second dimension is used to determine whether the real-time movement trajectory meets the preset location matching conditions, and the third dimension is used to determine whether the operation data meets the preset data content conditions.
[0008] The third service module generates a credibility label based on the judgment result.
[0009] The fourth service module determines whether to send an instruction to the mobile device to re-upload the job data based on the credibility tag.
[0010] The first service module, the second service module, the third service module, and the fourth service module interact with each other through a preset collaboration mechanism.
[0011] According to a second aspect of the present invention, a service module collaboration system is provided, the system comprising: The first service module is used to obtain the job data uploaded by the mobile device, the time information of the job data, the real-time movement trajectory of the mobile device, and the task information of the current task.
[0012] The second service module is used to determine the dynamic electronic fence area based on the task information.
[0013] The third service module is used to evaluate the credibility of the operation data based on at least the first, second and third dimensions, generate corresponding judgment results, and generate credibility labels based on the judgment results; wherein, the first dimension is used to determine whether the time information meets the preset time conditions, the second dimension is used to determine whether the real-time movement trajectory meets the location matching conditions related to the dynamic electronic fence area, and the third dimension is used to determine whether the operation data meets the preset data content conditions.
[0014] The fourth service module is used to send an instruction to the mobile device to re-upload the job data when the credibility tag is lower than a preset credibility threshold.
[0015] The first service module, the second service module, the third service module, and the fourth service module interact with each other through a preset collaboration mechanism.
[0016] The present invention has at least the following beneficial effects: This invention assesses the credibility of operational data from multiple dimensions by introducing dynamic electronic fences and the cumulative time ratio of real-time movement trajectories, combined with time consistency and data content authenticity analysis. Compared to existing methods that only focus on data integrity or preset route matching, this solution helps identify anomalies such as data forgery, post-entry data entry, or off-site data collection. Through asynchronous collaboration and adaptive adjustment mechanisms between modules, misjudgments caused by positioning errors or environmental changes can be reduced to some extent. Furthermore, collaborative detection of multiple mobile devices can provide a reference for identifying joint fraud or systemic anomalies. Overall, it improves the comprehensiveness of data credibility assessment and provides an implementable technical architecture for collaborative processing between service modules.
[0017] It should be understood that the description in this section is not intended to identify key or essential features of the embodiments of the present invention, nor is it intended to limit the scope of the invention. Other features of the invention will become readily apparent from the following description. Attached Figure Description
[0018] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.
[0019] Figure 1 A flowchart of a service module collaboration method provided in an embodiment of the present invention. Detailed Implementation
[0020] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.
[0021] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains. The terminology used herein in the description of this invention is for the purpose of describing particular embodiments only and is not intended to be limiting of the invention. The term "and / or" as used herein includes any and all combinations of one or more of the associated listed items.
[0022] It should be noted that some exemplary embodiments are described as processes or methods depicted as flowcharts. Although the flowcharts describe the steps as sequential processes, many of these steps can be performed in parallel, concurrently, or simultaneously. Furthermore, the order of the steps can be rearranged. A process can be terminated when its operation is complete, but it may also have additional steps not included in the figures. A process can correspond to a method, function, procedure, subroutine, subroutine, etc.
[0023] Example 1 This embodiment provides a service module collaboration method applicable to a cloud-based data processing system. This system includes multiple interconnected service modules and a task allocation unit. Specifically, the service modules include a first service module, a second service module, a third service module, and a fourth service module. The service modules interact with each other through a preset collaboration mechanism (e.g., sequential invocation or parallel triggering), forming a collaborative data processing link with sequential invocation or parallel triggering. The task allocation unit generates, stores, and manages task information for the current task and provides it to the first service module and other modules that need to obtain task information. This embodiment is particularly suitable for scenarios requiring credibility assessment of operational data uploaded from mobile devices, such as on-site inspections, data collection, and remote monitoring.
[0024] like Figure 1 As shown, the method includes the following steps: S100, the first service module obtains the job data uploaded by the mobile device, the time information of the job data, the real-time movement trajectory of the mobile device, and the task information of the current task.
