Environmental adaptive radio measurement calibration method and system

Abstract

The invention relates to an environment adaptive radio measurement calibration method and system. The method comprises the following steps: synchronously sampling temperature, humidity, air pressure and electromagnetic interference around radio equipment to obtain an environmental parameter vector; constructing a linear model and a nonlinear target model of a three-layer neural network structure according to the environmental parameter vector; solving an environmental parameter separation vector set by using the linear model and the nonlinear target model; executing one-dimensional iterative search optimization on the environment parameter separation vector set to obtain a target environment parameter vector; and inputting the target environment parameter vector into the nonlinear target model for calibration compensation calculation to obtain dynamic calibration compensation data. According to the method, the updating frequency is reduced when the environment is stable, the calibration precision is ensured, the system resource use is optimized, and the method is particularly suitable for wireless equipment with limited resources.

Description

Technical Field

[0001] The present invention relates to the technical field of radio metrology calibration, and in particular to an environment-adaptive radio metrology calibration method and system. Background Art

[0002] Radio metrology calibration is a key technical link to ensure stable equipment performance in the next generation of wireless communication systems. However, factors such as temperature, humidity, air pressure and electromagnetic interference in the actual environment have a significant impact on calibration accuracy. As the performance requirements of 5G and future communication technologies continue to increase, traditional calibration technology is difficult to adapt to the challenges brought by environmental variability. Especially in complex scenarios, the coupling effect and nonlinear influence of environmental factors cause the calibration error rate to increase sharply, resulting in reduced communication quality and system reliability.

[0003] Traditional radio metrology calibration methods mainly use three-dimensional (3D) multi-parameter compensation technology to compensate by establishing a mapping relationship between environmental factors and calibration parameters. However, this method has extremely high computational complexity, reaching O(N 3 ) level, which is difficult to achieve in resource-constrained real-time systems. Although the recently proposed two-dimensional (2D) search alternatives have reduced the computational burden to a certain extent, they still face the problem of high computational complexity in scenarios where environmental factors are interrelated, and the accuracy often cannot meet the requirements. Summary of the invention

[0004] The main purpose of the present invention is to provide an environment-adaptive radio metrology calibration method and system. The present invention reduces the update frequency when the environment is stable, which not only ensures the calibration accuracy but also optimizes the use of system resources. It is particularly suitable for wireless devices with limited resources.

[0005] To achieve the above object, the present invention provides an environment-adaptive radio measurement calibration method, comprising the following steps: Synchronously sample the temperature, humidity, air pressure and electromagnetic interference around the radio equipment to obtain an environmental parameter vector; Constructing a linear model and a nonlinear target model of a three-layer neural network structure according to the environmental parameter vector; Solving the environmental parameter separation vector set using the linear model and the nonlinear target model; Performing one-dimensional iterative search optimization on the environmental parameter separation vector set to obtain a target environmental parameter vector; The target environment parameter vector is input into the nonlinear target model to perform calibration compensation calculation to obtain dynamic calibration compensation data.

[0006] The present invention also provides an environment-adaptive radio metrology calibration system, comprising: A synchronous sampling module is used to synchronously sample the temperature, humidity, air pressure and electromagnetic interference around the radio equipment to obtain an environmental parameter vector; A construction module, used for constructing a linear model and a nonlinear target model of a three-layer neural network structure according to the environmental parameter vector; A solution module, used for solving a set of environmental parameter separation vectors using the linear model and the nonlinear target model; A search optimization module, used to perform one-dimensional iterative search optimization on the environmental parameter separation vector set to obtain a target environmental parameter vector; The calculation module is used to input the target environment parameter vector into the nonlinear target model to perform calibration compensation calculation to obtain dynamic calibration compensation data.

[0007] In summary, the technical solution provided by the present invention reduces the calibration calculation complexity from O(N) of the traditional 3D method to 3 ) is reduced to O(K·n·N), where K is the number of iterations, n is the number of environmental parameters, and N is the number of sampling points for each dimension search, which significantly reduces system resource consumption. The tilt projection operator technology is used to achieve accurate separation of the influence of different environmental factors. In particular, a special processing unit is designed for the interactive influence of temperature and humidity, so that the independent influence of each parameter in a complex environment can be accurately quantified, solving the problem that traditional methods are difficult to handle the coupling of environmental factors. The dual-model collaborative mechanism, in which the linear model provides initial estimates and the nonlinear target model performs accurate calculations, ensures both the calculation speed and the calibration accuracy, and significantly reduces the consumption of computing resources while maintaining similar performance to the most advanced methods. The dynamic calibration update cycle adjustment mechanism based on the environmental change rate automatically increases the update frequency when the environment changes rapidly and reduces the update frequency when the environment is stable, which not only ensures the calibration accuracy but also optimizes the use of system resources. It is particularly suitable for resource-constrained wireless devices. Through the sensitivity matrix analysis and the environmental parameter subset selection technology based on the information entropy criterion, the environmental parameters that have the most significant impact on the calibration accuracy are automatically identified and focused on, further reducing the computational burden while maintaining the calibration accuracy of key parameters. BRIEF DESCRIPTION OF THE DRAWINGS

[0008] Figure 1 It is a schematic diagram of the steps of an environment adaptive radio measurement calibration method in one embodiment of the present invention; Figure 2 It is a structural block diagram of an environment-adaptive radio metrology calibration system in one embodiment of the present invention.

[0009] The realization of the purpose, functional features and advantages of the present invention will be further explained in conjunction with embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0010] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0011] Reference Figure 1 This embodiment provides an environment-adaptive radio measurement calibration method, comprising the following steps: S1, synchronously sample the temperature, humidity, air pressure and electromagnetic interference around the radio equipment to obtain the environmental parameter vector; Among them, temperature sensors, humidity sensors, air pressure sensors and electromagnetic interference sensors are arranged at different positions of the radio equipment to build an environmental parameter acquisition network. The layout of the sensors needs to take into account the structural characteristics of the equipment and the environmental impact to ensure that the collected data can fully and accurately reflect the environmental status around the equipment, while avoiding local measurement errors caused by unreasonable sensor positions. A unified clock signal is set for the environmental parameter acquisition network so that all sensors collect data at a fixed sampling frequency to ensure the synchronization and time alignment characteristics of the data. In order to meet the real-time calibration requirements of the radio equipment, the setting of the sampling frequency needs to balance data accuracy and system resource consumption, and select a suitable fixed sampling frequency, such as ten times per second, to ensure accurate capture of the dynamic changes of environmental parameters and reduce calibration errors caused by asynchronous sampling. The collected raw environmental parameter data contains noise and outliers, and needs to be preprocessed to improve data quality. For temperature data, since the ambient temperature is affected by sudden fluctuations in a short period of time, low-pass filtering technology is used to remove high-frequency noise interference to ensure the stability and accuracy of temperature data. Select a suitable low-pass filter and set an appropriate cutoff frequency so that it can effectively filter out irrelevant high-frequency components while retaining the main trend of temperature changes. For humidity data and air pressure data, since these two types of environmental parameters are affected by factors such as air flow and equipment heat dissipation, filtering algorithms are applied to remove interference signals. Humidity data uses a specific filtering method to enable it to accurately reflect changes in environmental humidity, while the processing of air pressure data uses mean filtering to reduce errors caused by short-term fluctuations and make the data more stable and reliable. The processing of electromagnetic interference data is more complicated because there are many types of electromagnetic interference signals around radio equipment, including periodic interference, transient interference, and sudden abnormal signals. When processing electromagnetic interference data, a filtering method that can retain the main energy components is used to remove outliers and retain useful interference information, thereby avoiding unnecessary impact on radio calibration. After completing the filtering processing of different types of environmental parameters, all data are normalized to eliminate the differences in the numerical ranges between various parameters, so that they can be used for subsequent environmental impact modeling and compensation calculations under a unified scale. All parameters are mapped to the same numerical range to make them more suitable as input data. After normalization, temperature, humidity, air pressure, and electromagnetic interference data form standardized environmental parameter vectors.

