Smart testing method and system of an antenna based on its directional diagram

Abstract

The application discloses an intelligent testing method and system of an antenna based on an antenna directional diagram, and relates to the technical field of intelligent testing. An initial performance test diagram of the antenna is constructed according to characteristic forms of various performance characteristics and corresponding test time nodes. A plurality of abnormal test regions are determined based on the initial performance test diagram of the antenna. Test optimization measures are determined according to region positions, region forms and corresponding antenna test events of the plurality of abnormal test regions. Therefore, intelligent optimization events of the antenna are determined according to a plurality of test optimization items and the plurality of abnormal test regions, so as to determine optimized test regions. The initial performance test diagram of the antenna is updated according to the optimized test regions. An antenna directional diagram is determined based on the updated initial performance test diagram of the antenna, a previous test diagram of the antenna and a test scene of the antenna. Design optimization items of the antenna are determined according to recognition of the antenna directional diagram, and the accuracy of the design optimization items of the antenna is improved.

Description

Technical Field

[0001] This invention relates to the technical field of intelligent testing, and in particular to an intelligent testing method and system for antennas based on antenna patterns. Background Technology

[0002] With the development of technology, an antenna is a device used to radiate or receive electromagnetic waves in space. An antenna pattern is a graph showing how the relative field strength (normalized modulus) of the radiated field changes with direction at a certain distance from the antenna. It is usually used to describe the antenna's ability to transmit or receive electromagnetic waves in various directions in space. In the current technology, antenna testing requires manual testing and output of performance test graphs. However, these performance test graphs still contain many abnormal test areas, and the testing optimization is performed along a single dimension without fully considering the regional morphology of multiple abnormal test areas and the corresponding antenna test events. This affects the accuracy of the testing optimization measures, resulting in low accuracy of the antenna pattern and the antenna design optimization project. Summary of the Invention

[0003] The purpose of this invention is to overcome the shortcomings of the prior art. This invention provides an intelligent testing method and system for antennas based on antenna patterns.

[0004] This invention provides an intelligent testing method for antennas based on antenna radiation patterns, comprising: initializing various pulse parameters based on wave control commands and timing controllers from a host computer, and determining the rotation path of a turntable, wherein the antenna is mounted on the turntable; determining multiple performance combinations based on multiple performance parameters of the antenna, corresponding angles and shapes, and determining corresponding performance characteristics based on the identification of each performance combination; constructing an initial performance test map of the antenna based on the characteristic shapes of each performance characteristic and corresponding test time nodes, determining multiple abnormal test areas based on the initial performance test map of the antenna, and determining test optimization measures based on the regional locations, regional shapes and corresponding antenna test events of the multiple abnormal test areas; determining multiple test optimization items of the antenna based on the detection of test optimization measures, determining intelligent optimization events of the antenna based on the multiple test optimization items and multiple abnormal test areas, thereby determining the optimized test area; triggering the update of the initial performance test map of the antenna based on the optimized test area, determining the antenna radiation pattern based on the updated initial performance test map of the antenna, the antenna's previous test maps and the antenna's test scenario, and determining the design optimization items of the antenna based on the identification of the antenna radiation pattern.

[0005] This invention provides an intelligent testing system for antennas based on antenna patterns. The intelligent testing system for antennas based on antenna patterns is applied to the aforementioned intelligent testing method for antennas based on antenna patterns. The intelligent testing system for antennas based on antenna patterns includes:

[0006] The antenna rotation module is used to initialize various pulse parameters based on the wave control command and timing controller of the host computer, and to determine the rotation path of the turntable. At this time, the antenna is mounted on the turntable.

[0007] The performance characteristic module is used to determine multiple performance combinations based on multiple performance parameters of the antenna, corresponding angles and shapes, and to determine the corresponding performance characteristics based on the identification of each performance combination.

[0008] The test optimization measures module is used to construct the initial performance test map of the antenna based on the characteristic shape of each performance feature and the corresponding test time node, identify multiple abnormal test areas based on the initial performance test map of the antenna, and determine test optimization measures based on the location, shape and corresponding antenna test events of the multiple abnormal test areas.

[0009] The intelligent optimization event module is used to determine multiple test optimization items for the antenna based on the detection of test optimization measures, and to determine the intelligent optimization events for the antenna based on multiple test optimization items and multiple abnormal test areas, so as to determine the optimized test area.

[0010] The design optimization project module is used to trigger the update of the antenna's initial performance test map based on the optimized test area. Based on the updated antenna's initial performance test map, the antenna's previous test maps, and the antenna's test scenario, the antenna radiation pattern is determined, and the design optimization project for the antenna is determined based on the identification of the antenna radiation pattern.

[0011] Compared with the prior art, the beneficial effects of the present invention are:

[0012] In this embodiment of the invention, the method is used to initialize various pulse parameters based on the wave control command and timing controller of the host computer, and determine the rotation path of the turntable. At this time, the antenna is mounted on the turntable. Multiple performance combinations are determined based on multiple performance parameters of the antenna, corresponding angles and shapes, and corresponding performance characteristics are determined based on the identification of each performance combination. An initial performance test map of the antenna is constructed based on the characteristic shape of each performance characteristic and the corresponding test time node. Multiple abnormal test areas are determined based on the initial performance test map of the antenna. Test optimization measures are determined based on the regional location, regional shape and corresponding antenna test events of the multiple abnormal test areas. The introduction of the initial performance test map of the antenna is compatible with the overall consideration of the regional location, regional shape and corresponding antenna test events of multiple abnormal test areas, which improves the accuracy of test optimization measures and realizes the optimization of the antenna during the test process.

[0013] Therefore, multiple test optimization items for the antenna are determined based on the detection of test optimization measures. Intelligent optimization events for the antenna are then determined based on these multiple test optimization items and multiple abnormal test areas to identify the optimized test area. The optimized test area triggers the update of the antenna's initial performance test map. The antenna radiation pattern is determined based on the updated initial performance test map, previous test maps, and the antenna's test scenario. The antenna's design optimization items are then identified based on the antenna radiation pattern. This introduction of the optimized test area facilitates the update of the antenna's initial performance test map, achieving compatibility considerations between the updated initial performance test map, previous test maps, and the antenna's test scenario. This improves the accuracy of the antenna radiation pattern and the accuracy of the antenna's design optimization items. Attached Figure Description

[0014] Figure 1 This is a flowchart illustrating the intelligent testing method for antennas based on antenna patterns in an embodiment of the present invention.

[0015] Figure 2 This is a flowchart illustrating step S11 in the intelligent testing method for antennas based on antenna patterns according to an embodiment of the present invention.

[0016] Figure 3 This is a flowchart illustrating step S12 in the intelligent testing method for antennas based on antenna patterns according to an embodiment of the present invention.

[0017] Figure 4 This is a flowchart illustrating step S13 in the intelligent testing method for antennas based on antenna patterns according to an embodiment of the present invention.

[0018] Figure 5 This is a flowchart illustrating step S14 in the intelligent testing method for antennas based on antenna patterns in this embodiment of the invention.

[0019] Figure 6 This is a flowchart illustrating step S15 in the intelligent testing method for antennas based on antenna patterns in this embodiment of the invention.

[0020] Figure 7 This is a schematic diagram of the structural composition of an intelligent testing system for antennas based on antenna patterns, as described in an embodiment of the present invention. Detailed Implementation

[0021] The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention.

[0022] Please see Figures 1 to 7 A smart testing method for antennas based on antenna radiation patterns is proposed and applied to intelligent testing scenarios. The smart testing method for antennas based on antenna radiation patterns includes:

[0023] Step S11: Based on the wave control command and timing controller of the host computer, the initialization of each pulse parameter is triggered, and the rotation path of the turntable is determined. At this time, the antenna is mounted on the turntable.

[0024] Step S12: Determine multiple performance combinations based on the antenna's multiple performance parameters, corresponding angles and shapes, and determine the corresponding performance characteristics based on the identification of each performance combination;

[0025] Step S13: Construct an initial performance test map of the antenna based on the characteristic shape of each performance feature and the corresponding test time node. Based on the initial performance test map of the antenna, determine multiple abnormal test areas. Determine test optimization measures based on the location, shape and corresponding antenna test events of the multiple abnormal test areas.

[0026] Step S14: Based on the detection of test optimization measures, determine multiple test optimization items for the antenna, and determine the intelligent optimization events of the antenna according to the multiple test optimization items and multiple abnormal test areas, so as to determine the optimized test area;

[0027] Step S15: Trigger the update of the initial performance test map of the antenna based on the optimized test area. Determine the antenna radiation pattern based on the updated initial performance test map of the antenna, the previous test map of the antenna, and the test scenario of the antenna. Determine the design optimization items of the antenna based on the identification of the antenna radiation pattern.

