WO2022166534A1

WO2022166534A1 - Pre-distortion processing method and apparatus

Info

Publication number: WO2022166534A1
Application number: PCT/CN2022/071120
Authority: WO
Inventors: 关鹏飞
Original assignee: 大唐移动通信设备有限公司
Priority date: 2021-02-07
Filing date: 2022-01-10
Publication date: 2022-08-11
Also published as: CN114911837A

Abstract

A pre-distortion processing method and apparatus. The method comprises: obtaining a first continuous timing signal comprising a timing signal to be processed (S201); performing normalization processing on the first continuous timing signal to obtain a corresponding second continuous timing signal (S202); and inputting the second continuous timing signal into a pre-trained back propagation (BP) neural network model, and outputting a pre-distorted timing signal corresponding to the timing signal to be processed (S203), wherein the BP neural network model is an inverse behavior model of a corresponding target power amplifier (PA). Using the inverse behavior model based on the BP neural network, and performing normalization processing on the timing signal inputted into the model can better improve the non-linear factor of an overall system and improve the precision of pre-distortion processing.

Description

Predistortion processing method and device

technical field

The present application relates to the technical field of artificial intelligence, and in particular, the present application relates to a predistortion processing method and apparatus.

Background technique

A power amplifier (PA) is a device that amplifies the modulated frequency band signal to the required power value. As the device that consumes the most energy in a communication base station, how to improve its efficiency is particularly important. Due to the nonlinear characteristics and memory effect of the PA, it leads to in-band distortion and out-of-band spectrum spreading, which directly affects the efficiency of the base station and the quality of the transmitted signal.

Digital Pre-distortion (DPD) establishes an inverse behavior model of the PA, and then inserts the inverse behavior model into the PA in the link to preprocess the signal before it enters the PA. The power amplifier remains linear in the high power region, improving the quality of the transmitted signal. The traditional inverse behavioral model in the field of digital predistortion is obtained by using General Memory Polynomial (GMP), and the traditional GMP model is simplified based on the Volterra series model.

However, with the surge in the number of wireless communication users in recent years, especially the commercial use of 5G (5th Generation mobile networks, fifth-generation mobile communication technology), the bandwidth and frequency have also increased, which requires greater memory depth and nonlinear order and potential numerical instability. In this context, the predistortion processing accuracy of the traditional GMP model is low.

SUMMARY OF THE INVENTION

The purpose of this application is to solve at least one of the above-mentioned technical defects, and the technical solutions provided by the embodiments of this application are as follows:

In a first aspect, the present application provides a predistortion processing method, including:

obtaining the first continuous timing signal containing the timing signal to be processed;

Normalize the first continuous time series signal to obtain a corresponding second continuous time series signal;

Input the second continuous time series signal into the pre-trained back-propagation BP neural network model, and output the pre-distorted time series signal corresponding to the time series signal to be processed, wherein the BP neural network model is the target power used for power amplification of the time series signal to be processed. Inverse behavioral model of amplifier PA.

In an optional embodiment of the present application, the first continuous timing signal includes a first preset number of timing signals before the to-be-processed timing signal, and a second preset number of timing signals after the to-be-processed timing signal.

In an optional embodiment of the present application, performing normalization processing on the first continuous time series signal to obtain a corresponding second continuous time series signal, including:

Obtain a training data set, and obtain the mean and standard deviation of the time series data samples in the training data set, wherein the training data set includes a third preset number of PA input time series signal samples and corresponding PA output time series signal samples of the target PA;

Based on the mean value and the standard deviation, obtain the normalized time series signal corresponding to each time series signal in the first continuous time series signal;

Based on each normalized timing signal, a second continuous timing signal is obtained.

In an optional embodiment of the present application, it includes:

Through the input layer, the second continuous time series signal is received, and the second continuous time series signal is converted into an initial feature vector;

Through the hidden layer, based on the initial feature vector, the corresponding abstract feature vector is obtained;

Through the output layer, based on the abstract feature vector, the predistorted time series signal is output.

In an optional embodiment of the present application, the training process of the BP neural network model is as follows:

Use the training data set to train the BP neural network model, and obtain the pre-trained BP neural network model.

In an optional embodiment of the present application, the BP neural network model is trained by using the training data set, and the pre-trained BP neural network model is obtained, including:

Input each PA output timing signal sample into the BP neural network model, and output the corresponding predicted PA input timing signal;

Based on each predicted PA input timing signal and the corresponding PA input timing signal sample, obtain the corresponding mean square error;

Based on each mean square error, the network parameters of the BP neural network model are updated by the back-propagation algorithm, and then the pre-trained BP neural network model is obtained.

In an optional embodiment of the present application, the method further includes:

After acquiring the pre-distorted timing signals corresponding to the fourth preset number of timing signals to be processed, input each pre-distorted timing signal into the target PA respectively to obtain at least one corresponding PA output timing signal;

Based on each predistorted timing signal and the corresponding PA output timing signal, the network parameters of the BP neural network model are updated again.

