Method and system for testing multiple serial ports of 5G mainboard

Abstract

The invention relates to the technical field of 5G mainboard testing, and particularly discloses a 5G mainboard multi-serial port testing method and system, and the method comprises the steps: setting a stepping resolution, and collecting a space interference feature group of a 5G mainboard; analyzing the spatial feature group to extract a continuous null region, segmenting the space to obtain a null space body, predicting a dynamic migration path of the null space body, and extracting a core region and a diffusion boundary; predicting a null suppression radius based on the diffusion boundary coordinate set, generating a zeroing trap, and judging whether the updating speed of the zeroing trap is matched with the interference dynamic state or not; and if not, extracting a null secondary index and a beam change rate to determine an interference dynamic type, and constructing a scheduling decision matrix to optimize the stepping resolution. The implementation system comprises a feature extraction module, a null analysis module, an adaptation analysis module and a stepping optimization module. According to the invention, identification and dynamic tracking of the null region in the 5G mainboard multi-serial port test can be realized, test parameters are adjusted, and the test efficiency and accuracy are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of 5G motherboard testing, and in particular to a method and system for testing multiple serial ports of a 5G motherboard. Background Art

[0002] In the field of multi-serial port testing on 5G motherboards, as the number of serial ports increases and communication rates increase, problems such as spatial interference and dynamic interference caused by the collaborative work of multiple devices are becoming increasingly prominent.

[0003] Existing technologies for identifying null-sink areas ignore their dynamic migration characteristics caused by the movement of interference sources and environmental changes. They are unable to separate the core area from the diffusion boundary and have difficulty tracking the real-time changes in their spatial morphology, resulting in ambiguous positioning of the null-sink area and a lack of clear spatial basis for interference suppression.

[0004] Existing technologies mostly use fixed-parameter zero-trap designs in existing interference suppression methods, which cannot dynamically adjust the suppression range according to the migration path and interference intensity of the zero-trap area. When the interference source moves quickly or the intensity changes suddenly, the update speed of the zero-trap does not match the interference dynamics, which can easily lead to suppression lag or over-suppression, and the adaptability cannot meet the test needs.

[0005] To this end, the present invention provides a method and system for testing multiple serial ports of a 5G motherboard. Summary of the Invention

[0006] The purpose of the present invention is to provide a method and system for testing multiple serial ports of a 5G motherboard to solve the above-mentioned background problems.

[0007] The purpose of the present invention can be achieved through the following technical solutions:

[0008] A method for testing multiple serial ports of a 5G motherboard includes the following steps:

[0009] Set the step resolution, collect the spatial interference features of all 5G motherboards based on the step resolution, and establish a spatial feature group;

[0010] Analyze the spatial feature group and extract continuous null-sink areas. Perform spatial segmentation on the continuous null-sink areas to obtain null-sink space bodies. Predict the movement trend of the null-sink space bodies to obtain dynamic migration paths. Combined with the dynamic migration paths, extract the diffusion boundary coordinate set of the continuous null-sink areas.

[0011] Dynamically predict the diffusion boundary coordinate set to obtain the null suppression radius, formulate an adjustable null trap based on the null suppression radius, and perform dynamic adaptability analysis on the adjustable null trap to determine whether the update speed of the adjustable null trap matches the interference dynamics. If not, trigger an adaptive analysis signal;

[0012] If the adaptive analysis signal is triggered, the secondary zero-sink index and beam change rate are extracted to determine the interference dynamic type. The matching degree of different interference dynamic types is analyzed to construct a scheduling decision matrix, and an optimization strategy for the step resolution is formulated.

[0013] As a further technical solution of the present invention: the method of establishing the spatial feature group is:

[0014] The 5G motherboard is moved according to the step resolution. The field strength attenuation, bit error rate, signal-to-noise ratio, and step distance of the start and end coordinates are collected. Gradient calculations are performed to obtain the field strength step gradient, bit error step gradient, and signal-to-noise step gradient. Feature extraction is performed on the three step gradients to obtain device features and combination features.

[0015] Obtain the field intensity gradient variance and maximum error step gradient of all 5G motherboards in the area during the same monitoring period, and use the field intensity gradient variance and maximum error step gradient as interference features;

[0016] Obtain interference features, combination features, and device features and integrate them according to position coordinates to obtain spatial interference features and establish a spatial feature group.

[0017] As a further technical solution of the present invention: the method of obtaining the combination characteristics and device characteristics is:

[0018] Perform gradient correlation calculations on the field strength step gradient, bit error step gradient, and signal-to-noise step gradient of 5G motherboards i and j in the region, and obtain three gradient correlation coefficients, including the field strength step gradient, bit error step gradient, and signal-to-noise step gradient, as device characteristics.

[0019] Among them, i and j are the numbers of the 5G motherboard;

[0020] Based on the mean velocity vector of 5G motherboard i and j and the angle between the moving directions, interference coupling analysis is performed to extract the interference coupling factor, which is used as a combined feature.

