Electromagnetic environment test stationing optimization method based on big data analysis

Abstract

The invention discloses an electromagnetic environment test stationing optimization method based on big data analysis, relates to the technical field of electromagnetic environment test, and is used for solving the problems of low efficiency and distortion of test stationing. According to the method, an electromagnetic test data base is generated by collecting electromagnetic monitoring data and interference event data, credible hierarchical management is carried out on records through quality tags, space-time aggregation and partition clustering are executed on the basis to generate risk partitions and cover gaps, interpretable generation of candidate distribution points is driven, the candidate distribution points are worked into work orders, and the reliability of the work orders is improved. The work order comprises a measurement parameter set, a constraint parameter set, a replacement point set and a priority rule, linkage scheduling solving is carried out in combination with spatial constraint and a resource state, an execution plan comprising work order selection, resource allocation, a route sequence and time period scheduling is output, sampling is transmitted back in the execution process, consistency verification is carried out, and the work order is obtained. And when the trigger rule is met, a supplementary work order or a retest work order is generated, closed-loop iteration of stationing optimization is realized, and invalid field work and rework are reduced.

Description

Technical Field

[0001] This invention relates to the field of electromagnetic environment testing technology, and more specifically, to a method for optimizing electromagnetic environment testing site layout based on big data analysis. Background Technology

[0002] In scenarios such as electromagnetic compatibility assessments, radio regulation investigations, and electromagnetic environment surveys in industrial parks, field personnel often need to carry spectrum analyzers or field strength meters to conduct high-point measurements along roadsides, factory perimeters, and residential areas. They also need to arrange sampling points based on restricted areas, entry times, travel routes, and the availability of personnel and instruments. Current methods mostly rely on experience-based point selection and static table plans, often marking points at fixed intervals or in fixed areas, supplemented by manual temporary adjustments. When on-site issues such as road closures due to construction, nighttime restrictions, access restrictions in sensitive areas, or incompatible instrument capabilities arise, the planned point layout is difficult to quickly replace and rearrange, leading to a disconnect between point layout and resource scheduling. This results in situations where execution is impossible upon arrival, frequent route changes, and repeated rework.

[0003] However, existing technologies have limitations in deployment strategies. Field sampling data comes from different equipment and different batches of tasks, resulting in inconsistencies in recording caliber, calibration status, and sampling integrity. Furthermore, interference events are sudden and time-dependent, making it difficult to align historical measurement results under a unified time and spatial caliber. Under these circumstances, deployment decisions cannot accurately identify high-risk areas and coverage gaps, easily leading to insufficient points in key areas and over-sampling in low-risk areas. Moreover, deployment plans usually lack consistency verification and closed-loop supplementation mechanisms based on backhauled data. Once on-site disturbances cause measurement time shifts or abnormal results, it is difficult to trigger retesting or supplementary deployment in a timely manner. Ultimately, the test results are discontinuous and untraceable in terms of spatial distribution and temporal comparison, affecting the reliability of electromagnetic environment assessment and interference investigation conclusions. Summary of the Invention

[0004] To overcome the aforementioned shortcomings of the prior art, the following solution is proposed to address the problems of inefficient and distorted test point placement in the aforementioned background art.

[0005] To achieve the above objectives, the present invention provides the following technical solution: An optimization method for electromagnetic environment testing site layout based on big data analysis includes the following steps: S1. Collect electromagnetic monitoring data, interference event data, spatial constraint data and operational resource data, unify time reference and spatial coordinates, generate an electromagnetic test data base, and generate quality labels for the data; S2. Based on the electromagnetic test data base and quality label, perform spatiotemporal aggregation and partition clustering to generate risk partitions and coverage gaps; S3. Generate candidate deployment points based on risk zoning and coverage gaps, and encapsulate the candidate deployment points into a deployment work order set. The deployment work order includes a set of measurement parameters, a set of constraint parameters, a set of alternative points, and priority rules. S4. Based on the set of work orders and combined with spatial constraint data and operation resource data, perform linkage scheduling to solve the problem and generate an execution plan. The execution plan includes work order selection results, resource allocation results, route sequence and time period scheduling. S5. Perform sampling according to the execution plan and return the sampling summary and quality label. Perform consistency verification based on the returned data. When the triggering rules are met, generate supplementary work orders or retest work orders and update the execution plan and version records.

[0006] Furthermore, the specific steps of S1 include: Analyze the raw fields of electromagnetic monitoring data, interference event data, spatial constraint data, and operational resource data, and establish a unified data dictionary and field mapping rules; According to the field mapping rules, the original fields are formatted, missing items are filled in, and the scope is normalized to form standardized records; Based on a unified time reference, a sampling time identifier is written into the standardized record; based on a unified spatial coordinate, a sampling location identifier is written into the standardized record; and a frequency band identifier and a data source identifier are written into the standardized record. Convert spatial constraint data into constraint records and write constraint type identifier and constraint range identifier; convert job resource data into resource records and write resource category identifier and availability status identifier. Standardized records, constraint records, and resource records are stored in the database and indexed to generate an electromagnetic test data base.

[0007] Furthermore, the specific steps for generating the quality label for S1 include: Read the device status information and calibration status information associated with the device identifier, and write the device trusted status identifier; Perform integrity and continuity checks on the sampled sequences, extract stability features, and write them into the stability identifier; Perform consistency verification on duplicate sampling records at the same location or in the same partition, extract consistency features and write them into the consistency identifier; Availability level identifiers and trustworthiness level identifiers are generated based on the device's trusted status identifier, stability identifier, and consistency identifier; Write the availability level identifier and the reliability level identifier into the quality label, and establish a link between the quality label and the corresponding record of the electromagnetic test data base.

[0008] Furthermore, the specific steps for generating risk partitions and coverage gaps in S2 include: Perform spatiotemporal aggregation on the electromagnetic test data base according to spatial units and time windows, generate aggregation records and write aggregation identifiers; Extract field strength variation features, spectrum occupancy features, and interference event density features from the aggregated records to form a partitioned feature vector; Perform partition clustering based on partition feature vectors, output risk partitions, and write partition identifiers and trigger basis summaries for each risk partition. The coverage of spatial units and time windows is calculated based on the spatiotemporal aggregation results and preset coverage rules. If the coverage does not meet the preset coverage rules, a coverage gap identifier is written. Output a list of coverage gaps and associate the coverage gap identifiers with the corresponding risk zone identifiers.

[0009] Furthermore, the specific steps for generating candidate placement points for S3 include: Read the risk zones and coverage gaps, perform adjacency analysis on the spatial units of the risk zones, determine the zone boundary bands and generate candidate points for the boundary bands; Extract the spatial differences between field strength variation characteristics and spectrum occupancy characteristics of the spatial cells corresponding to the coverage gap, locate the variation zone and generate candidate points of the variation zone; Extract the location of sensitive objects and the boundary of restricted areas from spatial constraint data, and generate candidate points of constraint neighborhood around the location of sensitive objects and the boundary of restricted areas; Deduplication and reachability screening are performed on candidate points in the boundary zone, change zone, and constrained neighborhood to form a candidate point layout. Write a risk partition identifier, a coverage gap identifier, and a candidate source identifier for each candidate deployment point to generate a candidate deployment point set.

[0010] Furthermore, the specific steps of S3 in encapsulating candidate placement points into a placement work order set include: Generate a work order identifier for each candidate point in the candidate point set and write the point parameters; The measurement parameter set is determined based on the risk zoning identifier and the coverage gap identifier. The measurement parameter set includes the measurement time window, the scanning frequency band range, the sampling duration, and the sampling mode identifier. Based on spatial constraint data and operational resource data, a set of constraint parameters is determined. The set of constraint parameters includes entry window identifiers, restricted area constraint identifiers, access path constraint identifiers, instrument capability constraint identifiers, and personnel qualification constraint identifiers. An alternative point set is generated based on the adjacency relationship and accessibility constraints of points within the same risk zone. The alternative point set includes an alternative relationship identifier and an alternative activation condition identifier. Priority rules are generated based on risk zoning identifiers, coverage gap identifiers, and measurement parameter sets. The measurement parameter sets, constraint parameter sets, alternative point sets, and priority rules are then written into the site placement work order to form a site placement work order set.