[0025] Specifically, mobile devices (such as smartphones or handheld terminals carried by inspection personnel, or data acquisition modules installed on automated inspection robots) collect on-site data in real time during operations through built-in sensors, cameras, or external detection instruments, forming operational data. Operational data includes, but is not limited to, images, videos, temperature readings, vibration parameters, or equipment operating status values.
[0026] When collecting operational data, the mobile device simultaneously records the timestamp of the collection moment. After completing data collection or according to a preset reporting cycle, it encapsulates the operational data along with the timestamps of the collection and upload moments, and uploads it to the cloud data processing system via a network (such as 4G / 5G, Wi-Fi, or IoT communication protocols). The first service module, acting as the data entry point for the cloud data processing system, receives the aforementioned data packets.
[0027] The mobile device also reports its geographic location information in real time (e.g., obtained via GPS, BeiDou, or base station positioning), and the reporting frequency can be preset according to the scenario (as an optional implementation, it can be set to once every 5 seconds). The first service module stores this geographic location information as a real-time movement trajectory in chronological order. Specifically, the first service module receives each geographic location data point reported by the mobile device, and each data point includes a timestamp and corresponding latitude and longitude coordinates. The first service module stores these data points sequentially as an ordered sequence according to the order of the timestamps, forming the real-time movement trajectory. When out-of-order reported data points are received, the first service module reorders them according to the timestamps before storing them; when multiple data points exist at the same timestamp, the latest reported one is retained.
[0028] The task allocation unit pre-stores or generates task information for the current task in real time. This task information includes the task type (e.g., inspection, maintenance, testing), task area, expected operation duration, and data collection requirements. The first service module, while receiving operation data, requests and retrieves the task information for the current task from the task allocation unit. Alternatively, the mobile device retrieves task information from the task allocation unit before executing the task and then reports it; the first service module extracts the task information from the information reported by the mobile device.
[0029] After the first service module completes the data acquisition, it standardizes the above information (e.g., unifies the data format and fills in missing fields), and then passes it to the second service module according to the collaborative link for subsequent processing.
[0030] S200, the second service module determines the dynamic electronic fence area based on the task information.
[0031] Specifically, the second service module receives task information from the first service module and parses the task type and corresponding work area requirements. The work area requirements specify the geographical range within which the mobile device is allowed to execute the current task, such as an area within a preset radius centered on specific coordinates, or a specific polygonal area.
[0032] The second service module generates a virtual geographical boundary, called a dynamic electronic fence area, based on the requirements of the work area. The area within this boundary is considered a compliant area where the mobile device is allowed to perform the current task. In other words, the dynamic electronic fence area is a specific and computable implementation of the work area requirements.
[0033] The range parameters of the dynamic electronic fence area are determined according to the task type indicated by the task information; different task types correspond to different range parameters. As an optional implementation, for inspection tasks, it can be a circular area with a preset first radius (e.g., 5 meters) centered on the current location of the mobile device; for maintenance tasks, it can be a circular area with a preset second radius (e.g., 10 meters) centered on the current location of the mobile device. The dynamic electronic fence area can also be other shapes (e.g., rectangles or polygons), set according to the map of the actual work area.
[0034] The second service module can also adaptively adjust the range of the dynamic electronic fence area based on the real-time movement status of the mobile device or current environmental data. The real-time movement status includes movement speed or direction, and the current environmental data includes positioning accuracy or signal strength. The positioning accuracy refers to the error range of the geographical location information currently reported by the mobile device, and the signal strength refers to the quality of the wireless communication signal between the mobile device and the cloud data processing system. The specific adjustment rules are as follows: Movement speed adjustment rules: When the movement speed of the mobile device is greater than the preset speed threshold, the second service module expands the radius or side length of the dynamic electronic fence area according to the positive correlation between speed and range. The expanded range is equal to the base range multiplied by the speed adjustment coefficient, which is greater than 1 and increases monotonically with the increase of speed. When the movement speed is less than or equal to the preset speed threshold, the base range remains unchanged.