[0021] As a further technical solution of the present invention: the method of obtaining the dynamic migration path is:

[0022] Obtain the interference coupling factor in the spatial null body, build a migration model based on the null space gradient and center position coordinates, and combine the interference coupling factor to dynamically predict the position of the null space body;

[0023] If the dynamic coupling factor within the monitoring period is higher than or equal to the preset coupling threshold, the expansion amount of the null space migration direction is extracted by constructing an expansion equation along the migration direction of the predicted position and the center position of the null space;

[0024] A dynamic migration path of the null-sink space body is established based on the position of the null-sink space body and the expansion amount of the migration direction of the null-sink space body.

[0025] As a further technical solution of the present invention: the method of obtaining the spatial null body is:

[0026] Obtain the step distance moved by each 5G motherboard in the spatial feature group as a spatial grid;

[0027] Based on the interference characteristics of the spatial grid and the characteristics of the equipment, a continuous zero-sink criterion is established to extract the continuous zero-sink area;

[0028] The octree segmentation algorithm is used in combination with the interference coupling factor of the combined features to recursively divide the continuous null region into pixels and obtain the minimum spatial voxel.

[0029] Perform frequency analysis on the position corresponding to the minimum voxel in the monitoring period to obtain the voxel null rate;

[0030] The voxel null sink connectivity boundary is set. If the voxel null sink rate of the minimum voxel in the adjacent space is in the voxel null sink connectivity boundary, the minimum voxels in the adjacent space are merged to establish the null sink space volume.

[0031] As a further technical solution of the present invention, the method for determining whether the update speed of the adjustable zero trap matches the interference dynamics is:

[0032] Obtain the beam change rate when generating adjustable zero traps and the hysteresis zero trap coefficients in adjacent monitoring periods;

[0033] Determine whether the 5G motherboard test is in a hysteresis null state based on the hysteresis null coefficient. If it is in a hysteresis null state, obtain the null depth and null rate of the core area of ​​the continuous null area.

[0034] The deviation ratio between the zero-sink depth and the core area zero-sink rate is calculated to obtain the strength adaptation coefficient;

[0035] Different intensity adaptation levels are divided based on the intensity adaptation coefficient. If the intensity adaptation level is at the lowest adaptation level and is in a hysteresis zero sink state, it means that the update speed does not match the interference dynamics.

[0036] As a further technical solution of the present invention: the method of generating the adjustable zero trap is:

[0037] The antenna tilt angle and beam change rate of the 5G motherboard are collected during each monitoring period, and the null feature vector is constructed by combining the core area and diffusion boundary coordinate sets and the beam angle;

[0038] Combined with the null trap feature vector, the mapping relationship of the null trap suppression radius is established through the random forest algorithm to obtain the null trap suppression radius. Based on the null trap suppression radius, the dynamic null trap is generated through the beamforming algorithm and the dynamic null trap is used as an adjustable null trap.

[0039] As a further technical solution of the present invention: the method of formulating the optimization strategy of the step resolution is:

[0040] Construct a scheduling decision matrix and develop optimization strategies for different step resolutions;

[0041] Obtain the tracking error rate and suppression target achievement rate after the optimization strategy is formulated, and normalize the tracking error rate and the suppression target achievement rate respectively;

[0042] The normalized tracking error rate and suppression compliance rate are summed to obtain the adjustment effectiveness coefficient to determine whether the adjustment meets expectations. If not, adjustments or warnings are made.

[0043] As a further technical solution of the present invention: the method of constructing the scheduling decision matrix is:

[0044] Obtain the secondary null sink index and beam change rate, determine the interference dynamic type, and obtain the interference source movement rate and current step resolution under each interference dynamic type;

[0045] The motion matching degree is obtained by comparing the moving speed of the interference source with the current step resolution.

[0046] The ratio of the secondary index of the null sink to the current step resolution is processed to obtain the spatial gradient of the interference intensity change;

[0047] Obtain the critical spatial gradient, calculate the deviation ratio between the spatial gradient of the interference intensity change and the critical spatial gradient, and obtain the recognition matching degree;

[0048] The motion matching and recognition matching under different interference types are obtained, and the scheduling decision matrix for 5G motherboard testing is constructed in combination with physical constraints.

[0049] A system for testing multiple serial ports on a 5G motherboard includes the following modules:

[0050] Feature extraction module: used to set the step resolution, collect the spatial interference features of all 5G motherboards based on the step resolution, and establish a spatial feature group;

[0051] Null sink analysis module: used to analyze spatial feature groups and extract continuous null sink areas, perform spatial segmentation on the continuous null sink areas to obtain null sink space bodies, and predict the movement trend of the null sink space bodies to obtain dynamic migration paths. Combined with the dynamic migration paths, the diffusion boundary coordinate set of the continuous null sink areas is extracted.

[0052] Adaptation Analysis Module: This module is used to dynamically predict the loading boundary coordinate set to obtain the null suppression radius, formulate an adjustable zero trap based on the null suppression radius, and perform dynamic adaptability analysis on the adjustable zero trap to determine whether the update speed of the adjustable zero trap matches the interference dynamics. If not, an adaptation analysis signal is triggered.

[0053] Step optimization module: If the adaptive analysis signal is triggered, it is used to extract the secondary null sink index and beam change rate to determine the interference dynamic type, analyze the matching degree of different interference dynamic types to build a scheduling decision matrix, and formulate an optimization strategy for the step resolution.