[0028] refer to Figure 2 In step S11, the initialization of each pulse parameter is triggered based on the wave control command and timing controller of the host computer, and the rotation path of the turntable is determined. At this time, the antenna is mounted on the turntable.

[0029] In the specific implementation of this invention, the specific steps are as follows:

[0030] S111: Acquisition of antenna test requirements. The host computer generates a beam control command based on the test requirements. The timing controller receives the beam control command and triggers the initialization of each pulse parameter based on the beam control command. The initialization is confirmed to be complete by status readback. At this time, each pulse parameter includes pulse width, pulse repetition frequency and delay parameter.

[0031] S112: Mark multiple rotation position nodes of the turntable according to the initialization of each pulse parameter. Determine the rotation three-dimensional space of the antenna based on the multiple rotation position nodes of the turntable, the corresponding rotation attitude and the position of the antenna relative to the turntable. Determine the rotation path of the turntable based on the path planning of the rotation three-dimensional space of the antenna. At the same time, during the test, the host computer can dynamically optimize the rotation path of the turntable based on real-time data.

[0032] In the embodiments of this application, the acquisition of test requirements is the foundation for system startup, and it is necessary to clarify the antenna type, test objectives, and key parameter requirements. The host computer obtains user input through an interactive interface (such as a control panel developed with LabVIEW). The core acquisition items include: antenna type: such as microstrip antenna, phased array antenna, parabolic antenna, etc. Different antennas correspond to different test standards (e.g., microstrip antennas require attention to bandwidth, while phased array antennas require scanning beam pointing accuracy); test objectives: such as radiation pattern testing, VSWR (S11) testing, gain testing, etc. Different objectives directly affect subsequent parameter configuration (e.g., radiation pattern testing requires high angular resolution, and VSWR testing requires wide frequency band coverage); key parameters: including angular range (e.g., azimuth -90°~90°, elevation -30°~30°), frequency band range (e.g., 1.5~2.5GHz, 100MHz step), accuracy requirements (e.g., angular resolution 0.5°, data sampling rate 1kHz), and environmental conditions (e.g., temperature 25℃±2℃, humidity ≤60%).

[0033] The host computer translates the test requirements into machine-executable wave control instructions, which must adhere to a communication protocol (such as a custom binary protocol or Modbus TCP). The instruction structure must include a control word, a parameter word, and a checksum: Control word: Defines the operation type (such as "initialization", "angle scan", "frequency band switching") and device address (such as timing controller address 0x01, turntable address 0x02); Parameter word: Quantifies the test requirements into specific parameters, such as: Angle parameters: starting angle (0°), ending angle (60°), step interval (0.5°); Frequency band parameters: starting frequency (1.7GHz), ending frequency (1.9GHz), step interval (50MHz); Pulse parameters: pulse width (20ms), pulse repetition frequency (50Hz), delay parameter (5ms); Checksum: Uses CRC16 or cumulative checksum to ensure no data errors during instruction transmission (such as calculating the CRC value of the control word and parameter word and appending it to the end of the instruction).

[0034] After receiving the wave control command, the timing controller (such as an FPGA or STM32 microcontroller) parses and triggers the initialization of pulse parameters. The core parameters must match the hardware performance: Pulse width: controls the duration of a single test signal and must be greater than the vector network sampling time (e.g., if the vector network sampling requires 15ms, then the pulse width is set to 20ms to avoid data truncation); Pulse repetition frequency (PRF): determines the number of tests per unit time and must satisfy the Nyquist sampling theorem (e.g., if the antenna pattern requires 1° resolution and the turntable speed is 10° / s, then PRF≥10Hz, here it is set to 50Hz to leave margin); Delay parameter: compensates for signal transmission and mechanical response delay (e.g., if the turntable takes 3ms from receiving the command to starting and the vector network trigger response takes 2ms, then the total delay is set to 5ms to ensure that the angle and data are synchronized).

[0035] To ensure successful initialization, the timing controller needs to read back status information from the host computer. The readback mechanism must include: Status word: indicating the initialization result of each parameter (e.g., 0x00 indicates success, 0x01 indicates pulse width configuration failure, 0x02 indicates PRF exceeding limit, etc.); Actual parameter values: reading back the currently configured pulse width, PRF, and delay parameters for the host computer to compare (e.g., if the instruction requires a 20ms pulse width, and the readback value is 20ms, then it is confirmed to be correct); Timeout retry mechanism: if the host computer does not receive the readback within the specified time (e.g., 100ms), or the status word shows failure, it will automatically resend the instruction (up to 3 times) to avoid test interruption due to communication abnormalities.

[0036] Furthermore, the rotation position nodes are the key angle points that the turntable needs to stop or pass through during the test. Their marking needs to be combined with the pulse parameter initialization results and test requirements. Node generation basis: Based on the delay parameter (e.g., 5ms) and pulse repetition frequency (PRF, e.g., 50Hz) in the pulse parameters, calculate the time required for a single-point test (e.g., 1 / PRF = 20ms), and combine this with the angle range (e.g., -90 to 90) and step interval (e.g., 1) to determine the number of nodes (e.g., 181) and their positions (-90, -89, ..., 90). Node attribute marking: Each node needs to be associated with test attributes, including: Dwell time: determined based on pulse width and data acquisition rate (e.g., pulse width 20ms, acquisition rate 1kHz, then dwell time ≥ 20ms); Test mode: such as continuous scan (acquisition without dwell) or step scan (acquisition after dwell). Pattern testing usually uses step scan to ensure data accuracy; Node priority classification: mark key angles (e.g., 0 direction, main lobe region) with high priority, and prioritize acquisition or increase dwell time during testing; mark side lobe regions with low priority, and the number of sampling points can be appropriately reduced.

[0037] The rotating three-dimensional space refers to the three-dimensional spatial range covered by the antenna during the turntable's movement. It requires comprehensive modeling by combining the rotation position nodes, the turntable's attitude (e.g., the range of motion along the azimuth, elevation, and roll axes), and the antenna's installation position (e.g., offset and installation height). The range of motion (e.g., azimuth 0–360°, elevation -30°–90°) and rotation accuracy (e.g., 0.1) of each turntable axis are obtained. The turntable attitude (e.g., azimuth 30°, elevation 10°, roll 0°) corresponding to each rotation position node is then determined. Finally, calibration is performed using a laser tracker or mechanical calibration. Obtain the antenna offset relative to the turntable center (e.g., 5mm offset in the X direction, 3mm offset in the Y direction, and 100mm height in the Z direction), compensate for the offset in the spatial model, and ensure that the antenna phase center coincides with the turntable rotation center (error ≤ 0.5mm); use a three-dimensional coordinate system (e.g., the turntable center is the origin, X-axis is the azimuth direction, Y-axis is the elevation direction, and Z-axis is the height direction) to calculate the antenna phase center coordinates corresponding to each node, generate a spatial point cloud, and fit it into a continuous three-dimensional space through an interpolation algorithm (e.g., B-spline interpolation).

[0038] The rotation path is the trajectory of the turntable in three-dimensional space. Motion efficiency and testing accuracy need to be optimized based on the spatial model. The scanning path should be selected according to the testing requirements, such as: a serpentine path: azimuth angle scans line by line from -90 to 90 degrees, with pitch angle increments (suitable for planar radiation pattern testing); a spiral path: azimuth angle rotates continuously, with pitch angle increasing synchronously (suitable for spherical radiation pattern testing); an adaptive path: skips low-value areas (such as flat sidelobe areas) based on historical data; and a path that combines the turntable's maximum rotational speed (e.g., 60 / s) and acceleration (e.g., 30 / s²). 2 The system calculates the movement time between nodes (e.g., from 0 to 30, at a rotation speed of 60 / s, the movement time is 0.5s) to avoid mechanical vibration or data distortion caused by excessive speed. Based on the three-dimensional spatial model and the physical dimensions of the turntable (e.g., a turntable radius of 500mm), the system simulates the path movement process and detects the collision risk between the antenna and surrounding equipment (e.g., anechoic chamber absorbing materials, test brackets) (the safe distance must be ≥100mm).

[0039] During testing, the host computer dynamically adjusts the path based on real-time data (such as S11 parameters and signal strength) to improve testing efficiency. After collecting data from each node, the host computer calculates key indicators (e.g., S11 ≤ -10dB is acceptable, signal strength ≥ -70dBm is valid) and marks abnormal nodes (e.g., S11 > -10dB or sudden signal strength changes). When the proportion of abnormal nodes exceeds a threshold (e.g., 10%) or the data fluctuation in a certain area is too large (e.g., signal strength standard deviation > 5dBm), path optimization is triggered: encrypted sampling: in Temporary nodes are added around abnormal nodes (e.g., the original step 2 is adjusted to 1); skip invalid areas: for consecutive qualified nodes (e.g., S11 is ≤ -15dB), subsequent sampling is skipped and the next area is directly entered; adjust dwell time: the dwell time in high fluctuation areas is extended (e.g., from 10ms to 20ms); the host computer generates optimized path instructions and sends them to the turntable controller via Ethernet. The turntable automatically switches to the new path after the current node test is completed (e.g., the original plan was to move from 30 to 32, but after optimization, it is changed to 30→31→32).