[0035] Positioning accuracy adjustment rules: When the positioning accuracy error radius of the positioning signal is greater than the preset accuracy threshold, the second service module will increase the radius or side length of the dynamic electronic fence area by a compensation amount, which is equal to the absolute value of the difference between the positioning accuracy error radius and the accuracy threshold; when the positioning accuracy error radius of the positioning signal is less than or equal to the accuracy threshold, no compensation adjustment will be performed.
[0036] Signal strength adjustment rules: When the wireless communication signal strength of the mobile device is lower than the preset signal threshold, the second service module expands the radius or side length of the dynamic electronic fence area to tolerate the location reporting lag caused by signal delay; the expanded range is equal to the reference range multiplied by the signal compensation coefficient, which is negatively correlated with the signal strength, that is, the weaker the signal, the larger the coefficient.
[0037] The second service module obtains the real-time motion status and current environmental data from the first service module or the mobile device. After determining the dynamic electronic fence area, the second service module passes the area's range parameters (such as center coordinates and radius, or a list of polygon vertices) to the third service module for subsequent location matching.
[0038] Based on the above design of the dynamic electronic fence area, it has the following advantages: Task adaptability: Different task types (inspection, maintenance, testing, etc.) have different operational range requirements. This invention dynamically determines the range parameters of the electronic fence based on task information, ensuring that the location validity judgment matches the actual task requirements. This helps avoid misjudgments due to excessively small compliance areas or missed judgments due to excessively large compliance areas caused by using a uniform fence.
[0039] Dynamic responsiveness: Mobile devices may experience changes in movement speed or fluctuations in positioning signals during operation. This invention adaptively adjusts by incorporating real-time motion status (speed, direction) and environmental data (positioning accuracy, signal strength). When positioning is inaccurate or the device moves rapidly, the fence range is automatically expanded, helping to reduce misjudgments caused by technical errors; when high precision is required by the environment, the fence range is reduced, which helps to improve the rigor of the evaluation.
[0040] Configurability: The shape (circle, rectangle, polygon) and adjustment rules (speed adjustment coefficient, accuracy compensation amount, strictness coefficient) of the electronic fence can be preset or dynamically configured according to specific application scenarios, so that the present invention can be adapted to various industries (such as power inspection, security patrol, environmental monitoring, etc.) and positioning technologies with different precision (such as GPS, Beidou, UWB, Bluetooth beacon, etc.).
[0041] Reduce false alarm rate: The adaptive scaling mechanism helps reduce the number of mobile devices that are mistakenly identified as leaving the geofence due to positioning delay or signal jitter, thereby reducing unnecessary retransmission of instructions and improving the user experience and data processing efficiency of the system.
[0042] S300, the third service module evaluates the credibility of the job data based on at least the first, second and third dimensions, and generates corresponding judgment results.
[0043] Specifically, the third service module obtains the operation data, time information, and real-time movement trajectory from the first service module, and obtains the range parameters of the dynamic electronic fence area from the second service module. Then, it performs judgments according to the following three dimensions: The first dimension: Determine whether the time information meets the preset time conditions.
[0044] The third service module extracts the collection time and upload time from the time information and calculates the interval between the collection time and the upload time. A preset time condition is that this interval is less than or equal to a preset interval threshold (e.g., 30 minutes). If this condition is met, the judgment result for the first dimension is "compliant"; otherwise, it is "non-compliant". This judgment is used to identify whether there is suspected supplementary recording behavior with a long delay in uploading after collection.
[0045] The second dimension: Determine whether the real-time movement trajectory meets the location matching conditions related to the dynamic electronic fence area.
[0046] The third service module acquires all geographic location data points reported by the mobile device within a preset time window. The preset time window can be configured according to the actual scenario, for example, from a first preset duration before the collection time (as an optional implementation, such as 10 minutes) to a second preset duration after the upload time (as an optional implementation, such as 5 minutes).
[0047] The cumulative time spent by the mobile device within the dynamic electronic fence area within the time window is counted, and the ratio of this cumulative time to the total time of the time window is calculated to obtain the cumulative time percentage.
