Device Calibration Method and Device

Abstract

The present disclosure relates to the technical field of device calibration, and provides a device calibration method and apparatus. The method includes: obtaining training parameters, as well as obtaining first radio frequency parameters and environmental parameters of a target device; calibrating the training parameters by using a calibration device to obtain a calibration result; performing linear regression processing on the training parameters and the calibration result to obtain a regression result, and constructing a simulation model according to the regression result; extracting a first logic diagram from the simulation model, and optimizing the first logic diagram according to the calibration result; training the simulation model according to the optimized first logic diagram; inputting the environmental parameters into the simulation model to output second radio frequency parameters; when the difference between the second radio frequency parameters and the first radio frequency parameters is greater than a preset threshold, calibrating the target device based on the second radio frequency parameters and the first radio frequency parameters. By adopting the above technical means, the problems of high cost and low accuracy in calibrating devices in the prior art are solved.

Description

Technical Field

[0001] The present disclosure relates to the technical field of device calibration, and in particular, to a device calibration method and apparatus. Background Art

[0002] Before using a device, it is necessary to calibrate and debug the device. Among them, calibrating the device is to adjust the parameters of the device so that the device can work properly. Calibrating the device is very important throughout the process of using the device, and the quality of calibrating the device directly affects the working results of the device. Existing technologies often use dedicated hardware calibration instruments such as signal generators or spectrum analyzers to calibrate the parameters of the device. Because signal generators and spectrum analyzers are expensive, the cost of calibrating the device is very high. At the same time, the accuracy of calibrating the device using existing technologies also needs to be further improved.

[0003] In the process of implementing the concept of the present disclosure, the inventors found that there are at least the following technical problems in the related technologies: the problems of high cost and low accuracy in calibrating the device. Summary of the Invention

[0004] In view of this, embodiments of the present disclosure provide a device calibration method, apparatus, electronic device, and computer-readable storage medium to solve the problems of high cost and low accuracy in calibrating a device in the existing technology.

[0005] In a first aspect of the embodiments of the present disclosure, a device calibration method is provided, including: obtaining training parameters, and obtaining first radio frequency parameters and environmental parameters of a target device; calibrating the training parameters using a calibration device to obtain a calibration result; performing linear regression processing on the training parameters and the calibration result to obtain a regression result, and constructing a simulation model according to the regression result; extracting a first logic diagram from the simulation model, and optimizing the first logic diagram according to the calibration result; training the simulation model according to the optimized first logic diagram; inputting the environmental parameters into the simulation model, and outputting second radio frequency parameters; when the difference between the second radio frequency parameters and the first radio frequency parameters is greater than a preset threshold, calibrating the target device based on the second radio frequency parameters and the first radio frequency parameters.

[0006] In a second aspect of the embodiments of the present disclosure, a device calibration apparatus is provided, including: an acquisition module configured to acquire training parameters, and acquire a first radio frequency parameter and an environmental parameter of a target device; a calibration module configured to calibrate the training parameters by using a calibration device to obtain a calibration result; a construction module configured to perform linear regression processing on the training parameters and the calibration result to obtain a regression result, and construct a simulation model according to the regression result; an extraction module configured to extract a first logic diagram from the simulation model and optimize the first logic diagram according to the calibration result; a training module configured to train the simulation model according to the optimized first logic diagram; a model module configured to input the environmental parameter into the simulation model and output a second radio frequency parameter; a calibration module configured to, when the difference between the second radio frequency parameter and the first radio frequency parameter is greater than a preset threshold, calibrate the target device based on the second radio frequency parameter and the first radio frequency parameter.

[0007] In a third aspect of the embodiments of the present disclosure, an electronic device is provided, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the steps of the above method are implemented.

[0008] In a fourth aspect of the embodiments of the present disclosure, a computer-readable storage medium is provided. The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps of the above method are implemented.

[0009] The beneficial effects of the embodiments of the present disclosure compared with the prior art are as follows: acquiring training parameters, and acquiring a first radio frequency parameter and an environmental parameter of a target device; calibrating the training parameters by using a calibration device to obtain a calibration result; performing linear regression processing on the training parameters and the calibration result to obtain a regression result, and constructing a simulation model according to the regression result; extracting a first logic diagram from the simulation model and optimizing the first logic diagram according to the calibration result; training the simulation model according to the optimized first logic diagram; inputting the environmental parameter into the simulation model and outputting a second radio frequency parameter; when the difference between the second radio frequency parameter and the first radio frequency parameter is greater than a preset threshold, calibrating the target device based on the second radio frequency parameter and the first radio frequency parameter. By adopting the above technical means, the problems of high cost and low accuracy in calibrating a device in the prior art are solved, thereby reducing the cost of the calibration device and improving the accuracy of the calibration device. BRIEF DESCRIPTION OF THE DRAWINGS

[0010] In order to more clearly illustrate the technical solutions in the embodiments of the present disclosure, the following will briefly introduce the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings in the following description are only some embodiments of the present disclosure. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0011] Figure 1 is a schematic diagram of the application scenario of the embodiments of the present disclosure;

[0012] Figure 2 is a schematic flowchart of a device calibration method provided by the embodiments of the present disclosure;

[0013] Figure 3 is a schematic structural diagram of a device calibration apparatus provided by the embodiments of the present disclosure;

[0014] Figure 4 is a schematic structural diagram of an electronic device provided by the embodiments of the present disclosure. Detailed implementation manners

[0015] In the following description, specific details such as specific system architectures and technologies are presented for the purpose of illustration rather than limitation, so as to thoroughly understand the embodiments of the present disclosure. However, those skilled in the art should clearly understand that the present disclosure can also be implemented in other embodiments without these specific details. In other cases, detailed descriptions of well-known systems, devices, circuits, and methods are omitted to avoid unnecessary details from interfering with the description of the present disclosure.

[0016] A device calibration method and apparatus according to the embodiments of the present disclosure will be described in detail below with reference to the accompanying drawings.

[0017] Figure 1 is a schematic diagram of the application scenario of the embodiments of the present disclosure. This application scenario may include terminal devices 1, 2, and 3, a server 4, and a network 5.