[0012] S2, constructs a linear model and a nonlinear target model with a three-layer neural network structure according to the environmental parameter vector; Specifically, the environmental parameter vector is linearly transformed to construct a linear model structure and establish a linear mapping relationship between environmental parameters and calibration compensation. An environmental impact coefficient matrix and a bias vector are defined. By performing linear operations on the environmental parameter vector, the model can preliminarily estimate the impact of different environmental parameters on the calibration compensation value. Each element of the environmental impact coefficient matrix corresponds to the degree of influence of a certain environmental parameter on a specific calibration compensation value, and the bias vector is used to adjust the baseline value of the model output to compensate for the existing system error. The environmental impact coefficient matrix and bias vector of the linear model framework are optimized and trained by the least squares method to make the output of the model as close as possible to the actual compensation value of the historical calibration data. The optimization process of the least squares method solves the optimal coefficient matrix and bias vector by minimizing the sum of squares of the calibration error to obtain a linear model. A nonlinear target model with a three-layer neural network architecture is constructed according to the dimension of the environmental parameter vector to capture the complex nonlinear relationship between environmental parameters and calibration deviations. In order to ensure that the architecture of the neural network adapts to the characteristics of the environmental parameters, the number of nodes in the input layer of the neural network is set to the number of environmental parameters, so that each environmental parameter can be passed as an input signal to the subsequent layers of the network. The number of nodes in the hidden layer is set to twice the number of environmental parameters plus one to provide sufficient computing power while avoiding excessive complexity and waste of computing resources, while the number of nodes in the output layer is equal to the number of calibration parameters, so that the final output of the network is directly used for calibration compensation of radio equipment. In the process of constructing the nonlinear target model, the interaction between environmental parameters, especially the coupling effect of temperature and humidity on the calibration accuracy, is considered, so an additional interaction processing unit is introduced in the three-layer neural network architecture. By adding cross terms in the hidden layer, the model captures the nonlinear correlation between temperature and humidity and improves its adaptability to complex environmental conditions. The introduction of cross terms enables the network to not only learn the influence of a single environmental parameter, but also identify the trend of calibration deviation under the interaction of multiple parameters, thereby improving the accuracy of overall calibration compensation. The nonlinear target model framework is optimized by setting a fixed learning rate and training termination conditions so that the parameters of the model can accurately fit the historical calibration data of radio equipment. During the optimization process, a fixed learning rate is set to ensure that the model can converge stably during the training process, avoiding both oscillation caused by too high a learning rate and slow convergence caused by too low a learning rate. At the same time, a suitable training termination condition is set to prevent overfitting or insufficient training. The training termination conditions include the loss function change rate falling below a certain threshold, or the number of training rounds reaching the set maximum value. During the training process, the weights and biases of the neural network are adjusted through the batch gradient descent algorithm so that the output of the network matches the compensation value of the historical calibration data as much as possible, and finally the optimized nonlinear target model is obtained.

[0013] S3, using linear model and nonlinear target model to solve the environmental parameter separation vector set;It should be noted that the linear model is used to perform a preliminary mapping analysis on the environmental parameter vector to generate the initial independent impact estimate of the environmental parameters, and the basic structure of the oblique projection operator is constructed based on this. The linear model provides a quick preliminary estimate of the impact of the environmental parameters to reduce the computational complexity and accelerate the convergence speed of subsequent iterative optimization. After obtaining the preliminary estimate, the mapping characteristics of the nonlinear target model are combined to construct the oblique projection operator expression with residual terms to accurately describe the nonlinear impact of the environmental parameters. The oblique projection operator needs to use the output of the linear model as the initial value and correct it through nonlinear calculation to obtain a complete projection operator calculation framework so that it can more accurately describe the individual impact of the environmental parameters. In order to optimize the calculation accuracy of the oblique projection operator, the nonlinear target model is used to calculate the mapping difference before and after the environmental parameter projection, and the Lagrange multiplier method and conjugate gradient descent algorithm are used to minimize this difference to obtain the optimized residual term. The core of the optimization process is to ensure that the projection operator can accurately separate the impact of different environmental parameters on the calibration compensation, so that when multiple factors such as temperature, humidity, air pressure and electromagnetic interference are coupled with each other, the independent impact of a single environmental parameter can still be obtained. Since the nonlinear target model has stronger expressive power, it can capture more complex trends of environmental factors during the optimization process, and make necessary corrections to the initial estimate of the linear model to improve the accuracy of the final result. Based on the optimized residual term, the corresponding projection operators are constructed for temperature, humidity, air pressure and electromagnetic interference based on the nonlinear target model and the complete projection operator calculation framework. Since the linear model provides a preliminary estimate of the environmental parameters, the output of the linear model is used as the initial value in this process to accelerate the convergence process of the nonlinear projection operator. In this way, the computational complexity is reduced and the optimization process of the projection operator is ensured to be more stable. As the projection operators of different environmental parameters are constructed, a set of initial projection operators is formed, which contains a separate description of the influence of different environmental parameters and can reduce the mutual coupling between environmental parameters to a certain extent. The initial projection operator set is subjected to improved Gram-Schmidt orthogonalization to eliminate the mutual interference between different projection operators. The effect of orthogonalization is verified by using the nonlinear target model to ensure that all projection operators can maintain orthogonality so that they will not affect each other when applied. Through orthogonalization, the cross-influence between environmental parameters is reduced, so that each projection operator only retains the independent influence of specific environmental parameters without being disturbed by other parameters, thereby improving the accuracy of the final separation result. After orthogonalization, an orthogonalized oblique projection operator set is obtained. The environmental parameter vector is input into the orthogonalized oblique projection operator set, and the nonlinear target model is applied for projection mapping verification to ensure that the obtained separation vector accurately reflects the influence of a single environmental parameter on the calibration compensation.In this process, the nonlinear target model further corrects the projection mapping results through the feedback of calibration data, and ensures that each separation vector only contains the influence of the corresponding environmental parameters without being disturbed by other parameters. The independent influence of each environmental parameter is extracted, and a high-precision environmental parameter separation vector set is obtained.

