Method, device, terminal and storage medium for determining predistortion parameter value
By acquiring and iterating the predistortion parameter values of the RF path, the nonlinear distortion problem of the power amplifier is solved, the signal processing effect is improved, and better signal quality is achieved.
Patent Information
- Application Number
- CN202210682019.3
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2022-06-15
- Publication Date
- 2024-11-08
- Estimated Expiration
- 2042-06-15
AI Technical Summary
How to determine the predistortion parameters to offset the nonlinear distortion of the power amplifier in signal processing and improve the predistortion processing effect of the input signal.
通过获取多个射频通路的预失真参数值,进行样本信号的预失真处理和性能评估,重组和变异处理,迭代训练以确定射频性能最佳的参数值,用于对接收到的输入信号进行预失真处理。
The effect of predistortion processing is improved, so that the output signal can better match the radio frequency path, reduce the impact of nonlinear distortion, and improve the signal quality.
Smart Images

Figure CN115086121B_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the field of wireless communication technology, and in particular to a method, device, terminal and storage medium for determining a predistortion parameter value. Background Art
[0002] At present, a radio frequency path is provided in the terminal, and a power amplifier is provided in the radio frequency path, through which the input signal can be amplified. However, the power amplifier has nonlinear characteristics, and after processing the input signal, the processed output signal will cause nonlinear distortion. In order to offset the nonlinear distortion of the power amplifier, the input signal is generally pre-distorted by a pre-distortion parameter before the power amplifier amplifies the input signal. Therefore, how to determine the pre-distortion parameter has become an urgent problem to be solved. Summary of the invention
[0003] The embodiment of the present application provides a method, device, terminal and storage medium for determining a predistortion parameter value, which can improve the effect of predistortion processing on an input signal. The technical solution is as follows:
[0004] In one aspect, a method for determining a predistortion parameter value is provided, the method comprising:
[0005] Acquire multiple first parameter values of predistortion parameters corresponding to the first radio frequency path;
[0006] Based on the multiple first parameter values, predistortion processing is performed on the first sample input signals respectively to obtain multiple first sample radio frequency signals;
[0007] Based on the radio frequency performances respectively corresponding to the plurality of first sample radio frequency signals, determining a plurality of second parameter values having the highest radio frequency performance parameters from the plurality of first parameter values;
[0008] Performing recombination processing and mutation processing on the plurality of second parameter values to obtain a plurality of third parameter values;
[0009] Based on the RF performance of the second sample RF signal corresponding to the multiple third parameter values respectively, the multiple second parameter values and the multiple third parameter values, the predistortion parameter is iteratively trained to obtain a first target parameter value, and the first target parameter value is used for the first RF path to perform predistortion processing on the received first target input signal.
[0010] On the other hand, a device for determining a predistortion parameter value is provided, the device comprising:
[0011] An acquisition module, used to acquire a plurality of first parameter values of predistortion parameters corresponding to a first radio frequency path;
[0012] A processing module, configured to perform predistortion processing on the first sample input signals respectively based on the multiple first parameter values to obtain multiple first sample radio frequency signals;
[0013] A first determination module, configured to determine, based on the radio frequency performances respectively corresponding to the plurality of first sample radio frequency signals, a plurality of second parameter values having the highest radio frequency performance parameters from the plurality of first parameter values;
[0014] A processing module, used for performing recombination processing and mutation processing on the plurality of second parameter values to obtain a plurality of third parameter values;
[0015] A training module is used to iteratively train the predistortion parameter based on the RF performance of the second sample RF signal corresponding to the multiple third parameter values, the multiple second parameter values and the multiple third parameter values to obtain a first target parameter value, wherein the first target parameter value is used for the first RF path to perform predistortion processing on the received first target input signal.
[0016] On the other hand, a terminal is provided, the terminal comprising a processor and a memory, the memory storing at least one program code, the at least one program code being loaded and executed by the processor to implement the above-mentioned method for determining a pre-distortion parameter value.
[0017] On the other hand, a computer-readable storage medium is provided, wherein at least one program code is stored in the computer-readable storage medium, and the at least one program code is loaded and executed by a processor to implement the above-mentioned method for determining the pre-distortion parameter value.
[0018] On the other hand, a computer program product is provided. The computer program product stores at least one program code, and the at least one program code is loaded and executed by a processor to implement the above-mentioned method for determining the pre-distortion parameter value.
[0019] The beneficial effects of the technical solution provided by the embodiment of the present application are:
[0020] The embodiment of the present application provides a method for determining a pre-distortion parameter value, which first obtains a plurality of first parameter values, then determines a plurality of second parameter values with the highest RF performance parameters from the plurality of first parameter values, obtains a plurality of third parameter values by reorganizing and mutating the plurality of second parameter values, and iteratively trains the pre-distortion parameter based on the plurality of second parameter values and the plurality of third parameter values. Since the plurality of third parameter values are obtained based on the plurality of second parameter values, the second parameter value is equivalent to the parent parameter value of the third parameter value, and the third parameter value is equivalent to the child parameter value of the second parameter value. The pre-distortion parameter is trained together with the parent parameter value and the child parameter value, so that the first target parameter value obtained by training can better match the first RF path, thereby improving the effect of pre-distortion processing of the input signal by the first target parameter value.
[0021] It is to be understood that the foregoing general description and the following detailed description are exemplary only and are not restrictive of the present disclosure. BRIEF DESCRIPTION OF THE DRAWINGS
[0022] Figure 1 is a schematic diagram of an implementation environment of a method for determining a predistortion parameter value provided in an embodiment of the present application;
[0023] Figure 2 is a flow chart of a method for determining a predistortion parameter value provided by an embodiment of the present application;
[0024] Figure 3 is a flow chart of a method for determining a predistortion parameter value provided by an embodiment of the present application;
[0025] Figure 4 is a schematic diagram of an online training pre-distortion parameter provided in an embodiment of the present application;
[0026] Figure 5 is a schematic diagram of an online training pre-distortion parameter provided in an embodiment of the present application;
[0027] Figure 6 is a flow chart of a method for determining a predistortion parameter value provided by an embodiment of the present application;
[0028] Figure 7 is a schematic diagram of an offline training pre-distortion parameter provided in an embodiment of the present application;
[0029] Figure 8 is a schematic diagram of an offline training pre-distortion parameter provided in an embodiment of the present application;
[0030] Fig. 9 It is a structural schematic diagram of a device for determining a predistortion parameter value provided in an embodiment of the present application;
[0031] Fig.10It is a structural block diagram of a terminal provided in an embodiment of the present application. DETAILED DESCRIPTION
[0032] In order to make the technical solutions and advantages of the present application clearer, the implementation methods of the present application are described in further detail below.
[0033] The terms "first", "second", "third" and "fourth" etc. in the specification and claims of the present application and the drawings are used to distinguish different objects, rather than to describe a specific order. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions. For example, a process, method, system, product or device that includes a series of steps or units is not limited to the listed steps or units, but optionally includes steps or units that are not listed, or optionally includes other steps or units inherent to these processes, methods, products or devices.
