CN-121985350-A - Method, device, system and storage medium for optimizing probe layout of air interface test

CN121985350ACN 121985350 ACN121985350 ACN 121985350ACN-121985350-A

Abstract

A probe layout optimization method, device, system and storage medium for air interface test are provided, wherein the method comprises training a preset training model according to a reference test data set obtained in advance to obtain a reference model, obtaining antenna characteristic data and probe characteristic data of current tested equipment, inputting the antenna characteristic data and the probe characteristic data into the reference model to obtain target probe layout information, carrying out instruction conversion according to the target probe layout information to obtain an axis control instruction, and sending the axis control instruction to a mechanical movement mechanism to drive at least one test probe to move to a corresponding target position in the probe layout information. By utilizing the physical information neural network, the computing efficiency and the global convergence of the probe position optimization are improved. And the condition number of the channel matrix is reduced from the physical aspect, the problem of the pathological matrix is effectively improved, and the stability and the measurement accuracy of the wireless cable construction are improved. The probe position is changed by driving the mechanical mechanism, so that the testing efficiency is improved.

Inventors

WANG HANG
YANG JING
ZHANG JIANHUA
CHEN RUI
HU ZHENYUN
TIAN LEI

Assignees

中国汽车工程研究院股份有限公司
北京邮电大学
中汽院(江苏)汽车工程研究院有限公司

Dates

Publication Date: 20260505
Application Date: 20260203

Claims (9)

1. The probe layout optimization method for the air interface test is characterized by comprising the following steps of: Training a preset training model according to a pre-acquired reference test data set to obtain a reference model, wherein the reference test data set comprises antenna radiation patterns of various reference tested devices and probe radiation patterns of a reference test probe, and the various reference tested devices have different sizes, antenna numbers and arrangement modes; Acquiring antenna characteristic data and probe characteristic data of current tested equipment; inputting the antenna characteristic data and the probe characteristic data into the reference model to obtain target probe layout information; Performing instruction conversion according to the target probe layout information to obtain an axis control instruction; And sending the shaft control instruction to a mechanical movement mechanism, and driving at least one test probe to move to a corresponding target position in the probe layout information.
2. The method for optimizing a probe layout for air interface testing according to claim 1, wherein training a preset training model according to a pre-acquired reference test data set to obtain a reference model comprises: Obtaining the reference test data set through a random generation algorithm or parametric modeling; Performing iterative training on the preset training model according to the reference test data set to obtain predicted probe layout information corresponding to the current iteration times; Acquiring a theoretical transmission matrix according to the predicted probe layout information and a free space propagation model; constructing a composite loss function according to the theoretical transmission matrix and driving to update the training model; And if the composite loss function meets a preset convergence condition, determining that the updated training model is the reference model, otherwise, adding one to the current iteration number, and returning to the step of carrying out iterative training on the preset training model according to the reference test data set to obtain the predicted probe layout information corresponding to the current iteration number.
3. The method for optimizing probe layout for air interface test according to claim 1, wherein said antenna characteristic data comprises geometrical position coordinates and antenna radiation patterns of a plurality of antenna units in said current device under test; The probe characteristic data includes at least a probe radiation pattern.
4. The probe layout optimization method for air interface testing according to claim 1, further comprising: acquiring a transmission matrix according to the probe layout information; obtaining a calibration matrix according to the transmission matrix, and constructing a wireless cable; and performing performance test on the current tested equipment according to the air interface test requirement and the wireless cable.
5. A control apparatus, characterized by comprising: the training module is used for training a preset training model according to a pre-acquired reference test data set to obtain a reference model, wherein the reference test data set comprises antenna radiation patterns of various reference tested equipment and probe radiation patterns of a reference test probe, and the various reference tested equipment has different sizes, antenna numbers and arrangement modes; the data acquisition module is used for acquiring antenna characteristic data and probe characteristic data of the current tested equipment; the data processing module is used for inputting the antenna characteristic data and the probe characteristic data into the reference model to obtain target probe layout information; The command conversion module is used for carrying out command conversion according to the target probe layout information, obtaining an axis control command and sending the axis control command to the mechanical movement mechanism, wherein the axis control command is used for driving the mechanical movement mechanism to move at least one test probe to a corresponding target position in the probe layout information.
6. An air interface test system is characterized by comprising a tested device, a measuring probe, a mechanical movement mechanism and the control device according to claim 5; The mechanical movement mechanism is a mechanical arm or an arch system with a three-dimensional position adjusting function; The measuring probe is respectively connected with a flange or a sliding block at the tail end of the mechanical movement mechanism and can be independently adjusted by the mechanical movement mechanism; the control device is in communication connection with the mechanical movement mechanism and the tested equipment.
7. An electronic device comprising a processor, a memory and a program stored on the memory and executable on the processor, the program when executed by the processor implementing the steps of the probe layout optimization method of air interface testing as claimed in any one of claims 1 to 4.
8. A computer readable storage medium, characterized in that the computer readable storage medium has stored thereon a computer program which, when executed by a processor, implements the steps of the probe layout optimization method of the air interface test of any of claims 1 to 4.
9. A computer program product comprising computer instructions which, when executed by a processor, implement the steps of the probe layout optimization method of air interface testing of any of claims 1 to 4.