[0040] refer to Figure 3 In step S12, multiple performance combinations are determined based on multiple performance parameters of the antenna, corresponding angles and shapes, and corresponding performance characteristics are determined based on the identification of each performance combination.

[0041] In the specific implementation of this invention, the specific steps are as follows:

[0042] S121: The antenna rotates under the drive of the turntable and is tested. Multiple performance parameters of the antenna are collected, and the corresponding angles are marked. At the same time, the shape of the antenna is collected. Multiple performance combinations are determined based on the cross-matching of multiple performance parameters of the antenna, corresponding angles and shapes.

[0043] S122: In each performance combination, priority distribution information is determined based on the identification of the performance combination, performance detection mode is determined according to the priority distribution information and the combination content of the performance combination, and corresponding performance features are determined according to the performance combination and the corresponding performance detection mode, so as to collect multiple performance features and mark the test time nodes of each performance feature.

[0044] In the embodiments of this application, the antenna rotates under the drive of the turntable and is tested, collecting multiple performance parameters of the antenna, marking the corresponding angles, and simultaneously collecting the shape of the antenna. Based on the cross-matching of multiple performance parameters of the antenna, the corresponding angles, and the shape, multiple performance combinations are determined, thereby realizing the cross-matching of multiple performance parameters of the antenna, the corresponding angles, and the shape, and improving the accuracy of multiple performance combinations.

[0045] At this time, the antenna rotates along a preset path (such as helical scanning or grid scanning) driven by the turntable, synchronously triggering test equipment such as vector network analyzers to collect performance parameters; the turntable receives instructions from the host computer and rotates at a step angle (such as 1) and angular velocity (such as 10 / s), stopping or continuously scanning at each angle node (such as azimuth 0~360, elevation -10~60), the dwell time being determined by the delay value in the pulse parameters (such as 20ms); the timing controller generates a trigger signal according to the pulse repetition frequency (PRF, such as 50Hz), synchronously starting the vector network analyzer to collect the antenna's S-parameters (S11, S21), gain, directivity and other performance parameters, and the data is transmitted to the host computer in real time via GPIB or Ethernet; the host computer adds a timestamp (such as 2025-08-29 10:00:01.200) and angle label (such as azimuth 30, elevation 5) to each set of performance parameters to ensure that the data strictly corresponds to the spatial position.

[0046] The collected performance parameters are associated with their corresponding angles to construct a performance-angle mapping relationship, providing a foundation for subsequent cross-matching. Abnormal data (such as invalid values ​​where S11 suddenly jumps to -5dB due to turntable vibration) are removed, and the curve is smoothed using a moving average filter (such as a 5-point average). Angle labels are bound to each group of valid performance parameters in the format of "azimuth_pitch" (e.g., 30_0) and stored in a database (such as SQLite or MySQL), supporting queries by angle range (e.g., azimuth 25~35). The host computer plots the performance-angle curve in real time (e.g., gain versus azimuth curve), and the data rationality is verified manually or by algorithm (e.g., peak detection).

[0047] The antenna's physical morphology is acquired through visual or sensor technologies to identify structural anomalies (such as vibrator deformation or loose feed point). Visual acquisition: An industrial camera (e.g., 5 megapixels) captures images of the antenna at a fixed frame rate (e.g., 30fps), and image processing algorithms (e.g., edge detection, template matching) are used to extract morphological parameters (e.g., vibrator offset, tilt angle). Sensor acquisition: Strain gauges or laser displacement sensors monitor the displacement or stress of key antenna components (e.g., feed point) in real time, and the data is converted into digital signals (e.g., displacement of 0.1mm) via an ADC. Morphological marking: Morphological data is associated with angles in the format of angle_morphological parameter (e.g., 30_0_Offset0.2mm) and stored in a database.

[0048] Specifically, when testing a 5G base station antenna (3.5GHz band), the turntable rotated from azimuth 0° to 360° in 1-step increments, with the elevation angle fixed at 0°. The timing controller triggered the vector network analysis (VNA) at 50Hz PRF, collecting a set of data every 20ms. When the turntable reached azimuth 30°, the host computer recorded: timestamp 10:00:01.200, azimuth 30°, elevation 0°, S11-15dB, gain 8dBi, directivity 10dBi. The S11-15dB at azimuth 30° was marked as 30_0_S11, and the gain 8dBi was marked as 30_0_Gain. The host computer plotted a gain curve showing that the gain fluctuation was 0.5dB in the azimuth 30-35° range, which meets the design requirements (fluctuation ≤1dB), and the data passed verification.

[0049] During the test, the camera photographed the antenna at an azimuth angle of 30°. Image processing showed that the vibrator offset was 0.2mm (threshold ≤ 0.5mm), marked as 30_0_Offset0.2mm; the laser displacement sensor detected a feed point displacement of 0.05mm, marked as 30_0_Displacement0.05mm.

[0050] Cross-matching of performance parameters, angles, and morphological data forms performance combinations to identify abnormal or critical areas. Performance parameters (e.g., S11, gain) are associated with morphological parameters (e.g., offset) using angle labels to form triples (angle, performance parameter, morphological parameter); for example: (30_0, S11 -15dB, Offset 0.2mm). Triples are then classified into performance combinations based on rules or machine learning models: rule example: if S11 > -10dB and Offset > 0.3mm, it is classified as an abnormal combination (e.g., PC001); if gain > 7dBi and morphology is normal, it is classified as a normal combination (e.g., PC002). An SVM classifier is trained, inputting historical data (e.g., 1000 triples) and outputting combination types (abnormal / normal / critical). A unique ID (e.g., PC001) is assigned to each performance combination, storing combination records containing information such as angle range, performance parameter thresholds, and morphological anomaly type.

[0051] Specifically, in the 5G antenna test, the triplet at 30° azimuth (30_0, S11-15dB, Offset0.2mm) was classified as a normal combination (PC002); the triplet at 120° azimuth (120_0, S11-8dB, Offset0.4mm) was classified as an abnormal combination (PC001) because S11 exceeded the limit and the offset was too large; finally, 10 abnormal combinations and 50 normal combinations were generated to provide a basis for subsequent priority division.

[0052] Furthermore, within each performance combination, priority distribution information is determined based on the identification of the performance combination. The performance detection mode is determined according to the priority distribution information and the combination content of the performance combination. The corresponding performance features are determined according to the performance combination and the corresponding performance detection mode, so as to collect multiple performance features and mark the test time nodes of each performance feature. Multiple performance features and corresponding test time nodes are introduced.

[0053] At this point, priority distribution information is used to distinguish the urgency of performance combinations for testing. It needs to be comprehensively judged based on the degree of anomaly in performance parameters, the severity of morphological deviations, and the test objective. Priority classification is based on: Anomaly degree: graded according to the magnitude of performance parameter deviation from the threshold (e.g., S11 < -10dB is low priority, -10dB ≤ S11 < -5dB is medium priority, S11 ≥ -5dB is high priority); Morphological deviation: offset calculated using morphological sensor data (e.g., installation offset > 0.3mm is medium priority, > 0.5mm is high priority); Test objective: if the test focuses on band edge performance (e.g., the edge frequency point 3.4GHz of the 5G band 3.5GHz), the priority of the corresponding angle combination is increased; Priority output: a priority label (e.g., high / medium / low) is assigned to each performance combination, and a priority distribution table is generated (e.g., high priority accounts for 10%, medium priority accounts for 30%).

[0054] Performance testing modes are test strategies selected based on priority and combination of content, requiring a balance between test accuracy and efficiency. Modes include: Detailed Mode: Suitable for high-priority combinations, employing high-density sampling (e.g., angle step 0.5) and multiple averaging (e.g., 16 scans) to ensure data accuracy; Fast Mode: Suitable for low-priority combinations, employing low-density sampling (e.g., angle step 5) and single scan to reduce test time; Adaptive Mode: Suitable for medium-priority combinations, dynamically adjusting sampling density based on real-time data (e.g., switching to detailed mode when S11 fluctuation > 1dB).

[0055] Specifically, when testing a satellite communication antenna, performance combination PC001 (azimuth 120°, S11 -8dB, offset 0.4mm) was judged as high priority because S11 was close to the threshold (-5dB) and the offset exceeded the limit; PC002 (azimuth 30°, S11 -15dB, offset 0.2mm) was judged as low priority because the parameters were normal and the shape deviation was small; finally, 5 high-priority combinations, 15 medium-priority combinations and 40 low-priority combinations were generated.