[0034] It should be noted that the information (including but not limited to user device information, user personal information, etc.), data (including but not limited to data used for analysis, stored data, displayed data, etc.) and signals involved in this application are all authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data must comply with relevant laws, regulations and standards of relevant countries and regions. For example, the pre-distortion parameters, input signals, RF signals, output signals, etc. involved in this application are all obtained with full authorization.
[0035] Figure 1 is a schematic diagram of an implementation environment of a method for determining predistortion parameters provided in an embodiment of the present application, see Figure 1 The implementation environment includes: a terminal 101, a radio frequency path is provided in the terminal 101, and a predistorter and a power amplifier are provided in the radio frequency path. The terminal 101 can first perform predistortion processing on the input signal through the parameter value of the predistortion parameter of the predistorter to obtain a radio frequency signal, and then perform signal amplification processing on the radio frequency signal through the power amplifier to obtain an output signal, and finally transmit the output signal through an antenna.
[0036] In the embodiment of the present application, the output signal transmitted by the antenna can be received by the network device 102 or other devices, and the network device 102 is used as an example for description. Accordingly, the implementation environment also includes: the network device 102, the terminal 101 and the network device 102 can communicate with each other, the network device 102 receives the output signal, adjusts the resource allocation strategy or performs other operations according to the output signal, and no specific limitation is made to this.
[0037] The terminal 101 is a terminal 101 with wireless communication function. The terminal 101 can communicate with one or more core networks via a radio access network (RAN). The terminal 101 can be a mobile terminal 101, such as a mobile phone (or "cellular" phone) and a computer with a mobile terminal 101, for example, a portable, pocket-sized, handheld, computer-built-in or vehicle-mounted mobile device. The terminal 101 can be a mobile phone, a tablet computer, a computer with wireless communication function or a wearable device, etc. In the embodiments of the present application, this is not specifically limited.
[0038] The network device 102 is any network device 102 with wireless transceiver functions. For example, the network device 102 is a base station, an evolved Node B (eNB), a next generation Node B (gNB), an access point (AP) in a wireless fidelity (WIFI) system, a wireless relay node, a wireless backhaul node, a transmission point (TP), or a transmission and reception point (TRP).
[0039] Figure 2 is a flowchart of a method for determining predistortion parameters provided in an embodiment of the present application, see Figure 2 , the method comprising:
[0040] Step 201: Acquire a plurality of first parameter values of predistortion parameters corresponding to a first radio frequency path.
[0041] In an embodiment of the present application, the multiple first parameter values may be parameter values in multiple working states obtained by training the first RF path in an offline state, and one working state corresponds to one first parameter value. The terminal may perform online training on the predistortion parameter based on the multiple first parameter values, and obtain a first target parameter value matching the current working state according to the first parameter values corresponding to the multiple working states.
[0042] The multiple first parameter values may also be randomly initialized parameter values of the first RF path in a working state, and one working state corresponds to multiple first parameter values. The terminal may perform offline training on the predistortion parameter based on the multiple first parameter values to obtain a first target parameter value matching the working state.
[0043] In the embodiment of the present application, a first parameter value includes multiple sub-parameter values, and the number of the multiple sub-parameter values can be set and changed as needed, and there is no specific limitation on this.
[0044] Step 202: Based on a plurality of first parameter values, predistortion processing is performed on first sample input signals respectively to obtain a plurality of first sample radio frequency signals.
[0045] In this step, for any first parameter value, the first sample input signal can be predistorted by the first parameter value to obtain a first sample RF signal, and one first parameter value corresponds to one first sample RF signal.
[0046] In the embodiment of the present application, when training the pre-distortion parameters, the number of the first sample input signals is multiple, and the number can be set and changed as needed, and there is no specific limitation on this. However, when the first sample input signals are pre-distorted based on multiple first parameter values, the first sample input signals are the same sample input signal.
[0047] Step 203: Based on the radio frequency performances respectively corresponding to the plurality of first sample radio frequency signals, determine a plurality of second parameter values having the highest radio frequency performance parameters from the plurality of first parameter values.
[0048] In this step, the RF performance includes at least one of the adjacent channel leakage ratio (ACLR) and the error vector magnitude (EVM). The adjacent channel leakage ratio is mainly used to measure the influence of the RF path on the signal outside the main operating frequency, and the smaller the influence, the better. The error vector magnitude is a vector value with amplitude and angle, which is mainly used to measure the error between the actual output signal and the ideal output signal, and the smaller the error, the better.
[0049] If the radio frequency performance includes adjacent channel leakage ratio and error vector magnitude, the plurality of second parameter values with the highest radio frequency performance parameters are parameter values with the smallest adjacent channel leakage ratio and error vector magnitude.
[0050] Step 204: Perform recombination processing and mutation processing on the multiple second parameter values to obtain multiple third parameter values.
[0051] A parameter value includes multiple sub-parameter values. The reorganization processing is to exchange the multiple sub-parameter values respectively included in the multiple second parameter values. The mutation processing is to randomly process some sub-parameter values in the second parameter value. For example, based on the sub-parameter value, a numerical value is added, a numerical value is subtracted, or it is replaced with other numerical values.
[0052] Step 205: Based on the RF performance of the second sample RF signal corresponding to the plurality of third parameter values, the plurality of second parameter values and the plurality of third parameter values, iteratively train the predistortion parameter to obtain a first target parameter value.
[0053] The first target parameter value is used by the first radio frequency path to perform predistortion processing on a received first target input signal.
[0054] In an embodiment of the present application, if the pre-distortion parameters are trained online, the pre-distortion parameters can be trained online when the first target input signal is received, and after obtaining the first target parameter value, the first target input signal is pre-distorted by the first target parameter value. The pre-distortion parameters can also be trained online when the first RF path is online and idle, that is, before the first target input signal is received, to obtain the first target parameter value, and when the first target input signal is received, the first target input signal is directly pre-distorted by the first target parameter value.
[0055] If the pre-distortion parameters are trained offline, the first target parameter value is first obtained through training, and when the first target input signal is received, the trained first target parameter value is called to perform pre-distortion processing on the first target input signal.
[0056] The embodiment of the present application provides a method for determining a pre-distortion parameter value, which first obtains a plurality of first parameter values, then determines a plurality of second parameter values with the highest RF performance parameters from the plurality of first parameter values, obtains a plurality of third parameter values by reorganizing and mutating the plurality of second parameter values, and iteratively trains the pre-distortion parameter based on the plurality of second parameter values and the plurality of third parameter values. Since the plurality of third parameter values are obtained based on the plurality of second parameter values, the second parameter value is equivalent to the parent parameter value of the third parameter value, and the third parameter value is equivalent to the child parameter value of the second parameter value. The pre-distortion parameter is trained together with the parent parameter value and the child parameter value, so that the first target parameter value obtained by training can better match the first RF path, thereby improving the effect of pre-distortion processing of the input signal by the first target parameter value.