[0018] The terminal devices 1, 2, and 3 can be hardware or software. When the terminal devices 1, 2, and 3 are hardware, they can be various electronic devices with a display screen and supporting communication with the server 4, including but not limited to smart phones, tablet computers, laptop portable computers, and desktop computers, etc.; when the terminal devices 1, 2, and 3 are software, they can be installed in the above-mentioned electronic devices. The terminal devices 1, 2, and 3 can be implemented as multiple software or software modules, or can be implemented as a single software or software module, and the embodiments of the present disclosure do not limit this. Further, various applications can be installed on the terminal devices 1, 2, and 3, such as data processing applications, instant messaging tools, social platform software, search applications, shopping applications, etc.

[0019] Server 4 can be a server that provides various services. For example, it can be a background server that receives requests sent by terminal devices with which it establishes a communication connection. This background server can receive and analyze requests sent by terminal devices and generate processing results. Server 4 can be a single server, a server cluster composed of several servers, or a cloud computing service center. The embodiments of the present disclosure do not limit this.

[0020] It should be noted that Server 4 can be hardware or software. When Server 4 is hardware, it can be various electronic devices that provide various services for terminal devices 1, 2, and 3. When Server 4 is software, it can be multiple software or software modules that provide various services for terminal devices 1, 2, and 3, or a single software or software module that provides various services for terminal devices 1, 2, and 3. The embodiments of the present disclosure do not limit this.

[0021] Network 5 can be a wired network connected by coaxial cables, twisted pairs, and optical fibers, or a wireless network that can interconnect various communication devices without wiring, such as Bluetooth, Near Field Communication (NFC), Infrared, etc. The embodiments of the present disclosure do not limit this.

[0022] Users can establish a communication connection between terminal devices 1, 2, and 3 and Server 4 via Network 5 to receive or send information, etc. It should be noted that the specific types, quantities, and combinations of terminal devices 1, 2, and 3, Server 4, and Network 5 can be adjusted according to the actual requirements of the application scenario. The embodiments of the present disclosure do not limit this.

[0023] Figure 2 It is a schematic flowchart of a device calibration method provided by the embodiments of the present disclosure. Figure 2 The device calibration method can be executed by Figure 1 the terminal device or the server shown in Figure 2 As shown in, the device calibration method includes:

[0024] S201, obtain training parameters, and obtain the first radio frequency parameter and environmental parameter of the target device;

[0025] S202, calibrate the training parameters using a calibration device to obtain a calibration result;

[0026] S203, perform linear regression processing on the training parameters and the calibration result to obtain a regression result, and construct a simulation model according to the regression result;

[0027] S204. Extract the first logic diagram from the simulation model and optimize the first logic diagram according to the calibration result;

[0028] S205. Train the simulation model according to the optimized first logic diagram;

[0029] S206. Input the environmental parameters into the simulation model and output the second radio frequency parameter;

[0030] S207. When the difference between the second radio frequency parameter and the first radio frequency parameter is greater than the preset threshold, calibrate the target device based on the second radio frequency parameter and the first radio frequency parameter.

[0031] It should be noted that the environmental parameters of the target device include: the parameters of the actual environment where the target device is located and the parameters of the network environment, etc. The environmental parameters are used to describe the environmental conditions of the environment to which the target device belongs. The first radio frequency parameter is the current radio frequency parameter of the target device. The second radio frequency parameter of the target device corresponding to different environmental conditions is different. The second radio frequency parameter is the theoretically optimal radio frequency parameter of the target device under this environmental condition. The target device can be a device in any field. For example, if the target device is in the measurement and control field, it can be a mobile communication test instrument, a comprehensive tester, a program-controlled power supply, etc. Calibrating the target device means updating the existing first radio frequency parameter of the target device to the theoretically optimal second radio frequency parameter.

[0032] The training parameters include various environmental parameters, and each environmental parameter has been labeled with its corresponding second radio frequency parameter. Calibrating the training parameters using a calibration device can involve deleting or correcting some data in the training parameters that do not conform to the preset rules. For example, if the second radio frequency parameter corresponding to a certain environmental parameter is incorrect, then the second radio frequency parameter corresponding to that environmental parameter needs to be labeled. Linear regression processing can be any common fitting method, such as the least squares fitting method. Of course, linear regression processing can also be implemented with the help of some software, such as Excel or Matlab. Taking the environmental parameters as independent variables and the second radio frequency parameters as dependent variables, and inputting the environmental parameters and the second radio frequency parameters into the above software, the regression result can be output. For example, if the regression result is a function, then based on the regression result, a simulation model can be constructed. Taking this function as the main body, other requirements can be added to construct the simulation model. Other requirements, such as needing to display the data in an image, then the simulation model needs to add the conversion relationship from data to image on the basis of this function. The first logic diagram is the logic expressed by the simulation model, which can be understood as a mapping relationship or a function expression equivalent to the simulation model. The simulation model is constructed based on the regression result, and the first logic diagram is extracted from the simulation model. The first logic diagram can be obtained by adding something like the above-mentioned "other requirements" to the regression result. Optimizing the first logic diagram according to the calibration result can involve correcting the first logic diagram according to the calibration result. Training the simulation model according to the optimized first logic diagram can be understood as updating the model parameters of the simulation model by the backpropagation method according to the optimized first logic diagram.

[0033] Since the embodiments of the present disclosure can achieve the calibration of the target device by means of a model and no longer need to use a calibration instrument, the cost of the calibration device is reduced. At the same time, by improving the accuracy of the model, the accuracy of the calibration device can be improved.

[0034] According to the technical solution provided by the embodiments of the present disclosure, obtain training parameters, as well as the first radio frequency parameter and environmental parameter of the target device; calibrate the training parameters using a calibration device to obtain a calibration result; perform linear regression processing on the training parameters and the calibration result to obtain a regression result, and construct a simulation model according to the regression result; extract the first logic diagram from the simulation model, and optimize the first logic diagram according to the calibration result; train the simulation model according to the optimized first logic diagram; input the environmental parameter into the simulation model and output the second radio frequency parameter; when the difference between the second radio frequency parameter and the first radio frequency parameter is greater than a preset threshold, calibrate the target device based on the second radio frequency parameter and the first radio frequency parameter. By adopting the above technical means, the problems of high cost and low accuracy in calibrating devices in the prior art are solved, thereby reducing the cost of the calibration device and improving the accuracy of the calibration device.

