Digital pre-distortion method and device of multi-antenna satellite system and electronic equipment

Abstract

The invention provides a digital pre-distortion method and device of a multi-antenna satellite system and electronic equipment, and relates to the field of satellite communication, and the digital pre-distortion method is based on a multi-gating hybrid expert network, utilizes an expert network and a private network to learn common features and private features, and utilizes the gating network to adaptively select features for different tasks. By automatically adjusting parameterization between common features and private features, the predistortion model can effectively learn complex data distribution, improve linearization performance and reduce cross-channel nonlinear crosstalk. Compared with an existing pre-distortion model based on a neural network, the pre-distortion model provided by the invention can obtain a more accurate power amplifier inverse model under the training of the same number of iterations, so as to effectively compensate the distortion and crosstalk of the multi-path power amplifier. Compared with an existing polynomial machine learning method, the pre-distortion model has more excellent performance in the aspects of nonlinear distortion compensation and crosstalk suppression.

Description

Technical Field

[0001] The present application relates to the field of satellite communications, and in particular to a digital predistortion method, device and electronic equipment for a multi-antenna satellite system. Background Art

[0002] In satellite communication networks, as the demand for spectrum efficiency and the number of access devices continues to grow, multiple antennas will be deployed at the communication transceiver to improve the capacity, coverage and data transmission rate of satellite communication systems. However, like traditional communication systems, multi-antenna satellite systems also face linearization problems due to the presence of nonlinear devices such as power amplifiers in the channel. The nonlinear distortion of the power amplifier will reduce the signal quality and reduce the capacity of the system, especially when the power amplifier works in the nonlinear region, its saturation characteristics will cause the signal distortion to be further aggravated. Therefore, in order to ensure the efficiency and linearity of the power amplifier at the same time, efficient power amplifier linearization technology is also required in multi-antenna satellite systems. Digital pre-distortion technology has become the most widely used linearization technology due to its low cost, simple implementation and excellent performance. However, due to the coupling between antennas or the leakage of the shared local oscillator, the nonlinear behavior of each power amplifier may be affected by the behavior of adjacent channels. Therefore, the traditional one-to-one individual compensation linearization technology is not enough to meet the linearization requirements of multi-antenna systems with strong crosstalk. Summary of the invention

[0003] The purpose of the embodiments of the present application is to provide a digital predistortion method, device and electronic equipment for a multi-antenna satellite system, which is used to solve the above-mentioned problems existing in the prior art, correct the nonlinear distortion of the signal corresponding to the current antenna, and eliminate the crosstalk from other antennas.

[0004] In a first aspect, a digital predistortion method for a multi-antenna satellite system is provided, and the method may include: Acquire input data of multiple antennas to be processed; the input data includes I / Q components and signal amplitude of the input signal; The I / Q components and signal amplitudes of multiple antennas are spliced ​​to obtain a spliced ​​vector; The splicing vector is processed by using a trained pre-distortion model to obtain a target I / Q component of each antenna, and each target I / Q component is converted into a pre-distortion signal, so that a power amplifier corresponding to each antenna processes the corresponding pre-distortion signal and then transmits it; wherein the pre-distortion model includes a first number of expert networks, a second number of gated networks, and a second number of private networks.

[0005] In a possible implementation, for any expert network, the expert network is a fully connected network structure, including an expert input layer, an expert hidden layer and an expert output layer; The activation functions of the expert hidden layer and the expert output layer are both Tanh functions.

[0006] In a possible implementation, for any gated network, the gated network only includes a network layer; The activation function of the network layer is the Softmax function.

[0007] In a possible implementation, the private network is a fully connected network structure, including a private input layer, a private hidden layer and a private output layer; The activation functions of the private hidden layer and the private output layer are both Leaky ReLU functions.

[0008] In a possible implementation, a trained predistortion model is used to process the spliced ​​vector to obtain a target I / Q component of each antenna, including: Using a first number of expert networks in the predistortion model to process the spliced ​​vectors respectively, to obtain common feature vectors corresponding to the expert networks; and, using a second number of gated networks in the predistortion model to process the concatenated vector to obtain weights of each expert network; The second number of private networks in the predistortion model are used to process the weights of the expert networks and the common feature vectors corresponding to the expert networks to obtain target I / Q components of each antenna.

[0009] In a second aspect, a digital predistortion device for a multi-antenna satellite system is provided, and the device may include: An acquisition unit, used to acquire input data of multiple antennas to be processed; the input data includes I / Q components and signal amplitude of the input signal; A splicing unit, used for splicing I / Q components and signal amplitudes of multiple antennas to obtain a splicing vector; A processing unit is used to use a trained pre-distortion model to process the splicing vector to obtain a target I / Q component of each antenna, and convert each target I / Q component into a pre-distortion signal so that a power amplifier corresponding to each antenna processes the corresponding pre-distortion signal and then transmits it; wherein the pre-distortion model includes a first number of expert networks, a second number of gated networks, and a second number of private networks.

[0010] In a possible implementation, for any expert network, the expert network is a fully connected network structure, including an expert input layer, an expert hidden layer and an expert output layer; The activation functions of the expert hidden layer and the expert output layer are both Tanh functions.

[0011] In a possible implementation, for any gated network, the gated network only includes a network layer; The activation function of the network layer is the Softmax function.