[0057] In an embodiment of the present application, the multiple first parameter values can be multiple parameter values obtained by training the first RF channel in an offline state, or can be randomly initialized parameter values. Next, the multiple first parameter values are taken as an example to illustrate the multiple parameter values obtained by training in an offline state.
[0058] Figure 3 is a flowchart of a method for determining a predistortion parameter value provided by an embodiment of the present application, which is executed by a terminal, see Figure 3 , the method comprising:
[0059] Step 301: The terminal obtains a plurality of first parameter values of pre-distortion parameters of a first radio frequency path trained in an offline state.
[0060] The multiple first parameter values are respectively parameter values of the first RF path when being simulated in multiple working states, and the working state is used to reflect the working power, system bandwidth, battery voltage, etc. of the power amplifier. One working state corresponds to one first parameter value, and the first parameter value includes multiple sub-parameter values.
[0061] In the embodiment of the present application, a plurality of RF paths are provided in the terminal, and different RF paths correspond to different working frequency bands, and the parameter values of the corresponding pre-distortion parameters are also different. Based on this, the terminal can first determine the first RF path that is currently in an online state, and then obtain the plurality of first parameter values obtained by training the first RF path in an offline state from the non-volatile memory.
[0062] In an embodiment of the present application, the multiple first parameter values can be obtained by iterative training based on the signal error between the input signal and the output signal, as in the first implementation method below, or can be obtained by iterative training based on the RF performance of the RF signal corresponding to the input signal, as in the second implementation method below, which will be introduced separately below.
[0063] In the first implementation method, for any working state, the terminal amplifies the second sample input signal through the power amplifier in the first RF path to obtain a first sample output signal, and iteratively trains the pre-distortion parameters based on the signal error between the first sample output signal and the second sample input signal to obtain the first parameter value corresponding to the working state.
[0064] In this implementation, there are multiple second sample input signals, and the number can be set and changed as needed, and is not specifically limited here.
[0065] For any working state, the terminal directly amplifies the second sample input signal through the power amplifier in the first RF path to obtain the first sample output signal. The terminal determines the signal error between the first sample output signal and the second sample input signal, determines the fifth parameter value according to the signal error, and reduces the signal error by continuously adjusting the fifth parameter value to iteratively train the pre-distortion parameters until the signal error is within the preset error range or the number of iterations reaches the maximum number of iterations to obtain the first parameter value. For multiple working states, the terminal is trained in the above manner to obtain multiple first parameter values. Among them, the above offline training process can be performed in the baseband chip of the terminal, or in a PC (Personal Computer) device or other device.
[0066] See also Figure 4 , Figure 4In the above, offline training is performed in the first implementation manner to obtain multiple first parameter values, and then the multiple first parameter values are applied to the online training process of the predistortion parameters.
[0067] The second implementation method is that for any working state, the terminal obtains multiple fourth parameter values of the pre-distortion parameter corresponding to the first RF path, and based on the multiple fourth parameter values, pre-distorts the third sample input signal respectively to obtain multiple third sample RF signals, and iteratively trains the pre-distortion parameter based on the RF performance and multiple fourth parameter values corresponding to the multiple third sample RF signals to obtain the first parameter value.
[0068] In this implementation, for any working state, multiple fourth parameter values are respectively parameter values of the first RF path being simulated and randomly initialized in this working state, and the terminal performs pre-distortion processing on the third sample input signal through the multiple fourth parameter values to obtain multiple third sample RF signals, and then measures the RF performance of the multiple third sample RF signals through the RF performance self-test module, sorts the multiple fourth parameter values based on the RF performance corresponding to the multiple third sample RF signals, selects a preset number of fifth parameter values with the highest RF performance parameters from the multiple fourth parameter values, and obtains multiple fifth parameter values.
[0069] The terminal performs recombination and mutation processing on multiple fifth parameter values to obtain multiple sixth parameter values, and based on the multiple sixth parameter values, performs pre-distortion processing on the third sample input signal to obtain multiple fourth sample RF signals, and iteratively trains the pre-distortion parameters based on the RF performance corresponding to the multiple fourth sample RF signals, the multiple fifth parameter values, and the multiple sixth parameter values, until the number of iterations reaches the maximum number of iterations or the RF performance corresponding to the multiple fourth sample RF signals reaches the target value, and determines that the second convergence condition is met, and the terminal determines the fifth parameter value with the highest RF performance parameter corresponding to the second convergence condition as the first parameter value in the working state. For multiple working states, the terminal performs training in the above manner to obtain multiple first parameter values.
[0070] In this implementation, the process of the terminal obtaining multiple first parameter values based on multiple fourth parameter values is similar to steps 302-307, which will not be described in detail here.
[0071] For the above two implementation methods, after the terminal obtains multiple first parameter values, it can store the multiple first parameter values corresponding to the RF path in the non-volatile memory of the terminal according to the identification of the RF path. When the RF path is in an online state, the terminal directly calls the pre-stored multiple first parameter values corresponding to the RF path for online training.
[0072] See also Figure 5For the first RF path, the terminal trains the pre-distortion parameters when the first RF path is in an offline state to obtain multiple first parameter values, and then stores the multiple first parameter values in a non-volatile memory. When the pre-distortion parameters are trained online, the multiple first parameter values are obtained from the non-volatile memory.
[0073] It should be noted that if the initialized parameter values are used for training, the training time is long and it is difficult to implement online training. Therefore, in the embodiment of the present application, the parameter values obtained by offline training are used as the initialized parameter values for training, which can shorten the training time and implement online training. In addition to the above two implementation methods, the terminal can also perform offline training in other ways to obtain multiple first parameter values, which is not specifically limited.
[0074] Step 302: When the first radio frequency path is in an online state and in an idle state, the terminal performs predistortion processing on first sample input signals based on multiple first parameter values to obtain multiple first sample radio frequency signals.
[0075] A first parameter value corresponds to a first sample radio frequency signal.
[0076] In the embodiment of the present application, when training the predistortion parameters, the number of first sample input signals is multiple, but when predistortion processing is performed on the first sample input signals respectively based on multiple first parameter values, the first sample input signals are the same sample input signal.
[0077] When the first RF path is online and idle, for any first parameter value, the terminal performs predistortion processing on the first sample input signal through the first parameter value, outputs the predistortion processed signal through the power amplifier, and obtains the first sample RF signal.
[0078] In the embodiment of the present application, when the first RF path is in an online state and in an idle state, iterative training is performed on the predistortion parameters to ensure that the online training does not affect the normal operation of the transmitter of the first RF path.
[0079] Step 303: The terminal determines a plurality of second parameter values having the highest RF performance parameters from a plurality of first parameter values based on the RF performances respectively corresponding to the plurality of first sample RF signals.
[0080] The RF performance includes at least one of the adjacent channel leakage ratio and the error vector magnitude. The adjacent channel leakage ratio is mainly used to measure the influence of the RF path on the signal outside the main operating frequency. The smaller the influence, the better. The error vector magnitude is a vector value with amplitude and angle. It is mainly used to measure the error between the actual output signal and the ideal output signal. The smaller the error, the better.