[0048] The location matching condition associated with the dynamic electronic fence area is: the cumulative time percentage is greater than or equal to a preset coverage threshold (which can be set to 80% as an optional implementation). If this condition is met, the judgment result for the second dimension is "compliant"; otherwise, it is "non-compliant".
[0049] This judgment is used to confirm that the operator has stayed within the designated work area for a sufficient period of time, avoiding situations such as non-stop passage or data transmission outside the area. Non-stop passage refers to the mobile device briefly passing through the electronic fence area without performing any actual work; data transmission outside the area refers to work data actually being collected outside the electronic fence area and then uploaded after entering the area.
[0050] The third dimension: Determine whether the task data meets the preset data content conditions.
[0051] The preset data content condition is that the deviation between the statistical characteristics of the job data and the data distribution represented by the pre-trained authenticity discrimination model is less than a preset deviation threshold. This deviation is calculated by the third service module using the pre-trained authenticity discrimination model.
[0052] In this embodiment, the data distribution represented by the pre-trained authenticity discrimination model is constructed using either a One-Class Support Vector Machine (SVM) or a Gaussian Mixture Model (GMM). The following explanation uses the Gaussian Mixture Model as an example.
[0053] Model Structure: The Gaussian mixture model consists of a linear combination of K Gaussian distributions, where K is the preset number of components (e.g., K=3). Each Gaussian component contains a mean vector μ. k Covariance matrix Σ k and mixed weight πk (satisfies Σπ) k =1). The overall probability density function of the model is p(x)=Σ k=1 K (π) k ·N(x|μ k ,Σ k ), where x is the feature vector.
[0054] Model Training: Training data consists of manually labeled or verified reliable samples of historical real-world data. The sample size should be sufficient to reflect the distribution characteristics of the real data; for example, the higher the feature dimension, the larger the required sample size. During training, the Expectation-Maximization (EM) algorithm is used to iteratively optimize the mean, covariance matrix, and mixture weights of each component. The convergence condition can be set to the change in the log-likelihood function being less than a preset convergence threshold (e.g., 1e^(-1 / 2)). -5 After the model is trained, each training sample has a corresponding likelihood probability value.
[0055] Feature Extraction: The third service module performs data analysis on the operation data itself and extracts its statistical features. The statistical features are determined according to the data type. For example, for numerical sensor data (temperature, vibration, etc.), statistical quantities such as mean, variance, kurtosis, and skewness can be extracted; for image data, histogram of oriented gradients (HOG) features or deep feature vectors extracted by a pre-trained convolutional neural network can be extracted.
[0056] Deviation Calculation: Input the extracted feature vector x into the Gaussian mixture model, and calculate the likelihood probability L(x) = Σ under the Gaussian mixture model. k=1 K (π) k ·N(x|μ k ,Σ k Then the degree of deviation D = -ln(L(x) + ε), where ε is a very small positive number (e.g., 1e). -8 This deviation (D) is used to avoid the logarithm being infinite. The degree of deviation is a measure of the difference between the statistical characteristic and the actual data distribution model.
[0057] Deviation threshold determination and judgment: The preset deviation threshold can be determined based on the distribution of deviation degrees of the training samples. For example, the mean of the deviation degrees of all training samples can be added to a preset multiple, such as 3 times the standard deviation. When the deviation degree is less than the preset deviation threshold, the judgment result of the third dimension is "compliant" (i.e., the deviation degree of statistical features from the real data distribution model is within an acceptable range); otherwise, it is "non-compliant".
[0058] Alternative Model: If a single-class support vector machine is used, a Gaussian kernel function is used during model training. The kernel parameter and soft boundary parameter are determined through cross-validation (e.g., the kernel parameter is the reciprocal of the feature dimension, and the soft boundary parameter is 0.05). The model output function f(x) represents the decision function value for whether x is "normal". The deviation can be defined as max(0, -f(x)): when f(x) ≥ 0, the sample is judged as normal by the model, and the deviation is 0; when f(x) < 0, the sample is judged as abnormal, and the deviation is equal to -f(x) (a positive value). The preset condition is f(x) ≥ 0 (i.e., judged as normal by the model), at which point the judgment result of the third dimension is "compliant".