[0035] After performing linear regression on the training parameters and calibration results to obtain a regression result and constructing a simulation model based on the regression result, the method further includes: constructing a simulation network model using the simulation model and a neural network model; training the simulation network model using the training parameters; extracting a second logic diagram from the trained simulation network model and optimizing the second logic diagram according to the calibration result; retraining the simulation network model according to the optimized second logic diagram; and inputting environmental parameters into the simulation network model to output second radio frequency parameters.

[0036] Constructing a simulation network model using the simulation model and a neural network model can be that the simulation model is followed by the neural network model. In the embodiments of the present disclosure, the simulation network model is trained twice, thereby improving the accuracy of the trained model. The second logic diagram is similar to the first logic diagram, and optimizing the second logic diagram is similar to optimizing the first logic diagram. Since the simulation model constructed using a mathematical method has a limited accuracy improvement after reaching a certain level, while the neural network model does not have this problem. The neural network model can be trained through a large amount of machine learning to improve the accuracy of the model as much as possible. Therefore, the embodiments of the present disclosure use the neural network model to improve the accuracy of the simulation model. Specifically, in the simulation network model, the output of the simulation model is the input of the neural network model, and the neural network model outputs the second radio frequency parameters (although the simulation model is established based on the mapping relationship from environmental parameters to the second radio frequency parameters, in actual applications, the output of the simulation model may not be the second radio frequency parameters).

[0037] It should be noted that the neural network model can be any common neural network model, such as Faster-RCNN. The training methods in the present disclosure are all similar to deep learning training methods.

[0038] Retraining the simulation network model according to the optimized second logic diagram includes: determining the training data set corresponding to the training parameters according to the optimized second logic diagram; in the entire training process, the first round of training: training the simulation network model using the training data set while freezing the neural network model to update the parameters of the simulation model in the simulation network model; the second round of training: training the simulation network model using the training data set while freezing the simulation model to update the parameters of the neural network model in the simulation network model; the third round of training: training the simulation network model using the training data set to update the parameters of the simulation model and the neural network model in the simulation network model.

[0039] According to the optimized second logic diagram, the training data set corresponding to the training parameters is determined. For example, if the second radio frequency parameter corresponding to a certain environmental parameter is incorrect, then the optimized second logic diagram can be used to mark the correct second radio frequency parameter corresponding to a certain environmental parameter. Therefore, the training data set can be regarded as the updated training parameters. This step is similar to correcting the incorrect second radio frequency parameter corresponding to a certain environmental parameter during the calibration of the training parameters using a calibration device. However, since the second logic diagram is extracted from the trained simulation network model, the second logic diagram is superior to the calibration device.

[0040] Retrain the simulation network model according to the optimized second logic diagram, including three rounds of training. In the first round of training, with the neural network model frozen, use the training data set to train the simulation network model. This can be understood as only training the simulation model, and this step is used to adjust the parameters of the simulation model. In the second round of training: with the simulation model frozen, use the training data set to train the simulation network model. This can be understood as only training the neural network model, and this step is the most common training of the neural network model. In the third round of training: use the training data set to train the simulation network model, which is to train both the simulation model and the neural network model simultaneously, and this step is used to fine-tune the parameters of the simulation network model.

[0041] After retraining the simulation network model according to the optimized second logic diagram, the method further includes: inputting the environmental parameters into the neural network model in the simulation network model, and outputting the second radio frequency parameters.

[0042] In the embodiments of the present disclosure, the simulation model is only used in retraining to assist in adjusting the parameters of the neural network model (the embodiments of the present disclosure are different from the above embodiments. Here, the neural network model is the main body and the simulation model is the auxiliary, while in the above embodiments, the neural network model is the auxiliary and the simulation model is the main body. The advantage of this method is that it can help the neural network model converge as soon as possible. The network structure in the embodiments of the present disclosure can be to construct the simulation network model with the simulation model and the neural network model in parallel, so that the simulation model can be used to constrain the input and output of the neural network model). After completing the retraining of the simulation network model, extract the neural network model from the simulation network model, or remove the part of the simulation model in the simulation network model and only use the neural network model to map the environmental parameters.

[0043] That is, after obtaining the training parameters, the first radio frequency parameters, and the environmental parameters of the target device, the method further includes: training the neural network model with the training parameters so that the neural network model learns and stores the correspondence between the environmental parameters and the second radio frequency parameters; calibrating the training parameters using a calibration device to obtain a calibration result; extracting a third logic diagram from the neural network model and optimizing the third logic diagram according to the calibration result; retraining the neural network model according to the optimized third logic diagram; inputting the environmental parameters into the neural network model to output the second radio frequency parameters; when the difference between the second radio frequency parameters and the first radio frequency parameters is greater than a preset threshold, calibrating the target device based on the second radio frequency parameters and the first radio frequency parameters.

[0044] The third logic diagram is similar to the first logic diagram and will not be elaborated here. In the prior art, the radio frequency parameters of the target device are calibrated by a calibration device. The present disclosure first introduces a neural network model, and determines the second radio frequency parameters corresponding to the environmental parameters through the trained neural network model, and then calibrates the radio frequency parameters of the target device.

[0045] Before training the neural network model with the training parameters so that the neural network model learns and stores the correspondence between the environmental parameters and the second radio frequency parameters, the method further includes: obtaining the device information of multiple devices of the same type in the same application field as the target device; updating the model parameters of the neural network model according to the device information of the multiple devices.

[0046] If the neural network model is directly trained with the training parameters, the training time will be very long. To improve the training efficiency, before training the neural network model in the embodiments of the present disclosure, the model parameters of the neural network model are first updated according to the device information of multiple devices. Updating the model parameters of the neural network model according to the device information of multiple devices can be understood as updating the values of the model parameters of the neural network model to the interval corresponding to the target device. The device information includes: information on the application field of the device, the type of the device, and the working parameters of the device, such as the theoretical interval of the working frequency of a device. Through the above technical means, the convergence speed of the neural network model in training can be improved.