[0054] Beneficial effects of the present invention:

[0055] (1) By setting the step resolution, synchronously collecting multi-dimensional parameters such as field strength step gradient, error step gradient, signal-to-noise step gradient, and combining the equipment motion characteristics to construct a spatial interference feature group, a full range of quantitative characterization of the multi-serial port communication status of the 5G motherboard is achieved, providing high-precision basic data support for the subsequent null-sink area identification; by analyzing the spatial feature group to segment the null-sink space body, combining the dynamic migration path to accurately extract the core area and diffusion boundary, the spatial morphology and temporal motion trend of the null-sink area are characterized, reducing the problem of difficulty in tracking the dynamic changes of the null-sink area in the 5G motherboard multi-serial port test, and providing a clear spatial positioning basis for interference suppression.

[0056] (2) Based on the zero-sink suppression radius, an adjustable zero trap is generated, and its matching with the interference dynamics is judged through dynamic adaptability analysis, which realizes real-time and flexible suppression of interference, reduces the lack of adaptation of fixed traps to dynamic interference, and improves the response speed of the test system to complex interference environments.

[0057] (3) By identifying the dynamic type of interference and building a scheduling decision matrix, the step resolution is optimized in a targeted manner, so that the test parameters can adaptively match different interference scenarios and reduce invalid test operations. This is beneficial to improving the test accuracy and the efficiency of concurrent testing of multiple serial ports, and improving the reliability and stability of the test results. BRIEF DESCRIPTION OF THE DRAWINGS

[0058] The present invention will be further described below with reference to the accompanying drawings.

[0059] Figure 1 This is a flow chart of a method for testing multiple serial ports on a 5G motherboard according to the present invention;

[0060] Figure 2 This is a flow chart for extracting the expansion amount of the migration direction of the null-trapped space body in the present invention.

[0061] Figure 3 This is a module diagram of a system for testing multiple serial ports on a 5G motherboard according to the present invention. DETAILED DESCRIPTION

[0062] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making any creative efforts shall fall within the scope of protection of the present invention.

[0063] Example 1

[0064] See also Figure 1 As shown, the present invention is a method for testing multiple serial ports of a 5G motherboard, comprising the following steps:

[0065] S1. Set the step resolution, collect the spatial interference features of all 5G motherboards based on the step resolution, and establish a spatial feature group;

[0066] Among them, the method of collecting the spatial interference characteristics of multiple 5G motherboards based on step resolution is:

[0067] Preferably, a highly sensitive communication serial port of the 5G mainboard is extracted, and the bit error rate of the highly sensitive communication serial port is collected during the monitoring period;

[0068] Among them, the high-sensitivity communication serial port of the 5G motherboard can be RS-232, RS-485, or USB to serial port type;

[0069] Move the 5G motherboard according to the preset step resolution and record the moving position coordinates , collect the step distance of the 5G motherboard moving position coordinates during the monitoring period, as well as the field strength attenuation value and signal-to-noise ratio of the starting and ending coordinates during the monitoring period;

[0070] Calculate the field strength step gradient based on the field strength attenuation value and step distance of the start and end coordinates;

[0071] Calculate the error step gradient based on the error rate and step distance of the start and end coordinates;

[0072] Calculate the signal-to-noise step gradient based on the signal-to-noise ratio and step distance of the start and end coordinates;

[0073] Preferably, by the formula: Get the field strength step gradient G;

[0074] By formula: Get the error step gradient R;

[0075] By formula: Get the signal-to-noise step gradient N;

[0076] in, are the end point coordinates within the monitoring period Field strength attenuation value A , and the field strength attenuation value of the starting point coordinates ;

[0077] is the step distance, are the end point coordinates within the monitoring period Bit error rate , and the bit error rate of the starting coordinates ;

[0078] are the end point coordinates within the monitoring period Signal-to-noise ratio , and the signal-to-noise ratio of the starting coordinates

[0079] Collect the field strength step gradient, bit error step gradient, and signal-to-noise step gradient of multiple 5G motherboards in the area;

[0080] Perform gradient correlation calculations on the field strength step gradient, bit error step gradient, and signal-to-noise step gradient for 5G motherboard i and 5G motherboard j in the region. Three gradient correlation coefficients, including the field strength step gradient, bit error step gradient, and signal-to-noise step gradient, are obtained as device characteristics.

[0081] It can be understood that the Pearson correlation coefficient calculation method is used to measure the linear correlation between the three gradients (field strength, bit error, and signal-to-noise step gradient) of 5G motherboards i and j, and the corresponding correlation coefficient is obtained as the device feature;

[0082] Among them, i and j are the numbers of the 5G motherboard;

[0083] At the same time, obtain the speed mean vector and movement direction angle of 5G motherboard i and motherboard j during the monitoring period ;

[0084] Based on the mean velocity vector and the angle of motion direction of 5G motherboard i and 5G motherboard j , perform interference coupling analysis to extract interference coupling factors

[0085] Preferably, by the formula: Obtaining the interferometric coupling factor , and interfere with the coupling factor as a combination feature;

[0086] in, The communication wavelength when testing the 5G motherboard;

[0087] Obtain the field intensity gradient variance and maximum error step gradient of all 5G motherboards in the area during the same monitoring period, and use the field intensity gradient variance and maximum error step gradient as interference features;

[0088] Integrate interference features, combination features, and device features according to position coordinates to obtain spatial interference features;

[0089] It is understandable that the purpose of sorting out the spatial interference features is:

[0090] Objective 1: By integrating device characteristics such as field strength step gradient and error step gradient, as well as combined characteristics such as interference coupling factor, the spatial interference signature formed can be used to analyze spatial feature groups and extract continuous null regions, providing a data basis for subsequent segmentation and dynamic analysis of null space volumes.