[0059] This judgment is used to identify whether the data content differs significantly from the true pattern, thereby detecting possible forged or abnormal data.
[0060] The third service module combines the judgment results from the above three dimensions (each dimension being either "compliant" or "non-compliant") to generate the corresponding judgment result. For example, the judgment result can be represented as a combination of three Boolean values, such as a triple (compliant / non-compliant, compliant / non-compliant, compliant / non-compliant), for use in the subsequent generation of credibility labels.
[0061] In this embodiment, by combining the dynamic electronic fence with the cumulative time ratio of real-time movement trajectory, it helps to reflect the actual stay of the operator in the task area, and to a certain extent distinguishes between normal operation and non-stop passage or supplementary transmission outside the area, thereby improving the accuracy of data credibility assessment.
[0062] S400, the third service module generates a credibility label based on the judgment result, specifically including: S410, the judgment results of the first dimension, the second dimension and the third dimension are converted into a comprehensive metric value according to a preset mapping rule.
[0063] In this embodiment, the mapping rule adopts one of the following two methods: binary discrete mapping rule or weighted continuous fusion rule.
[0064] Method 1: Binary Discrete Mapping Rule When the time information meets preset time conditions, the real-time movement trajectory meets the location matching conditions, and the operation data meets preset data content conditions, the comprehensive metric value is a first value, such as 1; otherwise, the comprehensive metric value is a second value, such as 0. This rule is applicable to scenarios that require rapid decision-making and do not require differentiation of the degree of non-compliance.
[0065] Method 2: Weighted Continuous Fusion Rules The comprehensive metric is calculated by weighting and fusing the continuous penalty functions corresponding to the first, second, and third dimensions. The calculation formula is expressed as follows: S=α·e -λΔt ·f time +β·g(δ)·f position +γ·h(θ)·f data .
[0066] in: Δt represents the time interval between the time the job data is collected and the time it is uploaded.
[0067] e -λΔt λ is the time decay factor, which is a preset decay coefficient used to characterize the exponential penalty of time delay on data credibility, avoiding score abrupt changes near the threshold that may be caused by hard threshold.
[0068] f time ∈{0,1} represents the basic judgment result of the first dimension. It is 1 when Δt≤Tmax, and 0 otherwise. Tmax is the preset maximum allowable delay, for example, 30 minutes.
[0069] δ represents the cumulative time percentage deviation between the real-time movement trajectory and the dynamic electronic fence area, defined as: δ=∣(t inside / t window -p target |, where t inside t represents the cumulative time a mobile device spends within the electronic fence within a preset time window. window p represents the total duration of the window. target Set the target coverage (e.g., 0.8); g(δ) is the position matching penalty function, which can be taken as g(δ) = 1 / (1+k·δ). When δ = 0, g = 1, and it decreases as the deviation increases, reflecting the gradual impact of position compliance on credibility. k is the position matching penalty coefficient, which takes a value greater than 1. In an illustrative embodiment, it takes a value of 5.
[0070] f position ∈{0,1} represents the basic judgment result of the second dimension. It takes the value 1 when δ≤δmax, otherwise it takes the value 0, where δmax is the preset maximum acceptable deviation, for example, 0.2.
[0071] θ represents the degree of deviation between the operational data and the preset real data distribution model (e.g., Mahalanobis distance or reconstruction error).
[0072] h(θ) is the confidence function for the data content. Using h(θ) = e -μθThe sensitivity coefficient θ decays exponentially with increasing deviation, reflecting the smooth suppression of confidence by the degree of data anomaly. μ is a preset sensitivity coefficient (e.g., between 1 and 5) used to control the decay rate of the deviation θ on the confidence function of the data content. μ can be calibrated based on the distribution of deviation between normal and abnormal samples in historical data, so that h(θ) for normal samples is close to 1, and h(θ) for abnormal samples is significantly less than 1. In an illustrative embodiment, μ is 2.