[0014] S4, performing one-dimensional iterative search optimization on the environmental parameter separation vector set to obtain the target environmental parameter vector; Specifically, the calculation result of the linear model is set as the initial environmental parameter vector, and a reasonable search interval is set for each parameter in the initial vector to ensure that the search range covers the possible real values ​​of the environmental parameters. After setting the search interval, a fixed number of candidate values ​​are uniformly generated in the interval to form a parameter candidate value set, so that each candidate value represents a possible state of the current environmental parameter. In order to ensure the uniformity of the search, the candidate values ​​are uniformly distributed in the set search interval to cover enough possible values, while avoiding too sparse search precision or too dense calculation burden. After completing the construction of the parameter candidate value set, each candidate value is combined with other unchanged parameters and input into the nonlinear target model to calculate the corresponding calibration compensation value to obtain the candidate calibration result. The candidate calibration result is compared with the actual measured calibration deviation to evaluate the quality of each candidate value. The deviation degree is measured by calculating the mean square error corresponding to each candidate value to obtain the error evaluation result. The candidate value with smaller mean square error means that its corresponding environmental parameter value is closer to the actual situation. Therefore, in the next optimization process, the parameter candidate value with the smallest error is selected from the error evaluation result, and the current environmental parameter is updated to obtain the optimized environmental parameter vector after the current iteration. Since only one environmental parameter is optimized in each iteration, and other parameters are kept unchanged, the stability of the search process is guaranteed and the optimal solution is gradually approached. After completing the update of the environmental parameters, the optimized environmental parameter vector of the current iteration is terminated to determine whether to continue iterative optimization. The conditions for termination judgment include two aspects: one is whether the difference between the optimization results of the current iteration and the previous iteration is less than the preset threshold. If the optimized environmental parameter vector changes very little from the previous result, it is determined that it is close to the convergence state and the iteration is terminated; the second is to determine whether the maximum number of iterations has been reached. If the number of iterations has reached the set upper limit, the search is terminated to prevent infinite loops. In the iterative process, the search step size is reduced round by round to improve the search accuracy, so that a more refined search can be performed when approaching the optimal solution to obtain the target environmental parameter vector.

[0015] S5, inputting the target environment parameter vector into the nonlinear target model to perform calibration compensation calculation to obtain dynamic calibration compensation data.

[0016] Among them, the target environmental parameter vector is input into the nonlinear target model to calculate the compensation amount for the calibration deviation of the radio equipment. The nonlinear target model calculates the compensation values ​​of multiple calibration parameters, including frequency compensation value, power compensation value and phase compensation value, according to the different influence degrees of environmental parameters, combined with historical calibration data and optimized model weights, to form a calibration compensation vector. Since the error characteristics of radio equipment under different environmental conditions are different, when calculating the calibration compensation vector, the corresponding weight coefficient is assigned according to the importance of different calibration parameters, so that the compensation value output by the model can accurately match the actual working state of the equipment, and ensure that the adjustment range of different calibration parameters meets the equipment requirements, so as to achieve accurate compensation. In order to ensure that the compensation data adapts to the dynamic changes of the environment in real time, each parameter in the environmental parameter vector is continuously sampled, and its time derivative is calculated to obtain the change rate of the environmental parameter. By analyzing the time series data of the target environmental parameters, the temperature change rate, humidity change rate, air pressure change rate and electromagnetic interference change rate are calculated, and the environmental change rate vector is obtained, which reflects the instantaneous change trend of the environmental parameters and provides a basis for the dynamic adjustment of the calibration compensation value, so that the compensation strategy can be adaptively adjusted with environmental changes. The environmental change rate vector is compared with a preset change rate threshold to determine whether the calibration compensation value update condition is triggered, and an update trigger signal is generated. In order to avoid frequent adjustment of the calibration compensation value when the environment is relatively stable, a reasonable change rate threshold is set so that the compensation update is triggered only when the environmental change exceeds the set range, thereby ensuring the stability and computational efficiency of the system. If any item in the environmental change rate vector exceeds the threshold, the calibration update signal is triggered so that the compensation data can be adjusted according to the latest environmental parameters to ensure that the calibration accuracy of the radio equipment is not affected by environmental fluctuations. In the process of updating the calibration compensation value, the maximum relative change rate is calculated according to the environmental change rate vector, and an adaptive adjustment formula is constructed using an exponential decay function to determine the dynamic calibration update cycle. Since different environmental parameters have different change rates, when calculating the dynamic calibration update cycle, the maximum relative change rate of each parameter is comprehensively considered, and the compensation update frequency is adjusted by exponential decay, so that the update frequency is increased when the environment changes rapidly, and the update frequency is reduced when the environment is stable, thereby optimizing the system's computing resource allocation and avoiding unnecessary compensation adjustments. At the same time, through the construction of the adaptive adjustment formula, it is ensured that the update frequency of the calibration compensation data is always consistent with the environmental change rate, so that the adjustment of the compensation value is neither too frequent nor lagging behind the changes in the actual environment. After the dynamic calibration update cycle is determined, the calibration compensation vector is updated regularly according to the cycle, and the update frequency is automatically increased when the environment changes rapidly, and the update frequency is appropriately reduced when the environment is stable, so as to obtain timing-optimized calibration compensation data.Apply the timing optimized calibration compensation data to the calibration parameters of the radio equipment, and make real-time compensation adjustments to the frequency, power and phase to ensure that the equipment always maintains the optimal working state under different environmental conditions. Through the dynamic calibration compensation mechanism, the adaptability of radio equipment in complex environments is improved, while the measurement errors caused by environmental changes are reduced, the accuracy and reliability of radio measurement are improved, and the equipment always maintains a high-precision calibration state in long-term operation, and dynamic calibration compensation data is obtained.

[0017] The compensated radio parameter vector is normalized and compared with the standard parameter vector, and the ratio of the Euclidean distance between the two and the standard vector norm is calculated, which is used as a measure of the calibration error rate. The purpose of this step is to quantify the effect of the current calibration compensation so that subsequent optimization adjustments can be optimized based on specific error feedback, thereby improving the adaptability of the overall system. After calculating the calibration error rate, the change gradient of each calibration parameter in the calibration compensation vector relative to each environmental parameter is calculated, and the partial derivative absolute value matrix is ​​constructed based on this to quantify the impact of environmental factors on the calibration accuracy. This matrix is ​​the sensitivity matrix. The construction of the sensitivity matrix can provide environmental parameter impact analysis, so that the system can accurately identify which environmental factors have a greater impact on the radio calibration results, and how different calibration parameters change under the influence of environmental factors. After the sensitivity matrix is ​​calculated, the importance weights of the calibration parameters are weighted and summed with the corresponding elements in the sensitivity matrix to calculate the contribution of each environmental parameter to the overall calibration accuracy and obtain the environmental parameter impact factor. Since different calibration parameters have different importance to the stability and accuracy of radio equipment, the relative weight of calibration parameters is considered when calculating the environmental parameter influencing factors to ensure that the final calculation results can accurately reflect the impact of environmental factors on the overall calibration accuracy. The environmental parameter influencing factors are arranged in descending order, and the top-ranked environmental parameters are selected to form a candidate subset to obtain the preliminary selection results of the environmental parameters. Based on the preliminary selection results of the environmental parameters, information entropy analysis is performed to measure the amount of information contained in the candidate subset, and the optimization objective function is constructed in combination with the penalty term of the subset size to reduce unnecessary computational complexity while ensuring the calibration accuracy. The role of information entropy analysis is to evaluate the effective information content of environmental parameters and eliminate redundant information, while the introduction of the subset size penalty term is to prevent the selection of too many environmental parameters, which leads to an increase in computational complexity. After constructing the optimization objective function, the size of the optimal subset is determined by minimizing the overall evaluation index to obtain the optimal environmental parameter subset. The tilted projection operator design and one-dimensional iterative search optimization are re-executed for the optimal environmental parameter subset to construct a radio equipment target calibration system. By redesigning the tilt projection operator, the influence of each environmental parameter is ensured to be accurately separated in the nonlinear target model. At the same time, the one-dimensional iterative search optimization method is combined to improve the calibration accuracy and reduce the computational complexity.