In an optional embodiment of the present application, based on each predistorted timing signal and the corresponding PA output timing signal, the network parameters of the BP neural network are updated again, including:

Input each PA output signal into the BP neural network model, and output the corresponding predicted PA input timing signal;

Obtain the corresponding mean square error based on each predicted PA input timing signal and the corresponding predistorted timing signal;

The network parameters of the BP neural network model are updated based on each mean square error.

In a second aspect, the present application provides a predistortion processing device, including:

a timing signal acquisition module, used for acquiring the first continuous timing signal including the timing signal to be processed;

The normalization module is used for normalizing the first continuous time series signal to obtain the corresponding second continuous time series signal;

The pre-distortion processing module is used to input the second continuous time series signal into the pre-trained back-propagation BP neural network model, and output the pre-distorted time series signal corresponding to the time series signal to be processed, wherein the BP neural network model is used for processing the time series signal to be processed. Inverse behavioral model of the target power amplifier device PA used for power amplification.

In an optional embodiment of the present application, the normalization module is specifically used for:

Obtaining a training data set, and obtaining the mean and standard deviation of the time series data samples in the training data set, wherein the training data set includes a third preset number of PA input time series signal samples and corresponding PA output time series signal samples of the target PA;

In an optional embodiment of the present application, the predistortion processing module is specifically configured to:

In an optional embodiment of the present application, the device further includes a first training module for:

In an optional embodiment of the present application, the first training module is specifically used for:

In an optional embodiment of the present application, the device further includes a second training module for:

In an optional embodiment of the present application, the second training module is specifically used for:

In a third aspect, the present application provides an electronic device, including a memory and a processor;

A computer program is stored in the memory;

A processor, configured to execute a computer program to implement the method provided in the embodiment of the first aspect or any optional embodiment of the first aspect.

In a fourth aspect, the present application provides a computer-readable storage medium, characterized in that a computer program is stored on the computer-readable storage medium, and when the computer program is executed by a processor, an embodiment of the first aspect or any one of the first aspect is implemented Methods provided in alternative embodiments.

The beneficial effects brought by the technical solution provided by the application are:

Before amplifying the time series signal to be processed by the target PA, the time series signal to be processed is pre-distorted by the inverse behavior model based on BP neural network, and the time series signal input to the inverse behavior model based on BP neural network is processed before the pre-distortion process. The normalization process is carried out, the inverse behavior model based on BP neural network is used, and the time series signal of the input model is normalized, which can better improve the nonlinear factors of the overall system and improve the accuracy of pre-distortion processing.

Description of drawings

In order to illustrate the technical solutions in the embodiments of the present application more clearly, the following briefly introduces the accompanying drawings that need to be used in the description of the embodiments of the present application.

1 is a schematic diagram of a LUT table lookup process in the prior art;

FIG. 2 is a schematic flowchart of a predistortion processing method provided by an embodiment of the present application;

3 is a schematic structural diagram of a BP neural network model in an example of an embodiment of the present application;

4a is a schematic diagram of a comparison between a normalized IQ time series signal and an unnormalized time series signal in an example of an embodiment of the present application;

FIG. 4b is a schematic diagram of a comparison of NMSE curves in a process of training a model by a normalized IQ time series signal and an unnormalized time series signal, respectively, in an example of an embodiment of the present application;

Fig. 5 is the result schematic diagram of the BP neural network model in another example of the embodiment of the application;

FIG. 6 is a feature combination mode of the input layer in an example of the embodiment of the present application;

7 is a schematic diagram of a training process of a BP neural network model in an example of an embodiment of the present application;

FIG. 8 is a structural block diagram of a predistortion processing apparatus provided by an embodiment of the present application;

FIG. 9 is a schematic structural diagram of an electronic device according to an embodiment of the present application.

Detailed ways

The following describes in detail the embodiments of the present application, examples of which are illustrated in the accompanying drawings, wherein the same or similar reference numerals refer to the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain the present application, but not to be construed as a limitation on the present application.

It will be understood by those skilled in the art that the singular forms "a", "an", "the" and "the" as used herein can include the plural forms as well, unless expressly stated otherwise. It should be further understood that the word "comprising" used in the specification of this application refers to the presence of stated features, integers, steps, operations, elements and/or components, but does not preclude the presence or addition of one or more other features, Integers, steps, operations, elements, components and/or groups thereof. It will be understood that when we refer to an element as being "connected" or "coupled" to another element, it can be directly connected or coupled to the other element or intervening elements may also be present. Furthermore, "connected" or "coupled" as used herein may include wirelessly connected or wirelessly coupled. As used herein, the term "and/or" includes all or any element and all combination of one or more of the associated listed items.

In order to make the objectives, technical solutions and advantages of the present application clearer, the embodiments of the present application will be further described in detail below with reference to the accompanying drawings.