[0047] To improve the accuracy of the neural network model after training, the embodiments of the present disclosure retrain the neural network model. The calibration result is the second parameter value recommended by the calibration device.

[0048] In an alternative embodiment, the first radio frequency parameters include: the first working frequency, the first signal-to-noise ratio, the first signal gain, the first transmit signal power, the first transmit signal frequency, and the first s parameter; the second radio frequency parameters include: the second working frequency, the second signal-to-noise ratio, the second signal gain, the second transmit signal power, the second transmit signal frequency, and the second s parameter.

[0049] The s parameters include characteristic impedance, crosstalk, insertion loss, return loss, etc. For example, in the embodiments of the present disclosure, the first s parameter can be changed by adjusting the working circuit of the target device, and then the first radio frequency parameter can be calibrated based on the second radio frequency parameter. When the target device is working, different circuits can be selected, and thus the circuit with the second s parameter can be selected. Different circuits have different signal gains, and the most suitable circuit can be selected according to the second signal gain. The first transmission signal power can also be adjusted according to the second transmission signal power, etc.

[0050] It should be noted that any one of the first radio frequency parameter and the second radio frequency parameter in the embodiments of the present disclosure can correspond to a value within an interval. For example, the second transmission signal power can be the transmission signal power within an interval.

[0051] In an alternative embodiment, the environmental parameters of the target device are obtained; the light wave reflection constraint condition, the light wave scattering constraint condition, the light wave refraction constraint condition, and the light wave diffraction constraint condition are established respectively based on the light wave reflection principle, the light wave scattering principle, the light wave refraction principle, and the light wave diffraction principle; the light wave propagation energy function is established with the criterion of minimizing the energy loss of light wave propagation; a mathematical model is constructed based on the light wave reflection constraint condition, the light wave scattering constraint condition, the light wave refraction constraint condition, the light wave diffraction constraint condition, and the light wave propagation energy function; the environmental parameters are input into the mathematical model, and the second working wavelength of the target device is output; the second working frequency of the target device is determined according to the working wavelength; when the difference between the second working frequency and the first working frequency is greater than the preset threshold, the target device is calibrated based on the second working frequency and the first working frequency.

[0052] Among them, the second radio frequency parameter includes the second working frequency, and the first radio frequency parameter includes the first working frequency.

[0053] The embodiments of the present disclosure belong to the idea of mathematical modeling. Different wavelengths have different propagation capabilities in an environment. Such propagation differences include reflection differences, scattering differences, refraction differences, diffraction differences, etc. To enhance the propagation capability of the working wave of the target device in the environment and reduce the loss during propagation, the embodiments of the present disclosure establish light wave reflection constraint conditions based on the light wave reflection principle, establish light wave scattering constraint conditions based on the light wave scattering principle, establish light wave refraction constraint conditions based on the light wave refraction principle, and establish light wave diffraction constraint conditions based on the light wave diffraction principle. Through the above constraint conditions, the wavelength of the working wave of the target device can be controlled within a certain range, and then according to the light wave propagation energy function, the second working wavelength of the target device can be solved from this range. Since the wavelength and frequency satisfy certain conditions, the second working frequency of the target device can be determined according to the working wavelength. The environmental parameters include all the information of the environment where the target device is located, such as what obstacles in the environment where the target device is located affect the light wave propagation, the position and size of the obstacles, etc.

[0054] In an alternative embodiment, obtain the environmental parameters of the target device; according to the environmental parameters, determine at least one of the following parameters of the target device: optimal voltage, optimal current, and anti-interference information; based on at least one of the following parameters: optimal voltage, optimal current, and anti-interference information, determine the optimal transmission signal power of the target device, where the second radio frequency parameter includes the optimal transmission signal power; divide the optimal transmission signal power by the current transmission signal power to obtain a division value; determine the optimal signal gain according to the division value, where the second radio frequency parameter includes the optimal signal gain, and the first radio frequency parameter includes the current transmission signal power; determine the working circuit of the target device according to the optimal signal gain.

[0055] When the target device is in different environments or the operation of the target device is different, the allowed transmission signal power of the target device will be different, or considering the device power consumption, the optimal transmission signal power of the target device will be different. To solve the above technical problems, the embodiments of the present disclosure determine the optimal voltage, or optimal current, or anti-interference information of the target device through the environmental parameters. Since the internal resistance of the target device is known, the optimal transmission signal power of the target device can be determined according to the relationship between voltage, current, resistance, and power. The anti-interference information is used to indicate the anti-interference ability that the transmission signal power of the target device should have. In some cases, if the target device needs a strong anti-interference ability, then the transmission signal power of the target device should be enhanced; if the target device does not need a strong anti-interference ability, then the transmission signal power of the target device should be appropriately reduced according to the specific situation. Therefore, the optimal transmission signal power of the target device is also determined according to the anti-interference information.

[0056] Embodiments of the present disclosure can provide multiple gain circuits for a target device, thereby providing multiple gains for the target device. The division value obtained by dividing the optimal transmission signal power by the current transmission signal power can be understood as the gain required by the target device, that is, the optimal signal gain. The current signal gain is the gain currently used by the target device at the current moment. After the optimal signal gain, the working circuit of the target device can be determined according to the optimal signal gain. Determining the working circuit of the target device is also a method for calibrating the target device.

[0057] Any combination of the above all optional technical solutions can form an optional embodiment of the present application, which will not be elaborated here one by one.

[0058] The following are embodiments of the device of the present disclosure, which can be used to execute the method embodiments of the present disclosure. For details not disclosed in the embodiments of the device of the present disclosure, please refer to the method embodiments of the present disclosure.

[0059] Figure 3 It is a schematic diagram of a device calibration device provided by an embodiment of the present disclosure. As Figure 3 shown, the device calibration device includes:

[0060] An acquisition module 301, configured to acquire training parameters, and acquire first radio frequency parameters and environmental parameters of a target device;

[0061] A calibration module 302, configured to calibrate the training parameters by using a calibration device to obtain a calibration result;

[0062] A construction module 303, configured to perform linear regression processing on the training parameters and the calibration result to obtain a regression result, and construct a simulation model according to the regression result;

[0063] An extraction module 304, configured to extract a first logic diagram from the simulation model and optimize the first logic diagram according to the calibration result;

[0064] A training module 305, configured to train the simulation model according to the optimized first logic diagram;

[0065] A model module 306, configured to input the environmental parameters into the simulation model and output second radio frequency parameters;

[0066] A calibration module 307, configured to calibrate the target device based on the second radio frequency parameters and the first radio frequency parameters when the difference between the second radio frequency parameters and the first radio frequency parameters is greater than a preset threshold.