[0081] In an embodiment of the present application, if the RF performance includes an adjacent channel leakage ratio, the terminal measures the adjacent channel leakage ratios corresponding to the multiple first sample RF signals respectively, and obtains the RF performance corresponding to the multiple first sample RF signals respectively. If the RF performance includes an error vector magnitude, the terminal measures the error vector magnitudes corresponding to the multiple first sample RF signals respectively, and obtains the RF performance corresponding to the multiple first sample RF signals respectively. If the RF performance includes an adjacent channel leakage ratio and an error vector magnitude, the terminal measures the adjacent channel leakage ratios and the error vector magnitudes corresponding to the multiple first sample RF signals respectively, and obtains the RF performance corresponding to the multiple first sample RF signals respectively.
[0082] The RF path includes a RF performance self-test module, and the terminal can measure the RF performances respectively corresponding to the plurality of first sample RF signals through the RF performance self-test module.
[0083] In the embodiment of the present application, a first sample RF signal corresponds to a RF performance, and since a first parameter value corresponds to a first sample RF signal, a first parameter value corresponds to a RF performance. Based on this, the terminal can sort multiple first parameter values based on the level of RF performance, select a preset number of second parameter values with the highest RF performance parameters from the multiple first parameter values, and obtain multiple second parameter values.
[0084] The smaller the adjacent channel leakage ratio and the error vector magnitude, the higher the RF performance parameter. If the RF performance includes one of the adjacent channel leakage ratio and the error vector magnitude, the terminal can determine multiple second parameter values with the smallest adjacent channel leakage ratio or the error vector magnitude from multiple first parameter values. If the RF performance includes the adjacent channel leakage ratio and the error vector magnitude, the terminal can determine multiple second parameter values with the smallest adjacent channel leakage ratio and the error vector magnitude from multiple first parameter values.
[0085] The number of the second parameter values may be equal to the number of the first parameter values, or may be less than the number of the first parameter values, and there is no specific limitation on this.
[0086] Step 304: The terminal performs a recombination process and a mutation process on the multiple second parameter values to obtain multiple third parameter values.
[0087] In this step, for any second parameter value, the terminal exchanges the first part of the sub-parameter value of the second parameter value with the second part of the sub-parameter value of other second parameter values and mutates the third part of the sub-parameter value of the second parameter value to obtain a third parameter value.
[0088] A second parameter value includes multiple sub-parameter values. For any second parameter value, the terminal can exchange the first part of the sub-parameter values of the second parameter value with the second part of the sub-parameter values of one or more other second parameter values, and perform mutation processing on the third part of the sub-parameter values of the second parameter value to obtain a third parameter value.
[0089] For example, the number of second parameter values is 3, namely the first second parameter value, the second second parameter value and the third second parameter value, and one second parameter value includes 8 sub-parameter values. For the first second parameter value, the terminal exchanges 2 sub-parameter values in the first second parameter value with 2 sub-parameter values in the second second parameter value, and exchanges the other 2 sub-parameter values in the first second parameter value with 2 sub-parameter values in the third second parameter value, and performs mutation processing on the other 2 sub-parameter values in the first second parameter value, for example, adding a numerical value or subtracting a numerical value on the basis of the current sub-parameter value, or replacing the sub-parameter value with a preset numerical value. The method of the mutation processing is not specifically limited here, and finally the third parameter value is obtained.
[0090] One thing that needs to be explained is that, according to the above process, it can be seen that the third parameter value is obtained by reorganization and mutation based on the second parameter value. Therefore, the second parameter value is the parent parameter value of the third parameter value, and the third parameter value is the child parameter value of the second parameter value.
[0091] Step 305: The terminal performs predistortion processing on the first sample input signal based on multiple third parameter values to obtain multiple second sample RF signals.
[0092] A third parameter value corresponds to a second sample RF signal.
[0093] The process of the terminal performing predistortion processing on the first sample input signal based on multiple third parameter values is similar to the process of performing predistortion processing on the first sample input signal based on multiple first parameter values, which will not be repeated here.
[0094] Step 306: If the RF performance corresponding to the multiple second sample RF signals does not meet the first convergence condition, the terminal re-determines the multiple second parameter values with the highest RF performance parameters from the multiple second parameter values and the multiple third parameter values based on the RF performance corresponding to the multiple second parameter values and the RF performance corresponding to the multiple third parameter values.
[0095] The first convergence condition may be that the radio frequency performances respectively corresponding to the plurality of second sample radio frequency signals reach a target value or the number of iterations reaches a maximum number of iterations.
[0096] The terminal can measure the RF performance corresponding to multiple second sample RF signals respectively, and then determine whether the number of iterations reaches the maximum number of iterations or whether the RF performance corresponding to the multiple second sample RF signals respectively reaches the target value. If the number of iterations does not reach the maximum number of iterations and the RF performance corresponding to the multiple second sample RF signals respectively does not reach the target value, it is determined that the first convergence condition is not met. The terminal then re-determines multiple second parameter values with the highest RF performance parameters from the multiple second parameter values and the multiple third parameter values based on the RF performance corresponding to the multiple second parameter values respectively and the RF performance corresponding to the multiple third parameter values respectively.
[0097] When redetermining multiple second parameter values, the terminal can sort the RF performance corresponding to the multiple second parameter values and the RF performance corresponding to the multiple third parameter values based on the RF performance parameters, thereby redetermining multiple second parameter values with the highest RF performance parameters.
[0098] In the embodiment of the present application, the advantage of sorting together based on the RF performance corresponding to the second parameter value and the third parameter value is that even if the offspring obtained by this training, that is, the third parameter value, does not produce better RF performance, the parent parameter value, that is, the second parameter value, can still be used, and the overall performance of the parameter value of the pre-distortion parameter will not deteriorate.
[0099] Step 307: The terminal performs a step of reorganizing and mutating the multiple second parameter values based on the re-determined multiple second parameter values to obtain multiple third parameter values, until the RF performance corresponding to the multiple second sample RF signals respectively meets the first convergence condition, and the first target parameter value is obtained.
[0100] The first target parameter value is used for the first radio frequency path to perform predistortion processing on the first target input signal when the first radio frequency path is in an operating state.
[0101] The terminal reorganizes and mutates the re-determined multiple second parameter values based on the re-determined multiple second parameter values to obtain multiple third parameter values, and pre-distorts the first sample input signal based on the multiple third parameter values to obtain multiple second sample RF signals, and determines whether the number of iterations reaches the maximum number of iterations or whether the RF performances corresponding to the multiple second sample RF signals reach the target values. If the number of iterations does not reach the maximum number of iterations and the RF performances corresponding to the multiple second sample RF signals do not reach the target values, it is determined that the first convergence condition is not met, and the terminal re-executes steps 304 to 307 until the number of iterations reaches the maximum number of iterations or the RF performances corresponding to the multiple second sample RF signals reach the target values.
[0102] The terminal determines a plurality of second parameter values when the first convergence condition is satisfied, and determines a first target parameter value from the plurality of second parameter values.