[0018] In one example, the temperature, humidity, air pressure, and electromagnetic interference around the radio device are synchronously sampled to obtain an environmental parameter vector, including: Arrange temperature sensors, humidity sensors, air pressure sensors and electromagnetic interference sensors at different locations of the radio equipment to form an environmental parameter collection network; A unified clock signal is set for the environmental parameter collection network and a fixed sampling frequency is used for synchronous data collection to obtain the original environmental parameter data; Apply a low-pass filter to the temperature data in the original environmental parameter data to remove high-frequency noise interference and obtain temperature filtered data; Filtering algorithms are respectively executed on the humidity data and the air pressure data in the original environmental parameter data to remove interference signals, thereby obtaining humidity filter data and air pressure filter data; For the electromagnetic interference data in the original environmental parameter data, energy component preservation filtering is applied to remove abnormal values ​​to obtain electromagnetic interference filtering data; The temperature filter data, humidity filter data, air pressure filter data and electromagnetic interference filter data are normalized to obtain an environmental parameter vector.

[0019] In this example, an environmental parameter acquisition network is established, and the temperature sensor, humidity sensor, air pressure sensor, and electromagnetic interference sensor are reasonably arranged according to the structural characteristics, working characteristics, and spatial distribution characteristics of the radio equipment. For example, the temperature sensor is arranged near the power amplifier of the equipment to capture the heat changes generated during the operation of the equipment in real time; the humidity sensor is placed near the vent of the equipment to monitor the humidity change trend inside the equipment; the air pressure sensor is usually arranged on the top of the equipment or in the air intake area to monitor the instantaneous fluctuation of the air pressure; and the electromagnetic interference sensor is arranged near the signal input and output port or the sensitive circuit area to capture the electromagnetic environment changes that interfere with the equipment in real time. A unified high-precision clock signal source is introduced into the acquisition network, and the signal source is distributed to all sensor nodes in the network at the same time, so that all sensors are based on the same clock signal and perform precise and synchronous data acquisition according to a uniformly set fixed sampling frequency, ensuring the data timing alignment and consistency between environmental parameters, and obtaining the original environmental parameter data. Different parameter features in the original environmental parameter data are processed separately. For example, temperature data is affected by high-frequency interference (such as wind speed fluctuations or local instantaneous heating inside the equipment). Therefore, a high-order Butterworth low-pass filter algorithm is used to suppress the high-frequency noise components in the sensor raw data that exceed a specific cutoff frequency, and extract temperature filter data that can truly reflect the trend of ambient temperature changes. Similarly, for humidity and air pressure data, since they are susceptible to external random interference and unstable factors of the sensor itself, different algorithms such as Chebyshev filtering and sliding window filtering are used to retain the low-frequency change trend of the data itself, remove high-frequency random interference, and obtain stable humidity and air pressure filter data. Due to its inherent complexity, electromagnetic interference data manifests itself as irregular bursts of abnormal pulses or continuous periodic interference. Therefore, traditional filtering methods are difficult to effectively remove these abnormal interference signals. For such data, an energy component preservation filtering algorithm is used, such as decomposition and reconstruction through an improved wavelet transform method: in represents the data reconstruction result after electromagnetic interference filtering, Represents the first The wavelet coefficients of the layer, Indicates The energy level of the layer coefficient, represents the scale range that mainly contains energy after decomposition, The energy threshold obtained by training based on historical data is used to filter out abnormal wavelet components; this method can effectively extract the main electromagnetic interference signal and exclude occasional abnormal values ​​to obtain pure electromagnetic interference filtering data. The temperature filtering data, humidity filtering data, pressure filtering data and electromagnetic interference filtering data are normalized to unify the dimensions and eliminate the differences in numerical scales between different data. The dynamic adaptive normalization formula is used: in, Represents the normalized environmental parameters, It is The original value of the environmental parameter after filtering, and The historical data are The long-term mean and variance of the parameters, To prevent division by zero for small values, is an adaptive adjustment factor, which is determined through historical data training and is used to adjust the sensitivity of normalization. , humidity parameters , air pressure parameters and electromagnetic interference parameters Combined to form the final environmental parameter vector .

[0020] In one example, a linear model and a nonlinear target model of a three-layer neural network structure are constructed according to an environmental parameter vector, including: Performing linear transformation on the environmental parameter vector and constructing a linear model structure, establishing a linear mapping relationship between the environmental parameters and the calibration compensation by setting the environmental impact coefficient matrix and the bias vector, and obtaining a linear model framework; The environmental impact coefficient matrix and bias vector of the linear model framework are optimized and trained by the least square method to obtain a linear model; A three-layer neural network architecture is constructed according to the dimension of the environmental parameter vector, by setting the number of input layer nodes equal to the number of environmental parameters, the number of hidden layer nodes equal to two times the number of environmental parameters plus one, and the number of output layer nodes equal to the number of calibration parameters; Adding an interactive processing unit for temperature and humidity to the three-layer neural network architecture, and capturing the coupling effect between environmental parameters by setting additional cross terms in the three-layer neural network architecture, a nonlinear target model framework is obtained; The nonlinear target model framework is optimized by setting a fixed learning rate and training termination conditions to obtain a nonlinear target model.

[0021] In this example, a linear transformation method is used to establish a basic linear model framework. Assume that the environmental parameter vector Depend on Environmental parameters such as temperature, humidity, air pressure and electromagnetic interference are used to construct an environmental impact coefficient matrix. (Dimensions are ) and a bias vector (Dimensions are ), used to establish the calibration parameter vector The calculation formula is: in, represents the calibration compensation vector, each element of which corresponds to a different calibration parameter of the radio device, such as frequency compensation, power compensation, and phase compensation. Elements Indicates The environmental parameters The influence of the calibration compensation parameters, and the bias vector It is used to correct the system error that is not taken into account due to environmental factors. In order to optimize the linear model framework so that it can accurately reflect the impact of environmental parameters on calibration compensation, optimization training is performed based on the least squares method. Assume that the historical measurement data provides a set of known environmental parameters and its corresponding true calibration compensation , then the optimization goal is to minimize the error: in, represents the number of samples of historical data, Represents the Euclidean norm. By solving the optimization objective, the optimal environmental impact coefficient matrix is ​​obtained and the bias vector , forming an optimized linear model. After completing the construction of the linear model, considering that the influence of environmental parameters on calibration compensation has significant nonlinear characteristics, such as the coupling of temperature and humidity will cause complex changes in radio equipment parameters, a three-layer neural network architecture is constructed to establish a nonlinear target model. The number of nodes in the input layer of the neural network is equal to the number of environmental parameters. , the number of output layer nodes is equal to the number of calibration parameters , and the number of nodes in the hidden layer is set to , in order to provide sufficient nonlinear fitting capabilities while avoiding the increase in computational cost caused by excessive complexity of the model. In order to capture the interactive effects between environmental parameters such as temperature and humidity, cross terms are introduced in the hidden layer. Assuming that the temperature parameter and humidity parameters To have a synergistic effect, an additional cross term is introduced into the calculation formula of the hidden layer: in, Represents the hidden layer Node output, is the activation function (such as ReLU), is the weight vector, is the weight factor for the temperature-humidity interaction, is the bias term. This design can enhance the model's ability to express the nonlinear coupling effect of environmental factors, making the model's predictions more accurate under different environmental conditions. When training a neural network, in order to ensure the stability and convergence of the model, a fixed learning rate is set. And introduce training termination conditions, such as: in, is the calibration offset value predicted by the model, is the error tolerance set. By setting a reasonable learning rate And termination conditions are set to prevent gradient oscillation or overfitting problems during training, thereby ensuring that the final nonlinear target model can accurately characterize the impact of environmental parameters on radio equipment calibration compensation.