The digital predistortion modeling scheme in the existing base station can be expressed as the following formula using the GMP model:

In formula (1), x is the output signal of the power amplifier and y is the input signal of the power amplifier when establishing the reverse mode of the power amplifier (ie, the reverse behavior model), wherein the three parameters i, j, and k represent the signal memory depth and the signal modulus value respectively. Memory depth, nonlinear order of signal modulus value, b is the coefficient of the model. In the real-time system of the whole machine, for the consideration of speed and resources, the trained GMP model is usually implemented by a look-up table (Look-Up-Table, LUT). The LUT table looks up the table according to the signal amplitude. The value is the accumulated value of the multiplication of the distortion coefficient and the nonlinear term of the signal modulus value. The number of LUT tables is determined by the model memory depths i, j, each (i, j) combination corresponds to a LUT table, and the value stored in the table is calculated according to formula (2), where AMP is the quantized signal amplitude.

As shown in Figure 1, it is the LUT table lookup process with signal memory depth i and amplitude memory depth j. Z in the figure represents the time delay. For example, there is x(n) in a set of timing signals. After Z- ^j changes, x(nj) is obtained, abs represents the modulo value, and abs(x(nj)) is the formula ( 2) in |AMP|, combined with the delay Z ^-i to get x(ni), and then get:

Then, sum all y _ij(n) to get y(n).

The traditional GMP model is simplified based on the Volterra series model. With the increase of bandwidth and frequency, it often requires larger memory depth and nonlinear order, and has potential numerical instability. The modeling accuracy of the power amplifier model with strong memory effect and strong nonlinearity is not high. In other words, in this context, the pre-distortion processing accuracy of the traditional GMP model is low.

In view of the above problems, the embodiments of the present application provide a predistortion processor, an apparatus, an electronic device, and a computer-readable storage medium, and the solutions provided by the present application will be described in detail below.

FIG. 2 is a schematic flowchart of a predistortion processing method provided by an embodiment of the present application. As shown in FIG. 2 , the method may include the following steps:

Step S201, acquiring a first continuous timing signal including the timing signal to be processed.

Among them, the timing signal to be processed is the timing signal that is predistorted before entering the power amplifier. The continuous timing signal can be understood as a plurality of timing signals sampled at a certain time interval. If the timing signal to be processed corresponds to the current sampling time, the time series signal collected at the sampling time before the current sampling time can be called the time series signal before the to-be-processed time series signal, and the time series signal collected at the sampling time after the current sampling time can be called the time series signal after the to-be-processed time series signal , then the first continuous timing signal includes the timing signal to be processed, the timing signal before the timing signal to be processed, and the timing signal after the timing signal to be processed. The use of the BP neural network model in the subsequent steps needs to obtain the predistorted time series signal of the time series signal to be processed according to the characteristics of the time series signal before and after the time series signal to be processed, so the first continuous time series signal needs to be obtained here.

Step S202, performing normalization processing on the first continuous time series signal to obtain a corresponding second continuous time series signal.

Specifically, since the system involves a large amount of data in both training and online predistortion processing, and may also introduce magnitude order information, different features in the data will have different magnitudes of difference. When the coefficient is solved, the gradient weight of the large number of features will be larger, and the training speed will be reduced when the loss function is used to find the optimal value. In addition, for some activation functions commonly used in neural networks, such as sigmoid, tanh, etc., when the data is too large, it will appear in the saturation region of the activation function for derivation, resulting in extremely small parameter gradients, which is not conducive to error back propagation. Therefore, before the time series data is input into the input layer of the BP network model, the data needs to be normalized. Specifically, each timing signal included in the first continuous timing signal needs to be normalized to obtain the corresponding second continuous timing signal.

Step S203, input the second continuous time sequence signal into the pre-trained back propagation BP (Back Propagation) neural network model, and output the pre-distorted time sequence signal corresponding to the time sequence signal to be processed, wherein the BP neural network model is to be processed by the time sequence signal to be processed. The inverse behavioral model of the target power amplifier PA used for power amplification.

Among them, the structure of the BP neural network model is shown in Figure 3. It can be composed of three parts: the input layer, the hidden layer and the output layer. The input layer is the structured data required to build the inverse behavior model, and the number of hidden layers can be arbitrary. The hidden layer generates nonlinear features through nonlinear activation functions and linear weighting, and propagates useful feature information to the next layer in the form of dense feature vectors. The initial coefficients of the neural network are generally initialized with random values, and the training method is gradient descent. This training method requires multiple iterations on the data set for each model training.

Specifically, the normalized second continuous data is input into the pre-trained BP neural network model, and the corresponding pre-distorted time-series signal is output, and the corresponding pre-distorted time-series signal is input into the corresponding target PA to realize the pending processing. Amplification of timing signals.

In the solution provided by this application, before the target PA is used to amplify the time series signal to be processed, the inverse behavior model based on BP neural network is used to pre-distort the time series signal to be processed, and before the pre-distortion processing is performed, the input signal based on the BP neural network is pre-distorted. The time series signal of the inverse behavior model is normalized, using the inverse behavior model based on the BP neural network, and normalizing the time series signal of the input model, which can better improve the nonlinear factors of the overall system and improve the prediction performance. Accuracy of distortion handling.

The first preset number and the second preset number may be set according to actual requirements, for example, the first preset number and the second preset number may be set to be the same number.