[0011] Furthermore, the specific steps for generating the execution plan through S4's coordinated scheduling solution include: Read the set of work orders, spatial constraint data and operation resource data, construct work order feasibility judgment rules and filter out work orders that do not meet the work order feasibility judgment rules; Establish a matching table between work orders and resources. The matching table is associated with the instrument capability constraint identifier and personnel qualification constraint identifier of the work order, and is also associated with the availability status identifier of the resource record. The deployment work orders are sorted according to priority rules, and personnel and instrument resources are allocated to the deployment work orders based on the matching relationship table to generate resource allocation results. Based on the travel path constraint identifier and the entry window identifier, generate route candidate sequence and time period candidate sequence, perform conflict detection on route candidate sequence and time period candidate sequence and switch alternative points or rearrange work orders; After the conflict detection is passed, the execution plan is output. The execution plan includes the work order selection result, resource allocation result, route sequence and time period scheduling, and the execution plan version identifier.

[0012] Furthermore, the specific steps for S5's return sampling summary and quality labeling include: At the field terminal, retrieve the corresponding deployment work order according to the execution plan, write the work order identifier, and record the arrival time identifier and departure time identifier; Execute the sampling mode and scanning frequency band range corresponding to the measurement parameter set, and generate a sampling summary. The sampling summary includes a measurement result summary identifier, a frequency band coverage summary identifier, and an abnormal event identifier. Collect equipment status information and on-site operating condition information, and generate a feedback quality label. The feedback quality label includes equipment status identifier, sampling continuity identifier, sampling stability identifier and location consistency identifier. The sampling summary and the returned quality label are bound to the work order identifier to form the returned data and uploaded to the electromagnetic test data base.

[0013] Furthermore, the specific steps of S5 consistency verification include: Read the historical records in the base of the returned data and electromagnetic test data, and establish a comparison set according to the risk zone identifier and the coverage gap identifier; Perform consistency checks on adjacent points within the partition, consistency checks on retesting the same point, and consistency checks on trends within the time period on the comparison set, and generate consistency judgment labels; When the consistency judgment flag meets the judgment conditions corresponding to the triggering rule, locate the work order flag associated with the anomaly and extract the anomaly cause flag. The anomaly cause flag includes the equipment trusted status anomaly flag, the time period mismatch anomaly flag, and the spatial mutation anomaly flag. Write the consistency determination flag and the anomaly cause flag into the returned data and archive it to the version record.

[0014] Furthermore, the specific steps for S5 to generate supplementary or retest work orders and update the execution plan and version history include: Read the anomaly cause identifier and coverage gap list, determine the supplement type identifier or retest type identifier according to the triggering rules, and generate the corresponding supplement work order or retest work order; To add or retest work orders, write the measurement parameter set, constraint parameter set, alternative point set, and priority rules, and then merge the added or retest work orders into the point placement work order set. Based on the merged set of work orders, the linkage scheduling solution is re-executed to generate an updated execution plan and write the execution plan version identifier. Write the trigger rule identifier, exception reason identifier, supplement type identifier or retest type identifier, execution plan version identifier and execution plan difference record into the version record; Output a list of deployment points, execution plan version link records, and version records.

[0015] The technical effects and advantages of the electromagnetic environment testing site optimization method based on big data analysis of this invention are as follows: This invention constructs a unified time reference and unified spatial coordinate data base for electromagnetic monitoring data, interference event data, spatial constraint data, and operational resource data, and generates a traceable quality label for each record. This achieves consistency of measurement results from multiple sources, reliable hierarchical structure, and reproducible analysis, avoiding mis-aggregation and misjudgment caused by equipment status, sampling omissions, and differences in measurement standards. Building upon this foundation, spatiotemporal aggregation and partitioned clustering are performed based on electromagnetic test data and quality labels to form risk partitions and coverage gaps. This transforms site selection decisions from experience-based selection to interpretable generation oriented towards risk and coverage semantics. Furthermore, candidate sites are encapsulated into a set of site selection work orders containing measurement parameter sets, constraint parameter sets, alternative point sets, and priority rules. Combined with spatial constraints and work resource execution linkage scheduling, an integrated execution plan with work order selection, resource allocation, route sequence, and time slot scheduling is output, improving field organization efficiency and constraint satisfaction capabilities. Simultaneously, a closed-loop mechanism of sampling feedback, quality label feedback, and consistency verification is introduced. When trigger rules are met, supplementary work orders or retest work orders are automatically generated, and the execution plan and version records are updated. This allows test site selection to continuously iterate with environmental changes and on-site disturbances, reducing the risk of invalid fieldwork and rework, and improving the coverage integrity, result consistency, and task execution stability of electromagnetic environment testing. Attached Figure Description

[0016] Figure 1 This is a flowchart illustrating an electromagnetic environment testing site optimization method based on big data analysis according to the present invention. Detailed Implementation

[0017] 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.

[0018] In order to achieve the above objectives, Figure 1 A schematic diagram of the structure of an electromagnetic environment test site optimization method based on big data analysis according to the present invention is given, which specifically includes the following steps; S1. Collect electromagnetic monitoring data, interference event data, spatial constraint data and operational resource data, unify time reference and spatial coordinates, generate an electromagnetic test data base, and generate quality labels for the data; S2. Based on the electromagnetic test data base and quality label, perform spatiotemporal aggregation and partition clustering to generate risk partitions and coverage gaps; S3. Generate candidate deployment points based on risk zoning and coverage gaps, and encapsulate the candidate deployment points into a deployment work order set. The deployment work order includes a set of measurement parameters, a set of constraint parameters, a set of alternative points, and priority rules. S4. Based on the set of work orders and combined with spatial constraint data and operation resource data, perform linkage scheduling to solve the problem and generate an execution plan. The execution plan includes work order selection results, resource allocation results, route sequence and time period scheduling. S5. Perform sampling according to the execution plan and return the sampling summary and quality label. Perform consistency verification based on the returned data. When the triggering rules are met, generate supplementary work orders or retest work orders and update the execution plan and version records.

[0019] S1. Collect electromagnetic monitoring data, interference event data, spatial constraint data, and operational resource data; unify the time base and spatial coordinates; generate an electromagnetic test data base; and generate quality labels for the data. The specific implementation is as follows: First, multi-source input data corresponding to the area to be tested is acquired. The input data includes electromagnetic monitoring data, interference event data, spatial constraint data, and operational resource data. Electromagnetic monitoring data is used to characterize electromagnetic measurement information collected at different times and locations, and includes at least the sampling time field, sampling location field, frequency band field, and measurement result field. Interference event data is used to characterize event records related to electromagnetic anomalies, and includes at least the event occurrence time field, event location field, event type field, and event summary field. Spatial constraint data is used to characterize the accessibility and limitations of field measurements, and includes at least the constraint type field and constraint range field. The constraint type is used to distinguish between prohibited access constraints, access window constraints, access path constraints, and sensitive object constraints. The constraint range is used to describe the geographical range, time range, or object range of the corresponding constraint. Operational resource data is used to characterize the status of available resources in the field, and includes at least the resource category field and availability status field. The resource category is used to distinguish between personnel resources and instrument resources, and the availability status is used to distinguish between available, occupied, and unavailable states.

[0020] The recording scope and generation method of the measurement result field are as follows: The measurement result field includes a field strength measurement value field and a spectrum scan result field. The field strength measurement value field is used to record the electromagnetic field strength measurement value at the sampling location corresponding to the sampling time identifier within the time period corresponding to the sampling time identifier, and is recorded together with the measurement unit identifier to ensure consistency. The spectrum scan result field is used to record the scan results within the scanning frequency band. The spectrum scan result field at least includes the frequency point sequence and the measurement value sequence corresponding to each frequency point, and records the frequency step aperture and detector aperture of the spectrum scan, so that the spectrum occupancy characteristics can be repeatedly calculated based on the spectrum scan result field. If the electromagnetic monitoring data source is a fixed monitoring station or a mobile measurement terminal, the field strength measurement value field and the spectrum scan result field are directly mapped from the original acquisition results. If the source is a historical archive system, the field strength measurement value field and the spectrum scan result field are restored according to the original recording scope before being entered into the database.

[0021] The input data is parsed and standardized to form a standardized record based on a unified time base and spatial coordinates.