[0056] For the high-priority combination PC001 (azimuth 120°), the system selected detailed mode: scanning the 115°–125° range in 0.5 steps, and extracting the S11 parameter after averaging 16 scans; for the low-priority combination PC002 (azimuth 30°), the system used fast mode: scanning the 25°–35° range in 5 steps, and acquiring data in a single scan; the test time comparison showed that detailed mode took 2 minutes per combination, while fast mode took 10 seconds per combination.

[0057] Performance characteristics are key indicators extracted from performance combinations. They need to be collected specifically in conjunction with the detection mode, and time nodes need to be marked to support subsequent data analysis. The core steps include: Feature extraction: Detailed mode: Extract full-band S-parameters (such as S11 and S21 in 3.4-3.6GHz), 3dB beamwidth, sidelobe level and other features; Fast mode: Extract only the S11 and gain of the center frequency (such as 3.5GHz); Time node marking: Add a timestamp (such as 2025-08-29 10:05:30.000) to each feature to record its collection time for subsequent analysis of performance change trends during the testing process.

[0058] Specifically, in detailed mode, the features extracted by the PC001 combination include: 3.4GHz frequency point S11 = -8.2dB (timestamp 10:05:30); 3.5GHz frequency point S11 = -7.9dB (timestamp 10:05:31); 3dB beamwidth = 12 (timestamp 10:05:32); in fast mode, the PC002 combination only extracts: 3.5GHz frequency point S11 = -15.1dB (timestamp 10:06:40); all feature data are stored in the host computer database and associated with the performance combination ID (PC001 / PC002).

[0059] refer to Figure 4 In step S13, an initial performance test map of the antenna is constructed based on the characteristic shape of each performance feature and the corresponding test time node. Multiple abnormal test areas are determined based on the initial performance test map of the antenna. Test optimization measures are determined based on the location, shape and corresponding antenna test events of the multiple abnormal test areas.

[0060] In the specific implementation of this invention, the specific steps are as follows:

[0061] S131: Collect various performance characteristics and determine the characteristic shape and corresponding test time node of each performance characteristic based on the detection of each performance characteristic; at the same time, collect the antenna model, determine the first test pattern based on the characteristic shape of each performance characteristic and the antenna model, determine the second test pattern based on the test time node of each performance characteristic and the antenna model, and construct the initial performance test pattern of the antenna based on the first test pattern and the second test pattern.

[0062] S132: Based on the detection of the initial performance test map of the antenna, multiple sub-performance test areas are determined, and corresponding abnormal locations are determined based on the identification of each sub-performance test area. Multiple abnormal locations are collected, and multiple abnormal test areas are determined based on the multiple abnormal locations and the corresponding sub-performance test areas.

[0063] S133: Mark the location and shape of multiple abnormal test areas, determine the corresponding antenna test events based on the tracing of the location of multiple abnormal test areas, and determine test optimization measures based on the location, shape and corresponding antenna test events of multiple abnormal test areas.

[0064] In the embodiments of this application, various performance characteristics are collected, and the characteristic shape and corresponding test time node of each performance characteristic are determined based on the detection of each performance characteristic. At the same time, the antenna model is collected, and a first test pattern is determined based on the characteristic shape of each performance characteristic and the antenna model. A second test pattern is determined based on the test time node of each performance characteristic and the antenna model. An initial performance test pattern of the antenna is constructed based on the first test pattern and the second test pattern, thereby improving the accuracy of the initial performance test pattern of the antenna.

[0065] At this point, the antenna performance characteristics under various angles and configurations are obtained from S122, such as S11 parameters, gain, beamwidth, and sidelobe level. These characteristics are the basic data for subsequent analysis. Based on the variation patterns of the characteristics in the time or angle dimension, they are divided into different "configurations." Common configurations include: Stable type: small fluctuation range, such as gain changes within ±0.5dB; Fluctuating type: exhibiting periodic or non-periodic fluctuations, such as S11 fluctuating between ±1dB; Abrupt type: significant jumps, such as gain suddenly dropping by more than 3dB at a certain angle point; Test time node markers: each performance characteristic is associated with a test time node, for example: timestamp: 2025-08-29 07:15:22; turntable angle: azimuth 120°; rotation speed: 5° / s.

[0066] The host computer obtains the antenna model, such as SAT-5G-001, from the database or by manual input, and retrieves the reference performance parameters of that model (e.g., standard gain 15dBi, standard beamwidth 10°); First test graph (characteristic-performance graph): Horizontal axis: characteristic morphology (stable, fluctuating, abrupt change); Vertical axis: performance parameters (e.g., S11, gain); Function: To visually display the performance distribution under different morphologies, used to identify abnormal morphologies (e.g., abrupt changes concentrated in the low gain region); Second test graph (time-performance graph): Horizontal axis: test time node (e.g., 07:15:00~07:20:00); Vertical axis: performance parameters; Function: To display the performance trend over time, used to identify time-related anomalies (e.g., gain drops sharply in a specific time period).

[0067] Specifically, assuming the antenna collects the following characteristics at a 120° azimuth angle: S11 = -9.8dB; gain = 14.2dBi; beamwidth = 10.5°; test time node: 07:15:22; the system determines that the characteristic is "fluctuating" because S11 has periodic fluctuations in the 110° to 130° range, with a fluctuation amplitude of about 1dB.

[0068] For antenna model SAT-5G-001: The first test chart shows: stable characteristics: gain concentrated between 14.8 and 15.2 dBi; fluctuating characteristics: gain concentrated between 13.5 and 14.5 dBi; abrupt characteristics: gain concentrated between 12.0 and 13.0 dBi; The second test chart shows: 07:15:00 to 07:16:00: gain stabilized at 15 dBi; 07:16:30: gain suddenly dropped to 12.5 dBi; after 07:17:00: gain recovered to 14.8 dBi.

[0069] The initial performance test plot is a fusion of the first test plot (feature-performance) and the second test plot (time-performance), achieved through the following methods: associating feature morphology with time nodes, for example, "the abrupt feature appears at 07:16:30"; marking the location of outliers (such as gain drop points) in the angle-time-performance three-dimensional space; generating heatmaps or contour maps to visually display the performance distribution and outlier areas; key outputs: outlier area markers: such as the gain drop point at azimuth angle 120° and time 07:16:30; performance baseline: such as the standard gain of 15dBi for model SAT-5G-001; dynamic trend lines: such as the gain change curve over time.

[0070] Specifically, after integrating the first and second test charts, the initial performance test chart shows: Normal area (green): azimuth angle 0°~110°, 130°~180°, gain stable at 14.8~15.2dBi; Abnormal area (red): azimuth angle 120°, time 07:16:30, gain suddenly drops to 12.5dBi; Trend line: gain shows a "V-shaped" fluctuation at 07:16:30, which is related to the voltage fluctuation of the power supply network when the turntable rotates to 120°.

[0071] Furthermore, multiple sub-performance test areas are determined based on the detection of the initial performance test map of the antenna, and corresponding abnormal locations are determined based on the identification of each sub-performance test area. Multiple abnormal locations are collected, and multiple abnormal test areas are determined based on the multiple abnormal locations and the corresponding sub-performance test areas, thus introducing multiple abnormal test areas.

[0072] At this point, the initial performance test graph (constructed by S131) ​​contains multi-dimensional information such as performance characteristics, time nodes, angles, and shapes. The system first performs partitioning processing on the test graph, dividing areas with similar continuous performance changes into the same sub-performance test area. The partitioning criteria include: angle range: such as 0~30, 30~60; time range: such as 07:15:00~07:16:00; performance fluctuation amplitude: such as gain change within 1dB is the stable area, and more than 2dB is the fluctuation area.

[0073] Region partitioning methods include: clustering analysis (e.g., K-means): automatic grouping based on the similarity of performance characteristics; sliding window: partitioning regions according to a fixed angle / time window; threshold segmentation: marking abnormal regions based on performance parameter thresholds (e.g., S11>-10dB); generating multiple sub-performance test regions, each region containing: angle / time range; dominant performance characteristics (e.g., gain, S11); region type (stable region / fluctuating region / abrupt region).

[0074] Anomaly detection is performed on each sub-performance test area. The judgment criteria include: parameter exceeding limits: such as gain <14dBi (reference value 15dBi); morphological anomaly: such as S11 showing a spike (normal is a smooth curve); time synchronization anomaly: such as performance change and turntable rotation not being synchronized. The identified anomaly locations need to be recorded as follows: precise angle (such as 75°); time node (such as 07:17:05); anomalous parameter (such as gain 12.5dBi); anomaly type (such as gain drop); generate an anomaly location list, for example: anomaly location 1: 75°, 07:17:05, gain drop; anomaly location 2: 45°, 07:16:30, S11 spike.