The training data set is also used to train the pre-trained BP neural network model. In other words, at the same time as the pre-trained BP neural network model is trained, the mean and standard required for subsequent normalization of time series signals are determined. Difference.

The number of time series signal samples included in the training training data set may be set according to actual requirements, that is, the third preset number may be set according to actual requirements.

Specifically, the process of normalizing each timing signal included in the first continuous timing signal can be expressed as the following formula:

In the formula, x _k is the timing signal included in the first continuous timing signal, and the normalized timing signal corresponding to x′ _k is the timing signal included in the second continuous timing signal. μ is the mean of the sample data, and σ is the standard deviation of the sample data. The normalized data conforms to a normal distribution, that is, the mean is 0 and the standard deviation is 1.

Further, for a time series signal data set, each time series signal can be expressed as a real part (In-phase, may be referred to as I) component and an imaginary part (Quadrature, may be referred to as Q) component part, then the time series signal can also be It is called IQ timing signal, and can also be called IQ data or IQ format data. As shown in Figure 4a, the abscissas in the left and right graphs both represent the real component (ie, I) of the time series signal, and the ordinates both represent the imaginary component (ie, Q) of the time series signal, and the left image is not normalized The normalized data set, the picture on the right is the data set after normalization. Comparing the left and right pictures, we can see that the order of magnitude of the normalized data is significantly reduced (that is, the order of magnitude of the corresponding horizontal and vertical coordinates is significantly reduced), which is convenient for model processing. .

As shown in Figure 4b, the unnormalized dataset is used to train the BP neural network model (left image), and the normalized dataset is used to train the BP neural network model (right image). Specifically, , the data is divided into training set (corresponding to the curve 401 in the two figures), validation set (corresponding to the curve 402 in the two figures) and test set (corresponding to the curve 403 in the two figures), and the abscissas in the two figures represent The number of training times and the ordinate both identify the corresponding normalized mean square error (NMSE), and the corresponding curve can be called the NMSE curve. Comparing the left and right figures, we can see that using the normalized data set for training, the model converges faster, and the obtained model has better pre-distortion processing effect. To sum up, in the process of training and online predistortion processing, the time series signal may be normalized first, and then the normalized time series signal is used for training or online predistortion processing.

In an optional embodiment of the present application, the BP neural network includes an input layer, a third preset number of hidden layers, and an output layer, the second continuous time series signal is input into the pre-trained back-propagation BP neural network model, and the output The predistorted timing signal corresponding to the timing signal to be processed includes:

The BP neural network model is shown in Figure 5, including an input layer, a hidden layer (ie, a hidden layer) and an output layer. Specifically, a normalizer is also set before the input layer for normalizing the first continuous time series data. The input layer mainly involves how to combine the input time series signals into the input format required by the BP neural network model (ie, conversion from the initial feature vector). The main function of the hidden layer is to abstract the input information and discover hidden features, so that the input information can be represented as a feature vector (that is, to obtain an abstract feature vector). The number of hidden layers in this embodiment of the present application is 2, which are hidden layer 1 and hidden layer 2 in the figure respectively. The main function of the output layer is to convert the abstract feature information into the predistorted target IQ format, that is, to output the predistorted time sequence signal corresponding to the time sequence signal to be processed.

Specifically, the input layer converts the input time series signal into the input of the BP neural network. In this embodiment of the present application, the information of the input layer can be listed in the form of a matrix shown in The memory depth, the value of the signal at the current time, and the delay information, I and Q represent the imaginary part information and real part information corresponding to the time series signal, A^n represents the order information of the signal amplitude, and "other" is other scalable information to facilitate Extensions to model inputs. When using the BP neural network model, before the feature matrix enters the hidden layer, the matrix representation in Figure 6 needs to be converted into a preset vector form (ie, the initial feature vector), and the vector dimension is 1*n.

The depth and size of the hidden layer determine the learning upper limit of the BP network model, but simply increasing the depth and size cannot directly make the model better, but will make the model training difficult to converge. Through experimental tests, the present invention uses a two-layer hidden layer structure, and the calculation expression of each hidden layer is shown in (4):

h=f(X _input *W _h +b) (4)

Among them, X _input is the input of the layer, W _h is a matrix of n*m, n represents the output dimension of the upper layer, m represents the number of neurons in this layer, f(x) is the activation function, and b is the bias.

Activation function: The activation function is introduced to give the network model the ability to learn nonlinear functions. In this embodiment of the present application, the ReLU function can be used as the activation function, and the formula is shown in (5):

a=max(0,z) (5)

Among them, a is the output of the activation function, and z is the calculation result of the linear change of the neuron.

The output layer uses a fully connected linear unit for output, and the formula is shown in (6):

y _output = h*W _o +b (6)

Wo is a matrix of m*2, m represents the output dimension of the hidden layer, and 2 represents the 2-dimensional vector output of the real part and the imaginary part.