[0022] The unified time reference is implemented as follows: The unified time reference uses the reference clock specified in the task configuration as the reference time caliber, and maintains a time offset correction record for each device identifier. The time offset correction record is read and generated by the device identifier when data is accessed. The generation process includes extracting the alignment point between the device time and the reference clock time from the acquisition system log, and determining the time offset correction value of the device identifier in the current time period based on this. When performing format conversion on the original time field, the original time field is first parsed into the device time, and then the device time is converted into a time expression under the unified time reference according to the time offset correction record, and written into the unified time expression corresponding to the sampling time identifier and the event occurrence time field, so as to ensure that electromagnetic monitoring data and interference event data are comparable and traceable under the same time caliber.

[0023] Field parsing includes identifying the original field name, field type, unit scope, and missing value rules, and establishing a unified data dictionary and field mapping rules. The unified data dictionary specifies standard field names, data types, unit scopes, and value ranges, while the field mapping rules convert the original fields into standard fields defined by the unified data dictionary. Scope unification includes format conversion, missing value completion, and scope normalization. Format conversion converts the original time field into a time expression under a unified time base, converts latitude and longitude coordinates, projected coordinates, or numbered positions into location expressions under unified spatial coordinates, and converts frequency band start and end ranges, center frequencies, or scan lists into unified frequency band expressions. Missing value completion fills in the default values ​​or null markers of necessary fields according to preset completion rules. Scope normalization expresses fields with different units or dimensions in a unified unit. The unified spatial coordinates are implemented as follows: The unified spatial coordinates use the coordinate system specified in the task configuration as the spatial standard and maintain a location mapping table. This table maps station numbers, road station numbers, work area numbers, and other numbered locations to their corresponding locations in the unified spatial coordinates. When the raw data carries latitude and longitude coordinates, these coordinates are converted to unified spatial coordinates according to the coordinate system transformation rules. When the raw data carries projected coordinates, these coordinates are converted to unified spatial coordinates according to the coordinate system parameters. When the raw data carries numbered locations, the corresponding unified spatial coordinates are first obtained from the location mapping table, and then the unified spatial coordinates corresponding to the sampling location identifier and event location fields are written into the table. The location mapping table is generated from the constraint range fields of the travel path constraints and sensitive object constraints in the spatial constraint data, as well as the basic geographic data used in field planning, and maintains version consistency throughout the task execution.

[0024] Based on a unified time reference, a sampling time identifier is written to the standardized record. The sampling time identifier is used to uniquely identify the time point or time interval corresponding to a sampling action. Based on a unified spatial coordinate, a sampling location identifier is written to the standardized record. The sampling location identifier is used to uniquely identify the location point or spatial unit corresponding to a sampling action. At the same time, a frequency band identifier and a data source identifier are written. The frequency band identifier is used to identify the frequency band range or set of frequency bands covered by the record. The data source identifier is used to identify the data access channel and the source of the sampling equipment. The frequency band identifier generation rules are as follows: When electromagnetic monitoring data is expressed as a spectrum scan in terms of a scanning frequency band range, the frequency band identifier is jointly determined by the starting frequency expression, the ending frequency expression, and the frequency step size of the scanning frequency band range; when electromagnetic monitoring data is expressed as a set of frequency bands representing multiple discrete scanning segments, the frequency band identifier is determined by concatenating the starting frequency expression and the ending frequency expression of each scanning segment within the frequency band set according to a preset sorting rule; when electromagnetic monitoring data is a fixed-point frequency measurement, the frequency band identifier is determined by the fixed-point frequency expression and the corresponding bandwidth size. The above generation rules are used to ensure that risk zoning statistics and coverage gap determination under the same size can be reproduced.

[0025] Spatial constraint data is converted into constraint records and written with constraint type identifier and constraint range identifier. Operational resource data is converted into resource records and written with resource category identifier and availability status identifier. A correlation index is established between standardized records, constraint records and resource records. The correlation index includes at least the correlation index between sampling location identifier and constraint range, the correlation index between sampling time identifier and entry window constraint, and the correlation index between resource category identifier and resource record, thereby forming an electromagnetic test data base.

[0026] While forming the electromagnetic test data base, a quality label is generated for each standardized record. The quality label includes a device trust status identifier, a stability identifier, a consistency identifier, an availability level identifier, and a trust level identifier, and is associated with the corresponding standardized record. The device trust status identifier is generated by reading the device status information and calibration status information associated with the device identifier. The device status information includes self-test results, fault status, and operating mode. The calibration status information includes calibration validity period and calibration conclusion. When the device status information meets the requirements for normal operation and the calibration status information is within the validity period and passes the calibration conclusion, the device trust status identifier is recorded as trustworthy; otherwise, it is recorded as untrustworthy.

[0027] The stability identifier is generated by performing integrity and continuity checks on the sampled sequence. The integrity check is used to determine whether the sampled sequence covers the preset sampling duration and has no missing segments, and the continuity check is used to determine whether the time interval of the sampled sequence meets the preset sampling interval constraint. After the checks are passed, stability features are extracted. The stability features are obtained by combining the fluctuation amplitude, mutation number and abnormal event identifier of the sampled sequence, and written into the stability identifier according to the preset grading rules.

[0028] The consistency identifier is generated by performing consistency verification on repeated sampling records at the same location or in the same partition. Repeated sampling records are determined by matching the same sampling location identifier and satisfying the conditions of the same frequency band identifier and the same measurement aperture. The consistency verification is used to compare whether the measurement results of repeated sampling records fall within the preset consistency range. If they fall within the preset consistency range, the consistency identifier is written as a consistent state; otherwise, it is written as an inconsistent state.

[0029] Availability level identifiers and trustworthiness level identifiers are generated based on device trustworthiness status identifiers, stability identifiers, and consistency identifiers according to a hierarchical rule table. The hierarchical rule table takes a combination of the three types of identifiers as input and outputs the corresponding level results. The availability level identifier is used to determine whether the record participates in subsequent spatiotemporal aggregation and partition clustering, while the trustworthiness level identifier is used to perform trustworthiness stratification of the record in subsequent analysis.

[0030] The hierarchical rule table is maintained by the task configuration and includes at least the mapping relationship between the combination of three types of identifiers—equipment trust status identifier, stability identifier, and consistency identifier—and availability level identifier and trustworthiness level identifier. The hierarchical rule table is formulated based on field instrument calibration specifications, sampling process integrity requirements, and same-point retest consistency requirements, and remains unchanged under the same execution plan version identifier. When generating quality labels, the record is first judged based on the equipment trust status identifier to determine whether it meets the basic trust conditions, then based on the stability identifier to determine whether the sampling process meets the stable sampling conditions, and finally based on the consistency identifier to determine the comparability of the record with historical records of the same caliber. The three judgment results jointly trigger the corresponding mapping in the hierarchical rule table, output the availability level identifier and trustworthiness level identifier, and write them into the quality label, so that the subsequent spatiotemporal aggregation and partition clustering screening criteria are traceable and reproducible.

[0031] The aforementioned quality label generation process enables the records in the electromagnetic test data base to have a traceable and reliable state description, providing an input basis for subsequent steps to perform spatiotemporal aggregation, partition clustering, and candidate point generation based on the electromagnetic test data base and quality labels.

[0032] S2. Based on the electromagnetic test data base and quality labels, perform spatiotemporal aggregation and partition clustering to generate risk partitions and coverage gaps. The specific implementation is as follows: When performing spatiotemporal aggregation based on electromagnetic test data base and quality label, spatial units and time windows are first determined as aggregation dimensions. Spatial units are used to spatially discretize the area to be tested. Their values ​​can be grid units, road segment units, or spatial units divided by the neighborhood of sensitive objects. Spatial units are recorded in the electromagnetic test data base with spatial unit identifiers. The generation rules for spatial unit identifiers and time window identifiers are as follows: Spatial unit identifiers are obtained by discretizing the test area under a unified spatial coordinate system. The discretization is based on the restricted area constraints and access path constraints in the spatial constraint data to determine the reachable range boundary. Then, the reachable range is divided into several spatial units according to the preset division rules and spatial unit identifiers are generated to ensure that the spatial unit boundary is consistent with the field accessibility. Time window identifiers are generated by the operation period division rules specified in the task configuration. The operation period division rules include at least the allowed entry period and the prohibited entry period. For the allowed entry period, continuous time window identifiers are generated according to the preset window segmentation rules to ensure that the coverage gap and the time period scheduling are closed under the same time caliber.