[0022] In one example, solving a set of environmental parameter separation vectors using a linear model and a nonlinear target model includes: A linear model is used to perform a preliminary mapping analysis on the environmental parameter vector, generate the initial independent impact estimate of the environmental parameter, and obtain the basic structure of the oblique projection operator; Combining the mapping characteristics of the nonlinear target model, the expression of the oblique projection operator with residual terms is constructed, and the output of the linear model is introduced as the initial value into the nonlinear calculation to obtain a complete projection operator calculation framework; The nonlinear target model is used to calculate the mapping difference before and after the environmental parameters are projected, and the Lagrange multiplier method and conjugate gradient descent algorithm are used to minimize the difference to obtain the optimized residual term. Based on the nonlinear target model and the complete projection operator calculation framework, corresponding projection operators are constructed for temperature, humidity, air pressure and electromagnetic interference respectively. The initial estimate provided by the linear model is used to accelerate the convergence process and obtain the initial projection operator set. The improved Gram-Schmidt orthogonalization is performed on the initial projection operator set, and the orthogonal effect is verified using a nonlinear target model to eliminate the mutual interference between the projection operators and obtain an orthogonalized oblique projection operator set. The environmental parameter vector is input into the orthogonalized oblique projection operator set, and the nonlinear target model is applied to perform projection mapping verification, extract the independent influence of each environmental parameter, and obtain the environmental parameter separation vector set.

[0023] In this example, a linear model is used to perform a preliminary mapping analysis on the environmental parameter vector to generate initial independent impact estimates of the environmental parameters, and the basic structure of the oblique projection operator is established based on this. The environmental parameter vector is set to ,in represents temperature, Represents humidity, Represents air pressure, represents electromagnetic interference, and the calibration compensation vector Indicates the compensation value that the radio equipment needs to adjust, such as frequency, power, and phase. Use the linear mapping model: in, is the environmental influence coefficient matrix, which represents the linear influence of each environmental parameter on the calibration compensation. is the bias vector, which represents the compensation term of the system inherent error. Through this linear model, the preliminary impact estimate of each environmental parameter is obtained, forming a linear decomposition of the environmental parameter impact. However, since the impact of environmental parameters on radio calibration compensation has nonlinear characteristics, the tilt projection operator expression with residual terms is constructed in combination with the mapping characteristics of the nonlinear target model to optimize the independent impact estimate of the environmental parameters. The nonlinear target model is defined as: in, is a nonlinear mapping function, which indicates the nonlinear relationship between environmental parameters and calibration compensation, and is the residual term, which represents the nonlinear error part that the linear model cannot accurately describe. In order to use the output of the linear model as the initial value to introduce nonlinear calculation, the oblique projection operator is constructed: in, Representative The projection operator of the environmental parameters, is the independent influence value of environmental parameters estimated by the linear model, and represents the residual term of the environmental parameters in the nonlinear mapping, is the residual adjustment factor, which indicates the magnitude of the nonlinear correction. In order to optimize the oblique projection operator so that it can more accurately separate the independent effects of environmental parameters, the mapping difference before and after the environmental parameters are projected is calculated, and the Lagrange multiplier method and conjugate gradient descent algorithm are used to optimize it to minimize the projection error. Define the optimization objective: in, For the A sample of historical environmental parameters, represents the calculation result of the nonlinear target model after applying the projection operator, while Represents the calculation result under the original environmental parameter input. The optimization problem is solved by the Lagrange multiplier method and adjusted by the conjugate gradient descent algorithm. In order to minimize the error, the optimized residual term is obtained, and the calculation framework of the projection operator is further improved. Based on the optimized nonlinear projection operator, corresponding projection operator sets are constructed for temperature, humidity, air pressure and electromagnetic interference. Since the linear model provides a preliminary estimate of the environmental parameters, the linear model output is used as the initial projection value to accelerate the convergence of the nonlinear optimization process. This effectively reduces the computational complexity and enables the projection operator to converge to the optimal solution faster, forming an initial projection operator set. An improved Gram-Schmidt orthogonalization is performed on the initial projection operator set to eliminate the mutual influence between the projection operators. Assume that the initial projection operator set is ,in The projection operator representing a certain environmental parameter is processed by the following orthogonalization formula: in, represents the orthogonalized projection operator, and Represents the inner product relationship of the two projection operators. In this way, the orthogonality between different projection operators is ensured, thereby avoiding the cross-interference of the environmental parameters and improving the separation accuracy. After the orthogonalization is completed, the environmental parameter vector Input the orthogonalized oblique projection operator set and apply the nonlinear target model for projection mapping verification to ensure that the independent influence of each environmental parameter is correctly extracted. The extracted independent influence values ​​of environmental parameters constitute the environmental parameter separation vector set: in, They represent the independent impact values ​​of temperature, humidity, air pressure and electromagnetic interference respectively, and do not include the interference of other parameters.

[0024] In one example, a one-dimensional iterative search optimization is performed on the environmental parameter separation vector set to obtain a target environmental parameter vector, including: The calculation result of the linear model is set as the initial environment parameter vector, and a search interval is set for each parameter in the initial environment parameter vector, and a fixed number of uniformly distributed candidate values ​​are generated within the search interval to obtain a set of parameter candidate values; Input each candidate value in the parameter candidate value set into the nonlinear target model in combination with other unchanged parameters, calculate the corresponding calibration compensation value, and obtain a candidate calibration result; Compare the candidate calibration results with the actual measured calibration deviations, calculate the mean square error of each candidate value, and obtain the error evaluation result; Select the parameter candidate value corresponding to the minimum error from the error evaluation results, update the current environment parameters, and obtain the optimized environment parameter vector of the current iteration; The iteration termination judgment is performed on the optimized environment parameter vector of the current iteration. When the difference between the results of two adjacent iterations is less than the preset threshold or the maximum number of iterations is reached, the iteration is terminated. At the same time, the search step size is reduced round by round to obtain the target environment parameter vector.