[0067] It should be noted that the environmental parameters of the target device include: the parameters of the actual environment where the target device is located and the parameters of the network environment, etc. The environmental parameters are used to describe the environmental conditions of the environment to which the target device belongs. The first radio frequency parameter is the current radio frequency parameter of the target device. The second radio frequency parameter of the target device corresponding to different environmental conditions is different. The second radio frequency parameter is the theoretically optimal radio frequency parameter of the target device under this environmental condition. The target device can be a device in any field. For example, if the target device is in the measurement and control field, it can be a mobile communication test instrument, a comprehensive tester, a programmable power supply, etc. Calibrating the target device is to update the existing first radio frequency parameter of the target device to the theoretically optimal second radio frequency parameter.

[0068] The training parameters include various environmental parameters, and each environmental parameter has been labeled with its corresponding second radio frequency parameter. Using the calibration device to calibrate the training parameters can be to delete or correct some data that does not meet the preset rules in the training parameters. For example, if the second radio frequency parameter corresponding to a certain environmental parameter is incorrect, then the second radio frequency parameter corresponding to a certain environmental parameter needs to be labeled. Linear regression processing can be any common fitting method, such as the least squares fitting method. Of course, linear regression processing can also be implemented with the help of some software, such as excel and matlab. Taking the environmental parameters as independent variables and the second radio frequency parameters as dependent variables, and inputting the environmental parameters and the second radio frequency parameters into the above software, the regression result can be output. For example, if the regression result is a function, then according to the regression result, a simulation model is constructed. Based on this function, other requirements can be added to construct the simulation model. Other requirements, such as the need to display the data in an image, then the simulation model needs to add the conversion relationship from data to image on the basis of this function. The first logic diagram is the logic expressed by the simulation model, which can be understood as a mapping relationship or a function expression equivalent to the simulation model. The simulation model is constructed according to the regression result, and the first logic diagram is extracted from the simulation model. The first logic diagram can be something like adding the above-mentioned "other requirements" on the basis of the regression result. Optimizing the first logic diagram according to the calibration result can be to correct the first logic diagram according to the calibration result. Training the simulation model according to the optimized first logic diagram can be understood as updating the model parameters of the simulation model by the backpropagation method according to the optimized first logic diagram.

[0069] Since the embodiments of the present disclosure can achieve the calibration of the target device by means of a model and no longer need to use a calibration instrument, the cost of the calibration device is reduced. At the same time, by improving the accuracy of the model, the accuracy of the calibration device can be improved.

[0070] According to the technical solution provided by the embodiments of the present disclosure, training parameters are obtained, and the first radio frequency parameters and environmental parameters of the target device are obtained; the calibration device is used to calibrate the training parameters to obtain a calibration result; linear regression processing is performed on the training parameters and the calibration result to obtain a regression result, and a simulation model is constructed according to the regression result; the first logic diagram is extracted from the simulation model, and the first logic diagram is optimized according to the calibration result; the simulation model is trained according to the optimized first logic diagram; the environmental parameters are input into the simulation model, and the second radio frequency parameters are output; when the difference between the second radio frequency parameters and the first radio frequency parameters is greater than a preset threshold, the target device is calibrated based on the second radio frequency parameters and the first radio frequency parameters. By adopting the above technical means, the problems of high cost and low accuracy in calibrating devices in the prior art are solved, thereby reducing the cost of the calibration device and improving the accuracy of the calibration device.

[0071] Optionally, the construction module 303 is further configured to construct a simulation network model using the simulation model and the neural network model; train the simulation network model using the training parameters; extract the second logic diagram from the trained simulation network model, and optimize the second logic diagram according to the calibration result; re-train the simulation network model according to the optimized second logic diagram; input the environmental parameters into the simulation network model, and output the second radio frequency parameters.

[0072] Constructing a simulation network model using the simulation model and the neural network model may be that the simulation model is followed by the neural network model. In the embodiments of the present disclosure, the simulation network model is trained twice, thereby improving the accuracy of the trained model. The second logic diagram is similar to the first logic diagram, and optimizing the second logic diagram is similar to optimizing the first logic diagram. Because the accuracy of the simulation model constructed using mathematical methods is difficult to improve further after reaching a certain level, but this problem does not exist in the neural network model. The neural network model can improve the accuracy of the model as much as possible through a large number of machine learning trainings. Therefore, the embodiments of the present disclosure use the neural network model to improve the accuracy of the simulation model. Specifically, in the simulation network model, the output of the simulation model is the input of the neural network model, and the neural network model is the second radio frequency parameter (although the simulation model is established based on the mapping relationship from the environmental parameters to the second radio frequency parameter, in actual applications, the output of the simulation model may not be the second radio frequency parameter).

[0073] It should be noted that the neural network model can be any common neural network model, such as Faster-RCNN. The training methods in the present disclosure are all similar to the methods of deep learning training.

[0074] Optionally, the construction module 303 is further configured to determine a training data set corresponding to the training parameters according to the optimized second logic diagram; during the entire training process, in the first round of training: in the case of freezing the neural network model, use the training data set to train the simulation network model to update the parameters of the simulation model in the simulation network model; in the second round of training: in the case of freezing the simulation model, use the training data set to train the simulation network model to update the parameters of the neural network model in the simulation network model; in the third round of training: use the training data set to train the simulation network model to update the parameters of the simulation model and the neural network model in the simulation network model.

[0075] According to the optimized second logic diagram, determine a training data set corresponding to the training parameters. For example, if the second radio frequency parameter corresponding to a certain environmental parameter is incorrect, then the optimized second logic diagram can be used to mark the correct second radio frequency parameter corresponding to a certain environmental parameter. Therefore, the training data set can be regarded as the updated training parameters. This step is similar to correcting the second radio frequency parameter corresponding to an incorrect environmental parameter in calibrating the training parameters using a calibration device. However, since the second logic diagram is extracted from the trained simulation network model, the second logic diagram is superior to the calibration device.