[0033] The time window is used to segment and express a continuous time period. Its value can be a fixed duration window or a window divided according to the work period. The time window is recorded in the electromagnetic test data base with the time window identifier. Subsequently, the standardized records in the electromagnetic test data base are merged according to the spatial unit identifier and the time window identifier to generate aggregate records and write aggregate identifiers. The aggregate identifier is used to uniquely identify the combination of spatial unit and time window corresponding to an aggregate record. During the merging, the quality labels are constrained and screened. The screening rules must include at least the availability level identifier meeting the conditions for participating in the aggregation, and the equipment trust status identifier meeting the trust status, so as to ensure that the data entering the aggregate record and its quality label are traceable under the same caliber.

[0034] The minimum set of fields for the aggregated record includes the aggregate identifier, spatial unit identifier, time window identifier, set of sampling location identifiers participating in the aggregation, set of sampling time identifiers participating in the aggregation, set of frequency band identifiers participating in the aggregation, quality label filtering information for the aggregation, and a summary view of measurement result fields used for feature extraction. The summary view of measurement result fields consists of a sequence of field fields of field strength measurement values ​​and a sequence of field fields of spectral scan results within the aggregation range, maintaining a correspondence with their respective sampling location identifiers and sampling time identifiers to support subsequent repeated calculations of field strength variation features and spectral occupancy features.

[0035] Based on the aggregated records, field strength variation features, spectrum occupancy features, and interference event density features are extracted to form a partitioned feature vector, specifically including: Field strength variation characteristics are used to characterize the changes in measurement results over time under the same aggregate identifier. They are obtained by statistically analyzing the measurement result sequences sorted by sampling time identifiers within the aggregate record, extracting three types of statistics: fluctuation level, degree of abrupt change, and trend direction. The fluctuation level is obtained from the dispersion of the measurement result sequence; the degree of abrupt change is obtained from the amplitude and frequency of changes in measurement results corresponding to adjacent sampling time identifiers; and the trend direction is obtained from the overall upward or downward tendency of the measurement result sequence within the time window. Spectrum occupancy characteristics are used to characterize the occupancy status corresponding to frequency band identifiers under the same aggregate identifier. They are obtained by merging the spectrum scan results within the aggregate record, extracting the set of occupied frequency bands, occupancy persistence, and occupancy stability. The three types of statistics include the occupied frequency band set, which is obtained by merging frequency band segments determined to have signals during spectrum scanning. The determination method for having signals is as follows: Within the same scan frequency band, the noise baseline aperture is first extracted from the spectrum scan result field. The noise baseline aperture is generated from historical scan records of idle periods specified in the task configuration or from the set of low occupied frequency points within the same time window. Then, the measured value sequence corresponding to the frequency point sequence of each scan is compared with the noise baseline aperture. When the measured value of a certain frequency point meets the detection threshold rule configured in the task, the frequency point is marked as a frequency point with a signal. Adjacent frequency points with signals are merged into frequency band segments according to the frequency continuity rule, and the frequency band segments are merged into an occupied frequency band set according to the overlap merging rule. The aforementioned detection threshold rules, frequency continuity rules, and overlap merging rules are all fixed by the task configuration, thereby ensuring that the set of occupied frequency bands can be reproduced under the same configuration; the occupancy persistence is obtained by the occurrence span of the frequency band segment within the time window, and the occupancy stability is obtained by the consistency of the frequency band segment in repeated scans; the interference event density feature is used to characterize the occurrence intensity of interference events under the same aggregation identifier. It is obtained by counting the number of event records that match the spatial unit identifier and time window identifier in the interference event data, and grouping and counting the event records in combination with the event type field to obtain the event density expression.

[0036] The three types of features mentioned above are written into the partition feature vector in a unified order and associated with the aggregation identifier. The partition feature vector serves as the input for subsequent partition clustering.

[0037] Partition clustering is performed based on partition feature vectors, outputting risk partitions and writing partition identifiers and trigger basis summaries to these risk partitions. Partition clustering includes four sub-processes: determining the clustering input set, determining the clustering similarity caliber, performing clustering partitioning, and generating clustering results. The clustering input set consists of partition feature vectors corresponding to all cluster identifiers. The clustering similarity caliber measures the closeness between feature vectors of different partitions, and is implemented using a segmented consistency determination method. For any two partition feature vectors, first compare whether the field strength change features simultaneously satisfy the fluctuation level consistency rule, the abrupt change degree consistency rule, and the trend direction consistency rule; then compare whether the spectrum occupancy features simultaneously satisfy the occupied frequency band set consistency rule, the occupancy persistence consistency rule, and the occupancy stability consistency rule; finally, compare whether the interference event density features satisfy the event density consistency rule. Only when the above three types of features satisfy their respective consistency rules are the two partition feature vectors determined to satisfy the similarity criteria. Each consistency rule is given by the task configuration, and the corresponding inputs can be repeatedly calculated from the aggregated records. An agglomerative clustering strategy is adopted when performing clustering partitioning. Each cluster identifier's corresponding partition feature vector is initialized as an initial cluster. In each round of merging, all cluster pairs are traversed. If the representative vectors of two clusters satisfy the similarity criteria, the two clusters are merged into a new cluster. The representative vector is selected by counting the number of times each partition feature vector in the cluster satisfies the similarity criteria with other partition feature vectors in the cluster, and taking the partition feature vector with the highest number of satisfactions as the representative vector. Merging stops when no cluster pair satisfies the similarity criteria. Finally, each cluster is output as a risk partition and written with the partition identifier and trigger basis summary.

[0038] Furthermore, the field strength variation characteristics, spectrum occupancy characteristics, and interference event density characteristics are aligned and compared under the same caliber; when performing clustering, cluster identifiers that meet the aggregation conditions are assigned to the same cluster, and each cluster corresponds to a risk partition; The partition identifier is used to uniquely identify the risk partition, and the trigger basis summary is used to record the key basis for the formation of the risk partition. The trigger basis summary includes at least the dominant feature type, the dominant feature source aggregation identifier, and the quality label constraint information corresponding to the dominant feature. The dominant feature type is determined from the field strength change feature, spectrum occupancy feature, and interference event density feature. The dominant feature source aggregation identifier is used to indicate the set of aggregated records within the risk partition that meet the dominant judgment condition for the clustering results. The quality label constraint information is used to indicate the screening criteria of the availability level identifier and the device trust status identifier used when the risk partition is generated, thereby ensuring that the risk partition has traceability. The dominant judgment condition is that the constituent elements of the representative vector in the clustering input set meet the judgment condition given by the task configuration a certain number of times within the risk partition.

[0039] Based on the spatiotemporal aggregation results and preset coverage rules, the coverage of spatial units and time windows is calculated, and a list of coverage gaps is generated and the association between coverage gap identifiers and corresponding risk zone identifiers is established.

[0040] The preset coverage rules are used to specify the coverage conditions that a combination of a spatial unit identifier and a time window identifier should meet in the current task. The coverage conditions include at least the sampling record existence condition and the sampling record validity condition. The sampling record existence condition is used to limit the number of sampling records that should be included in the aggregated record. The sampling record validity condition is used to limit the sampling records that enter the coverage calculation to meet the availability level identifier participation condition and the device trust status identifier to be trust status. Coverage is used to represent the degree to which the combination of spatial unit identifier and time window identifier meets the preset coverage rules. The calculation process involves verifying each aggregated record item by item: first verifying the existence of the sampled records, then verifying the validity of the sampled records, and summarizing the verification results into a coverage expression. When the coverage does not meet the preset coverage rules, a coverage gap identifier is written for the combination of spatial unit identifier and time window identifier, and the coverage gap identifier is added to the coverage gap list. The specific steps for determining coverage are as follows: First, perform a sampling record existence condition verification on the aggregated records. The verification content is whether there is a standardized record in the aggregated records that matches the spatial unit identifier and time window identifier, and whether the sampling duration of the standardized record meets the sampling duration requirements corresponding to the measurement parameter set. Then, perform a sampling record validity condition verification on the standardized records that pass the existence verification. The verification content is whether the availability level identifier of the record meets the conditions for participating in the aggregation and the device trust status identifier is trustworthy. When both the existence condition verification and the validity condition verification pass, the coverage is determined to meet the preset coverage rules; otherwise, the coverage is determined not to meet the preset coverage rules and a coverage gap identifier is written.