[0103] In an embodiment of the present application, after the terminal obtains multiple second parameter values, the multiple second parameter values can be stored in a storage unit (Memory) or a non-volatile memory (NVM). If the terminal stores multiple second parameter values in the storage unit, then when the terminal is powered on again after power failure, the terminal re-iteratively trains the pre-distortion parameters based on the multiple first parameter values, which can save storage space of the non-volatile memory. If the terminal stores multiple second parameter values in the non-volatile memory, then when the terminal is powered on again after power failure, the terminal can determine that the multiple second parameter values stored before power failure are multiple first parameter values, and re-iteratively train the pre-distortion parameters through the above steps, which can shorten the training time.
[0104] Continue to see Figure 5 The terminal uses the obtained multiple first parameter values as the initialization parameter values, and then uses the result of the RF performance self-test as the fitness function. Then, based on the RF performance self-test result, multiple second parameter values, i.e., the parent parameter values, are determined, and the multiple second parameter values are reorganized and mutated to obtain multiple third parameter values, i.e., the child parameter values. The RF performance of the child parameter values is measured, and it is determined whether the first convergence condition is met according to the test result. If the first convergence condition is met, the parent parameter value when the first convergence condition is met is stored in the storage unit. If the first RF path is continuously in the online state, the parent parameter value is used as the initialization parameter value for the next online training, i.e., the first parameter value. If the convergence condition is not met, the parent parameter value is re-determined, and iterative training continues.
[0105] In the embodiment of the present application, the timing at which the terminal determines the first target parameter value from the multiple second parameter values can be implemented in any of the following ways:
[0106] In a first implementation manner, before receiving the first target input signal, the terminal determines a first target parameter value from a plurality of second parameter values, and when receiving the first target input signal, directly performs pre-distortion processing on the first target input signal using the first target parameter value.
[0107] In a second implementation method, the terminal iteratively trains the pre-distortion parameters before receiving the first target input signal, determines the first target parameter value from multiple second parameter values when receiving the first target input signal, and then pre-distorts the first target input signal using the first target parameter value.
[0108] In a third implementation, when receiving a first target input signal, the terminal iteratively trains the pre-distortion parameters to obtain multiple second parameter values, then determines a first target parameter value from the multiple second parameter values, and performs pre-distortion processing on the first target input signal using the first target parameter value.
[0109] In an embodiment of the present application, the process of the terminal determining the first target parameter value may be: the terminal determines the second parameter value with the highest RF performance parameter from multiple second parameter values as the first target parameter value, or determines the second parameter value corresponding to the current working state as the first target parameter value.
[0110] If the radio frequency performance includes one of an adjacent channel leakage ratio and an error vector magnitude, the terminal determines a second parameter value with the smallest adjacent channel leakage ratio or error vector magnitude from a plurality of second parameter values as the first target parameter value. If the radio frequency performance includes an adjacent channel leakage ratio and an error vector magnitude, the terminal determines a second parameter value with the smallest adjacent channel leakage ratio and error vector magnitude from a plurality of second parameter values as the first target parameter value.
[0111] One point that needs to be explained is that the terminal periodically executes the above steps 302-307 when the first RF path is online and idle, until the cycle reaches the working cycle or working time of the first RF path. The terminal determines the first target parameter value from the multiple second parameter values obtained from the last cycle training, and pre-distorts the first target input signal through the first target parameter value.
[0112] It can be seen that when the RF path is always online and the RF path remains unchanged, the online training of the pre-distortion parameters can be continued within a preset period. When the working state of the terminal changes, especially when the antenna load changes and the ambient temperature changes, the performance of the transmitter in the RF path can be greatly improved. Generally speaking, after applying online memory pre-distortion, the voltage standing wave ratio (VSWR) at the load end of the power amplifier can be within 2:1 to obtain good transmission performance, which greatly reduces the difficulty of amplifier matching circuit design and improves transmission performance. Therefore, the method for online training of pre-distortion parameters provided in the embodiment of the present application can make the performance of the transmitter better and better with use.
[0113] Furthermore, the method does not require training of predistortion parameters under high and low temperature conditions, thus saving training time.
[0114] If the RF path currently in an online state changes, for example, from the first RF path to the second RF path, the terminal obtains multiple first parameter values of the pre-distortion parameters of the second RF path trained in the offline state, and then iteratively trains the pre-distortion parameters of the second RF path by executing the above steps 302-307 to obtain multiple second parameter values of the second RF path, determines a second target parameter value from the multiple second parameter values, and performs pre-distortion processing on the second target input signal by using the second target parameter value.
[0115] In the embodiment of the present application, the genetic algorithm is very suitable for solving nonlinear problems such as pre-distortion parameters. The genetic algorithm is combined with the results of the RF performance self-test as its fitness function to iteratively train the pre-distortion parameters. It does not require a large amount of computing resources. It only needs to apply the first target parameter value determined from multiple second parameter values to the pre-distorter, which greatly reduces the complexity of system implementation, improves the stability of system performance, and makes it easier to implement in the RF chip. It is no longer necessary to send the signal received by the feedback receiver back to the baseband chip for processing, which saves the high-speed digital signal interface resources of the RF chip and the baseband chip and the need for a large amount of computing resources of the baseband chip, thereby reducing the cost and power consumption of pre-distortion parameter training.
[0116] The embodiment of the present application provides a method for determining a pre-distortion parameter value, which first obtains a plurality of first parameter values, then determines a plurality of second parameter values with the highest RF performance parameters from the plurality of first parameter values, obtains a plurality of third parameter values by reorganizing and mutating the plurality of second parameter values, and iteratively trains the pre-distortion parameter based on the plurality of second parameter values and the plurality of third parameter values. Since the plurality of third parameter values are obtained based on the plurality of second parameter values, the second parameter value is equivalent to the parent parameter value of the third parameter value, and the third parameter value is equivalent to the child parameter value of the second parameter value. The pre-distortion parameter is trained together with the parent parameter value and the child parameter value, so that the first target parameter value obtained by training can better match the first RF path, thereby improving the effect of pre-distortion processing of the input signal by the first target parameter value.
[0117] Next, a description is given by taking a plurality of first parameter values as randomly initialized parameter values as an example.
[0118] Figure 6 is a flow chart of a method for determining a predistortion parameter value provided by an embodiment of the present application, wherein the terminal, referring to Figure 6 , the method comprising:
[0119] Step 601: The terminal obtains a plurality of initialized first parameter values.
[0120] The multiple first parameter values are randomly initialized parameter values corresponding to a working state.
[0121] Step 602: The terminal performs predistortion processing on first sample input signals based on multiple first parameter values to obtain multiple first sample RF signals.
[0122] Step 603: The terminal determines a plurality of second parameter values having the highest RF performance parameters from a plurality of first parameter values based on the RF performances respectively corresponding to the plurality of first sample RF signals.
[0123] Step 604: The terminal performs a recombination process and a mutation process on the multiple second parameter values to obtain multiple third parameter values.
[0124] Step 605: The terminal performs predistortion processing on the first sample input signal based on multiple third parameter values to obtain multiple second sample RF signals.