[0076] Retrain the simulation network model according to the optimized second logic diagram, including three rounds of training. In the first round of training, in the case of freezing the neural network model, use the training data set to train the simulation network model, which can be understood as only training the simulation model. This step is used to adjust the parameters of the simulation model; in the second round of training: in the case of freezing the simulation model, use the training data set to train the simulation network model, which can be understood as only training the neural network model. This step is the most common training of the neural network model; in the third round of training: use the training data set to train the simulation network model, which is to train the simulation model and the neural network model simultaneously. This step is used to fine-tune the parameters of the simulation network model.

[0077] Optionally, the construction module 303 is further configured to input the environmental parameters into the neural network model in the simulation network model and output the second radio frequency parameters.

[0078] In the embodiments of the present disclosure, the simulation model is only used in retraining to assist in adjusting the parameters of the neural network model (the embodiments of the present disclosure are different from the above embodiments in that the neural network model is the main body and the simulation model is the auxiliary, while in the above embodiments, the neural network model is the auxiliary and the simulation model is the main body. The advantage of this method is that it can help the neural network model converge as soon as possible. The network structure of the embodiments of the present disclosure can be to construct a simulation network model in parallel with the simulation model and the neural network model, so that the simulation model can be used to constrain the input and output of the neural network model). After completing the retraining of the simulation network model, the neural network model is extracted from the simulation network model, or the part of the simulation model is removed from the simulation network model, and only the neural network model is used to map the environmental parameters.

[0079] Optionally, the first calibration module 302 is further configured to train the neural network model using the training parameters, so that the neural network model learns and stores the correspondence between the environmental parameters and the second radio frequency parameters; use the calibration device to calibrate the training parameters to obtain a calibration result; extract the third logic diagram from the neural network model, and optimize the third logic diagram according to the calibration result; retrain the neural network model according to the optimized third logic diagram; input the environmental parameters into the neural network model and output the second radio frequency parameters; when the difference between the second radio frequency parameter and the first radio frequency parameter is greater than a preset threshold, calibrate the target device based on the second radio frequency parameter and the first radio frequency parameter.

[0080] The third logic diagram is similar to the first logic diagram and will not be elaborated here. In the prior art, the radio frequency parameters of the target device are calibrated by a calibration device. The present disclosure first introduces a neural network model, and determines the second radio frequency parameter corresponding to the environmental parameter through the trained neural network model, and then calibrates the radio frequency parameter of the target device.

[0081] Optionally, the first calibration module 302 is further configured to obtain the device information of multiple devices of the same type in the same application field as the target device; update the model parameters of the neural network model according to the device information of the multiple devices.

[0082] If the neural network model is directly trained using the training parameters, the training time will be very long. To improve the training efficiency, before training the neural network model in the embodiments of the present disclosure, the model parameters of the neural network model are first updated according to the device information of multiple devices. Updating the model parameters of the neural network model according to the device information of multiple devices can be understood as updating the value of the model parameters of the neural network model to the interval corresponding to the target device. The device information includes: information on the application field of the device, the type of the device, and the working parameters of the device, such as the theoretical interval of the working frequency of a device. Through the above technical means, the convergence speed of the neural network model in training can be improved.

[0083] To improve the accuracy of the neural network model after training, the embodiments of the present disclosure retrain the neural network model. The calibration result is the second parameter value recommended by the calibration device.

[0084] In an alternative embodiment, the first radio frequency parameters include: the first operating frequency, the first signal-to-noise ratio, the first signal gain, the first transmit signal power, the first transmit signal frequency, and the first s-parameter; the second radio frequency parameters include: the second operating frequency, the second signal-to-noise ratio, the second signal gain, the second transmit signal power, the second transmit signal frequency, and the second s-parameter.

[0085] The s-parameters include: characteristic impedance, crosstalk, insertion loss, return loss, etc. For example, the embodiments of the present disclosure can change the first s-parameter by adjusting the working circuit of the target device, and then calibrate the first radio frequency parameters based on the second radio frequency parameters. The target device can select different circuits during operation, and thus select the circuit with the second s-parameter. Different circuits have different signal gains, and the most suitable circuit can be selected according to the second signal gain. The first transmit signal power can also be adjusted according to the second transmit signal power, etc.

[0086] It should be noted that any one of the first radio frequency parameters and the second radio frequency parameters in the embodiments of the present disclosure can correspond to a value within an interval. For example, the second transmit signal power can be a transmit signal power within an interval.

[0087] Optionally, the obtaining module 301 is further configured to obtain the environmental parameters of the target device; respectively establish the light wave reflection constraint condition, the light wave scattering constraint condition, the light wave refraction constraint condition, and the light wave diffraction constraint condition based on the light wave reflection principle, the light wave scattering principle, the light wave refraction principle, and the light wave diffraction principle; establish a light wave propagation energy function with the criterion of minimizing the energy loss of light wave propagation; construct a mathematical model based on the light wave reflection constraint condition, the light wave scattering constraint condition, the light wave refraction constraint condition, the light wave diffraction constraint condition, and the light wave propagation energy function; input the environmental parameters into the mathematical model, and output the second operating wavelength of the target device; determine the second operating frequency of the target device according to the operating wavelength; when the difference between the second operating frequency and the first operating frequency is greater than a preset threshold, calibrate the target device based on the second operating frequency and the first operating frequency.

[0088] Among them, the second radio frequency parameter includes the second operating frequency, and the first radio frequency parameter includes the first operating frequency.

[0089] Embodiments of the present disclosure belong to the idea of mathematical modeling. Different wavelengths have different propagation capabilities in an environment, and such propagation differences include reflection differences, scattering differences, refraction differences, diffraction differences, etc. To enhance the propagation capability of the working wave of the target device in the environment and reduce the loss during propagation, embodiments of the present disclosure establish a light wave reflection constraint condition based on the light wave reflection principle, establish a light wave scattering constraint condition based on the light wave scattering principle, establish a light wave refraction constraint condition based on the light wave refraction principle, and establish a light wave diffraction constraint condition based on the light wave diffraction principle. Through the above constraint conditions, the wavelength of the working wave of the target device can be controlled within a certain range, and then according to the light wave propagation energy function, the second working wavelength of the target device is solved from this range. Since the wavelength and frequency satisfy certain conditions, the second working frequency of the target device can be determined according to the working wavelength. The environmental parameters include all the information of the environment where the target device is located, such as what obstacles affect the light wave propagation in the environment where the target device is located, the position and size of the obstacles, etc.