[0091] Objective 2: Support the prediction of the movement trend of the null trap space. The spatial interference features include the position coordinates, gradient changes, and interference coupling relationship of the 5G motherboard. Based on these features, a migration model can be constructed to predict the dynamic migration path of the null trap space, and extract the core area and diffusion boundary to provide a spatial range basis for the formulation of adjustable null traps.

[0092] Objective 3: Provide a basis for step resolution optimization. Spatial interference characteristics, such as null rate changes and beam change rates, can be used to determine the dynamic type of interference, such as strong dynamic interference and mobile-dominated interference. By combining these types of interference, we can analyze the matching degree and construct a scheduling decision matrix, ultimately supporting the formulation of step resolution optimization strategies.

[0093] Obtain the spatial interference features of all 5G motherboards in the area and establish a spatial feature group.

[0094] S2. Analyze the spatial feature group and extract continuous null-sink areas. Perform spatial segmentation on the continuous null-sink areas to obtain null-sink space bodies. Predict the movement trend of the null-sink space bodies to obtain dynamic migration paths. Combined with the dynamic migration paths, extract the core area and diffusion boundary of the continuous null-sink areas.

[0095] The method for analyzing the spatial feature group and extracting continuous null regions is as follows:

[0096] Obtain the step distance moved by each 5G motherboard in the spatial feature group as a spatial grid;

[0097] Based on the interference characteristics of the spatial grid and the characteristics of the equipment, a continuous zero-sink criterion is established to extract the continuous zero-sink area;

[0098] The octree segmentation algorithm is used in combination with the interference coupling factor of the combined features to recursively divide the continuous null region into pixels and obtain the minimum spatial voxel.

[0099] Those skilled in the art will understand that the test area is divided into spatial grids based on the step distance, and each grid corresponds to the relevant data of the 5G motherboard at that location. Among them, the interference characteristics include the field intensity gradient variance and the maximum error step gradient within the area, and the device characteristics are the correlation coefficients of the three step gradients of field intensity, error, and signal-to-noise between different motherboards. By setting the thresholds of the interference characteristics, combined characteristics, and device characteristics, a continuous nulling criterion is formed;

[0100] When the interference characteristics and device characteristics of a certain grid meet the threshold conditions in the criterion at the same time, the grid is marked as a zero sink point, and then the adjacent zero sink points are connected to form a continuous zero sink area;

[0101] Using the octree segmentation algorithm, the entire continuous null region is considered as an initial cube. The interference coupling factor in the combined features is combined to determine whether the current cube needs to be segmented: if the difference in the interference coupling factors of the various parts of the cube is higher than the preset threshold, indicating that the null characteristics of the region are uneven, it is then divided into eight smaller cubes on average. For each newly generated small cube, the above judgment and segmentation process is repeated until the difference in the interference coupling factors within all small cubes is lower than the preset threshold. The smallest cube obtained at this time is the minimum spatial voxel.

[0102] Perform frequency analysis on the position corresponding to the minimum voxel in the monitoring period to obtain the voxel null rate;

[0103] The frequency analysis method is as follows:

[0104] Obtain the number of times the position coordinates corresponding to the minimum voxel in space are in the continuous null zone and the number of monitoring times within N monitoring cycles;

[0105] The number of times the spatial minimum voxel is in the continuous null zone is ratioed to the number of monitoring times to obtain the voxel null rate.

[0106] Obtain the voxel null sink rate of the minimum voxel in the adjacent space and set the voxel null sink connectivity boundary;

[0107] If the voxel null sink rate of the minimum voxel in the adjacent space is at the voxel null sink connected boundary, the minimum voxels in the adjacent space are merged to establish the null sink space volume;

[0108] It can be understood that the voxel null connectivity boundary (i.e., the threshold range of the null rate) is then set to determine whether adjacent voxels have continuity in the null characteristics. If the null rates of two adjacent minimum voxels are both within this connectivity boundary, it indicates that the two have consistent null characteristics and are merged into a whole unit. By recursively merging all adjacent voxels that meet the conditions, a null space with spatial continuity is finally formed, reflecting the spatial morphology of the null area in the 5G motherboard multi-serial port test.