Specifically, as shown in FIG. 7 is a schematic diagram of the online real-time system corresponding to the embodiment of the application, the training of the BP neural network model can be divided into two stages, one of which is the offline training stage, and the pre-trained data obtained in the basic application The second stage of the BP neural network model is the step-by-step tuning stage in the online use process. Each training process requires a large amount of iterative calculations. Considering that the first training time of the base station after power-on is too long, the application embodiment adopts the timing of the above two-stage training, and selects a large batch of data with different powers as the training time. Pre-training data set, a model is trained offline, and the online real-time system will use the pre-training model as the training starting point to gradually optimize. Among them, the structure of the pre-trained BP neural network model can be set according to different power amplifier models.

It should be noted that in the offline training stage, the parameters of the normalizer are determined at the same time, in other words, it can be understood as determining the mean and standard deviation in the normalization process, then in the subsequent online use process and online update stage normalizer parameters remain unchanged.

Further, in the offline training stage, the BP neural network model is trained by using the pre-acquired training data set, wherein the training data used are the PA input time series signal samples of the target PA and the corresponding PA output time series signal samples. Specifically, each PA output timing signal sample is input into the BP neural network model, and the corresponding predicted PA input timing signal is output; based on each predicted PA input timing signal and the corresponding PA input timing signal sample, the corresponding mean square error is obtained. ; Based on each mean square error, the network parameters of the BP neural network model are updated by the back propagation algorithm, and then the pre-trained BP neural network model is obtained.

Specifically, the above process is the second-stage training process. Specifically, each PA output signal is input into the BP neural network model, and the corresponding predicted PA input time series signal is output; based on each predicted PA input time series signal and the corresponding predicted PA input time series signal The distorted time series signal is obtained, and the corresponding mean square error is obtained; the network parameters of the BP neural network model are updated based on each mean square error. It can be understood that the obtained PA output signal can also be referred to as the feedback data in the figure, that is, using multiple feedback data to update the network parameters of the BP neural network model, wherein the update frequency of the network parameters can be set according to actual needs. For example, the network parameters can be updated every preset time interval. It can be understood that, the fourth preset number can be set according to actual requirements.

It should be noted that the objective function used in the two-stage training process can be to minimize the mean square error, and the formula is shown in (7):

y represents the IQ data before over-amplification (that is, the PA input timing signal sample or the predistorted timing signal), that is, the expected value obtained by the inverse behavior model of the power amplifier, and y _output is the output of the BP neural network model (the predicted PA input timing signal) . Since the gradient descent algorithm is used to solve the BP network, the cut-off condition for training an available model iteration is: when the number of iterations exceeds the preset maximum number of iterations or the objective function does not decrease within the specified number of iterations.

FIG. 8 is a structural block diagram of a predistortion processing apparatus provided by an embodiment of the present application. As shown in FIG. 8 , the apparatus 800 may include: a timing signal acquisition module 801, a normalization module 802, and a predistortion processing module 803, wherein :

The timing signal acquisition module 801 is configured to acquire the first continuous timing signal including the timing signal to be processed;

The normalization module 802 is configured to perform normalization processing on the first continuous time series signal to obtain a corresponding second continuous time series signal;

The predistortion processing module 803 is used to input the second continuous time series signal into the pretrained back-propagation BP neural network model, and output the predistorted time series signal corresponding to the time series signal to be processed, wherein the BP neural network model is to perform the processing of the time series signal to be processed. Inverse behavioral model of the target power amplifier device PA used for power amplification.

In the solution provided by this application, before the target PA is used to amplify the time series signal to be processed, the inverse behavior model based on the BP neural network is used to pre-distort the time series signal to be processed, and before the pre-distortion processing is performed, the input signal based on the BP neural network is pre-distorted. The time series signal of the inverse behavior model is normalized, using the inverse behavior model based on the BP neural network, and normalizing the time series signal of the input model, which can better improve the nonlinear factors of the overall system and improve the prediction performance. Accuracy of distortion handling.

Referring next to FIG. 9 , it shows a schematic structural diagram of an electronic device (eg, a terminal device or a server that executes the method shown in FIG. 2 ) 900 suitable for implementing an embodiment of the present application. The electronic devices in the embodiments of the present application may include, but are not limited to, such as mobile phones, notebook computers, digital broadcast receivers, PDAs (personal digital assistants), PADs (tablets), PMPs (portable multimedia players), in-vehicle terminals (such as mobile terminals such as in-vehicle navigation terminals), wearable devices, etc., and stationary terminals such as digital TVs, desktop computers, and the like. The electronic device shown in FIG. 9 is only an example, and should not impose any limitations on the functions and scope of use of the embodiments of the present application.

The electronic device includes: a memory and a processor, where the memory is used to store a program for executing the methods described in the above method embodiments; the processor is configured to execute the program stored in the memory. The processor here may be referred to as the processing device 901 described below, and the memory may include at least one of a read-only memory (ROM) 902, a random access memory (RAM) 903, and a storage device 908, which are specifically as follows shown:

As shown in FIG. 9 , an electronic device 900 may include a processing device (eg, a central processing unit, a graphics processor, etc.) 901 that may be loaded into random access according to a program stored in a read only memory (ROM) 902 or from a storage device 908 Various appropriate actions and processes are executed by the programs in the memory (RAM) 903 . In the RAM 903, various programs and data necessary for the operation of the electronic device 900 are also stored. The processing device 901, the ROM 902, and the RAM 903 are connected to each other through a bus 904. An input/output (I/O) interface 905 is also connected to bus 904 .