[0025] In this example, a linear model is used to calculate an initial environment parameter vector to provide a starting point for the optimization search. Assume that the environment parameter vector is ,in represents temperature, Represents humidity, Represents air pressure, represents electromagnetic interference, and its initial estimate is calculated by a linear model: in, is the environmental influence coefficient matrix, which represents the influence of environmental parameters on calibration compensation. is the bias term used to compensate for system errors. Through this linear model, the initial environmental parameter vector is obtained , and use this as the starting point for optimization iteration. After obtaining the initial environmental parameter vector, set the search interval for each parameter in the vector to ensure that the optimization process is carried out within a reasonable range. The method of setting the search interval is based on the distribution characteristics of historical data and the range of environmental parameter changes allowed by the system. For example, if the temperature parameter If the change in the past data range is small, the search interval is set narrower. If the change is large, a wider search interval is set. In order to generate uniformly distributed candidate values ​​within the search interval, the following formula is used: in, Indicates The environmental parameters Candidate values, is the estimated value of the parameter in the previous iteration, is the search step length, is the number of candidate values. This formula ensures that the candidate values ​​are evenly distributed within the set interval to provide comprehensive optimization options. After generating a set of parameter candidate values, each candidate value is combined with other unchanged parameters and input into the nonlinear target model to calculate the corresponding calibration compensation value. The nonlinear target model is expressed as: in, is the calibration compensation vector, represents a nonlinear target model that can accurately describe the impact of environmental parameters on radio calibration compensation. After the calculation, a set of candidate calibration results are obtained. In order to evaluate the quality of these candidate calibration results, the mean square error between them and the actual measured calibration deviation is calculated. The mean square error is defined as: in, is the actual measured calibration compensation value, Candidate calibration values ​​calculated for the nonlinear target model, is the number of calibration parameters. It reflects the degree of fit of the candidate environmental parameters to the calibration compensation. The smaller the error, the closer the environmental parameter value is to the actual situation. After completing the error evaluation, select the candidate value with the smallest mean square error from the error evaluation results, and update the current environmental parameters to make them converge to the optimal solution. The update formula is: in, is the environmental parameter value after the current iteration update, which corresponds to the candidate value with the smallest mean square error. After each round of iteration, the optimization result is terminated to decide whether to continue the optimization. The termination condition includes two aspects. One is whether the difference between the environmental parameter vector of the current iteration and the result of the previous iteration is less than the preset threshold: in, is the termination threshold, which is used to control the optimization accuracy. If this condition is met, the iteration is stopped and the final optimization environment parameter vector is output. Another termination condition is that the number of iterations reaches the maximum value. , to prevent the optimization process from falling into an infinite loop. In order to improve the optimization accuracy, the search step size is gradually reduced in each iteration, so that the optimization process can search a larger range in the early stage and make more precise adjustments in the later stage. The update strategy of the search step size adopts exponential decay: in, For the The search step size during round iteration, is the initial search step length, is the attenuation factor. This method ensures that the search step gradually converges in the later stage of optimization, thereby improving the accuracy of the final environmental parameter vector. Through the above steps, the target environmental parameter vector is obtained, which can accurately reflect the real environmental state around the radio equipment and provide the optimal compensation parameters for radio measurement calibration.

[0026] In one example, the target environment parameter vector is input into the nonlinear target model for calibration compensation calculation to obtain dynamic calibration compensation data, including: The target environment parameter vector is input into the nonlinear target model for compensation calculation, and a frequency compensation value, a power compensation value and a phase compensation value are generated according to different weight coefficients of multiple calibration parameters to obtain a calibration compensation vector; Calculate the time derivative of the continuous sampling value of each parameter in the environmental parameter vector, generate the temperature change rate, humidity change rate, air pressure change rate and electromagnetic interference change rate, and obtain the environmental change rate vector; Compare the environmental change rate vector with a preset change rate threshold to determine whether a calibration compensation value update condition is triggered, and obtain an update trigger signal; The maximum relative change rate is calculated based on the environmental change rate vector, and an adaptive adjustment formula is constructed using an exponential decay function to obtain a dynamic calibration update cycle. The calibration compensation vector is updated regularly according to the dynamic calibration update cycle, the update frequency is increased when the environment changes rapidly, and the update frequency is reduced when the environment is stable, so as to obtain the calibration compensation data with optimized timing; The timing-optimized calibration compensation data is applied to the calibration parameters of the radio equipment, and the frequency, power and phase are compensated and adjusted in real time to obtain dynamic calibration compensation data.

[0027] In this example, the target environment parameter vector is input into the nonlinear target model to calculate the calibration compensation value and obtain the calibration compensation vector for radio equipment calibration. The optimized target environment parameter vector is set to ,in represents the optimized temperature, Represents the optimized humidity, Indicates the optimized air pressure, Represents the optimized electromagnetic interference. Nonlinear target model The calibration compensation value is calculated by a deep neural network or a nonlinear mapping method, where the different weight coefficients of each calibration parameter determine its contribution to the final compensation value. The calculation formula is: in, Represents the calibration compensation vector, including frequency compensation , Power compensation and phase compensation is the calibration weight matrix whose elements Indicates The environmental parameters The degree of influence of the calibration compensation value; is the bias vector used to correct system errors. Since the relationship between environmental parameters and calibration compensation is usually nonlinear, using a nonlinear model for compensation calculation can improve compensation accuracy and ensure the stability of radio equipment under complex environmental conditions. After calculating the calibration compensation vector, in order to ensure that the calibration compensation can respond to environmental changes in a timely manner, the environmental parameter vector Perform dynamic monitoring and calculate its time derivative to obtain the environmental change rate vector By performing time differentiation on the continuous sampling values ​​of the environmental parameters, we obtain: in, represents the rate of change of temperature, represents the rate of change of humidity, is the rate of change of air pressure, Represents the rate of change of electromagnetic interference. The sliding window method is used to calculate the environmental change rate to reduce the impact of short-term fluctuations on the calculation results. The environmental change rate vector is compared with the preset change rate threshold vector Compare to determine whether it is necessary to trigger the update of the calibration compensation value. The trigger condition is: When this condition is met, an update trigger signal is generated, indicating that the current environmental change has exceeded the threshold set by the system, indicating that the current calibration compensation is no longer applicable to the current environment and a new calibration calculation is required. After determining the calibration compensation update trigger signal, in order to optimize the timing of the calibration update, it is necessary to calculate the maximum relative change rate and construct an adaptive adjustment formula based on the exponential decay function to calculate the dynamic calibration update cycle. The maximum relative change rate is defined as: Based on this, the dynamic calibration update period is calculated: in, is the maximum update period, which indicates the longest calibration interval when the environment changes very slowly; is the minimum update period, indicating the shortest calibration interval when the environment changes drastically; is an adjustment factor that controls the speed of change of the calibration update frequency. This adaptive update strategy ensures that the calibration compensation is frequently updated when the environment changes rapidly, while reducing unnecessary consumption of computing resources when the environment is stable. For example, if a sudden change in temperature and humidity is detected, the calibration compensation value should be adjusted quickly, and when the environment is in a stable state, the calibration update frequency is reduced to reduce energy consumption and computing burden. After determining the dynamic calibration update cycle, the calibration compensation vector is updated regularly according to the cycle to optimize the calibration timing. When the environment changes rapidly, the system automatically increases the update frequency of the calibration compensation so that the radio equipment can quickly adapt to environmental changes; when the environment is stable, the system reduces the update frequency to reduce unnecessary compensation adjustments and ensure stable operation of the system. The optimized calibration compensation data is applied to the calibration parameters of the radio equipment for real-time compensation adjustments of frequency, power and phase. For frequency compensation, the following dynamic calibration formula is used: in, is the frequency after compensation, is the original frequency, is the temperature-humidity interaction coefficient, is the compensation adjustment factor. Similarly, the power compensation and phase compensation are adjusted using corresponding nonlinear calculation formulas to ensure that the radio equipment can maintain high-precision measurement performance under different environmental conditions.