Description

Method, device, system and storage medium for optimizing probe layout of air interface test Technical Field The application relates to The technical field of wireless communication testing, in particular to a probe layout optimization method, device and system for an Over-The-Air (OTA) test and a storage medium. Background As mobile communication technology evolves toward the fifth generation new air interface (Fifth Generation New Radio,5G NR) and more advanced systems, the hardware architecture of wireless devices has been radically transformed. Modern wireless devices generally adopt an integrated design of a Radio Frequency (RF) front end and an antenna array, and no conventional RF test port is reserved. Therefore, the air interface test has become the only feasible way to verify the radio frequency performance of the wireless equipment and ensure the product quality. The physical nature of OTA testing is to interact with the device under test (Device Under Test, DUT) by means of spatial radiation using a measurement probe disposed in a dark room or shielded box. In a multiple-input multiple-output (Multiple Input Multiple Output, MIMO), i.e., multiple-probe, multiple-antenna, test scenario, this complex spatial propagation process is mathematically described by a transmission matrix that characterizes the signal coupling characteristics from the transmitting end antenna elements to the receiving end probes. The mathematical properties of the transmission matrix, in particular its condition number, directly determine the numerical stability and measurement accuracy of the OTA test system. A good state (low condition number) transmission matrix means that the system has good linearity independence in the spatial dimension, which is critical for implementing a "wireless cable" connection technology. Wireless cable technology aims to construct a "virtual cable" equivalent to a physical wire in the air through spatial signal processing, thereby realizing accurate signal injection and measurement of a specific antenna port of a DUT. Its core algorithm relies on inverting (or pseudo-inverting) the transmission matrix to compute the spatial decoupling weights. However, if the transmission matrix exhibits morbidity (i.e., condition number is too large), meaning that the matrix is nearly singular, there is a high degree of correlation between rows or columns. In this case, the matrix inversion process is very unstable and any small noise or mechanical error in the measurement environment is strongly amplified in the operation. This will directly lead to the calculated decoupling weights containing a huge deviation, so that the constructed "virtual connection" cannot realize the expected port isolation, and the technical scheme cannot meet the requirements of the communication test on the signal-to-noise ratio and the isolation of the signal. The prior art improvement in the number of transmission matrix conditions relies primarily on mechanically moving the probe or DUT position, which is essentially a blind random search. Due to the lack of efficient utilization of DUT antenna array placement and scatterer geometry, the system cannot predict the correct direction to improve condition numbers, and can only be forced to take a large number of "trial and error" measurements using monte carlo tests (i.e., trial and error measurements by iteratively adjusting DUT positions, rotating turret angles, or randomly selecting a subset from a large number of redundant probes) or violent exhaustion. The lack of purposeful random movement not only greatly increases the time consumption of mechanical execution and data processing and seriously reduces the test efficiency, but also is difficult to ensure convergence to the optimal layout in a limited time, and cannot meet the high-efficiency test requirement of a large-scale automatic production line. Disclosure of Invention At least one embodiment of the application provides a probe layout optimization method, device and system for air interface test and a storage medium, which are used for solving the problems that the lack of purposeful random movement in the prior art not only greatly increases the time consumption of mechanical execution and data processing, seriously reduces the test efficiency and cannot meet the high-efficiency test requirement of a large-scale automatic production line. In order to solve the technical problems, the application is realized as follows: In a first aspect, an embodiment of the present application provides a probe layout optimization method for air interface testing, including: Training a preset training model according to a pre-acquired reference test data set to obtain a reference model, wherein the reference test data set comprises antenna radiation patterns of various reference tested devices and probe radiation patterns of a reference test probe, and the various reference tested devices have different sizes, antenna numbers and arrang