[0109] Obtain the center position coordinates of the null space body in N monitoring cycles and construct the center position sequence ;

[0110] Preferably, N ≥ 10;

[0111] Among them, the method of predicting the movement trend of the null-sink space body and obtaining the dynamic migration path is as follows:

[0112] Obtain the average voxel null sink rate of all spatial minimum voxels in the null sink space to obtain the spatial null sink rate;

[0113] It can be understood that the physical significance of constructing a null trap volume is to transform the discrete null trap signal characteristics in 5G motherboard multi-serial port testing into a three-dimensional entity with spatial continuity. Its physical significance lies in characterizing the spatial form and distribution pattern of the null trap area through octree segmentation and voxel merging. This provides a quantitative carrier for dynamically tracking the migration path of the null trap area and distinguishing the core area from the diffusion boundary. This in turn supports the precise design of adjustable null traps and the dynamic optimization of the interference suppression radius, ensuring the effective capture and targeted suppression of complex spatial interference in multi-serial port communications, and improving the accuracy and reliability of the test.

[0114] Obtain the difference in spatial null sink rates between two adjacent monitoring periods within N monitoring periods, as well as the displacement distance of the null sink space body between two adjacent monitoring periods;

[0115] The null spatial gradient of the null space body is calculated based on the difference in spatial null rate and the displacement distance of the null space body in two adjacent monitoring periods. ;

[0116] Obtain the interference coupling factor in the spatial null body based on the null spatial gradient , center position coordinates , and combined with the interference coupling factor to build a migration model to dynamically predict the position of the null space body ;

[0117] Preferably, by establishing the migration equation: Build a migration model to dynamically predict the location of the null space body ;

[0118] in, is the interference coupling factor, is the time interval of the monitoring cycle, is the zero-sink spatial gradient ;

[0119] like Figure 2 As shown in the figure, if the dynamic coupling factor in the monitoring period is higher than or equal to the preset coupling threshold, the null space body - The migration direction of the null-sink space body is extracted by constructing an expansion equation;

[0120] If the dynamic coupling factor during the monitoring period is lower than the preset coupling threshold, the dynamic coupling factor is continuously monitored;

[0121] Preferably, the coupling threshold is greater than 0.5, as shown by the formula: Get the expansion amount of the migration direction of the null space body ;

[0122] Wherein, k is the sensitivity proportional coefficient used to calibrate the expansion amount;

[0123] Preferably, in order to reduce interference suppression redundancy caused by over-expansion, k is preferably in the range of 0.3 to 0.7, with k = 0.5 being a typical value. The value range of k balances the sensitivity to migration trends and computational stability;

[0124] Position based on null space The dynamic migration path of the null-sink space body is established by combining the expansion amount of the migration direction of the null-sink space body;

[0125] Based on the dynamic migration path of the null-sink space body, the core area and diffusion boundary of the continuous null-sink area composed of the null-sink space body are extracted;

[0126] Extracting the position coordinates of the diffusion boundary and establishing a diffusion boundary coordinate set;

[0127] It can be understood that based on the dynamic migration trajectory of the null-sink space body, the dense area with a high null-sink rate and a continuous and stable existence during migration is defined as the core area (the core carrier of the null-sink characteristic), and the edge voxels outside the core area with a gradual change in null-sink rate (connecting the core and the external defect area) and dynamically expanding or contracting with migration are defined as the diffusion boundary; the three-dimensional coordinates of the diffusion boundary voxels are extracted and integrated into a diffusion boundary coordinate set according to the spatial adjacency relationship to capture the dynamically changing edge morphology of the null-sink area.

[0128] Example 2

[0129] like Figure 1 As shown, the present invention is a method for testing multiple serial ports of a 5G motherboard, which also includes the following steps:

[0130] S3. Dynamically predict the diffusion boundary coordinate set to obtain a null suppression radius, formulate an adjustable null trap based on the null suppression radius, perform dynamic adaptability analysis on the adjustable null trap, determine whether the update speed of the adjustable null trap matches the interference dynamics, and trigger an adaptive analysis signal if they do not match;

[0131] The method for dynamically predicting the diffusion boundary coordinate set is:

[0132] The antenna tilt angle and beam change rate of the 5G motherboard are collected during each monitoring period, and the null feature vector is constructed by combining the core area and diffusion boundary coordinate sets and the beam angle;

[0133] Combined with the null feature vector, the null suppression radius is obtained by establishing a mapping relationship between the null suppression radius and the null suppression radius through the random forest algorithm. ;

[0134] It will be understood by those skilled in the art that, within each monitoring period, the angular parameter of the antenna radiation direction is synchronously obtained as the antenna tilt angle, the rate of change of the beam pointing over time is obtained as the beam change rate, and the extracted core area and diffusion boundary coordinate sets are obtained, and the angle between the beam and the null zone is calculated to reflect the directional association between the beam coverage and the null zone;

[0135] Integrate the inclination angle, rate of change, core area center coordinates, diffusion boundary perimeter and area, and beam angle into a high-dimensional null feature vector;

[0136] Using null-sink feature vectors and actual null-sink suppression radius in historical monitoring data as sample pairs, a random forest algorithm is trained to learn the nonlinear mapping relationship between features and suppression radius. The predicted null-sink suppression radius is output through model inference to guide the dynamic adjustment of subsequent interference suppression strategies.

[0137] By formula: Establish a propagation constraint model to suppress the null radius Perform optimization to obtain the optimized null suppression radius r;

[0138] Among them, the method of formulating an adjustable zero trap based on the zero trap suppression radius is:

[0139] Based on the null suppression radius, a dynamic null trap is generated by a beamforming algorithm and used as an adjustable null trap.