[0125] Step 606: If the RF performance corresponding to the multiple second sample RF signals does not meet the first convergence condition, the terminal re-determines the multiple second parameter values with the highest RF performance parameters from the multiple second parameter values and the multiple third parameter values based on the RF performance corresponding to the multiple second parameter values and the RF performance corresponding to the multiple third parameter values.
[0126] Step 607: The terminal performs a step of reorganizing and mutating the multiple second parameter values based on the re-determined multiple second parameter values to obtain multiple third parameter values, until the RF performance corresponding to the multiple second sample RF signals respectively meets the first convergence condition, and the first target parameter value is obtained.
[0127] In an embodiment of the present application, the terminal determines multiple second parameter values when the RF performance corresponding to the multiple second sample RF signals respectively meets the first convergence condition. The multiple second parameter values can be applied to the pre-distortion processing of the input signal, and can also be used as Figure 3 In the illustrated embodiment, multiple first parameter values obtained by offline training are directly applied to online training. If multiple second parameter values are applied to the pre-distortion processing of the input signal, after the terminal determines the multiple second parameter values, it determines the first target parameter value from the multiple second parameter values, and when the first RF path is online and in working state, the first target input signal is pre-distorted by the first target parameter value. Alternatively, the terminal may also determine the first target parameter value from the multiple second parameter values when receiving the first target input signal, and then pre-distort the first target input signal by the first target parameter value.
[0128] The terminal may determine, from a plurality of second parameter values, a second parameter value with the highest radio frequency performance parameter as a first target parameter value, where the first target parameter value corresponds to the parameter value input to the first target in the working state.
[0129] For multiple working states, the terminal is trained in the above manner to obtain target parameter values corresponding to the multiple working states respectively, and then when the first radio frequency is actually in the working state, the target input signal is predistorted by the target parameter value corresponding to the working state.
[0130] Steps 602-607 are similar to steps 302-307 and will not be repeated here.
[0131] In the embodiment of the present application, the RF performance test result is used as the fitness function. Since each fitness calculation measures the RF performance instead of calculating the signal error using the output signal and the input signal, the training of the pre-distortion parameters does not require a large amount of computing resources, does not require the output signal and the input signal to have good time synchronization, does not require the calculation of the signal gain, and is not sensitive to the noise of the feedback receiver.
[0132] Furthermore, since the pre-distortion parameters are trained using the test results of the RF performance, the feedback receiver of the RF transceiver no longer needs to oversample the feedback received signal, which reduces the bandwidth requirements of the RF feedback receiver and further reduces the cost and power consumption of the RF chip feedback receiver circuit. In addition, the pre-distortion parameter training method gets rid of the dependence on the signal delay, signal gain and noise performance of the feedback receiver, and improves the stability of the system performance.
[0133] See also Figure 7 ,from Figure 7 It can be seen from the figure that: the terminal obtains multiple initialized first parameter values, then uses the result of the RF performance self-test as the fitness function, determines multiple second parameter values, i.e., the parent parameter values, based on the RF performance self-test result, performs recombination and mutation processing on the multiple second parameter values, obtains multiple third parameter values, i.e., the child parameter values, measures the RF performance of the child parameter values, and determines whether the convergence condition is met according to the test result. If the convergence condition is not met, the training is continued. If the convergence condition is met, the first target parameter value is obtained, and the first target parameter value is applied to the pre-distortion processing of the input signal. See Figure 8 , Figure 8 This is a schematic diagram of applying the first target parameter value obtained by the above process to the predistortion processing of the input signal.
[0134] The embodiment of the present application provides a method for determining a pre-distortion parameter value, which first obtains a plurality of first parameter values, then determines a plurality of second parameter values with the highest RF performance parameters from the plurality of first parameter values, obtains a plurality of third parameter values by reorganizing and mutating the plurality of second parameter values, and iteratively trains the pre-distortion parameter based on the plurality of second parameter values and the plurality of third parameter values. Since the plurality of third parameter values are obtained based on the plurality of second parameter values, the second parameter value is equivalent to the parent parameter value of the third parameter value, and the third parameter value is equivalent to the child parameter value of the second parameter value. The pre-distortion parameter is trained together with the parent parameter value and the child parameter value, so that the first target parameter value obtained by training can better match the first RF path, thereby improving the effect of pre-distortion processing of the input signal by the first target parameter value.
[0135] Fig. 9 is a schematic diagram of a structure of a device for determining a predistortion parameter value provided in an embodiment of the present application, see Fig. 9 , the device comprises:
[0136] An acquisition module 901 is used to acquire a plurality of first parameter values of predistortion parameters corresponding to a first radio frequency path;
[0137] A first processing module 902 is used to perform predistortion processing on the first sample input signal based on multiple first parameter values to obtain multiple first sample RF signals;
[0138] A first determination module 903 is configured to determine, based on the radio frequency performances respectively corresponding to the plurality of first sample radio frequency signals, a plurality of second parameter values having the highest radio frequency performance parameters from the plurality of first parameter values;
[0139] A second processing module 904 performs a recombination process and a mutation process on the plurality of second parameter values to obtain a plurality of third parameter values;
[0140] The training module 905 is used to iteratively train the pre-distortion parameters based on the RF performance of the second sample RF signal corresponding to the multiple third parameter values, the multiple second parameter values and the multiple third parameter values, so as to obtain the first target parameter value, and the first target parameter value is used for the first RF path to perform pre-distortion processing on the received first target input signal.
[0141] In one possible implementation, the training module 905 is used to perform pre-distortion processing on the first sample input signal based on multiple third parameter values to obtain multiple second sample RF signals; if the RF performance corresponding to the multiple second sample RF signals does not meet the first convergence condition, based on the RF performance corresponding to the multiple second parameter values and the RF performance corresponding to the multiple third parameter values, re-determine the multiple second parameter values with the highest RF performance parameters from the multiple second parameter values and the multiple third parameter values; based on the re-determined multiple second parameter values, reorganize and mutate the multiple second parameter values to obtain multiple third parameter values, until the RF performance corresponding to the multiple second sample RF signals meets the first convergence condition to obtain the first target parameter value.
[0142] In another possible implementation, the device includes:
[0143] The second determination module 906 is used to determine multiple second parameter values when the RF performance corresponding to the multiple second sample RF signals meets the first convergence condition; and determine the second parameter value with the highest RF performance parameter as the first target parameter value from the multiple second parameter values.
[0144] In another possible implementation, a parameter value includes multiple sub-parameter values;
[0145] The second processing module 904 is used for exchanging the first part of the sub-parameter values of the second parameter value with the second part of the sub-parameter values of other second parameter values and performing mutation processing on the third part of the sub-parameter values of the second parameter value for any second parameter value to obtain a third parameter value.
[0146] In another possible implementation, the multiple first parameter values are parameter values in multiple working states obtained by training the first radio frequency path in an offline state, and one working state corresponds to one first parameter value;
[0147] The acquisition module 901 is used to perform signal amplification processing on the second sample input signal through a power amplifier in the first RF path to obtain a first sample output signal; based on the signal error between the first sample output signal and the second sample input signal, iteratively train the predistortion parameter to obtain a first parameter value.