[0075] To expand anomaly locations into regions, the following should be considered: Spatial expansion: Expand to both sides of the anomaly location by a certain angle (e.g., ±5) from the center; Temporal expansion: Expand the time window before and after the anomaly time node (e.g., ±10 seconds) from the center; Performance correlation: If similar anomalies exist in adjacent regions (e.g., gains of 70-80 are all <14dBi), they are merged into the same anomaly test region; Avoid duplicate marking, for example: the distance between anomaly location 1 (75) and anomaly location 2 (80) is <5, so they are merged into anomaly test region 1 (70-85); anomaly location 3 (120) is isolated and marked as anomaly test region 2 (115-125); Generate multiple anomaly test regions, each region containing: angle / time range; dominant anomaly type (e.g., gain drop); and a list of associated anomaly locations.

[0076] Specifically, the initial performance test results for a 5G base station antenna (model: ANT-5G-001) showed the following: 0~30: stable gain (14.5~15dBi); 30~60: gain fluctuation (13~15dBi); 60~90: sudden drop in gain (15dBi→12dBi); 90~120: S11 spike (-9dB→-6dB).

[0077] Divide the performance test areas into: Area A (0-30, stable area); Area B (30-60, fluctuating area); Area C (60-90, sudden change area); Area D (90-120, abnormal area); Identify abnormal locations: Area C: gain 12.5dBi at 75 (abnormal location 1); Area D: S11 -6dB at 105 (abnormal location 2); Determine the abnormal test areas: Abnormal location 1 (75) expands to Area 1 (70-85, gain drop); Abnormal location 2 (105) expands to Area 2 (100-110, S11 spike); Output results: Two abnormal test areas: Area 1: gain drop at 70-85, the power supply network needs to be checked; Area 2: S11 spike at 100-110, impedance matching needs to be checked.

[0078] Therefore, by marking the location and shape of multiple abnormal test areas, and tracing the location of these areas to determine the corresponding antenna test events, test optimization measures are determined based on the location, shape, and corresponding antenna test events of the multiple abnormal test areas. This improves the accuracy of the test optimization measures. At the same time, an initial performance test map of the antenna is introduced, which takes into account the location, shape, and corresponding antenna test events of multiple abnormal test areas as a whole, further improving the accuracy of the test optimization measures and realizing the optimization of the antenna during the testing process.

[0079] At this point, the abnormal test areas identified in S132 are marked to clarify their specific range in the spatial or temporal dimensions. Marking methods include: angular range: such as azimuth 70°~85°, pitch -5°~0°; time range: such as 07:15:30~07:16:00; coordinate mapping: if the test chart uses a Cartesian coordinate system, it is marked as (X1,Y1)~(X2,Y2). Based on the variation pattern of performance parameters within the region, the region shape is defined. Common types include: spike type: parameters vary within a local range... Rapid changes within the range (e.g., S11 suddenly rises from -10dB to -5dB); Gradual decrease: parameters continuously decrease (e.g., gain gradually decreases from 15dBi to 12dBi); Oscillating: parameters fluctuate repeatedly within the range (e.g., sidelobe levels oscillate between -20dB and -15dB); Plateau: parameters are stable but exceed the threshold (e.g., noise figure is always >3dB); Generate a list of abnormal regions, each region containing: location label (e.g., A1: 70°~85°); morphology label (e.g., spike type); dominant abnormal parameter (e.g., S11).

[0080] Test events refer to specific operational or environmental conditions that cause abnormal areas, which need to be determined through location backtracking test logs. Common events include: turntable motion events: such as the turntable accelerating from 60° to 90°; beam switching events: such as the beam switching from state 1 to state 2; environmental interference events: such as temperature fluctuations caused by the start-up of the test room air conditioner; equipment triggering events: such as the vector network detector switching test band; time synchronization matching: aligning the time range of the abnormal area with the event timestamps in the test log; angle correlation analysis: if the abnormal area corresponds to a specific angle range, checking the turntable control commands under that angle; parameter cross-validation: combining other parameters (such as temperature, voltage) to determine whether it is an environmental or equipment event; matching a corresponding test event for each abnormal area, for example: A1 (70°~85°) → Event E1: the turntable rotates at a constant speed of 10° / s; A2 (100°~110°) → Event E2: the beam switches from a wide beam to a narrow beam.

[0081] Specifically, assume that S132 outputs two abnormal regions: Region A1: azimuth angle 70°~85°, gain decreases from 14dBi to 12dBi (gradual decrease); Region A2: azimuth angle 100°~110°, S11 increases abruptly from -10dB to -6dB (sharp increase); The results are labeled as follows: A1: position (70°~85°), shape (gradual decrease), parameter (gain); A2: position (100°~110°), shape (sharp increase), parameter (S11).

[0082] Meanwhile, the time range for region A1 (70°~85°) is 07:15:30~07:15:45. The test log shows: 07:15:30: The turntable starts rotating from 60° to 90° at 10° / s; 07:15:45: The turntable stops at 90°; Trace event: E1 (turntable high-speed rotation); The time range for region A2 (100°~110°) is 07:16:10~07:16:20. The log shows: 07:16:10: The beam control command switches from wide beam (state 1) to narrow beam (state 2); 07:16:20: Switching complete; Trace event: E2 (beam switching).

[0083] Based on the location (range of impact), morphology (pattern of change), and test events (triggering causes) of the anomaly areas, targeted optimization measures are formulated. Core principles include: Location priority: Prioritize anomalies at core angles (e.g., main beam direction); Morphological targeting: Spiral type → increase test node density; Gradual descent type → check hardware aging; Oscillating type → shield against environmental interference; Event correlation: Turntable events → optimize motion control algorithms; Beam events → adjust switching timing; Environmental events → add shielding measures. A schematic table of optimization measures is provided, as shown in Table 1.

[0084] Table 1. Schematic Diagram of Optimization Measures

[0085] Exception types Optimization measures Spiral + Beam Switching Extend the handover stabilization time and increase the number of waiting test nodes after the handover. Slow-descent type + turntable movement Reduce turntable speed and increase angle sampling density Oscillation + Environmental Interference The electromagnetic shielding room was activated, and environmental parameters (temperature and humidity) were recorded simultaneously. Platform type + device triggering Calibrate the vector network and replace aging cables.

[0086] Generate a list of optimization measures, specifying: optimization objectives (e.g., eliminating the S11 spike in A2); specific measures (e.g., increasing the settling time by 100ms after beam switching); and verification methods (e.g., retesting the A2 region to confirm that S11 < -10dB).

[0087] Specifically, for region A1 (gradual descent type + high-speed turntable rotation): optimization measures: reduce the turntable speed to 5° / s, and sample once every 2° within the range of 70° to 85° (originally 5°); for region A2 (peak type + beam switching): optimization measures: add 100ms stabilization time after beam switching, and densify the test nodes within the range of 100° to 110° (sample once every 1°).

[0088] refer to Figure 5 In step S14, multiple test optimization items for the antenna are determined based on the detection of test optimization measures, and intelligent optimization events for the antenna are determined based on the multiple test optimization items and multiple abnormal test areas to determine the optimized test area.

[0089] In the specific implementation of this invention, the specific steps are as follows:

[0090] S141: Collect test optimization measures, determine the test optimization procedure list based on the detection of the test optimization measures, determine multiple test optimization items for the antenna based on the identification of the test optimization procedure list, and mark the test optimization content of each test optimization item;

[0091] S142: Collect the current position of the antenna, determine the intelligent optimization event of the antenna based on the current position of the antenna, the test optimization content of each test optimization item and multiple abnormal test areas; determine multiple optimization test nodes based on the analysis of the intelligent optimization event of the antenna, and determine the optimized test area based on the multiple optimization test nodes and the performance characteristics corresponding to the antenna.

[0092] S143: Determine the corresponding matching coefficient based on the matching between the optimized test area and the corresponding abnormal test area. If the matching coefficient exceeds the preset matching coefficient threshold, the optimized test area will replace the existing abnormal test area.

[0093] In the embodiments of this application, test optimization measures are collected, a test optimization process list is determined based on the detection of the test optimization measures, multiple test optimization items of the antenna are determined based on the identification of the test optimization process list, and the test optimization content of each test optimization item is marked, thus introducing multiple test optimization items of the antenna and corresponding test optimization content.

[0094] At this point, the system collects and optimizes test measures. Based on the dependencies and execution order of the optimization measures, it generates a test optimization procedure table, including: procedure number (e.g., OP01, OP02); procedure content (e.g., "adjust turntable control parameters"); preconditions (e.g., "turntable calibration must be completed"); execution time (e.g., "5 minutes before test"); responsible person (e.g., "turntable control module"); dependency analysis: OP01 (turntable calibration) must be executed before OP02 (adjust turntable speed); OP03 (vector network bandwidth adjustment) can be executed independently; the test optimization procedure table is shown in Table 2.