[0041] At the same time, the coverage gap identifier is associated with the corresponding risk zone identifier. The association method is to use the spatial unit identifier and the time window identifier as bridging keys to map the entries in the coverage gap list to the zone identifiers containing the same aggregation identifier within the risk zone. This provides a consistent input caliber and traceability for generating candidate sites based on risk zones and coverage gaps.

[0042] S3. Generate candidate deployment points based on risk zoning and coverage gaps, and encapsulate these candidate deployment points into a deployment work order set. Each deployment work order includes a set of measurement parameters, a set of constraint parameters, a set of alternative points, and priority rules. The specific implementation is as follows: When generating candidate points based on risk zones and coverage gaps, the process first reads the zone identifier of the risk zone and the coverage gap identifier from the coverage gap list. Spatial unit identifiers are used as spatial range constraints for candidate point generation, and time window identifiers are used as temporal range constraints for candidate point association. Adjacency analysis is then performed on the spatial units corresponding to the risk zone. This adjacency analysis determines whether spatial units share boundaries or have direct connectivity, thereby determining the zone boundary band. The zone boundary band refers to the set of spatial units adjacent to other risk zones or non-risk areas. Within the partition boundary zone, candidate points for the boundary zone are generated. The method for generating candidate points for the boundary zone is to select at least one representative location point in each boundary space unit as the candidate point location, and write the risk partition identifier and the candidate source identifier as the boundary zone source for the candidate point. Spatial differences between field strength variation features and spectrum occupancy features are extracted from the spatial units corresponding to the coverage gap to determine the range of the change zone. The spatial difference is determined by comparing the differences in the partition feature vectors of the spatial unit with the coverage gap and its neighboring spatial units, and the set of adjacent boundaries that meet the difference determination conditions is defined as the range of the change zone. Candidate points of the change zone are generated within the range of the change zone, and the coverage gap identifier and the candidate source identifier are written into the candidate points of the change zone as the source of the change zone. The difference determination conditions are given by the consistency rules in the task configuration and are obtained by comparing the partition feature vectors item by item. The locations of sensitive objects and the boundaries of restricted areas are extracted from the spatial constraint data. The locations of sensitive objects are used to represent the location points or spatial ranges of objects or areas that need to be covered, and the boundaries of restricted areas are used to represent the boundaries of areas that cannot be entered in the field. Constraint neighborhood candidate points are generated around the locations of sensitive objects and the boundaries of restricted areas. The constraint neighborhood candidate points are generated by selecting reachable location points within the neighborhood of the location of the sensitive object according to the travel path constraint, and simultaneously selecting reachable location points outside the boundary of the restricted area according to the safety distance constraint. The candidate source identifier is written as the constraint neighborhood source for the constraint neighborhood candidate points.

[0043] Deduplication is performed on candidate points in the boundary zone, change zone, and constraint neighborhood, and filtering is conducted based on accessibility constraints. Deduplication removes candidate points with the same sampling location identifier or whose spatial distance is less than the preset merging distance. The preset merging distance is determined by the task configuration, based on the spatial resolution requirements of field measurements and the minimum reachable unit scale of the travel path constraints. The spatial distance is calculated using a plane distance caliber under a unified spatial coordinate system. If a candidate point comes from a numbered location mapping, the unified spatial coordinates are obtained first from the location mapping table before spatial distance determination. By fixing the preset merging distance, the candidate point set can be stably generated under the same risk zone identifier and coverage gap identifier input, avoiding inconsistencies in the number of candidate points due to different personnel operations. Accessibility constraint filtering removes candidate points that fall within the restricted area constraint range or do not meet the travel path constraints, ultimately forming a candidate point set. Each candidate point in the candidate point set is written with a risk zone identifier, coverage gap identifier, and candidate source identifier to ensure that the candidate point set can be traced back to the risk zone and coverage gap.

[0044] When encapsulating candidate points into a point deployment work order set, a work order identifier is generated for each candidate point in the set, and point parameters are written into it. Point parameters at least include a sampling location identifier, a spatial unit identifier, and a candidate source identifier. Subsequently, a measurement parameter set is determined based on the risk zone identifier and coverage gap identifier. This measurement parameter set specifies the measurement configuration for the work order during field execution. The measurement parameter set at least includes a measurement time window, scanning frequency band range, sampling duration, and sampling mode identifier. The measurement time window is taken from the time window identifier associated with the candidate point or determined by the time window identifier corresponding to the coverage gap. The scanning frequency band range is determined by the statistical results of the frequency band identifier corresponding to the risk zone or the frequency band identifier associated with the coverage gap. The sampling duration is determined by the sampling requirements of the preset coverage rules for the spatial unit. The sampling method is identified by the type of identification, such as fixed-point measurement, mobile measurement, or multiple re-measurement. Then, based on the spatial constraint data and the operation resource data, the constraint parameter set is determined. The constraint parameter set is used to specify the constraints under which the work order can be executed. The constraint parameter set includes at least the entry window identifier, the restricted area constraint identifier, the passage path constraint identifier, the instrument capability constraint identifier, and the personnel qualification constraint identifier. Among them, the entry window identifier is determined by the constraint range identifier of the entry window constraint, the restricted area constraint identifier is determined by the constraint range identifier of the restricted area constraint, the passage path constraint identifier is determined by the constraint range identifier of the passage path constraint, the instrument capability constraint identifier is determined by the instrument capability requirements corresponding to the scanning frequency band range and the sampling mode identifier, and the personnel qualification constraint identifier is determined by the qualification requirements for specific scenarios or specific equipment in the operation rules.

[0045] When generating alternative point sets, candidate points within the same risk zone identifier are used as the candidate set. Alternative relationship identifiers are generated by combining the adjacency relationships of points filtered by accessibility constraints. These identifiers indicate alternative candidate points that can be switched to when the primary candidate point is unreachable or there is an entry window conflict. Alternative activation condition identifiers describe the conditions that trigger alternatives, including at least restricted area constraint triggering, travel path constraint conflict triggering, and entry window identifier conflict triggering. Priority rules are generated based on risk zone identifiers, coverage gap identifiers, and measurement parameter sets. These priority rules determine the ranking of the point deployment work order set in the coordinated scheduling solution. The generation method assigns higher priority to spatial unit identifiers and time window identifiers associated with coverage gap identifiers, higher priority to spatial units in the risk zone with more significant triggering criteria summary indications, and higher priority to candidate points within the neighborhood of sensitive object locations. Higher priority means that the point is ranked earlier in the priority rule comparison order. Priority rules are written into the site selection work order in the form of a comparable rule sequence. Finally, the measurement parameter set, constraint parameter set, alternative point set, and priority rules are written into the site selection work order bound to the work order identifier to form a site selection work order set. The site selection work order set is then associated with the electromagnetic test data base through sampling location identifiers, risk zone identifiers, and coverage gap identifiers, providing executable input for the coordinated scheduling solution to generate the execution plan.

[0046] S4. Based on the set of work orders and combined with spatial constraint data and job resource data, perform a coordinated scheduling solution to generate an execution plan. The execution plan includes work order selection results, resource allocation results, route sequence, and time slot scheduling. The specific implementation is as follows: When performing coordinated scheduling based on the work order set and combined with spatial constraint data and operation resource data, the work order identifier, point parameters, measurement parameter set, constraint parameter set, alternative point set and priority rules in the work order set are read first. At the same time, the entry window identifier, restricted area constraint identifier and passage path constraint identifier in the spatial constraint data are read, as well as the resource category identifier and availability status identifier in the operation resource data are read. For each deployment work order, a work order feasibility determination is performed. The work order feasibility determination is used to determine whether the work order has the conditions for execution within the current scheduling cycle. The determination process includes determining whether the point parameters fall within the restricted area constraint range based on the restricted area constraint identifier, determining whether the measurement time window falls within the allowed entry time period based on the entry window identifier, and determining whether the point parameters have a reachable path based on the access path constraint identifier. When any determination is not met, the work order is marked as an infeasible work order and removed from the solution input set, or the alternative activation condition identifier in the alternative point set is triggered and the corresponding alternative point is switched to form an alternative work order. After the work order feasibility determination, a set of feasible work orders participating in the linkage scheduling solution is formed, and each feasible work order is still associated with its risk partition identifier and coverage gap identifier to ensure that the risk and coverage semantics of the deployment are not lost during the scheduling solution process.