[0028] In one example, the environment adaptive radio metrology calibration method further includes: The radio parameter vector measured after compensation is normalized and compared with the standard parameter vector, and the ratio of the Euclidean distance between the two and the norm of the standard vector is calculated to obtain the calibration error rate; The change gradient of each calibration parameter in the calibration compensation vector relative to each environmental parameter is calculated, and the absolute value matrix of partial derivatives between the calibration parameters and environmental factors is constructed to obtain the sensitivity matrix; The importance weights of the calibration parameters are weighted and summed with the corresponding elements of the sensitivity matrix, and the contribution of each environmental parameter to the overall calibration accuracy is calculated to obtain the environmental parameter influencing factor; Arrange the environmental parameter influencing factors in descending order, select the top-ranked environmental parameters to form a candidate subset, and obtain the preliminary selection results of the environmental parameters; Based on the preliminary results of environmental parameters, information entropy analysis is performed, and the optimization objective function is constructed in combination with the subset size penalty term. The optimal subset size is determined by minimizing the overall evaluation index, and the optimal environmental parameter subset is obtained. The tilt projection operator design and one-dimensional iterative search optimization are re-executed for the optimal subset of environmental parameters to construct a target calibration system for radio equipment.

[0029] In this example, the compensated radio parameter vector is normalized and compared with the standard parameter vector, and the ratio of the Euclidean distance between the two and the standard vector norm is calculated to obtain the calibration error rate. The measured radio parameter vector is assumed to be ,in represents the frequency compensation value, represents the power compensation value, represents the phase compensation value, and the standard parameter vector Represents the compensation value that the system should have under ideal conditions. In order to measure the compensation error, normalization is performed: in, and Represent the mean and standard deviation of the standard parameter vector respectively. Calculate the normalized Euclidean distance between the two: Calculate the calibration error rate: in, It reflects the degree of deviation between the current calibration parameters and the standard parameters. The smaller the value, the higher the calibration accuracy. After calculating the calibration error rate, the influence of environmental parameters on the calibration compensation is analyzed. Therefore, the change gradient of each calibration parameter in the calibration compensation vector relative to each environmental parameter is calculated, and the absolute value matrix of partial derivatives between the calibration parameters and environmental factors is constructed, that is, the sensitivity matrix . Set the environmental parameter vector to ,in represents temperature, Represents humidity, Represents air pressure, represents electromagnetic interference, then the elements of the sensitivity matrix are defined as: in, Representative The environmental parameters The sensitivity of the calibration compensation value is calculated. By calculating this matrix, the influence of different environmental parameters on the calibration compensation is determined. In order to analyze the contribution of different environmental parameters to the calibration accuracy, the importance weight of the calibration parameters is With the sensitivity matrix Perform a weighted sum operation to calculate the impact factor of each environmental parameter: in, Indicates The contribution of each environmental parameter to the overall calibration accuracy is calculated. After calculating the environmental parameter influencing factors, they are sorted in descending order, and the top-ranked environmental parameters are selected to form a candidate subset to obtain the preliminary selection results of the environmental parameters. After determining the preliminary environmental parameters, in order to optimize the selection of the subset, an optimization objective function is constructed based on information entropy analysis and combined with the subset size penalty term to determine the optimal subset size. The candidate subset is set to , then the information entropy is defined as: in, Representative The probability of contribution of each environmental parameter to the calibration error. The optimization goal is to minimize the overall evaluation index: in, is the subset size penalty factor, which is used to control the size of the environmental parameter subset so that it can reduce the computational complexity while ensuring the calibration accuracy. By solving this optimization problem, the optimal environmental parameter subset is determined. After determining the optimal subset of environmental parameters, the tilt projection operator design and one-dimensional iterative search optimization are re-executed for the subset to construct the final radio equipment target calibration system. The optimization goal of the tilt projection operator is to ensure that the independent impact of environmental parameters on calibration compensation is maximized, so its expression is: in, Representative The projection operator of the environmental parameters, is the residual term, is the projection weight. In order to optimize the tilted projection operator, the Lagrange multiplier method is used to minimize the mean square error of the residual term, and the Gram-Schmidt orthogonalization method is combined to keep the projection operators of different environmental parameters orthogonal, thereby reducing the mutual interference between parameters. After the projection operator is optimized, the target environmental parameter vector is optimized using a one-dimensional iterative search. , optimize one parameter each iteration while keeping other parameters unchanged, and select the optimal solution by minimizing the mean square error criterion: in, represents the iteration round, Represents the true calibration compensation value. The iteration termination condition is: Or the maximum number of iterations is reached Through the above optimization process, the target calibration system of radio equipment is finally constructed, and it is ensured that it can maintain the optimal calibration accuracy under different environmental conditions.

[0030] Reference Figure 2 This embodiment provides an environment-adaptive radio metrology calibration system, including: The synchronous sampling module 1 is used to synchronously sample the temperature, humidity, air pressure and electromagnetic interference around the radio equipment to obtain an environmental parameter vector; Construction module 2, used to construct a linear model and a nonlinear target model of a three-layer neural network structure according to the environmental parameter vector; A solution module 3, used for solving the environmental parameter separation vector set by using a linear model and a nonlinear target model; Search optimization module 4, used for performing one-dimensional iterative search optimization on the environmental parameter separation vector set to obtain the target environmental parameter vector; The calculation module 5 is used to input the target environment parameter vector into the nonlinear target model to perform calibration compensation calculation to obtain dynamic calibration compensation data.

[0031] In this embodiment, for the specific implementation of each unit in the above system embodiment, please refer to the above method embodiment, which will not be repeated here.

[0032] It should be noted that, in this article, the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, system, article or method including a series of elements includes not only those elements, but also includes other elements not explicitly listed, or also includes elements inherent to such process, system, article or method. In the absence of further restrictions, an element defined by the sentence "comprises a ..." does not exclude the presence of other identical elements in the process, system, article or method including the element.

[0033] The above description is only a preferred embodiment of the present invention, and does not limit the patent scope of the present invention. Any equivalent structure or equivalent process transformation made by using the contents of the present invention specification and drawings, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present invention.

Claims

1. An environment-adaptive radio measurement calibration method, characterized in that: include: Synchronously sample the temperature, humidity, air pressure and electromagnetic interference around the radio equipment to obtain an environmental parameter vector; Constructing a linear model and a nonlinear target model of a three-layer neural network structure according to the environmental parameter vector; Solving the environmental parameter separation vector set using the linear model and the nonlinear target model; Performing one-dimensional iterative search optimization on the environmental parameter separation vector set to obtain a target environmental parameter vector; The target environment parameter vector is input into the nonlinear target model to perform calibration compensation calculation to obtain dynamic calibration compensation data.

2. The environmental adaptive radio measurement calibration method according to claim 1, characterized in that: The temperature, humidity, air pressure and electromagnetic interference around the radio equipment are synchronously sampled to obtain an environmental parameter vector, including: Arrange temperature sensors, humidity sensors, air pressure sensors and electromagnetic interference sensors at different locations of the radio equipment to form an environmental parameter collection network; Setting a unified clock signal for the environmental parameter acquisition network and using a fixed sampling frequency to perform synchronous data acquisition to obtain original environmental parameter data; Applying a low-pass filter to the temperature data in the original environmental parameter data to remove high-frequency noise interference, thereby obtaining temperature filtered data; Filtering the humidity data and the air pressure data in the original environmental parameter data to remove interference signals, thereby obtaining humidity filtered data and air pressure filtered data; Applying energy component preservation filtering to the electromagnetic interference data in the original environmental parameter data to remove abnormal values, thereby obtaining electromagnetic interference filtering data; The temperature filter data, the humidity filter data, the air pressure filter data, and the electromagnetic interference filter data are normalized to obtain an environmental parameter vector.