[0090] Optionally, the obtaining module 301 is further configured to obtain the environmental parameters of the target device; determine at least one of the following parameters of the target device according to the environmental parameters: the optimal voltage, the optimal current, and the anti-interference information; determine the optimal transmission signal power of the target device based on at least one of the following parameters: the optimal voltage, the optimal current, and the anti-interference information, where the second radio frequency parameter includes the optimal transmission signal power; divide the optimal transmission signal power by the current transmission signal power to obtain a division value; determine the optimal signal gain according to the division value, where the second radio frequency parameter includes the optimal signal gain and the first radio frequency parameter includes the current transmission signal power; determine the working circuit of the target device according to the optimal signal gain.

[0091] When the target device is in different environments or the operation of the target device is different, the allowed transmission signal power of the target device will be different, or considering the device power consumption, the optimal transmission signal power of the target device will be different. To solve the above technical problems, embodiments of the present disclosure determine the optimal voltage, or the optimal current, or the anti-interference information of the target device through the environmental parameters. Since the internal resistance of the target device is known, the optimal transmission signal power of the target device can be determined according to the relationship between voltage, current, resistance, and power. The anti-interference information is used to indicate the anti-interference ability that the transmission signal power of the target device should have. In some cases, if the target device needs a strong anti-interference ability, then the transmission signal power of the target device should be enhanced; if the target device does not need a strong anti-interference ability, then the transmission signal power of the target device should be appropriately reduced according to the specific situation. Therefore, the optimal transmission signal power of the target device is also determined according to the anti-interference information.

[0092] Embodiments of the present disclosure can provide multiple gain circuits for a target device, thereby providing multiple gains for the target device. The division value obtained by dividing the optimal transmission signal power by the current transmission signal power can be understood as the gain required by the target device, that is, the optimal signal gain. The current signal gain is the gain currently used by the target device at the current moment. After the optimal signal gain, the working circuit of the target device can be determined according to the optimal signal gain. Determining the working circuit of the target device is also a method of calibrating the target device.

[0093] It should be understood that the magnitudes of the sequence numbers of the steps in the above embodiments do not mean the order of execution. The order of execution of each process should be determined according to its function and internal logic, and should not constitute any limitation to the implementation process of the embodiments of the present disclosure.

[0094] Figure 4 is a schematic diagram of the electronic device 4 provided by the embodiments of the present disclosure. As Figure 4 shown, the electronic device 4 of this embodiment includes: a processor 401, a memory 402, and a computer program 403 stored in the memory 402 and executable on the processor 401. When the processor 401 executes the computer program 403, the steps in the above-mentioned various method embodiments are implemented. Alternatively, when the processor 401 executes the computer program 403, the functions of each module / unit in the above-mentioned various device embodiments are implemented.

[0095] Exemplarily, the computer program 403 can be divided into one or more modules / units. The one or more modules / units are stored in the memory 402 and executed by the processor 401 to complete the present disclosure. The one or more modules / units can be a series of computer program instruction segments capable of performing specific functions, and the instruction segments are used to describe the execution process of the computer program 403 in the electronic device 4.

[0096] The electronic device 4 can be a desktop computer, a notebook, a palm computer, a cloud server, and other electronic devices. The electronic device 4 can include, but is not limited to, the processor 401 and the memory 402. Those skilled in the art can understand that Figure 4 merely examples of the electronic device 4 do not constitute a limitation to the electronic device 4. It can include more or fewer components than shown in the figure, or combine some components, or different components. For example, the electronic device can also include input / output devices, network access devices, buses, etc.

[0097] The processor 401 may be a Central Processing Unit (CPU), or may be other general-purpose processors, Digital Signal Processors (DSPs), Application Specific Integrated Circuits (ASICs), Field-Programmable Gate Arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or the processor may also be any conventional processor, etc.

[0098] The memory 402 may be an internal storage unit of the electronic device 4. For example, the hard disk or memory of the electronic device 4. The memory 402 may also be an external storage device of the electronic device 4. For example, a plug-in hard disk equipped on the electronic device 4, a Smart Media Card (SMC), a Secure Digital (SD) card, a Flash Card, etc. Further, the memory 402 may also include both the internal storage unit and the external storage device of the electronic device 4. The memory 402 is used to store computer programs and other programs and data required by the electronic device. The memory 402 may also be used to temporarily store data that has been output or is to be output.

[0099] Those skilled in the art can clearly understand that for the convenience and brevity of description, only the above division of each functional unit and module is used as an example. In actual applications, the above functions can be allocated to different functional units and modules according to needs, that is, the internal structure of the device is divided into different functional units or modules to complete all or part of the functions described above. Each functional unit and module in the embodiments can be integrated into one processing unit, or each unit can exist physically alone, or two or more units can be integrated into one unit. The above integrated unit can be implemented in the form of hardware or in the form of a software functional unit. In addition, the specific names of each functional unit and module are only for the convenience of mutual distinction and do not limit the protection scope of this application. The specific working processes of the units and modules in the above system can refer to the corresponding processes in the foregoing method embodiments and will not be elaborated herein.

[0100] In the above embodiments, the descriptions of the respective embodiments have their own emphases. For parts not detailed or recorded in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0101] Those of ordinary skill in the art can realize that the units and algorithm steps of each example described in combination with the embodiments disclosed herein can be implemented by electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are executed in a hardware or software manner depends on the specific application and design constraints of the technical solution. Professional technicians can use different methods for each specific application to implement the described functions, but such implementation should not be considered to exceed the scope of this disclosure.