[0148] In another possible implementation, the multiple first parameter values are parameter values in multiple working states obtained by training the first radio frequency path in an offline state, and one working state corresponds to one first parameter value;
[0149] The acquisition module 901 is used to obtain multiple fourth parameter values of the pre-distortion parameter corresponding to the first RF path; based on the multiple fourth parameter values, the third sample input signal is pre-distorted to obtain multiple third sample RF signals; based on the RF performance corresponding to the multiple third sample RF signals and the multiple fourth parameter values, the pre-distortion parameter is iteratively trained to obtain the first parameter value.
[0150] In another possible implementation, the first processing module 902 is used to perform pre-distortion processing on the first sample input signal based on multiple first parameter values to obtain multiple first sample RF signals when the first RF path is in an online state and in an idle state; wherein the pre-distortion processing on the first sample input signal based on multiple first parameter values to obtain multiple first sample RF signals is performed periodically.
[0151] In another possible implementation, the radio frequency performance includes: at least one of an adjacent channel leakage ratio and an error vector magnitude;
[0152] The device also includes:
[0153] The measuring module is used to measure at least one of an adjacent channel leakage ratio and an error vector magnitude of the first sample radio frequency signal to obtain radio frequency performance of the first sample radio frequency signal.
[0154] The embodiment of the present application provides a device for determining a pre-distortion parameter value, which first obtains a plurality of first parameter values, then determines a plurality of second parameter values with the highest RF performance parameters from the plurality of first parameter values, obtains a plurality of third parameter values by reorganizing and mutating the plurality of second parameter values, and iteratively trains the pre-distortion parameter based on the plurality of second parameter values and the plurality of third parameter values. Since the plurality of third parameter values are obtained based on the plurality of second parameter values, the second parameter value is equivalent to the parent parameter value of the third parameter value, and the third parameter value is equivalent to the child parameter value of the second parameter value. The pre-distortion parameter is trained together with the parent parameter value and the child parameter value, so that the first target parameter value obtained by training can better match the first RF path, thereby improving the effect of pre-distortion processing of the input signal by the first target parameter value.
[0155] Please refer to Fig.10 , which shows a structural block diagram of a terminal 1000 shown in an exemplary embodiment of the present application. The terminal 1000 may be a device having the function of controlling other devices, such as a smart phone and a tablet computer. The terminal 1000 in the present application may include one or more of the following components: a processor 1010 and a memory 1020.
[0156] The processor 1010 may include one or more processing cores. The processor 1010 uses various interfaces and lines to connect the various parts of the entire terminal 1000, and executes various functions and processes data of the terminal 1000 by running or executing program codes, programs, code sets or program code sets stored in the memory 1020, and calling data stored in the memory 1020. Optionally, the processor 1010 can be implemented in at least one hardware form of digital signal processing (DSP), field programmable gate array (FPGA), and programmable logic array (PLA). The processor 1010 can integrate one or more combinations of a central processing unit (CPU), a graphics processing unit (GPU), a neural network processor (NPU), and a modem. Among them, the CPU mainly processes the operating system, user interface, and application programs; the GPU is responsible for rendering and drawing the content to be displayed on the display screen; the NPU is used to implement artificial intelligence (AI) functions; and the modem is used to process wireless communications. It is understandable that the above-mentioned modem may not be integrated into the processor 1010, but may be implemented by a separate chip.
[0157] The memory 1020 may include a random access memory (RAM) or a read-only memory (ROM). Optionally, the memory 1020 includes a non-transitory computer-readable storage medium. The memory 1020 may be used to store program codes, programs, codes, code sets, or program code sets. The memory 1020 may include a program storage area and a data storage area, wherein the program storage area may store program codes for implementing an operating system, program codes for at least one function (such as a touch function, a sound playback function, an image playback function, etc.), program codes for implementing the above-mentioned various method embodiments, etc.; the data storage area may store data (such as audio data, a phone book), etc., created according to the use of the terminal 1000.
[0158] In addition, those skilled in the art can understand that the structure of the terminal 1000 shown in the above drawings does not constitute a limitation on the terminal 1000, and the terminal 1000 may include more or fewer components than shown, or combine certain components, or arrange the components differently. For example, the terminal 1000 also includes a microphone, a speaker, a radio frequency circuit, an input unit, a sensor, an audio circuit, a wireless fidelity (Wireless Fidelity, Wi-Fi) module, a power supply, a Bluetooth module and other components, which will not be repeated here.
[0159] In an exemplary embodiment, a computer-readable medium is further provided. The computer-readable medium stores at least one program code. The at least one program code is loaded and executed by a processor to implement the method for determining the pre-distortion parameter value in the above embodiment.
[0160] In an exemplary embodiment, a computer program product is further provided. The computer program product stores at least one program code. The at least one program code is loaded and executed by a processor to implement the method for determining the pre-distortion parameter value in the above embodiment.
[0161] In some embodiments, the computer program involved in the embodiments of the present application may be deployed and executed on a computer device, or on multiple computer devices located at one location, or on multiple computer devices distributed at multiple locations and interconnected by a communication network. Multiple computer devices distributed at multiple locations and interconnected by a communication network may constitute a blockchain system.
[0162] A person skilled in the art will understand that all or part of the steps to implement the above embodiments may be accomplished by hardware or by instructing related hardware through a program, and the program may be stored in a computer-readable storage medium, and the above-mentioned storage medium may be a read-only memory, a disk or an optical disk, etc.
[0163] The above description is only for the purpose of facilitating the technical solution of the present application to be understood by those skilled in the art, and is not intended to limit the present application. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application shall be included in the protection scope of the present application.
Claims
1. A method for determining a predistortion parameter value, characterized in that: The method comprises: Acquire multiple first parameter values of predistortion parameters corresponding to the first radio frequency path; Based on the multiple first parameter values, predistortion processing is performed on the first sample input signals respectively to obtain multiple first sample radio frequency signals; Based on the radio frequency performances respectively corresponding to the plurality of first sample radio frequency signals, determine a plurality of second parameter values having the highest radio frequency performance parameters from the plurality of first parameter values, wherein one second parameter value includes a plurality of sub-parameter values; Performing recombination processing and mutation processing on the plurality of second parameter values to obtain a plurality of third parameter values; Based on the multiple third parameter values, pre-distortion processing is performed on the first sample input signal to obtain multiple second sample RF signals; if the RF performances corresponding to the multiple second sample RF signals do not meet the first convergence condition, based on the RF performances corresponding to the multiple second parameter values and the RF performances corresponding to the multiple third parameter values, multiple second parameter values with the highest RF performance parameters are re-determined from the multiple second parameter values and the multiple third parameter values; based on the re-determined multiple second parameter values, the steps of reorganizing and mutating the multiple second parameter values to obtain multiple third parameter values are performed until the RF performances corresponding to the multiple second sample RF signals meet the first convergence condition to obtain a first target parameter value, which is used by the first RF path to perform pre-distortion processing on the received first target input signal; The step of performing recombination and mutation processing on the plurality of second parameter values to obtain a plurality of third parameter values includes: For any second parameter value, the first part of the sub-parameter values of the second parameter value is exchanged with the second part of the sub-parameter values of other second parameter values, and the third part of the sub-parameter values of the second parameter value is mutated to obtain the third parameter value.