[0073] f data ∈{0,1} represents the basic judgment result of the third dimension. It is 1 when θ≤θmax, and 0 otherwise. θmax is the preset maximum allowable deviation, for example, the mean of the deviation of all training samples plus 3 times the standard deviation.
[0074] α, β, γ are preset weighting coefficients, and satisfy α+β+γ=1; the weighting coefficients can be dynamically adjusted according to the task type or scenario, for example, the location weight is increased for inspection tasks, and the data weight is increased for maintenance tasks.
[0075] In this embodiment, the interval duration, positional deviation, and data deviation degree are all dimensionless or adjusted by coefficients to ensure that each variable in the formula is a pure numerical value. The basic judgment result f for each dimension... time f position f data As a hard constraint, it ensures that when any dimension seriously fails to meet the preset conditions, the contribution of the corresponding item is zero, thereby achieving the unity of hard constraints and soft penalties.
[0076] The weighted continuous fusion rule is suitable for scenarios that require fine-grained differentiation of the credibility levels of operational data. For example, when the system needs to sort multiple batches of data based on credibility scores, prioritize higher-risk data, or adopt differentiated processing strategies for data with different credibility levels in subsequent processes. Compared to the binary discrete mapping rule (which only outputs "credible / untrustworthy"), this rule can output continuous score values, providing a basis for refined data quality management.
[0077] S420, when the comprehensive metric value is greater than or equal to the preset credibility threshold, a first credibility tag is generated; when the comprehensive metric value is less than the preset credibility threshold, a second credibility tag is generated.
[0078] The first credibility label indicates credibility, and the second credibility label indicates untrustworthiness. The preset credibility threshold can be set based on actual needs, for example, to 0.6.
[0079] S500, the fourth service module determines whether to send an instruction to the mobile device to re-upload the job data (hereinafter referred to as "re-upload instruction") based on the credibility tag.
[0080] Specifically, the fourth service module receives the trustworthiness tag from the third service module. If the trustworthiness tag is the first trustworthiness tag (indicating trustworthiness), it is determined that the job data does not need to be re-uploaded, and no retransmission instruction is issued; if the trustworthiness tag is the second trustworthiness tag (indicating untrustworthiness), it is determined that the job data needs to be re-collected or re-uploaded, and the fourth service module generates a retransmission instruction accordingly and sends it to the corresponding mobile device via the network.
[0081] The retransmission instruction includes at least an identifier (e.g., data packet ID or collection timestamp) for the job data to be retransmitted, instructing the mobile device to locate and resend the corresponding job data. Upon receiving the instruction, the mobile device can automatically retrieve the data from its local cache and re-upload it, or prompt the operator to re-collect the data and then upload it.
[0082] As an optional implementation, the fourth service module may also include retransmission reason information (such as "time delay is too long", "location matching fails" or "data content is abnormal") when sending retransmission instructions, for reference by mobile devices or operators.
[0083] Furthermore, the first service module, the second service module, the third service module, and the fourth service module interact asynchronously with each other through a message queue.
[0084] Specifically, each service module writes the data to be processed and the processing results into a message queue in the form of messages. Downstream modules pull messages from the message queue for processing, thereby achieving decoupling between modules and traffic smoothing.
[0085] When the time taken for the third service module to generate the credibility tag exceeds the preset response time threshold (e.g., 2 seconds), it indicates that the current processing load is high or the model calculation is complex. In order to avoid blocking the subsequent data processing flow, the third service module temporarily stores the current job data and its corresponding credibility tag in a cache queue and immediately returns to the idle state to receive the next batch of data.
[0086] The cache queue adopts a first-in, first-out (FIFO) structure to temporarily store data records corresponding to retransmission instructions to be sent. When the third service module detects that the data processing link is idle, it triggers the fourth service module to retrieve at least one record from the cache queue. The criteria for determining that the data processing link is idle may include: the number of messages to be processed in the message queue is lower than a preset backlog threshold, and the CPU utilization rate of the cloud data processing system is lower than a preset load threshold.