[0047] Establish a matching table between work orders and resources. The matching table is used to align the execution requirements of the deployment work orders with the capability status of the work resources. The matching table uses the work order identifier as an index to read the instrument capability constraint identifier and personnel qualification constraint identifier of the work order, and uses the resource record as an index to read the resource category identifier and availability status identifier. When the resource category is identified as an instrument resource and its capability range covers the scanning frequency band range and meets the sampling mode identification requirements, the instrument resource record is written into the optional resource set corresponding to the work order identifier. When a resource is identified as a personnel resource and its qualifications meet the personnel qualification constraint requirements and are available, the personnel resource record is written to the optional resource set corresponding to the work order identifier. Subsequently, the feasible work order set is sorted according to priority rules. The comparison order of the priority rules is used to determine the order among work orders associated with coverage gap identifiers, work orders associated with risk zone identifiers, and work orders in the neighborhood of sensitive objects. Within the same type of work order, they are sorted according to the urgency of the measurement time window. Based on the sorting results and the matching relationship table, resource allocation is performed. Resource allocation includes selecting a set of personnel resources and a set of instrument resources for each work order. After selection, the available status identifier of the corresponding resource record is updated to occupied status to avoid the same resource being repeatedly allocated in the same time period, thereby generating the resource allocation result.

[0048] When generating route order and time period scheduling, the traversability relationship between points is first constructed based on the traversable path constraint identifier, forming a traversable graph structure. In the traversable graph structure, the nodes correspond to the sampling position identifiers in the point parameters, and the edges correspond to the reachable connections that satisfy the traversable path constraint identifiers. For each set of work orders with completed resource allocation, a time period candidate sequence is generated based on the measurement time period window. The time period candidate sequence is used to give the execution time arrangement of each work order. The time period candidate sequence is filtered according to the entry window identifier. After filtering, the candidate time periods that fall within the allowed entry time period are retained. Then, a route candidate sequence is generated based on the traversable graph structure. The route candidate sequence is used to give the point access order for work order execution. The arrival time is generated as follows: taking the starting point of the route candidate sequence as the starting time, the corresponding travel path is retrieved in the travelable map structure according to the adjacent sampling position identifiers of the route candidate sequence, and the path length and travel time restriction are obtained according to the constraint range field of the travel path constraint. Then, the travel time between adjacent points is obtained by combining the travel time estimation rules given in the task configuration. After obtaining the travel time, the travel time and sampling time are gradually accumulated in the order of the route to form the arrival time, planned start time and planned end time of each work order identifier. The time result is used for entry window conflict detection and time period scheduling.

[0049] When generating route candidate sequences, work orders associated with coverage gap identifiers are prioritized and inserted into the routes, then arranged adjacently with the same risk zone identifier and adjacent spatial unit identifiers to reduce time wastage caused by cross-zone travel. Subsequently, conflict detection is performed on the route candidate sequences and time period candidate sequences. Conflict detection includes at least entry window conflict detection and resource occupancy conflict detection. Entry window conflict detection is used to determine whether the arrival time under the route sequence falls within the allowable range of the measurement time period window and entry window identifier. Resource occupancy conflict detection is used to determine whether the same personnel resource or the same instrument resource is assigned to multiple work orders in the same time period. When a conflict is detected, the alternative point is switched according to the alternative activation condition identifier in the alternative point set, or the work order is rearranged according to the priority rules and the route candidate sequence and time period candidate sequence are regenerated until the conflict detection is passed.

[0050] Once the conflict detection passes, an execution plan is output. The execution plan includes the work order selection results, resource allocation results, route sequence, and time slot schedule, and also includes a version identifier for subsequent updates and traceability. The work order selection results record the set of work order identifiers participating in this plan and their corresponding location parameters. The resource allocation results record the personnel and instrument resource records associated with each work order identifier. The route sequence records the execution order of each work order identifier. The time slot schedule records the start and end times of the plan for each work order identifier.

[0051] In a real-world example, there are multiple spatial units associated with coverage gaps around a city road, and some of these units fall within the nighttime access window constraint. In this case, during the work order feasibility assessment phase, work orders that cannot be accessed at night are marked as infeasible during the corresponding time period, and an alternative point set selection is triggered to select alternative candidate points accessible from outside the road. Simultaneously, if only one available instrument can cover a certain scanning frequency band, the matching table only provides that instrument as an optional resource set for work orders that require that frequency band. During resource allocation, the instrument will be prioritized for work orders associated with coverage gaps and with more urgent measurement time windows, while other work orders will be scheduled to subsequent time periods when the instrument is available. This forms a route sequence and time period schedule that satisfies the access window constraint and resource capability constraint.

[0052] S5. Perform sampling according to the execution plan and return the sampling summary and quality label. Perform consistency verification based on the returned data. When the triggering rules are met, generate a supplementary work order or a retest work order and update the execution plan and version record. The specific implementation is as follows: When performing sampling according to the execution plan and transmitting the sampling summary and quality label back, the field personnel carry the assigned instrument resources and arrive at the corresponding sampling location marker in the execution plan in the order of the route. They retrieve the deployment work order bound to the sampling location marker at the field terminal and write the work order marker. At the same time, they record the arrival time marker and departure time marker. The field personnel execute the sampling method corresponding to the sampling mode marker according to the measurement parameter set in the deployment work order, and complete the spectrum scanning or fixed-point measurement according to the scanning frequency band range. The sampling duration meets the sampling duration requirements.

[0053] After sampling is completed, a sampling summary is generated. The sampling summary must include at least a measurement result summary identifier, a frequency band coverage summary identifier, and an anomaly event identifier. The generation rules for the measurement result summary identifier are as follows: Extract the field strength measurement value field sequence that is consistent with the measurement parameter set from the sampling sequence, and use the representative value of the field strength measurement value field sequence as the core summary value of the measurement result summary identifier. The representative value adopts the median value of the sampling sequence within the sampling duration coverage range, and writes the median value and the measurement unit identifier into the measurement result summary identifier to ensure comparability between different device identifiers.

[0054] The generation rules for the frequency band coverage summary identifier are as follows: Read the scan frequency band range and spectrum scan result fields, verify whether the frequency point sequence within the scan frequency band range completely covers the frequency step entry diameter and scan segment boundary specified in the task configuration. If the coverage is complete, write the coverage completion identifier. If there are missing scan segments, write the missing scan segment identifier and record the frequency band segment corresponding to the missing segment to support subsequent consistency verification and supplementary work order generation. Among them, the measurement result summary identifier is used to characterize the summary expression of the key measurement results of this sampling, the frequency band coverage summary identifier is used to characterize the coverage status of the completed scan within the scan frequency band range, and the abnormal event identifier is used to characterize the type and time period of abnormal events that occur during the sampling process.

[0055] Synchronously collect equipment status information and on-site operating condition information to generate feedback quality tags. These tags must include at least an equipment status identifier, a sampling continuity identifier, a sampling stability identifier, and a location consistency identifier. The equipment status identifier is obtained from the instrument's self-test results and operating mode records. The sampling continuity identifier is determined by the presence of missing segments in the sampling sequence. The sampling stability identifier is determined by the fluctuation level and number of abrupt changes in the sampling sequence. The location consistency identifier is obtained by matching the sampling location identifier with the positioning record. The sampling summary and feedback quality tags are bound to the work order identifier to form feedback data, which is then uploaded to the electromagnetic test data base. The feedback data is associated with the original site deployment work order, risk zone identifier, and coverage gap identifier to ensure that subsequent consistency verification uses the same traceability standard.

[0056] When performing consistency verification based on the backhaul data, the backhaul data and its associated historical records are first read from the electromagnetic test data base, and a comparison set is established according to the risk partition identifier and the coverage gap identifier. The comparison set is used to organize the range of records to be compared. The risk partition identifier is used to limit the comparison records to come from the same risk partition, and the coverage gap identifier is used to limit the comparison records to come from the spatial unit identifier and time window identifier corresponding to the same coverage gap. Then, the comparison set is sequentially subjected to the consistency verification of adjacent points within the partition, the consistency verification of the same point retest, and the consistency verification of the trend within the time period, and a consistency judgment identifier is generated.

[0057] The consistency verification of adjacent points within a zone is used to compare whether the differences in the measurement result summary identifiers between backed data that are adjacent in spatial unit identifiers and have the same set of measurement parameters within the same risk zone fall within a preset consistency range; the consistency verification of retests at the same point is used to compare whether the differences in the measurement result summary identifiers between multiple backed data that have the same sampling location identifier and the same scanning frequency band range fall within a preset consistency range; the consistency verification of trends within a time period is used to compare whether the direction and magnitude of change of the measurement result summary identifier of the same spatial unit identifier within adjacent time windows are consistent with historical trends.