[0140] Those skilled in the art will understand that, based on the null suppression radius, a beamforming algorithm is used to adjust the phase and amplitude of the 5G motherboard antenna array: by performing targeted modulation on the beam pattern within the spatial range defined by the null suppression radius, a null area with signal attenuation is formed in the interference direction corresponding to the core area and the diffusion boundary, and the spatial range of the null area can dynamically change with the null suppression radius, for example, the radius can be adjusted in real time due to changes in interference intensity and position. The null area thus generated with adjustable range is called an adjustable null trap, which achieves dynamic suppression of interference in a specific space.

[0141] Perform dynamic adaptability analysis on the adjustable zero trap to determine whether the update speed of the adjustable zero trap matches the interference dynamics. If not, generate an adaptive analysis signal.

[0142] The dynamic adaptability analysis of the adjustable zero trap is performed to determine whether the update speed of the adjustable zero trap matches the interference dynamics.

[0143] Obtain the rate of change of the beam pointing speed when the adjustable zero trap is generated in adjacent monitoring cycles to obtain the beam change rate ;

[0144] Obtain the null steering update frequency, perform ratio processing on the beam change rate and the null steering update frequency, and obtain the hysteresis null steering coefficient;

[0145] If the hysteresis zero sink coefficient is higher than the preset hysteresis zero sink threshold, it is determined that the 5G motherboard test is in the hysteresis zero sink state;

[0146] If it is in the hysteresis nulling state, the nulling depth and nulling rate of the core area corresponding to the continuous nulling area are obtained;

[0147] It will be understood by those skilled in the art that the zero trap update frequency refers to the number of times the adjustable zero trap is updated per unit time, which is obtained by recording the total number of times the adjustable zero trap is updated during the monitoring period and dividing the result by the monitoring duration;

[0148] The null depth is the degree of signal attenuation in the core area, which is calculated by measuring the difference between the field strength attenuation value in the core area and the field strength value in the normal communication area.

[0149] The core area zero sink rate is the frequency ratio of the smallest voxel in the core area in the zero sink area. The intensity adaptation coefficient is obtained by calculating the deviation ratio between the zero sink depth and the core area zero sink rate by counting the number of times each voxel in the core area is in the zero sink area during the monitoring period and averaging the ratio to the total monitoring number.

[0150] Different intensity adaptation levels are divided based on the intensity adaptation coefficient. If the intensity adaptation level is at the lowest adaptation level and is in a hysteresis zero-sink state, that is, the update speed does not match the interference dynamics, the adaptation analysis signal is triggered;

[0151] It should be explained that when the 5G motherboard test is in the delayed zero trap state and the intensity adaptation level is at the lowest adaptation level, it means that the interference source may move or change in intensity, and it is determined that the update speed of the adjustable zero trap does not match the interference dynamics.

[0152] S4. If the adaptive analysis signal is triggered, extract the secondary null sink index and beam change rate to determine the interference dynamic type, analyze the matching degree of different interference dynamic types to build a scheduling decision matrix, and formulate an optimization strategy for the step resolution;

[0153] Among them, the method of extracting the secondary index of null sink and the beam change rate to determine the interference dynamic type is:

[0154] If the adaptive analysis signal is triggered, the change rate of the core area zero sink rate is calculated to obtain the zero sink secondary index. ;

[0155] Based on the zero-sink secondary index and beam change rate , determine the disturbance dynamic type;

[0156] For example, if the beam change rate >5° / s and >0.1 / s, it is a strong dynamic interference;

[0157] like >5° / s and ≤0.1 / s, indicating mobile-dominated dynamic interference;

[0158] like ≤5° / s and >0.1 / s, intensity-dominated dynamic interference;

[0159] Obtain the interference source movement rate and current step resolution under each interference dynamic type;

[0160] The motion matching degree is obtained by comparing the moving speed of the interference source with the current step resolution.

[0161] The ratio of the secondary index of the null sink to the current step resolution is processed to obtain the spatial gradient of the interference intensity change;

[0162] Obtain the spatial gradient of the minimum interference intensity change in the 5G motherboard test in historical data and obtain the critical spatial gradient;

[0163] The deviation ratio of the spatial gradient of the interference intensity change and the critical spatial gradient is calculated to obtain the recognition matching degree;

[0164] Obtain motion matching and recognition matching under different interference types, and build a scheduling decision matrix for 5G motherboard testing based on physical constraints;

[0165] Formulate optimization strategies with different step resolutions based on the scheduling decision matrix;

[0166] Preferably, the method of formulating the optimization strategy of different step resolutions is: obtaining the tracking error rate and the suppression standard rate after formulating the optimization strategy, and normalizing the tracking error rate and the suppression standard rate respectively;

[0167] The normalized tracking error rate and the suppression target achievement rate are summed to obtain the adjustment effectiveness coefficient;

[0168] Based on the adjustment effectiveness coefficient, it is determined whether the adjustment has achieved the expected result. If it has not achieved the expected result, a second adjustment is made. If the adjustment has not achieved the expected result for M consecutive times, a hardware warning signal is triggered.