[0095] Table 2 Test Optimization Procedure Table

[0096] Process number Process content Prerequisites Execution time OP01 Turntable calibration none 10 minutes before the test OP02 Adjust the turntable speed to 5° / s OP01 completed 5 minutes before the test OP03 Adjust the vector network scanning bandwidth to 1 GHz none 5 minutes before the test

[0097] Based on the process list, the system identifies multiple test optimization items. Each item includes: item number (e.g., P1, P2); optimization content (e.g., "turntable speed optimization"); range of influence (e.g., "azimuth angle 70-85°"); and expected effect (e.g., "gain fluctuation <0.5dB"). The system marks the specific optimization content for each item, for example: P1: turntable speed optimization (5° / s), area of ​​influence A1 (70-85°); P2: beam switching wait optimization (100ms), area of ​​influence A2 (90-100°); P3: vector network bandwidth optimization (1GHz), affecting the entire system, and generates a list of optimization items.

[0098] Furthermore, the current position of the antenna is collected, and intelligent optimization events for the antenna are determined based on the current position of the antenna, the test optimization content of each test optimization project, and multiple abnormal test areas. Multiple optimization test nodes are determined based on the analysis of the intelligent optimization events of the antenna, and the optimized test area is determined based on the performance characteristics corresponding to the multiple optimization test nodes and the antenna, thereby improving the accuracy of the optimized test area.

[0099] At this time, the system obtains the current position of the antenna in real time through the turntable encoder, for example: current angle: 75; current beam status: beam 2; current test time: 2025-08-29 08:30:00; the system generates intelligent optimization events based on the following factors: test optimization items (from S141): P1: turntable speed optimization (5 / s), affected area A1 (70~85); P2: encrypted test nodes (every 2 samples), affected area A1 (70~85); abnormal test area (from S132): A1: 70~85, gain fluctuation >1dB; A2: 150~165, sidelobe level exceeds limit; current antenna position: the current angle 75 is located in the A1 area; the system generates optimization events according to the following rules: if the current position is located in the abnormal area, the optimization event is triggered; the optimization event includes information such as optimization items, affected area, and execution time.

[0100] The system analyzes the intelligent optimization events and extracts the following key information: Optimization project: P1: Turntable speed optimization (5 / s); P2: Encrypted test nodes (every 2 samples); Affected area: A1: 70~85; The system generates test nodes according to the optimization project: Turntable speed optimization (P1): Original node interval: 5; Optimized node interval: 2; Generated nodes: 70,72,74,76,78,80,82,84; Encrypted test nodes (P2): Based on the above nodes, add intermediate nodes: Generated nodes: 70,71,72,73,74,75,76,77,78,79,80,81,82,83,84,85.

[0101] The system will optimize the test nodes and associate them with performance characteristics (from S131). For example: Node 70: S11: -25dB; Gain: 15dBi; Beamwidth: 30; Node 71: S11: -24dB; Gain: 14.8dBi; Beamwidth: 31. The system generates optimized test areas according to the following rules: Merging of consecutive nodes: If the performance characteristics of adjacent nodes are similar (e.g., gain fluctuation <0.5dB), they are merged into one area; if the performance characteristics of adjacent nodes are significantly different (e.g., gain fluctuation >0.5dB), they are divided into different areas; Adjustment of area boundaries: The range of the optimized area is adjusted according to the boundaries of the abnormal test areas (A1: 70~85).

[0102] Therefore, the matching coefficient is determined based on the matching between the optimized test area and the corresponding abnormal test area. If the matching coefficient exceeds the preset matching coefficient threshold, the optimized test area replaces the existing abnormal test area, thus realizing the replacement of the optimized test area.

[0103] At this time, the system obtains the following data: Optimized test area (from S142): Area A1: 70~85, containing nodes [70,71,…,85]; Performance characteristics: gain 14.8~15dBi, S11 -25~-24dB; Abnormal test area (from S132): Area A1: 70~85, gain fluctuation >1dB; Area A2: 150~165, sidelobe level exceeds limit.

[0104] The system calculates the matching coefficient through the following steps: Region overlap calculation: Calculate the angular overlap ratio between the optimized region and the abnormal region; Formula: Overlap = (Optimized region ∩ Abnormal region) / (Optimized region ∪ Abnormal region); Performance feature similarity calculation: Compare the performance feature differences between the optimized region and the abnormal region; Formula: Similarity = 1 - (|Optimized gain - Abnormal gain| / Abnormal gain); Comprehensive matching coefficient calculation: Formula: Matching coefficient = 0.6 × Overlap + 0.4 × Similarity.

[0105] The system presets a matching coefficient threshold of 0.9, meaning that if the matching coefficient is ≥0.9, the match is considered successful; if the matching coefficient is <0.9, the match is considered unsuccessful. The system makes judgments based on the following rules: Successful match: The optimized region completely covers the abnormal region; the performance characteristics are highly similar to the abnormal region; Unsuccessful match: The optimized region does not completely cover the abnormal region; the performance characteristics differ significantly from the abnormal region.

[0106] Replacement conditions: The replacement operation is only performed when a match is successful; the system completes the replacement through the following steps: Update the list of abnormal test areas: mark the original abnormal area A1 as optimized; add the optimized area A1' to the new list of abnormal test areas; update the test plan: subsequent tests will be executed based on the optimized area A1'; the original abnormal area A1 will no longer participate in the test.

[0107] refer to Figure 6 In step S15, the initial performance test map of the antenna is updated based on the optimized test area. The antenna pattern is determined based on the updated initial performance test map of the antenna, the previous test map of the antenna, and the test scenario of the antenna. The design optimization items of the antenna are determined based on the identification of the antenna pattern.

[0108] In the specific implementation of this invention, the specific steps are as follows:

[0109] S151: Collect the optimized test area and synthesize it with the antenna's initial performance test map to replace the existing abnormal test area. At this time, during the replacement process of the optimized test area, the end nodes of the optimized test area and the connection nodes corresponding to the antenna's initial performance test map are connected in a transitional manner, and the connection status of the optimized test area at the end nodes is gradually improved, and the antenna's initial performance test map is updated.

[0110] S152: Collect the antenna test scenario, determine the antenna's previous test pattern based on the antenna test scenario and antenna model, and determine the antenna radiation pattern by combining the updated antenna initial performance test pattern and the antenna's previous test pattern.

[0111] S153: Based on the detection of the antenna pattern, multiple key parameter regions are determined. Based on the identification of each key parameter region, the antenna sub-optimization design events are determined, and the design parameter changes corresponding to the antenna sub-optimization design events are marked. Based on each sub-optimization design event, the antenna shape, and the antenna test scenario, the antenna design optimization items are determined.

[0112] In the embodiments of this application, an optimized test area is collected and synthesized with the initial performance test map of the antenna to replace the existing abnormal test area. At this time, during the replacement process of the optimized test area, the end nodes of the optimized test area and the connection nodes corresponding to the initial performance test map of the antenna are connected in a transitional manner, and the connection status of the optimized test area at the end nodes is gradually improved, thereby realizing the update of the initial performance test map of the antenna.

[0113] At this point, the optimized test area is derived from the output of S143 and typically includes the following information: angle range (e.g., 70°~85°); performance parameters (e.g., gain, sidelobe level, beamwidth, etc.); test time node (e.g., 2025-08-29 08:45:00); the system reads the optimized test area data from the database or real-time cache to prepare for subsequent synthesis operations.

[0114] The initial performance test plot is typically plotted with angle on the horizontal axis and performance parameters on the vertical axis. Abnormal regions in the initial test plot (e.g., gain values ​​of 13.0–12.0 dBi in the 70°–85° range) are removed; optimized regions are inserted (e.g., gain values ​​of 14.8–15.0 dBi in the 70°–85° range); and the endpoints of the optimized regions (e.g., 70°, 85°) are smoothly transitioned to the connecting nodes in the initial test plot (e.g., 69°, 86°). The formula is: Transition node value = (Optimized node value + Connecting node value) / 2; for example: 69° gain: 14.0. dBi, 70° gain: 14.8dBi → 69.5° gain: (14.0+14.8) / 2=14.4dBi; 85° gain: 15.0dBi, 86° gain: 11.0dBi → 85.5° gain: (15.0+11.0) / 2=13.0dBi; Gradually adjust the performance characteristics of the transition nodes to ensure curve continuity; Update the initial performance test chart and generate the updated test chart; Write the synthesized test chart to the database or real-time cache; Mark the update time (e.g., 2025-08-29 08:47:00); Generate an update log to record the replaced areas, transition nodes, and performance changes.