Typically, the following devices can be connected to the I/O interface 905: input devices 906 including, for example, a touch screen, touchpad, keyboard, mouse, camera, microphone, accelerometer, gyroscope, etc.; including, for example, a liquid crystal display (LCD), speakers, vibration An output device 907 such as a computer; a storage device 908 including, for example, a magnetic tape, a hard disk, etc.; and a communication device 909 . The communication means 909 may allow the electronic device 900 to communicate wirelessly or by wire with other devices to exchange data. While FIG. 9 illustrates an electronic device having various means, it should be understood that not all of the illustrated means are required to be implemented or available. More or fewer devices may alternatively be implemented or provided.

In particular, according to embodiments of the present application, the processes described above with reference to the flowcharts may be implemented as computer software programs. For example, embodiments of the present application include a computer program product comprising a computer program carried on a non-transitory computer readable medium, the computer program containing program code for performing the method illustrated in the flowchart. In such an embodiment, the computer program may be downloaded and installed from the network via the communication device 909, or from the storage device 908, or from the ROM 902. When the computer program is executed by the processing device 901, the above-mentioned functions defined in the methods of the embodiments of the present application are executed.

It should be noted that the computer-readable storage medium mentioned above in the present application may be a computer-readable signal medium or a computer-readable storage medium, or any combination of the above two. The computer-readable storage medium can be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus or device, or a combination of any of the above. More specific examples of computer readable storage media may include, but are not limited to, electrical connections with one or more wires, portable computer disks, hard disks, random access memory (RAM), read only memory (ROM), erasable Programmable read only memory (EPROM or flash memory), optical fiber, portable compact disk read only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination of the above. In this application, a computer-readable storage medium can be any tangible medium that contains or stores a program that can be used by or in conjunction with an instruction execution system, apparatus, or device. In this application, however, a computer-readable signal medium may include a data signal propagated in baseband or as part of a carrier wave, carrying computer-readable program code therein. Such propagated data signals may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the foregoing. A computer-readable signal medium can also be any computer-readable medium other than a computer-readable storage medium that can transmit, propagate, or transport the program for use by or in connection with the instruction execution system, apparatus, or device . Program code embodied on a computer readable medium may be transmitted using any suitable medium including, but not limited to, electrical wire, optical fiber cable, RF (radio frequency), etc., or any suitable combination of the foregoing.

In some embodiments, the client and server can use any currently known or future developed network protocol such as HTTP (HyperText Transfer Protocol) to communicate, and can communicate with digital data in any form or medium Communication (eg, a communication network) interconnects. Examples of communication networks include local area networks ("LAN"), wide area networks ("WAN"), the Internet (eg, the Internet), and peer-to-peer networks (eg, ad hoc peer-to-peer networks), as well as any currently known or future development network of.

The above-mentioned computer-readable medium may be included in the above-mentioned electronic device; or may exist alone without being assembled into the electronic device.

The above-mentioned computer-readable medium carries one or more programs, and when the above-mentioned one or more programs are executed by the electronic device, the electronic device:

Obtain the first continuous time series signal including the to-be-processed time series signal; perform normalization processing on the first continuous time series signal to obtain the corresponding second continuous time series signal; input the second continuous time series signal into the pre-trained back-propagation BP neural network The model outputs the predistorted time series signal corresponding to the time series signal to be processed, wherein the BP neural network model is an inverse behavior model of the target power amplifier PA used for power amplification of the time series signal to be processed.

Computer program code for performing the operations of the present application may be written in one or more programming languages, including but not limited to object-oriented programming languages—such as Java, Smalltalk, C++, and This includes conventional procedural programming languages - such as the "C" language or similar programming languages. The program code may execute entirely on the user's computer, partly on the user's computer, as a stand-alone software package, partly on the user's computer and partly on a remote computer, or entirely on the remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer through any kind of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (eg, using an Internet service provider through Internet connection).

The flowchart and block diagrams in the Figures illustrate the architecture, functionality, and operation of possible implementations of systems, methods and computer program products according to various embodiments of the present application. In this regard, each block in the flowchart or block diagrams may represent a module, segment, or portion of code that contains one or more logical functions for implementing the specified functions executable instructions. It should also be noted that, in some alternative implementations, the functions noted in the blocks may occur out of the order noted in the figures. For example, two blocks shown in succession may, in fact, be executed substantially concurrently, or the blocks may sometimes be executed in the reverse order, depending upon the functionality involved. It is also noted that each block of the block diagrams and/or flowchart illustrations, and combinations of blocks in the block diagrams and/or flowchart illustrations, can be implemented in dedicated hardware-based systems that perform the specified functions or operations , or can be implemented in a combination of dedicated hardware and computer instructions.