[0169] Preferably, M ≥ 3;

[0170] It's understandable that the adjustment process implements closed-loop optimization based on the adjustment effectiveness coefficient: The tracking error rate and the suppression compliance rate are first normalized (mapped to the interval [0,1]), and the two are summed to obtain the adjustment effectiveness coefficient, which is used to quantitatively evaluate the adjustment effect. If the coefficient does not reach the preset threshold, the system initiates a secondary adjustment, specifically by optimizing the step resolution, dynamically correcting the null suppression radius, or adjusting the beam pointing parameters to improve tracking accuracy and interference suppression capabilities. If the adjustment effectiveness coefficient still does not meet the standard after M consecutive adjustments (M ≥ 3), the system determines that the hardware performance is abnormal, triggering a hardware warning signal and prompting equipment maintenance.

[0171] Example 3

[0172] like Figure 3 As shown, the present invention is a system for testing multiple serial ports of a 5G motherboard, including the following modules:

[0173] Feature extraction module: used to set the step resolution, collect the spatial interference features of all 5G motherboards based on the step resolution, and establish a spatial feature group;

[0174] Null-sink analysis module: used to analyze spatial feature groups and extract continuous null-sink areas, perform spatial segmentation on the continuous null-sink areas to obtain null-sink spatial bodies, and predict the movement trend of the null-sink spatial bodies to obtain dynamic migration paths. Combined with the dynamic migration paths, the core areas and diffusion boundaries of the continuous null-sink areas are extracted.

[0175] Adaptive analysis module: used to dynamically predict the diffusion boundary coordinate set to obtain the null suppression radius, formulate an adjustable null trap based on the null suppression radius, perform dynamic adaptability analysis on the adjustable null trap, and determine whether the update speed of the adjustable null trap matches the interference dynamics. If not, an adaptive analysis signal is triggered;

[0176] Step optimization module: If the adaptive analysis signal is triggered, it is used to extract the secondary null sink index and beam change rate to determine the interference dynamic type, analyze the matching degree of different interference dynamic types to build a scheduling decision matrix, and formulate an optimization strategy for the step resolution.

[0177] The above is a detailed description of an embodiment of the present invention. However, the content is only a preferred embodiment of the present invention and should not be considered to limit the scope of the present invention. All equivalent changes and improvements made within the scope of the present invention should still fall within the scope of the present invention.

Claims

1. A method for testing multiple serial ports on a 5G motherboard, characterized by: The steps include: Set the step resolution, collect the spatial interference features of all 5G motherboards based on the step resolution, and establish a spatial feature group; Analyze the spatial feature group and extract continuous null-sink areas. Perform spatial segmentation on the continuous null-sink areas to obtain null-sink space bodies. Predict the movement trend of the null-sink space bodies to obtain dynamic migration paths. Combined with the dynamic migration paths, extract the diffusion boundary coordinate set of the continuous null-sink areas. Dynamically predict the diffusion boundary coordinate set to obtain the null suppression radius, formulate an adjustable null trap based on the null suppression radius, and perform dynamic adaptability analysis on the adjustable null trap to determine whether the update speed of the adjustable null trap matches the interference dynamics. If not, trigger an adaptive analysis signal; If the adaptive analysis signal is triggered, the secondary zero-sink index and beam change rate are extracted to determine the interference dynamic type. The matching degree of different interference dynamic types is analyzed to construct a scheduling decision matrix, and an optimization strategy for the step resolution is formulated.

2. A method for testing multiple serial ports on a 5G motherboard according to claim 1, characterized in that: The spatial feature group is established as follows: The 5G motherboard is moved according to the step resolution. The field strength attenuation, bit error rate, signal-to-noise ratio, and step distance of the start and end coordinates are collected. Gradient calculations are performed to obtain the field strength step gradient, bit error step gradient, and signal-to-noise step gradient. Feature extraction is performed on the three step gradients to obtain device features and combination features. Obtain the field intensity gradient variance and maximum error step gradient of all 5G motherboards in the area during the same monitoring period, and use the field intensity gradient variance and maximum error step gradient as interference features; Obtain interference features, combination features, and device features and integrate them according to position coordinates to obtain spatial interference features and establish a spatial feature group.

3. A method for testing multiple serial ports of a 5G motherboard according to claim 2, characterized in that: The method for obtaining the combined features and device features is as follows: Perform gradient correlation calculations on the field strength step gradient, bit error step gradient, and signal-to-noise step gradient of 5G motherboards i and j in the region, and obtain three gradient correlation coefficients, including the field strength step gradient, bit error step gradient, and signal-to-noise step gradient, as device characteristics. Among them, i and j are the numbers of the 5G motherboard; Based on the mean velocity vector of 5G motherboard i and j and the angle between the moving directions, interference coupling analysis is performed to extract the interference coupling factor, which is used as a combined feature.