[0087] The fourth service module packages the data corresponding to the retrieved retransmission instructions into a composite message. The packaging method involves encapsulating the identifier of at least one retransmission instruction, the job data identifier, and optional retransmission reason information in the same data packet according to a predetermined format (e.g., JSON array or TLV format). Then, the fourth service module sends this composite message to the corresponding mobile device in a single transaction to reduce network connection overhead and the number of signaling interactions.
[0088] After receiving the composite message, the mobile device parses out each retransmission instruction and performs the corresponding data retransmission operation according to the content of each instruction.
[0089] Furthermore, this embodiment also includes a multi-mobile device collaborative detection mechanism.
[0090] When at least two mobile devices are present within the same dynamic electronic fence area, the third service module acquires the operation data uploaded by each of the at least two mobile devices. For numerical operation data (e.g., temperature, vibration), the third service module constructs a feature vector from the data uploaded by each device; for image-based operation data, it extracts image feature vectors (e.g., HOG features or depth features). Then, it calculates the pairwise similarity between each set of operation data.
[0091] Similarity can be calculated using cosine similarity or Euclidean distance. When there are two or more mobile devices, the pairwise similarity between all device pairs is calculated, and the minimum (or average) value is taken as the overall similarity to characterize the consistency of data uploaded by multiple devices in the region. As an optional implementation, the cosine similarity threshold can be preset to 0.85.
[0092] When the overall similarity exceeds the preset similarity threshold, and the credibility labels corresponding to each task data are all second credibility labels (i.e., all are judged as "untrustworthy"), the third service module determines that there may be a systemic failure or joint fraud behavior in the electronic fence area, and marks each task data as multi-source consistency anomaly.
[0093] The third service module then generates a joint verification alarm, which includes: the time of the anomaly, the electronic fence area identifier, a list of mobile devices involved, a summary of the operation data uploaded by each device, and the similarity calculation results. The third service module then sends the alarm to the supervisory authority (e.g., the supervisory center of the cloud data processing system or the monitoring terminal of the operations and maintenance personnel) for further manual verification.
[0094] Example 2 This example provides a service module collaboration system, the system comprising: The first service module is used to obtain the job data uploaded by the mobile device, the time information of the job data, the real-time movement trajectory of the mobile device, and the task information of the current task.
[0095] The second service module is used to determine the dynamic electronic fence area based on the task information.
[0096] The third service module is used to evaluate the credibility of the operation data based on at least the first, second and third dimensions, generate corresponding judgment results, and generate credibility labels based on the judgment results; wherein, the first dimension is used to determine whether the time information meets the preset time conditions, the second dimension is used to determine whether the real-time movement trajectory meets the location matching conditions related to the dynamic electronic fence area, and the third dimension is used to determine whether the operation data meets the preset data content conditions.
[0097] The fourth service module is used to send an instruction to the mobile device to re-upload the job data when the credibility tag is lower than a preset credibility threshold.
[0098] The first service module, the second service module, the third service module and the fourth service module interact with each other through a preset coordination mechanism, such as synchronous calling, asynchronous message queue or sequential / parallel triggering.
[0099] The specific functional implementation, parameter configuration, data processing flow, and optional implementation methods of each module in this embodiment can all refer to the corresponding steps in Embodiment 1, and will not be repeated here.
[0100] This invention also provides an electronic device, including: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, the instructions being configured to perform the method described in this invention.
[0101] This invention also provides a computer-readable storage medium storing computer-executable instructions for performing the methods described in this invention.
[0102] It should be understood that the various forms of processes shown above can be used to reorder, add, or delete steps. For example, the steps described in this invention can be executed in parallel, sequentially, or in different orders, as long as the desired result of the technical solution disclosed in this invention can be achieved, and this is not limited herein.
[0103] The specific embodiments described above do not constitute a limitation on the scope of protection of this invention. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this invention should be included within the scope of protection of this invention.