[0058] The preset consistency range is given by the task configuration. It is defined as the maximum range of difference that the measurement result summary identifier is allowed to appear between the comparison records when the device trust status identifier is a trust status and the sampling stability identifier meets the stability condition. This is to avoid misjudging normal fluctuations as abnormalities. The specific method for determining the preset consistency range is as follows: During the task startup phase, historical records with trusted device status identifiers and sampling stability identifiers meeting stability conditions are selected from the electromagnetic test data base. These records are grouped according to sampling location identifiers and scanning frequency band ranges. The natural fluctuation range of the measurement result summary identifiers of each group under normal operating conditions is statistically analyzed, and this natural fluctuation range is solidified as the preset consistency range in the task configuration. When a new area covered by the task lacks historical records, an initial preset consistency range is generated using baseline records of similar scenarios. After subsequent retest work orders are returned, the version of the preset consistency range is updated with the returned data that has passed consistency verification.

[0059] If the consistency judgment identifier meets the judgment conditions corresponding to the triggering rule, the work order identifier associated with the anomaly is located and the anomaly cause identifier is extracted. The anomaly cause identifier includes at least the equipment trusted status anomaly identifier, the time period mismatch anomaly identifier, and the spatial mutation anomaly identifier. The equipment trusted status anomaly identifier is used to indicate that the equipment status identifier or calibration status information in the returned quality label does not meet the trusted status. The time period mismatch anomaly identifier is used to indicate that the arrival time identifier or sampling time identifier does not fall within the allowable range of the measurement time period window or the entry window identifier. The spatial mutation anomaly identifier is used to indicate that there is a mutation difference exceeding the preset consistency range in the consistency verification of adjacent points within the partition or the trend consistency verification within the time period. The consistency judgment identifier and the anomaly cause identifier are written into the returned data and archived to the version record. The version record is used to record the input range, judgment result, and triggering basis of this consistency verification.

[0060] In one embodiment, when the triggering rule is met, a supplementary work order or a retest work order is generated and the execution plan and version record are updated. Specifically, the abnormal cause identifier and the coverage gap list are read, the supplementary type identifier or the retest type identifier is determined according to the triggering rule, and the corresponding work order is generated. The triggering rule is used to specify under what combination of consistency judgment identifier and abnormal cause identifier the supplementary work order or the retest work order is generated. The supplementary type identifier is used to characterize the type of work order that adds candidate points to eliminate coverage gaps or improve partition boundary resolution, and the retest type identifier is used to characterize the type of work order that repeats sampling of existing sampling location identifiers to confirm the results. The triggering rules are fixed using a combination of judgment methods: When the anomaly cause identifier includes an equipment trusted status anomaly identifier, a retest type identifier is triggered and a retest work order is generated; when the anomaly cause identifier includes a time period mismatch anomaly identifier, a retest type identifier is triggered and a retest work order is generated, and the measurement time period window of the retest work order must meet the entry window identifier; when the anomaly cause identifier includes a spatial mutation anomaly identifier and the coverage gap list shows that the corresponding spatial unit identifier still has a coverage gap identifier under the corresponding time window identifier, an supplement type identifier is triggered and a supplement work order is generated; when the spatial mutation anomaly identifier and the equipment trusted status anomaly identifier appear simultaneously, the retest type identifier is triggered first to generate a retest work order, and if the spatial mutation anomaly identifier still exists after the retest is passed, the supplement type identifier is triggered to generate a supplement work order.

[0061] The minimum set of fields in the version record includes the execution plan version identifier, trigger rule identifier, anomaly cause identifier, supplement type identifier or retest type identifier, execution plan difference record, set of affected work order identifiers, set of switched sampling location identifiers, and risk partition identifiers and coverage gap identifiers associated with the difference items. The execution plan difference record is used to identify at least new work order identifiers, replacement work order identifiers, route sequence changes, and time period scheduling changes, so that the execution plan version link record can be traced back to the trigger cause and specific changes from any version.

[0062] When the anomaly cause identifier includes a time period mismatch anomaly identifier or a device trusted status anomaly identifier, a retest work order is generated according to the retest type identifier, and the retest work order is written with the same set of measurement parameters and constraint parameters as the original work order. At the same time, the measurement time period window can be adjusted to the available time period that meets the entry window identifier. When the anomaly cause identifier includes a spatial mutation anomaly identifier and the spatial unit identifier still has a coverage gap identifier, a supplementary work order is generated according to the supplementary type identifier, and new point parameters are generated around the original sampling location identifier based on the alternative point set and the travel path constraint identifier.

[0063] Supplementary or retest work orders are written into the measurement parameter set, constraint parameter set, alternative point set, and priority rules, and then merged into the deployment work order set. Based on the merged deployment work order set, the linkage scheduling solution is re-executed to generate an updated execution plan, which is then written to the execution plan version identifier. The trigger rule identifier, exception reason identifier, supplementary or retest type identifier, execution plan version identifier, and execution plan difference record are written to the version record. The execution plan difference record is used to identify specific differences in new work order identifiers, replacement work order identifiers, changed route order, and changed time period scheduling.

[0064] In a real-world example, a sampling work order near the boundary of an industrial park was scheduled to be completed within a specific measurement window in the execution plan. However, when the field team arrived, temporary road closures caused the access route constraint markers to be unmet. The field team had to take a detour, and their arrival time markers fell outside the planned time period. The returned data showed inconsistent results with historical trends during the trend consistency verification within the time period. At the same time, the returned quality labels recorded abnormal location consistency markers or time period mismatch markers. Based on this, a retest work order was generated, adjusting the measurement time window to a time period where the access window markers allowed and the road was reopened. Points accessible outside the park were selected as alternatives from the set of alternative points. In another example, near a residential area, the consistency verification of adjacent points within a zone revealed an abnormal spatial change marker, and the coverage gap list showed that the spatial unit marker still lacked valid sampling records in the adjacent time window. A supplementary work order was generated, adding new candidate points between the original point and adjacent points. The supplementary work order was then prioritized according to the priority rules and the execution plan was regenerated. Thus, through supplementation and retesting, a closed-loop processing of coverage gaps and abnormal changes was achieved.

[0065] This invention constructs a unified time reference and unified spatial coordinate data base for electromagnetic monitoring data, interference event data, spatial constraint data, and operational resource data, and generates a traceable quality label for each record. This achieves consistency in the caliber of multi-source measurement results, reliable hierarchical structure, and reproducible analysis, avoiding mis-aggregation and misjudgment caused by equipment status, sampling omissions, and caliber differences.

[0066] Building upon this foundation, spatiotemporal aggregation and partitioned clustering are performed based on electromagnetic test data and quality labels to form risk partitions and coverage gaps. This transforms site selection decisions from experience-based selection to interpretable generation oriented towards risk and coverage semantics. Furthermore, candidate sites are encapsulated into a set of site selection work orders containing measurement parameter sets, constraint parameter sets, alternative point sets, and priority rules. Combined with spatial constraints and work resource execution linkage scheduling, an integrated execution plan with work order selection, resource allocation, route sequence, and time slot scheduling is output, improving field organization efficiency and constraint satisfaction capabilities. Simultaneously, a closed-loop mechanism of sampling feedback, quality label feedback, and consistency verification is introduced. When trigger rules are met, supplementary work orders or retest work orders are automatically generated, and the execution plan and version records are updated. This allows test site selection to continuously iterate with environmental changes and on-site disturbances, reducing the risk of invalid fieldwork and rework, and improving the coverage integrity, result consistency, and task execution stability of electromagnetic environment testing.

[0067] The above embodiments can be implemented, in whole or in part, by software, hardware, firmware, or any other combination thereof. When implemented using software, the above embodiments can be implemented, in whole or in part, in the form of a computer program product.

[0068] Those skilled in the art will recognize that the modules and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0069] In addition, the functional modules in the various embodiments of this application can be integrated into one processing module, or each module can exist physically separately, or two or more modules can be integrated into one module.