[0115] Specifically, antenna model: 5G base station antenna; optimized test area: A1': 70°~85°, gain 14.8~15.0dBi; A2': 150°~165°, sidelobe level -18.0~-17.5dB; initial performance test chart: 70°~85° gain: 13.0~12.0dBi (abnormal); 150°~165° sidelobe level: -16.0~-15.5dB (abnormal); system reads data from A1' and A2'; abnormal area replacement: shift Except for the gain values ​​of 13.0 to 12.0 dBi between 70° and 85°; the gain value of inserting A1' is 14.8 to 15.0 dBi; end node transition connection: 69° gain: 14.0 dBi, 70° gain: 14.8 dBi → 69.5° gain: 14.4 dBi; 85° gain: 15.0 dBi, 86° gain: 11.0 dBi → 85.5° gain: 13.0 dBi; connection improvement: adjust the gain values ​​at 69.5° and 85.5° to ensure a smooth curve.

[0116] Update initial performance test plot: Write the synthesized test plot to the database; generate an update log to record the replaced areas and performance changes; Optimization effects: Gain improvement: Gain at 70°–85° increased from 13.0–12.0 dBi to 14.8–15.0 dBi; Sidelobe level reduction: Sidelobe level at 150°–165° decreased from -16.0–-15.5 dBi to -18.0–-17.5 dBi; Smooth transition: Gain values ​​at 69.5° and 85.5° transition naturally, improving curve continuity.

[0117] Furthermore, the antenna's test scenario is collected, and the antenna's previous test patterns are determined based on the test scenario and antenna model. The antenna radiation pattern is then determined by combining the updated initial performance test pattern and the previous test patterns, thus improving the accuracy of the antenna radiation pattern.

[0118] At this point, the system obtains the antenna test scenario through the following methods: manual input: such as "5G base station antenna" or "satellite communication antenna"; automatic identification: automatically matching the scenario based on the test task number or antenna model; database query: extracting similar scenarios from historical test records; and collecting a scenario classification table, as shown in Table 3.

[0119] Table 3 Scene Classification Table

[0120] Scene type Application scenarios Test focus 5G base station City coverage Gain, beamwidth, sidelobe level satellite communications Space Transmission High gain, low noise, directivity radar system Target detection | Impulse response, main lobe width

[0121] The system determines past test images through the following steps: Scene matching: Filter historical test records based on the current scene (e.g., "5G base station"); Model matching: Further filter test images of the same model antenna (e.g., "ANT-5G-001"); Time filtering: Prioritize test data within the last 6 months; The system retrieves matching past test images through database queries and arranges them in reverse chronological order.

[0122] The system synthesizes the test pattern through the following steps: aligning the updated initial performance test pattern with previous test patterns by angle; weighting the performance parameters at the same angle (with more recent data having higher weight); smoothing the synthesized data to ensure curve continuity; example of the synthesis formula: for gain data: Gsynthesis(θ) = 0.7 × Gupdate(θ) + 0.3 × Ghistory(θ); where: Gupdate(θ): updated gain data; Ghistory(θ): average value of historical gain data; 0.7 and 0.3 are weighting coefficients (which can be adjusted according to actual conditions) to generate the final antenna pattern.

[0123] Specifically, the antenna model is ANT-5G-001; the test scenario is a 5G base station; the updated initial performance test chart is 70-85 with a gain of 14.8-15.0 dBi; previous test charts are H001 (2025-02-15) and H002 (2025-05-20).

[0124] Input scenario: "5G base station"; The system automatically matches historical test images of the same model and scenario; Query results: H001: 70~85 gain 13.0~14.0dBi; H002: 70~85 gain 13.5~14.5dBi; Calculate historical average value: 70~85 gain 13.25~14.25dBi.

[0125] Data alignment: Align the updated data with historical data by angle; Weighted average: 70 gain: 0.7×14.8+0.3×13.25=14.335dBi; 85 gain: 0.7×15.0+0.3×14.25=14.775dBi; Smoothing: Interpolate the synthesized gain data. Value and smoothing; Final radiation pattern: 70-85 Gain: 14.335-14.775 dBi; Sidelobe level: -17.5--17.0 dB (updated data); Curve continuity: Smooth transition is ensured through interpolation; Optimization effect: Gain stability: The synthesized gain data is between the updated and historical data, avoiding extreme values; Radiation pattern accuracy: Combining historical trends improves the reliability of the radiation pattern; Test efficiency: Automatic synthesis reduces manual intervention and improves test efficiency.

[0126] Therefore, based on the detection of the antenna pattern, multiple key parameter regions are determined. Sub-optimization design events for the antenna are identified according to the recognition of each key parameter region, and the design parameter changes corresponding to these sub-optimization design events are marked. Based on each sub-optimization design event, the antenna's shape, and the antenna's test scenario, the antenna's design optimization projects are determined, improving the accuracy of these projects and further optimizing the antenna design. Simultaneously, an optimized test area is introduced, facilitating the updating of the antenna's initial performance test pattern. This achieves compatibility considerations between the updated initial performance test pattern, previous test patterns, and the antenna's test scenario, further improving the accuracy of the antenna pattern and the accuracy of the antenna's design optimization projects.

[0127] At this point, the input is the antenna pattern generated by S152, which includes: angle range (e.g., 0-360°); performance parameters (e.g., gain, sidelobe level, beamwidth, etc.); test scenario (e.g., 5G base station antenna); the system determines the key parameter regions through the following steps: parameter threshold setting: set the thresholds for key parameters according to the scenario, such as: gain threshold: ≥14dBi; sidelobe level threshold: ≤-17dB; beamwidth threshold: ≤30; region division: divide the pattern into multiple regions according to angle; parameter detection: perform parameter detection on each region to identify regions that exceed the threshold.

[0128] Based on the detection results of key parameter areas, the system generates sub-optimization design events: Event type: The event type is determined according to the abnormal parameters, such as: insufficient gain → gain optimization event; sidelobe level exceeding the standard → sidelobe suppression event; beamwidth exceeding the standard → beamwidth optimization event; Parameter change marker: Each event is marked with a suggestion for changes in design parameters, such as: gain optimization event: increase the number of antenna elements or adjust the feed network; sidelobe suppression event: optimize array weighting or adjust the reflector shape; beamwidth optimization event: adjust the array spacing or phase distribution.

[0129] The system generates design optimization projects based on sub-optimization design events, antenna configurations, and test scenarios. It then selects suitable optimization methods based on antenna configuration (e.g., array antennas, reflector antennas), adapts to different scenarios (e.g., 5G base stations), and integrates multiple sub-optimization events into a complete design optimization project.

[0130] Specifically, antenna model: ANT-5G-001; test scenario: 5G base station antenna; key parameter area detection results: 70~80: gain 13.8dBi (abnormal, threshold ≥14dBi); 70~80: beamwidth 32 (abnormal, threshold ≤30).

[0131] Key parameter region detection: Detected gain and beamwidth anomalies in the 70-80 range; generated a list of key parameter regions; Sub-optimization design event generation: Gain optimization event: marked by increasing the number of antenna elements and adjusting the feed network; Beamwidth optimization event: marked by adjusting array spacing and optimizing phase distribution; Design optimization project generation: Integrated gain and beamwidth optimization events; generated a 5G base station antenna gain and beamwidth optimization project; clarified optimization methods and expected improvements; Design optimization project: Project name: 5G base station antenna gain and beamwidth optimization; Optimization method: Increase the number of antenna elements, adjust array spacing, optimize phase distribution; Expected improvement: Gain increased to ≥14dBi, beamwidth reduced to ≤30; Optimization effect: Performance improvement: Through optimized design, the antenna gain and beamwidth in the 70-80 range both meet the threshold requirements; Design guidance: Provides clear optimization directions and parameter adjustment suggestions for subsequent antenna designs.

[0132] Please see Figure 7 , Figure 7 This is a schematic diagram of the structural composition of an intelligent testing system for antennas based on antenna patterns according to an embodiment of the present invention; the intelligent testing system for antennas based on antenna patterns includes:

[0133] The antenna rotation module 21 is used to initialize each pulse parameter based on the wave control command and timing controller of the host computer, and determine the rotation path of the turntable. At this time, the antenna is mounted on the turntable.

[0134] The performance characteristic module 22 is used to determine multiple performance combinations based on multiple performance parameters of the antenna, corresponding angles and shapes, and to determine the corresponding performance characteristics based on the identification of each performance combination.

[0135] The test optimization measures module 23 is used to construct the initial performance test map of the antenna based on the characteristic shape of each performance feature and the corresponding test time node, determine multiple abnormal test areas based on the initial performance test map of the antenna, and determine test optimization measures based on the area location, area shape and corresponding antenna test events of the multiple abnormal test areas.

[0136] The intelligent optimization event module 24 is used to determine multiple test optimization items of the antenna based on the detection of test optimization measures, and to determine the intelligent optimization event of the antenna based on the multiple test optimization items and multiple abnormal test areas, so as to determine the optimized test area.

[0137] The design optimization project module 25 is used to trigger the update of the antenna's initial performance test map based on the optimized test area, determine the antenna radiation pattern based on the updated antenna's initial performance test map, the antenna's previous test map and the antenna's test scenario, and determine the antenna's design optimization project based on the identification of the antenna radiation pattern.