[0102] In the embodiments provided by this disclosure, it should be understood that the disclosed device / electronic device and method can be implemented in other ways. For example, the device / electronic device embodiments described above are only illustrative. For example, the division of modules or units is only a logical function division. In actual implementation, there may be other division methods. Multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the displayed or discussed coupling or direct coupling or communication connection between each other can be through some interfaces. The indirect coupling or communication connection of the device or unit can be in an electrical, mechanical or other form.

[0103] The units described as separate components may or may not be physically separated. The components shown as units may or may not be physical units, that is, they can be located in one place, or can be distributed to multiple network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0104] In addition, the functional units in each embodiment of this disclosure can be integrated into one processing unit, or each unit can exist physically alone, or two or more units can be integrated into one unit. The above integrated units can be implemented in the form of hardware or in the form of software functional units.

[0105] When an integrated module / unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, to implement all or part of the processes in the above-described embodiment methods of the present disclosure, it can also be completed by a computer program instructing relevant hardware. The computer program can be stored in a computer-readable storage medium. When the computer program is executed by a processor, the steps of the above-described various method embodiments can be implemented. The computer program can include computer program code, and the computer program code can be in the form of source code, object code, executable file, or some intermediate form, etc. The computer-readable medium can include: any entity or device capable of carrying the computer program code, recording medium, USB flash drive, mobile hard disk, magnetic disk, optical disc, computer memory, read-only memory (ROM), random access memory (RAM), electrical carrier signal, telecommunication signal, and software distribution medium, etc. It should be noted that the content included in the computer-readable medium can be appropriately increased or decreased according to the requirements of legislation and patent practice within the jurisdiction. For example, in some jurisdictions, according to legislation and patent practice, the computer-readable medium does not include electrical carrier signals and telecommunication signals.

[0106] The above embodiments are only used to illustrate the technical solutions of the present disclosure, rather than to limit them; although the present disclosure has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that: they can still modify the technical solutions recorded in the foregoing embodiments, or perform equivalent replacements for some of the technical features; and these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the various embodiments of the present disclosure, and should all be included in the protection scope of the present disclosure.

Claims

1. A device calibration method, characterized in that, Including: Obtain training parameters, and obtain the first radio frequency parameters and environmental parameters of the target device; Use a calibration device to calibrate the training parameters to obtain a calibration result; Perform linear regression processing on the training parameters and the calibration result to obtain a regression result, and construct a simulation model according to the regression result; Extract a first logic diagram from the simulation model, and optimize the first logic diagram according to the calibration result; Train the simulation model according to the optimized first logic diagram; Input the environmental parameters into the simulation model to output second radio frequency parameters; When the difference between the second radio frequency parameters and the first radio frequency parameters is greater than a preset threshold, calibrate the target device based on the second radio frequency parameters and the first radio frequency parameters; After performing linear regression processing on the training parameters and the calibration result to obtain a regression result and constructing a simulation model according to the regression result, the method further includes: Use the simulation model and the neural network model to construct a simulation network model; Train the simulation network model using the training parameters; Extract a second logic diagram from the trained simulation network model, and optimize the second logic diagram according to the calibration result; Retrain the simulation network model according to the optimized second logic diagram; Input the environmental parameters into the simulation network model to output the second radio frequency parameters; The retraining the simulation network model according to the optimized second logic diagram includes: Determine the training data set corresponding to the training parameters according to the optimized second logic diagram; During the entire training process, the first round of training: in the case of freezing the neural network model, use the training data set to train the simulation network model to update the parameters of the simulation model in the simulation network model; The second round of training: in the case of freezing the simulation model, use the training data set to train the simulation network model to update the parameters of the neural network model in the simulation network model; The third round of training: use the training data set to train the simulation network model to update the parameters of the simulation model and the neural network model in the simulation network model.

2. The method according to claim 1, wherein After retraining the simulation network model according to the optimized second logic diagram, the method further includes: Input the environmental parameters into the neural network model in the simulation network model to output the second radio frequency parameters.

3. The method according to claim 1, characterized in that, After obtaining the training parameters and obtaining the first radio frequency parameters and environmental parameters of the target device, the method further includes: Train the neural network model using the training parameters so that the neural network model learns and stores the corresponding relationship between the environmental parameters and the second radio frequency parameters; Use a calibration device to calibrate the training parameters to obtain a calibration result; Extract a third logic diagram from the neural network model, and optimize the third logic diagram according to the calibration result; Retrain the neural network model according to the optimized third logic diagram; Input the environmental parameters into the neural network model, and output the second radio frequency parameter; When the difference between the second radio frequency parameter and the first radio frequency parameter is greater than a preset threshold, calibrate the target device based on the second radio frequency parameter and the first radio frequency parameter.

4. The method according to claim 3, wherein Before using the training parameters to train the neural network model so that the neural network model learns and stores the corresponding relationship between the environmental parameters and the second radio frequency parameter, the method further includes: Obtain the device information of multiple devices of the same type in the same application field as the target device; Update the model parameters of the neural network model according to the device information of the multiple devices.

5. The method according to claim 1, characterized in that, Include: The first radio frequency parameter includes: the first operating frequency, the first signal-to-noise ratio, the first signal gain, the first transmitted signal power, the first transmitted signal frequency, and the first s parameter; The second radio frequency parameter includes: the second operating frequency, the second signal-to-noise ratio, the second signal gain, the second transmitted signal power, the second transmitted signal frequency, and the second s parameter.

6. An apparatus calibration device for performing the method according to any one of claims 1 to 5, characterized in that Include: An acquisition module, configured to acquire training parameters, and acquire the first radio frequency parameter and environmental parameters of the target device; A first calibration module, configured to calibrate the training parameters using a calibration device to obtain a calibration result; A construction module, configured to perform linear regression processing on the training parameters and the calibration result to obtain a regression result, and construct a simulation model according to the regression result; An extraction module, configured to extract a first logic diagram from the simulation model, and optimize the first logic diagram according to the calibration result; A training module, configured to train the simulation model according to the optimized first logic diagram; A model module, configured to input the environmental parameters into the simulation model and output the second radio frequency parameter; A second calibration module, configured to calibrate the target device based on the second radio frequency parameter and the first radio frequency parameter when the difference between the second radio frequency parameter and the first radio frequency parameter is greater than a preset threshold.

7. An electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1 to 5.

8. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 5.

Images