2. The method according to claim 1, characterized in that The process of determining the first target parameter value includes: Determine a plurality of second parameter values when the radio frequency performances respectively corresponding to the plurality of second sample radio frequency signals meet the first convergence condition; A second parameter value having the highest radio frequency performance parameter is determined from the multiple second parameter values as the first target parameter value.
3. The method according to claim 1, characterized in that: The multiple first parameter values are parameter values in multiple working states obtained by training the first radio frequency path in an offline state, and one working state corresponds to one first parameter value; For any working state, the process of obtaining a first parameter value corresponding to the working state includes: A second sample input signal is amplified by a power amplifier in the first radio frequency path to obtain a first sample output signal; The predistortion parameter is iteratively trained based on a signal error between the first sample output signal and the second sample input signal to obtain the first parameter value.
4. The method according to claim 1, characterized in that: The multiple first parameter values are parameter values in multiple working states obtained by training the first radio frequency path in an offline state, and one working state corresponds to one first parameter value; For any working state, the process of obtaining a first parameter value corresponding to the working state includes: Acquire multiple fourth parameter values of the predistortion parameters corresponding to the first radio frequency path; Based on the multiple fourth parameter values, predistortion processing is performed on the third sample input signal respectively to obtain multiple third sample radio frequency signals; Based on the radio frequency performances respectively corresponding to the multiple third sample radio frequency signals and the multiple fourth parameter values, the predistortion parameter is iteratively trained to obtain the first parameter value.
5. The method according to claim 3 or 4, characterized in that: The pre-distortion processing is performed on the first sample input signal based on the multiple first parameter values to obtain multiple first sample radio frequency signals, including: When the first radio frequency path is in an online state and in an idle state, pre-distortion processing is performed on the first sample input signals based on the multiple first parameter values to obtain multiple first sample radio frequency signals; The step of performing predistortion processing on the first sample input signal based on the multiple first parameter values to obtain multiple first sample radio frequency signals is performed periodically.
6. The method according to claim 1, characterized in that The radio frequency performance includes: at least one of an adjacent channel leakage ratio and an error vector magnitude; The process of determining the radio frequency performance of the first sample radio frequency signal includes: At least one of an adjacent channel leakage ratio and an error vector magnitude (EVM) of the first sample RF signal is measured to obtain RF performance of the first sample RF signal.
7. A device for determining a predistortion parameter value, characterized in that: The device comprises: An acquisition module, used to acquire a plurality of first parameter values of predistortion parameters corresponding to a first radio frequency path; A processing module, configured to perform predistortion processing on the first sample input signals respectively based on the multiple first parameter values to obtain multiple first sample radio frequency signals; A first determination module, configured to determine, based on the radio frequency performances respectively corresponding to the plurality of first sample radio frequency signals, a plurality of second parameter values having the highest radio frequency performance parameters from the plurality of first parameter values, wherein one second parameter value includes a plurality of sub-parameter values; A processing module, used for performing recombination processing and mutation processing on the plurality of second parameter values to obtain a plurality of third parameter values; A training module, configured to perform pre-distortion processing on the first sample input signal based on the multiple third parameter values to obtain multiple second sample RF signals; if the RF performances respectively corresponding to the multiple second sample RF signals do not meet the first convergence condition, based on the RF performances respectively corresponding to the multiple second parameter values and the RF performances respectively corresponding to the multiple third parameter values, re-determine multiple second parameter values with the highest RF performance parameters from the multiple second parameter values and the multiple third parameter values; based on the re-determined multiple second parameter values, perform the steps of performing recombination processing and mutation processing on the multiple second parameter values to obtain multiple third parameter values, until the RF performances respectively corresponding to the multiple second sample RF signals meet the first convergence condition, and obtain a first target parameter value, the first target parameter value being used for the first RF path to perform pre-distortion processing on the received first target input signal; The processing module is used for exchanging the first part of the sub-parameter values of any second parameter value with the second part of the sub-parameter values of other second parameter values and performing mutation processing on the third part of the sub-parameter values of the second parameter value to obtain a third parameter value.
8. The device according to claim 7, characterized in that The second determination module is used to determine multiple second parameter values when the radio frequency performance corresponding to the multiple second sample radio frequency signals respectively meets the first convergence condition; and determine the second parameter value with the highest radio frequency performance parameter from the multiple second parameter values as the first target parameter value.
9. The device according to claim 7, characterized in that The multiple first parameter values are parameter values in multiple working states obtained by training the first radio frequency path in an offline state, and one working state corresponds to one first parameter value; The acquisition module is used to perform signal amplification processing on the second sample input signal through the power amplifier in the first radio frequency path to obtain a first sample output signal; The predistortion parameter is iteratively trained based on a signal error between the first sample output signal and the second sample input signal to obtain the first parameter value.
10. The device according to claim 7, characterized in that The multiple first parameter values are parameter values in multiple working states obtained by training the first radio frequency path in an offline state, and one working state corresponds to one first parameter value; The acquisition module is used to acquire a plurality of fourth parameter values of the predistortion parameter corresponding to the first RF path; based on the plurality of fourth parameter values, predistortion processing is performed on the third sample input signal respectively to obtain a plurality of third sample RF signals; Based on the radio frequency performances respectively corresponding to the multiple third sample radio frequency signals and the multiple fourth parameter values, the predistortion parameter is iteratively trained to obtain the first parameter value.
11. The device according to claim 9 or 10, characterized in that The first processing module is used to perform pre-distortion processing on the first sample input signal based on the multiple first parameter values to obtain multiple first sample RF signals when the first RF path is in an online state and in an idle state; wherein the step of performing pre-distortion processing on the first sample input signal based on the multiple first parameter values to obtain multiple first sample RF signals is performed periodically.
12. The device according to claim 7, characterized in that The radio frequency performance includes: at least one of an adjacent channel leakage ratio and an error vector magnitude; The device also includes: The measuring module is used to measure at least one of an adjacent channel leakage ratio and an error vector magnitude of the first sample radio frequency signal to obtain radio frequency performance of the first sample radio frequency signal.
13. A terminal, characterized in that: The terminal includes a processor and a memory, wherein at least one program code is stored in the memory, and the at least one program code is loaded and executed by the processor to implement the method for determining a predistortion parameter value according to any one of claims 1 to 6.
14. A computer-readable storage medium, characterized in that: At least one program code is stored in the computer-readable storage medium, and the at least one program code is loaded and executed by the processor to implement the method for determining the pre-distortion parameter value according to any one of claims 1 to 6.
15. A computer program product, characterized in that At least one program code is stored in the computer program product, and the at least one program code is loaded and executed by a processor to implement the method for determining a pre-distortion parameter value according to any one of claims 1 to 6.