The modules or units involved in the embodiments of the present application may be implemented in a software manner, and may also be implemented in a hardware manner. Wherein, the name of the module or unit does not constitute a limitation of the unit itself under certain circumstances, for example, the timing signal acquisition module can also be described as a "module for acquiring timing signals".

The functions described herein above may be performed, at least in part, by one or more hardware logic components. For example, without limitation, exemplary types of hardware logic components that may be used include: Field Programmable Gate Arrays (FPGAs), Application Specific Integrated Circuits (ASICs), Application Specific Standard Products (ASSPs), Systems on Chips (SOCs), Complex Programmable Logical Devices (CPLDs) and more.

In the context of this application, a machine-readable medium may be a tangible medium that may contain or store the program for use by or in connection with the instruction execution system, apparatus or device. The machine-readable medium may be a machine-readable signal medium or a machine-readable storage medium. Machine-readable media may include, but are not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, devices, or devices, or any suitable combination of the foregoing. More specific examples of machine-readable storage media would include one or more wire-based electrical connections, portable computer disks, hard disks, random access memory (RAM), read only memory (ROM), erasable programmable read only memory (EPROM or flash memory), fiber optics, compact disk read only memory (CD-ROM), optical storage, magnetic storage, or any suitable combination of the foregoing.

The apparatuses provided in the embodiments of the present application may implement at least one module among the multiple modules through the AI model. AI-related functions may be performed by non-volatile memory, volatile memory, and a processor.

The processor may include one or more processors. At this point, the one or more processors may be general-purpose processors, such as central processing units (CPUs), application processors (APs), etc., or pure graphics processing units, such as graphics processing units (GPUs), visual processing Units (VPUs), and/or AI-specific processors, such as Neural Processing Units (NPUs).

The one or more processors control the processing of input data according to predefined operating rules or artificial intelligence (AI) models stored in non-volatile memory and volatile memory. Provides predefined operating rules or artificial intelligence models through training or learning.

Here, providing by learning refers to obtaining a predefined operation rule or an AI model with desired characteristics by applying a learning algorithm to a plurality of learning data. This learning may be performed in the apparatus itself in which the AI according to an embodiment is performed, and/or may be implemented by a separate server/system.

The AI model can contain multiple neural network layers. Each layer has multiple weight values, and the calculation of a layer is performed by the calculation result of the previous layer and the multiple weights of the current layer. Examples of neural networks include, but are not limited to, Convolutional Neural Networks (CNN), Deep Neural Networks (DNN), Recurrent Neural Networks (RNN), Restricted Boltzmann Machines (RBM), Deep Belief Networks (DBN), Bidirectional Recurrent Deep Neural Networks (BRDNN), Generative Adversarial Networks (GAN), and Deep Q-Networks.

A learning algorithm is a method of training a predetermined target device (eg, a robot) using a plurality of learning data to cause, allow or control the target device to make a determination or prediction. Examples of such learning algorithms include, but are not limited to, supervised learning, unsupervised learning, semi-supervised learning, or reinforcement learning.

Those skilled in the art can clearly understand that, for the convenience and brevity of description, for the specific method implemented when the computer-readable medium described above is executed by an electronic device, reference may be made to the corresponding process in the foregoing method embodiment, which is not repeated here. Repeat.

The technical solutions provided in the embodiments of the present application can be applied to various systems, especially 5G systems. For example, the applicable system may be a global system of mobile communication (GSM) system, a code division multiple access (CDMA) system, a wideband code division multiple access (Wideband Code Division Multiple Access, WCDMA) general packet Wireless service (general packet radio service, GPRS) system, long term evolution (long term evolution, LTE) system, LTE frequency division duplex (frequency division duplex, FDD) system, LTE time division duplex (time division duplex, TDD) system, Long term evolution advanced (LTE-A) system, universal mobile telecommunication system (UMTS), worldwide interoperability for microwave access (WiMAX) system, 5G New Radio (New Radio, NR) system, etc. These various systems include terminal equipment and network equipment. The system may also include a core network part, such as an evolved packet system (Evloved Packet System, EPS), a 5G system (5GS), and the like.

The terminal device involved in the embodiments of the present application may be a device that provides voice and/or data connectivity to a user, a handheld device with a wireless connection function, or other processing device connected to a wireless modem. In different systems, the name of the terminal device may be different. For example, in the 5G system, the terminal device may be called user equipment (User Equipment, UE). Wireless terminal equipment can communicate with one or more core networks (Core Network, CN) via a radio access network (Radio Access Network, RAN). "telephone) and computers with mobile terminal equipment, eg portable, pocket-sized, hand-held, computer-built or vehicle-mounted mobile devices, which exchange language and/or data with the radio access network. For example, Personal Communication Service (PCS) phones, cordless phones, Session Initiated Protocol (SIP) phones, Wireless Local Loop (WLL) stations, Personal Digital Assistants (Personal Digital Assistants), PDA) and other devices. Wireless terminal equipment may also be referred to as system, subscriber unit, subscriber station, mobile station, mobile station, remote station, access point , a remote terminal device (remote terminal), an access terminal device (access terminal), a user terminal device (user terminal), a user agent (user agent), and a user device (user device), which are not limited in the embodiments of the present application.