Claims
1. A service module coordination method characterized by, The method includes the following steps: The first service module obtains the job data uploaded by the mobile device, the time information of the job data, the real-time movement trajectory of the mobile device, and the task information of the current task; The second service module determines the dynamic electronic fence area based on the task information; The third service module evaluates the credibility of the operation data based on at least the first, second, and third dimensions and generates corresponding judgment results. The first dimension is used to determine whether the time information meets the preset time conditions, the second dimension is used to determine whether the real-time movement trajectory meets the location matching conditions related to the dynamic electronic fence area, and the third dimension is used to determine whether the operation data meets the preset data content conditions. The third service module generates a credibility label based on the judgment result; The fourth service module determines whether to send an instruction to the mobile device to re-upload the job data based on the credibility tag. The first service module, the second service module, the third service module, and the fourth service module interact with each other through a preset collaboration mechanism.
2. The method according to claim 1, characterized in that, The range parameters of the dynamic electronic fence area are determined according to the task type indicated by the task information; different task types correspond to different range parameters.
3. The method according to claim 1, characterized in that, The location matching conditions include: within a preset time window, the cumulative time percentage of the mobile device being located within the dynamic electronic fence area is greater than or equal to a preset coverage threshold.
4. The method according to claim 1, characterized in that, The preset time conditions include: the interval between the time of data collection and the time of data upload is less than or equal to a preset interval threshold.
5. The method according to claim 1, characterized in that, The preset data content conditions include: the degree of deviation between the statistical characteristics of the job data and the data distribution represented by the pre-trained authenticity discrimination model is less than a preset deviation threshold.
6. The method according to claim 1, characterized in that, The third service module generates a credibility label based on the evaluation results, specifically including: The judgment results of the first dimension, the second dimension, and the third dimension are converted into a comprehensive metric value according to a preset mapping rule; the mapping rule is a binary discrete mapping rule or a weighted continuous fusion rule. When the comprehensive metric value is greater than or equal to the preset credibility threshold, a first credibility tag is generated; when the comprehensive metric value is less than the preset credibility threshold, a second credibility tag is generated.
7. The method according to claim 1, characterized in that, After determining the dynamic electronic fence area based on the task information, the second service module also adaptively adjusts the range of the dynamic electronic fence area based on the real-time motion status of the mobile device or the current environmental data; the real-time motion status includes movement speed or movement direction, and the current environmental data includes positioning accuracy or signal strength.
8. The method according to claim 1, characterized in that, The first, second, third, and fourth service modules interact asynchronously via message queues. When the third service module takes longer than a preset response time threshold to generate a credibility tag, the third service module temporarily stores the current job data and the credibility tag in a cache queue and triggers the fourth service module to assemble at least one re-upload instruction temporarily stored in the cache queue into a data packet and send it to the mobile device in a single transaction.
9. The service module collaboration method according to claim 1, characterized in that, Also includes: When at least two mobile devices exist within the same dynamic electronic fence area, the third service module obtains the operation data uploaded by the at least two mobile devices respectively and calculates the similarity between the operation data. When the similarity exceeds a preset similarity threshold and the credibility labels corresponding to each operation data are all lower than the preset credibility threshold, the third service module marks each operation data as having multi-source consistency anomaly and sends a joint verification alarm to the regulatory end.
10. A service module collaboration system, characterized in that, The system includes: The first service module is used to obtain the job data uploaded by the mobile device, the time information of the job data, the real-time movement trajectory of the mobile device, and the task information of the current task. The second service module is used to determine the dynamic electronic fence area based on the task information; The third service module is used to evaluate the credibility of the operation data based on at least the first, second and third dimensions, generate corresponding judgment results, and generate credibility labels based on the judgment results; wherein, the first dimension is used to determine whether the time information meets the preset time conditions, the second dimension is used to determine whether the real-time movement trajectory meets the location matching conditions related to the dynamic electronic fence area, and the third dimension is used to determine whether the operation data meets the preset data content conditions. The fourth service module is used to send an instruction to the mobile device to re-upload the job data when the credibility tag is lower than a preset credibility threshold. The first service module, the second service module, the third service module, and the fourth service module interact with each other through a preset collaboration mechanism.