3. The environmental adaptive radio measurement calibration method according to claim 1, characterized in that: The step of constructing a linear model and a nonlinear target model of a three-layer neural network structure according to the environmental parameter vector includes: Performing a linear transformation on the environmental parameter vector and constructing a linear model structure, establishing a linear mapping relationship between the environmental parameter and the calibration compensation by setting an environmental impact coefficient matrix and a bias vector, and obtaining a linear model framework; The environmental impact coefficient matrix and the bias vector of the linear model framework are optimized and trained by the least square method to obtain a linear model; Constructing a three-layer neural network architecture according to the dimension of the environmental parameter vector, by setting the number of input layer nodes equal to the number of environmental parameters, the number of hidden layer nodes equal to two times the number of environmental parameters plus one, and the number of output layer nodes equal to the number of calibration parameters; Adding an interactive effect processing unit for temperature and humidity to the three-layer neural network architecture, and capturing the coupling effect between environmental parameters by additionally setting cross terms in the three-layer neural network architecture to obtain a nonlinear target model framework; The nonlinear target model framework is optimized by setting a fixed learning rate and a training termination condition to obtain a nonlinear target model.

4. The environmentally adaptive radio measurement calibration method according to claim 1, characterized in that: The method of solving the environmental parameter separation vector set by using the linear model and the nonlinear target model includes: Using the linear model to perform preliminary mapping analysis on the environmental parameter vector, generate initial independent impact estimates of the environmental parameters, and obtain the basic structure of the oblique projection operator; Combining the mapping characteristics of the nonlinear target model, constructing an oblique projection operator expression with a residual term, introducing the output of the linear model as an initial value into the nonlinear calculation, and obtaining a complete projection operator calculation framework; The nonlinear target model is used to calculate the mapping difference before and after the environmental parameter projection, and the Lagrange multiplier method and the conjugate gradient descent algorithm are used to minimize the difference to obtain an optimized residual term; Based on the nonlinear target model and the complete projection operator calculation framework, corresponding projection operators are constructed for temperature, humidity, air pressure and electromagnetic interference, and the initial estimation provided by the linear model is used to accelerate the convergence process to obtain an initial projection operator set; Performing an improved Gram-Schmidt orthogonalization process on the initial projection operator set, verifying the orthogonal effect using the nonlinear target model, eliminating mutual interference between projection operators, and obtaining an orthogonalized oblique projection operator set; The environmental parameter vector is input into the orthogonalized oblique projection operator set, the nonlinear target model is applied to perform projection mapping verification, the independent influence of each environmental parameter is extracted, and the environmental parameter separation vector set is obtained.

5. The environmentally adaptive radio measurement calibration method according to claim 1, characterized in that: The performing one-dimensional iterative search optimization on the environmental parameter separation vector set to obtain a target environmental parameter vector includes: The calculation result of the linear model is set as an initial environment parameter vector, and a search interval is set for each parameter in the initial environment parameter vector, and a fixed number of uniformly distributed candidate values ​​are generated within the search interval to obtain a parameter candidate value set; Input each candidate value in the parameter candidate value set into the nonlinear target model in combination with other unchanged parameters, calculate the corresponding calibration compensation value, and obtain a candidate calibration result; Comparing the candidate calibration results with the actual measured calibration deviations, calculating the mean square error of each candidate value, and obtaining an error evaluation result; Selecting a parameter candidate value corresponding to the minimum error from the error evaluation result, updating the current environment parameter, and obtaining an optimized environment parameter vector of the current iteration; The optimization environment parameter vector of the current iteration is subjected to an iteration termination judgment. When the difference between two adjacent iteration results is less than a preset threshold or the maximum number of iterations is reached, the iteration is terminated, and the search step is reduced round by round to obtain the target environment parameter vector.

6. The environmentally adaptive radio measurement calibration method according to claim 1, characterized in that: The step of inputting the target environment parameter vector into the nonlinear target model to perform calibration compensation calculation to obtain dynamic calibration compensation data includes: Inputting the target environment parameter vector into the nonlinear target model for compensation calculation, generating a frequency compensation value, a power compensation value and a phase compensation value according to different weight coefficients of a plurality of calibration parameters, and obtaining a calibration compensation vector; Calculating the time derivative of the continuous sampling values ​​of each parameter in the environmental parameter vector, generating the temperature change rate, humidity change rate, air pressure change rate and electromagnetic interference change rate, and obtaining the environmental change rate vector; Comparing the environmental change rate vector with a preset change rate threshold, determining whether a calibration compensation value update condition is triggered, and obtaining an update trigger signal; Calculate the maximum relative change rate based on the environmental change rate vector, use the exponential decay function to construct an adaptive adjustment formula, and obtain a dynamic calibration update period; The calibration compensation vector is updated regularly according to the dynamic calibration update cycle, the update frequency is increased when the environment changes rapidly, and the update frequency is reduced when the environment is stable, so as to obtain calibration compensation data with optimized timing; The timing-optimized calibration compensation data is applied to calibration parameters of radio equipment, and real-time compensation adjustment is performed on frequency, power and phase to obtain dynamic calibration compensation data.

7. The environmentally adaptive radio measurement calibration method according to claim 6, characterized in that: The environmental adaptive radio metrology calibration method further comprises: The radio parameter vector measured after compensation is normalized and compared with the standard parameter vector, and the ratio of the Euclidean distance between the two and the norm of the standard vector is calculated to obtain the calibration error rate; Calculating the change gradient of each calibration parameter in the calibration compensation vector relative to each environmental parameter, and constructing an absolute value matrix of partial derivatives between the calibration parameters and environmental factors to obtain a sensitivity matrix; The importance weights of the calibration parameters are weighted and summed with the corresponding elements of the sensitivity matrix, and the contribution of each environmental parameter to the overall calibration accuracy is calculated to obtain the environmental parameter influencing factor; Arrange the environmental parameter influencing factors in descending order, select the top-ranked environmental parameters to form a candidate subset, and obtain a preliminary selection result of the environmental parameters; Based on the preliminary selection results of the environmental parameters, information entropy analysis is performed, and an optimization objective function is constructed in combination with a subset size penalty term, and the optimal subset size is determined by minimizing the overall evaluation index to obtain the optimal environmental parameter subset; The tilt projection operator design and one-dimensional iterative search optimization are re-executed for the optimal environmental parameter subset to construct a target calibration system for the radio device.

8. An environment-adaptive radio metrology calibration system, characterized in that: For implementing the steps of the method according to any one of claims 1 to 7, the environment adaptive radio metrology calibration system comprises: A synchronous sampling module is used to synchronously sample the temperature, humidity, air pressure and electromagnetic interference around the radio equipment to obtain an environmental parameter vector; A construction module, used for constructing a linear model and a nonlinear target model of a three-layer neural network structure according to the environmental parameter vector; A solution module, used for solving a set of environmental parameter separation vectors using the linear model and the nonlinear target model; A search optimization module, used to perform one-dimensional iterative search optimization on the environmental parameter separation vector set to obtain a target environmental parameter vector; The calculation module is used to input the target environment parameter vector into the nonlinear target model to perform calibration compensation calculation to obtain dynamic calibration compensation data.

Images