4. The method for testing multiple serial ports of a 5G motherboard according to claim 1, wherein: The dynamic migration path is obtained as follows: Obtain the interference coupling factor in the spatial null body, build a migration model based on the null space gradient and center position coordinates, and combine the interference coupling factor to dynamically predict the position of the null space body; If the dynamic coupling factor within the monitoring period is higher than or equal to the preset coupling threshold, the expansion amount of the null space migration direction is extracted by constructing an expansion equation along the migration direction of the predicted position and the center position of the null space; A dynamic migration path of the null-sink space body is established based on the position of the null-sink space body and the expansion amount of the migration direction of the null-sink space body.

5. A method for testing multiple serial ports of a 5G motherboard according to claim 4, characterized in that: The method of obtaining the spatial null body is: The step distance of each 5G motherboard movement in the spatial feature group is obtained as a spatial grid. Based on the interference characteristics and device characteristics of the spatial grid, a continuous zero-sink criterion is established to extract the continuous zero-sink area. The octree segmentation algorithm is used in combination with the interference coupling factor of the combined features to recursively divide the continuous null region into pixels and obtain the minimum spatial voxel. Perform frequency analysis on the position corresponding to the minimum voxel in the monitoring period to obtain the voxel null rate; The voxel null sink connectivity boundary is set. If the voxel null sink rate of the minimum voxel in the adjacent space is in the voxel null sink connectivity boundary, the minimum voxels in the adjacent space are merged to establish the null sink space volume.

6. The method for testing multiple serial ports of a 5G motherboard according to claim 1, wherein: The method for determining whether the update speed of the adjustable zero trap matches the interference dynamics is: Obtain the beam change rate when generating adjustable zero traps and the hysteresis zero trap coefficients in adjacent monitoring periods; Determine whether the 5G motherboard test is in a hysteresis null state based on the hysteresis null coefficient. If it is in a hysteresis null state, obtain the null depth and null rate of the core area of ​​the continuous null area. The deviation ratio between the zero-sink depth and the core area zero-sink rate is calculated to obtain the strength adaptation coefficient; Different intensity adaptation levels are divided based on the intensity adaptation coefficient. If the intensity adaptation level is at the lowest adaptation level and is in a hysteresis zero sink state, it means that the update speed does not match the interference dynamics.

7. A method for testing multiple serial ports on a 5G motherboard according to claim 6, characterized in that: The adjustable zero trap is generated in the following way: The antenna tilt angle and beam change rate of the 5G motherboard are collected during each monitoring period, and the null feature vector is constructed by combining the core area and diffusion boundary coordinate sets and the beam angle; Combined with the null trap feature vector, the mapping relationship of the null trap suppression radius is established through the random forest algorithm to obtain the null trap suppression radius. Based on the null trap suppression radius, the dynamic null trap is generated through the beamforming algorithm and the dynamic null trap is used as an adjustable null trap.

8. The method for testing multiple serial ports of a 5G motherboard according to claim 1, wherein: The optimization strategy for the step resolution is formulated as follows: Construct a scheduling decision matrix and formulate optimization strategies with different step resolutions. Obtain the tracking error rate and suppression compliance rate after formulating the optimization strategy, and normalize the tracking error rate and suppression compliance rate respectively. The normalized tracking error rate and suppression compliance rate are summed to obtain the adjustment effectiveness coefficient to determine whether the adjustment meets expectations. If not, adjustments or warnings are made.

9. A method for testing multiple serial ports on a 5G motherboard according to claim 8, characterized in that: The scheduling decision matrix is ​​constructed as follows: Obtain the secondary null sink index and beam change rate, determine the interference dynamic type, and obtain the interference source movement rate and current step resolution under each interference dynamic type; The motion matching degree is obtained by comparing the moving speed of the interference source with the current step resolution. The ratio of the secondary index of the null sink to the current step resolution is processed to obtain the spatial gradient of the interference intensity change; Obtain the critical spatial gradient, calculate the deviation ratio between the spatial gradient of the interference intensity change and the critical spatial gradient, obtain the recognition matching degree, obtain the motion matching degree and recognition matching degree under different interference types, and construct the scheduling decision matrix for 5G motherboard testing in combination with physical constraints.

10. A system for testing multiple serial ports of a 5G motherboard, for implementing the method for testing multiple serial ports of a 5G motherboard according to any one of claims 1 to 9, characterized in that: Includes the following modules: Feature extraction module: used to set the step resolution, collect the spatial interference features of all 5G motherboards based on the step resolution, and establish a spatial feature group; Null sink analysis module: used to analyze spatial feature groups and extract continuous null sink areas, perform spatial segmentation on the continuous null sink areas to obtain null sink space bodies, and predict the movement trend of the null sink space bodies to obtain dynamic migration paths. Combined with the dynamic migration paths, the diffusion boundary coordinate set of the continuous null sink areas is extracted. Adaptation Analysis Module: This module is used to dynamically predict the loading boundary coordinate set to obtain the null suppression radius, formulate an adjustable zero trap based on the null suppression radius, and perform dynamic adaptability analysis on the adjustable zero trap to determine whether the update speed of the adjustable zero trap matches the interference dynamics. If not, an adaptation analysis signal is triggered. Step optimization module: If the adaptive analysis signal is triggered, it is used to extract the secondary null sink index and beam change rate to determine the interference dynamic type, analyze the matching degree of different interference dynamic types to build a scheduling decision matrix, and formulate an optimization strategy for the step resolution.

Images