[0138] The technical features of the above embodiments can be combined arbitrarily. For the sake of brevity, not all combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

Claims

1. A smart testing method for antennas based on antenna radiation patterns, characterized in that, include: Based on the wave control command and timing controller of the host computer, the initialization of each pulse parameter is triggered and the rotation path of the turntable is determined. At this time, the antenna is mounted on the turntable. Multiple performance combinations are determined based on the antenna's various performance parameters, corresponding angles, and shapes. The corresponding performance characteristics are then determined based on the identification of each performance combination. An initial performance test map of the antenna is constructed based on the characteristic shape of each performance feature and the corresponding test time node. Multiple abnormal test areas are identified based on the initial performance test map of the antenna. Test optimization measures are determined based on the location, shape and corresponding antenna test events of the multiple abnormal test areas. Based on the detection of test optimization measures, multiple test optimization items for the antenna are determined. Based on the multiple test optimization items and multiple abnormal test areas, intelligent optimization events for the antenna are determined to determine the optimized test area. The initial performance test map of the antenna is updated based on the optimized test area. The antenna pattern is determined based on the updated initial performance test map, the previous test map of the antenna, and the test scenario of the antenna. The design optimization items of the antenna are determined based on the identification of the antenna pattern. The process involves constructing an initial performance test map of the antenna based on the characteristic morphology of each performance feature and the corresponding test time node; identifying multiple abnormal test areas based on the initial performance test map; and determining test optimization measures based on the location, morphology, and corresponding antenna test events of the multiple abnormal test areas, including: Collect various performance characteristics, and determine the characteristic shape and corresponding test time node of each performance characteristic based on the detection of each performance characteristic; at the same time, collect the antenna model, determine the first test pattern based on the characteristic shape of each performance characteristic and the antenna model, determine the second test pattern based on the test time node of each performance characteristic and the antenna model, and construct the initial performance test pattern of the antenna based on the first test pattern and the second test pattern. The process of constructing an initial performance test map of the antenna based on the characteristic morphology of each performance feature and the corresponding test time node, identifying multiple abnormal test areas based on the initial performance test map, and determining test optimization measures based on the location, morphology, and corresponding antenna test events of the multiple abnormal test areas, further includes: Multiple sub-performance test areas are determined based on the detection of the initial performance test map of the antenna. The corresponding abnormal locations are determined based on the identification of each sub-performance test area. Multiple abnormal locations are collected. Multiple abnormal test areas are determined based on the multiple abnormal locations and the corresponding sub-performance test areas. Mark the location and shape of multiple abnormal test areas, determine the corresponding antenna test events based on the tracing of the location of multiple abnormal test areas, and determine test optimization measures based on the location, shape and corresponding antenna test events of multiple abnormal test areas. The process of updating the initial performance test map of the antenna based on the optimized test area, determining the antenna radiation pattern based on the updated initial performance test map, previous test maps of the antenna, and the test scenario of the antenna, and determining the design optimization items of the antenna based on the identification of the antenna radiation pattern includes: The optimized test area is collected and combined with the initial performance test map of the antenna to replace the existing abnormal test area. During the replacement process of the optimized test area, the end nodes of the optimized test area and the connection nodes corresponding to the initial performance test map of the antenna are connected in a transitional manner, and the connection status of the optimized test area at the end nodes is gradually improved, thereby updating the initial performance test map of the antenna. The step of updating the initial performance test map of the antenna based on the optimized test area, determining the antenna radiation pattern based on the updated initial performance test map, previous test maps of the antenna, and the test scenario of the antenna, and determining the design optimization items of the antenna based on the identification of the antenna radiation pattern, further includes: The test scenario of the antenna is collected, and the previous test pattern of the antenna is determined based on the test scenario and the antenna model. The antenna radiation pattern is determined by combining the updated initial performance test pattern of the antenna and the previous test pattern of the antenna. Based on the detection of the antenna pattern, multiple key parameter regions are determined. Sub-optimization design events of the antenna are determined according to the identification of each key parameter region, and the changes in design parameters corresponding to the sub-optimization design events are marked. The design optimization items of the antenna are determined according to each sub-optimization design event, the antenna shape, and the antenna test scenario.

2. The intelligent testing method for antennas based on antenna patterns according to claim 1, characterized in that, The host computer-based wave control commands and timing controller trigger the initialization of each pulse parameter and determine the rotation path of the turntable. At this time, the antenna is mounted on the turntable, including: To collect the antenna's test requirements, the host computer generates a beam control command based on the test requirements. The timing controller receives the beam control command and triggers the initialization of each pulse parameter based on the beam control command. The initialization is confirmed to be complete by status readback. At this time, each pulse parameter includes the pulse width, pulse repetition frequency, and delay parameter. Multiple rotation position nodes of the turntable are marked based on the initialization of each pulse parameter. The rotation three-dimensional space of the antenna is determined based on the multiple rotation position nodes of the turntable, the corresponding rotation attitude, and the position of the antenna relative to the turntable. The rotation path of the turntable is determined based on the path planning of the rotation three-dimensional space of the antenna. At the same time, during the test, the host computer can dynamically optimize the rotation path of the turntable based on real-time data.

3. The intelligent testing method for antennas based on antenna patterns according to claim 1, characterized in that, The process of determining multiple performance combinations based on multiple performance parameters of the antenna, corresponding angles, and shapes, and determining corresponding performance characteristics based on the identification of each performance combination, includes: The antenna rotates under the drive of the turntable and is tested. Multiple performance parameters of the antenna are collected, and the corresponding angles are marked. At the same time, the shape of the antenna is collected. Multiple performance combinations are determined based on the cross-matching of multiple performance parameters, corresponding angles and shapes of the antenna. In each performance combination, priority distribution information is determined based on the identification of the performance combination. The performance detection mode is determined according to the priority distribution information and the combination content of the performance combination. The corresponding performance feature is determined according to the performance combination and the corresponding performance detection mode, so as to collect multiple performance features and mark the test time node of each performance feature.

4. The intelligent testing method for antennas based on antenna patterns according to claim 1, characterized in that, The detection based on test optimization measures determines multiple test optimization items for the antenna. Based on these multiple test optimization items and multiple abnormal test areas, intelligent optimization events for the antenna are determined to identify the optimized test area, including: Collect test optimization measures, determine the test optimization procedure list based on the detection of the test optimization measures, identify multiple test optimization items for the antenna based on the identification of the test optimization procedure list, and mark the test optimization content of each test optimization item.

5. The intelligent testing method for antennas based on antenna patterns according to claim 4, characterized in that, The method of determining multiple test optimization items for the antenna based on test optimization measures, and determining intelligent optimization events for the antenna based on these multiple test optimization items and multiple abnormal test areas to determine the optimized test area, further includes: The system acquires the current position of the antenna, determines the intelligent optimization events of the antenna based on the current position of the antenna, the test optimization content of each test optimization item, and multiple abnormal test areas; determines multiple optimization test nodes based on the analysis of the intelligent optimization events of the antenna, and determines the optimized test area based on the multiple optimization test nodes and the performance characteristics corresponding to the antenna. The matching coefficient is determined based on the matching between the optimized test area and the corresponding abnormal test area. If the matching coefficient exceeds the preset matching coefficient threshold, the optimized test area will replace the existing abnormal test area.

6. An intelligent testing system for antennas based on antenna radiation patterns, characterized in that, The intelligent testing system for antennas based on antenna patterns is applied to the intelligent testing method for antennas based on antenna patterns as described in any one of claims 1 to 5, wherein the intelligent testing system for antennas based on antenna patterns includes: The antenna rotation module is used to initialize various pulse parameters based on the wave control command and timing controller of the host computer, and to determine the rotation path of the turntable. At this time, the antenna is mounted on the turntable. The performance characteristic module is used to determine multiple performance combinations based on multiple performance parameters of the antenna, corresponding angles and shapes, and to determine the corresponding performance characteristics based on the identification of each performance combination. The test optimization measures module is used to construct the initial performance test map of the antenna based on the characteristic shape of each performance feature and the corresponding test time node, identify multiple abnormal test areas based on the initial performance test map of the antenna, and determine test optimization measures based on the location, shape and corresponding antenna test events of the multiple abnormal test areas. The intelligent optimization event module is used to determine multiple test optimization items for the antenna based on the detection of test optimization measures, and to determine the intelligent optimization events for the antenna based on multiple test optimization items and multiple abnormal test areas, so as to determine the optimized test area. The design optimization project module is used to trigger the update of the antenna's initial performance test map based on the optimized test area. Based on the updated antenna's initial performance test map, the antenna's previous test maps, and the antenna's test scenario, the antenna radiation pattern is determined, and the design optimization project for the antenna is determined based on the identification of the antenna radiation pattern.

Images