It should be understood that although the various steps in the flowchart of the accompanying drawings are sequentially shown in the order indicated by the arrows, these steps are not necessarily executed in sequence in the order indicated by the arrows. Unless explicitly stated herein, the steps are performed in no strict order and may be performed in other orders. Moreover, at least a part of the steps in the flowchart of the accompanying drawings may include multiple sub-steps or multiple stages, and these sub-steps or stages are not necessarily executed at the same time, but may be executed at different times, and the execution sequence is also It does not have to be performed sequentially, but may be performed alternately or alternately with other steps or at least a portion of sub-steps or stages of other steps.

The above are only part of the embodiments of the present application. It should be pointed out that for those skilled in the art, some improvements and modifications can be made without departing from the principles of the present application. These improvements and modifications should also be regarded as The protection scope of this application.

Claims

A predistortion processing method, the method comprising:

obtaining the first continuous timing signal containing the timing signal to be processed;

normalizing the first continuous timing signal to obtain a corresponding second continuous timing signal;

Inputting the second continuous time series signal into a pre-trained back-propagation BP neural network model, and outputting a pre-distorted time series signal corresponding to the to-be-processed time-series signal, wherein the BP neural network model is for the to-be-processed time series The inverse behavior model of the target power amplifier PA used for power amplification of the signal.
The method of claim 1 , wherein the first continuous timing signal comprises a first preset number of timing signals before the to-be-processed timing signal and a second preset number after the to-be-processed timing signal number of timing signals.
The method according to claim 1, wherein the normalizing the first continuous time series signal to obtain the corresponding second continuous time series signal comprises:

Obtain a training data set, and obtain the mean and standard deviation of time series data samples in the training data set, wherein the training data set includes a third preset number of PA input time series signal samples of the target PA and corresponding PA outputs timing signal samples;

obtaining, based on the mean value and the standard deviation, a normalized time sequence signal corresponding to each time sequence signal in the first continuous time sequence signal;

The second continuous timing signal is obtained based on each normalized timing signal.
The method of claim 1, wherein the BP neural network comprises an input layer, a third preset number of hidden layers, and an output layer, and the second continuous time series signal is input into a pretrained backpropagation BP The neural network model outputs the predistorted time series signal corresponding to the time series signal to be processed, including:

Through the input layer, the second continuous time series signal is received, and the second continuous time series signal is converted into an initial feature vector;

Through the hidden layer, based on the initial feature vector, obtain the corresponding abstract feature vector;

Through the output layer, the predistorted time series signal is output based on the abstract feature vector.
The method according to claim 3, wherein, the training process of the BP neural network model is as follows:

The BP neural network model is trained by using the training data set to obtain the pre-trained BP neural network model.
The method according to claim 5, wherein the training of the BP neural network model by using the training data set to obtain the pre-trained BP neural network model comprises:

Input each PA output timing signal sample into the BP neural network model, and output the corresponding predicted PA input timing signal;

Based on each predicted PA input timing signal and the corresponding PA input timing signal sample, obtain the corresponding mean square error;

Based on each mean square error, the network parameters of the BP neural network model are updated by using the back-propagation algorithm, and then the pre-trained BP neural network model is obtained.
The method of claim 1, further comprising:

After acquiring the pre-distorted timing signals corresponding to the fourth preset number of timing signals to be processed, input each pre-distorted timing signal into the target PA respectively to obtain at least one corresponding PA output timing signal;

Based on each predistorted timing signal and the corresponding PA output timing signal, the network parameters of the BP neural network model are updated again.
The method according to claim 7, wherein, the network parameters of the BP neural network are updated again based on the timing signals of each predistortion and the corresponding PA output timing signals, including:

Input each PA output signal into the BP neural network model, and output the corresponding predicted PA input timing signal;

Obtain the corresponding mean square error based on each predicted PA input timing signal and the corresponding predistorted timing signal;

The network parameters of the BP neural network model are updated based on each mean square error.
A predistortion processing device, the device comprising:

a timing signal acquisition module, used for acquiring the first continuous timing signal including the timing signal to be processed;

a normalization module, configured to perform normalization processing on the first continuous time series signal to obtain a corresponding second continuous time series signal;

A predistortion processing module, configured to input the second continuous time series signal into a pretrained back-propagation BP neural network model, and output a predistorted time series signal corresponding to the to-be-processed time series signal, wherein the BP neural network model An inverse behavioral model of the target power amplifier device PA used for power amplifying the to-be-processed timing signal.
An electronic device including a memory and a processor;

A computer program is stored in the memory;

The processor for executing the computer program to implement the method of any one of claims 1 to 8.
A computer-readable storage medium storing a computer program on the computer-readable storage medium, when the computer program is executed by a processor, implements the method of any one of claims 1 to 8.