[0070] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

[0071] In conclusion, the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for optimizing electromagnetic environment testing site layout based on big data analysis, characterized in that: The specific steps include: S1, collecting electromagnetic monitoring data, interference event data, spatial constraint data and operation resource data, unifying time base and spatial coordinates, generating electromagnetic test data base, and generating quality label for data; S2, based on the electromagnetic test data base and the quality label, performing time and space aggregation and partition clustering, generating risk partition and coverage gap; S3, based on the risk partition and the coverage gap, generating candidate layout, encapsulating the candidate layout into a layout work order set, the layout work order including a set of measurement parameters, a set of constraint parameters, a set of alternative points and priority rules; S4, based on the layout work order set and combined with the spatial constraint data and the operation resource data, performing linkage scheduling solution, generating execution plan, the execution plan including work order selection result, resource allocation result, route sequence and time period scheduling; S5, executing sampling according to the execution plan and returning sampling summary and quality label, performing consistency check based on the returned data, generating supplementary work order or retest work order when the trigger rule is met, and updating the execution plan and version record. 2.The electromagnetic environment test site optimization method based on big data analysis of claim 1, wherein: The specific steps of S1 include: Analyzing the original fields of electromagnetic monitoring data, interference event data, spatial constraint data and operation resource data, establishing a unified data dictionary and field mapping rule; Performing format conversion, missing item filling and caliber normalization on the original fields according to the field mapping rule to form standardized records; According to the unified time base, write the sampling time identifier to the standardized records, according to the unified spatial coordinates, write the sampling position identifier to the standardized records, and write the frequency band identifier and data source identifier to the standardized records; Convert the spatial constraint data into constraint records and write the constraint type identifier and constraint range identifier, convert the operation resource data into resource records and write the resource category identifier and available state identifier; Store the standardized records, constraint records and resource records and establish an index relationship to generate the electromagnetic test data base. 3.The electromagnetic environment test site optimization method based on big data analysis of claim 1, wherein: In the step S1, the specific steps of generating the quality label include: Read the device state information and calibration state information associated with the device identifier, and write the device trusted state identifier; Perform integrity verification and continuity verification on the sampling sequence, extract stability features and write stability identifier; Perform consistency verification on repeated sampling records at the same location or in the same partition, extract consistency features and write consistency identifier; Generate availability level identifier and trustworthiness level identifier based on device trusted state identifier, stability identifier and consistency identifier; Write the availability level identifier and the trustworthiness level identifier into the quality label, and associate the quality label with the corresponding record of the electromagnetic test data base.

4. The electromagnetic environment test site optimization method based on big data analysis of claim 1, wherein: In the step S2, the specific steps of generating risk partition and coverage gap include: Performing time and space aggregation on the electromagnetic test data base according to spatial unit and time window, generating aggregation records and writing aggregation identifier; Extracting field strength variation features, spectrum occupation features and interference event density features from the aggregation records to form partition feature vectors; Performing partition clustering based on the partition feature vectors, outputting risk partition and writing partition identifier and trigger basis summary for the risk partition; Calculate the coverage degree of the spatial unit and the time window based on the spatio-temporal aggregation result and the preset coverage rule, and write the coverage gap identifier when the coverage degree does not meet the preset coverage rule; Output the coverage gap list and associate the coverage gap identifier with the corresponding risk partition identifier.

5. The electromagnetic environment test site optimization method based on big data analysis of claim 1, wherein: In S3, the specific steps for generating candidate deployment points include: Read the risk partition and coverage gap, perform adjacency relationship analysis on the spatial unit of the risk partition, determine the partition boundary zone and generate boundary zone candidate points; Extract the spatial difference of field strength change characteristics and frequency spectrum occupation characteristics corresponding to the coverage gap spatial unit, locate the change zone and generate change zone candidate points; Extract the sensitive object location and no-entry area boundary from the spatial constraint data, and generate constraint neighborhood candidate points around the sensitive object location and no-entry area boundary; De-duplicate and reachability filter the boundary zone candidate points, change zone candidate points and constraint neighborhood candidate points to form candidate deployment points; Write risk partition identifier, coverage gap identifier and candidate source identifier for each candidate deployment point to generate candidate deployment point set.

6. The electromagnetic environment test site optimization method based on big data analysis according to claim 5, characterized in that: In S3, the specific steps for encapsulating candidate deployment points into deployment point work order set include: Generate a work order identifier for each candidate deployment point in the candidate deployment point set and write the point location parameters; Determine the measurement parameter set according to the risk partition identifier and the coverage gap identifier, which includes the measurement time period window, the scanning frequency range, the sampling time length and the sampling mode identifier; Determine the constraint parameter set according to the spatial constraint data and the job resource data, which includes the approach window identifier, the no-entry area constraint identifier, the traffic path constraint identifier, the instrument capability constraint identifier and the personnel qualification constraint identifier; Generate a replacement point set based on the point adjacency relationship and reachability constraints within the same risk partition, which includes the replacement relationship identifier and the replacement enable condition identifier; Generate a priority rule based on the risk partition identifier, the coverage gap identifier and the measurement parameter set, and write the measurement parameter set, the constraint parameter set, the replacement point set and the priority rule into the deployment point work order to form the deployment point work order set.

7. The electromagnetic environment test site optimization method based on big data analysis of claim 1, wherein: In S4, the specific steps for generating an execution plan by linkage scheduling solution include: Read the deployment point work order set, spatial constraint data and job resource data, construct work order feasibility judgment rules and exclude work orders that do not meet the work order feasibility judgment rules; Establish a matching relationship table between the work order and the resource, which associates the instrument capability constraint identifier and the personnel qualification constraint identifier of the work order with the available state identifier of the resource record; Sort the deployment point work order set according to the priority rule, assign personnel resources and instrument resources to the deployment point work order based on the matching relationship table, and generate a resource allocation result; Generate route candidate sequence and time period candidate sequence based on the traffic path constraint identifier and the approach window identifier, perform conflict detection on the route candidate sequence and the time period candidate sequence, and switch to the replacement point or rearrange the work order; Output the execution plan after the conflict detection passes, which writes the work order selection result, the resource allocation result, the route sequence and the time period scheduling, and writes the execution plan version identifier. 8.The electromagnetic environment test site optimization method based on big data analysis of claim 1, wherein: In S5, the specific steps for returning the sampling summary and quality label include: The field terminal calls the corresponding deployment work order according to the execution plan, writes the work order identifier, and records the arrival time identifier and the departure time identifier; The sampling mode and the scanning frequency range corresponding to the measurement parameter set are executed, a sampling summary is generated, and the sampling summary includes a measurement result summary identifier, a frequency band coverage summary identifier, and an abnormal event identifier; Device state information and field working condition information are collected to generate a backhaul quality label, and the backhaul quality label includes a device state identifier, a sampling continuity identifier, a sampling stability identifier, and a location consistency identifier; The sampling summary and the backhaul quality label are bound with the work order identifier to form backhaul data and uploaded to the electromagnetic test data base.

9. The electromagnetic environment test site optimization method based on big data analysis according to claim 8, characterized in that: In S5, the specific steps of consistency verification include: Read the backhaul data and the historical records in the electromagnetic test data base, and establish a comparison set according to the risk partition identifier and the coverage gap identifier; Perform in-partition adjacent point consistency verification, same point retest consistency verification, and time period trend consistency verification on the comparison set to generate a consistency determination identifier; When the consistency determination identifier meets the determination condition corresponding to the trigger rule, locate the work order identifier associated with the abnormality and extract the abnormal reason identifier, which includes a device trust state abnormal identifier, a time period mismatch abnormal identifier, and a spatial mutation abnormal identifier; Write the consistency determination identifier and the abnormal reason identifier into the backhaul data and archive them to the version record.

10. The electromagnetic environment test site optimization method based on big data analysis of claim 9, wherein: In S5, the specific steps of generating an additional work order or a retest work order and updating the execution plan and the version record include: Read the abnormal reason identifier and the coverage gap list, determine the additional type identifier or the retest type identifier according to the trigger rule, and generate the corresponding additional work order or retest work order; Write the measurement parameter set, the constraint parameter set, the replacement point set, and the priority rule for the additional work order or the retest work order, and integrate the additional work order or the retest work order into the deployment work order set; Based on the integrated deployment work order set, re-execute the linkage scheduling solution to generate an updated execution plan and write the execution plan version identifier; Write the trigger rule identifier, the abnormal reason identifier, the additional type identifier or the retest type identifier, the execution plan version identifier, and the execution plan difference record into the version record; Output the deployment list, the execution plan version link record, and the version record.

Images