[0012] In a third aspect, an electronic device is provided, the electronic device comprising a processor, a communication interface, a memory and a communication bus, wherein the processor, the communication interface and the memory communicate with each other via the communication bus; Memory, used to store computer programs; The processor is used to implement any method step described in the first aspect when executing the program stored in the memory.

[0013] In a fourth aspect, a computer-readable storage medium is provided, wherein a computer program is stored in the computer-readable storage medium, and when the computer program is executed by a processor, any method step described in the first aspect is implemented.

[0014] The present application provides a digital predistortion method for a multi-antenna satellite system, which is based on a multi-feedback loop architecture and uses a multi-gated hybrid expert network to model the predistortion model of the multi-antenna system. Among them, the multi-gated hybrid expert network uses an expert network and a private network to learn common features and private features, and uses a gated network to adaptively select features for different tasks. By automatically adjusting the parameterization between common features and private features, the predistortion model can effectively learn complex data distributions, improve linearization performance, and reduce nonlinear crosstalk across channels. Compared with the existing neural network-based predistortion model, the predistortion model of the present application can obtain a more accurate power amplifier inverse model under the same number of iterations of training to effectively compensate for multi-channel power amplifier distortion and crosstalk. Compared with the existing polynomial machine learning method, the predistortion model has better performance in both nonlinear distortion compensation and crosstalk suppression. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings required for use in the embodiments of the present application will be briefly introduced below. It should be understood that the following drawings only show certain embodiments of the present application and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without paying creative work.

[0016] Figure 1 A system architecture diagram of a digital predistortion method applied to a multi-antenna satellite system provided in an embodiment of the present application; Figure 2 A schematic flow chart of a digital predistortion method for a multi-antenna satellite system provided in an embodiment of the present application; Figure 3A schematic diagram of a single DPD module provided in an embodiment of the present application; Figure 4 A schematic diagram of an application process of a digital predistortion method for a multi-antenna satellite system provided in an embodiment of the present application; Figure 5 A schematic structural diagram of a digital predistortion device for a multi-antenna satellite system provided in an embodiment of the present application; Figure 6 A schematic diagram of the structure of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0017] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments of the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0018] The digital predistortion method for a multi-antenna satellite system provided in the embodiment of the present application can be applied in Figure 1 In the system architecture shown in Figure 1 As shown, the system may include: a server and a multi-antenna satellite system. The server may be a physical server, or a server cluster or distributed system composed of multiple physical servers, or a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, content delivery networks (CDN), and big data and artificial intelligence platforms. The multi-antenna satellite system and the server may be directly or indirectly connected via wired or wireless communication methods, which is not limited in this application.

[0019] A multi-antenna satellite system is used to send input data of each antenna in a time period to be processed to a server; The server is used to obtain input data of each antenna to execute a digital predistortion method for a multi-antenna satellite system provided in the present application.

[0020] In satellite communication networks, as the demand for spectrum efficiency and the number of access devices continues to grow, multiple antennas will be deployed at the communication transceiver to improve the capacity, coverage and data transmission rate of satellite communication systems. However, like traditional communication systems, multi-antenna satellite systems also face linearization problems due to the presence of nonlinear devices such as power amplifiers in the channel. The nonlinear distortion of the power amplifier will reduce the signal quality and reduce the capacity of the system, especially when the power amplifier works in the nonlinear region, its saturation characteristics will cause the signal distortion to be further aggravated. Therefore, in order to ensure the efficiency and linearity of the power amplifier at the same time, efficient power amplifier linearization technology is also required in multi-antenna satellite systems. Digital pre-distortion technology has become the most widely used linearization technology due to its low cost, simple implementation and excellent performance. However, due to the coupling between antennas or the leakage of the shared local oscillator, the nonlinear behavior of each power amplifier may be affected by the behavior of adjacent channels. Therefore, the traditional one-to-one individual compensation linearization technology is not enough to meet the linearization requirements of multi-antenna systems with strong crosstalk.

[0021] At present, there are two main types of multi-antenna digital pre-distortion architectures: single DPD architecture and multiple DPD architecture. Among them, the single DPD architecture only uses one digital pre-distortion module to linearize the main lobe beam direction of the synthesized signal of each power amplifier, and its implementation complexity is greatly reduced. However, this method cannot guarantee the linearity of all power amplifiers, resulting in distortion in directions other than the main beam direction. The multiple DPD architecture deploys a dedicated DPD module on each RF link, that is, each DPD module performs one-to-one individual compensation for each power amplifier. This method can ensure the linearity of all power amplifiers and compensate for signal distortion in each direction. However, the complexity of the system will increase exponentially with the increase in the number of antennas, and it is only applicable to the full digital architecture.

[0022] Therefore, the present application provides a digital predistortion method for a multi-antenna satellite system to solve the above-mentioned problems existing in the prior art, which can correct the nonlinear distortion of the signal corresponding to the current antenna and eliminate the crosstalk from other antennas.

[0023] The preferred embodiments of the present application are described below in conjunction with the drawings in the specification. It should be understood that the preferred embodiments described herein are only used to illustrate and explain the present application, and are not used to limit the present application. In addition, the embodiments and features in the embodiments of the present application may be combined with each other if there is no conflict.

[0024] Figure 2 The present invention provides a flowchart of a digital predistortion method for a multi-antenna satellite system. Figure 2 As shown, the method may include: Step S210: Obtain input data of multiple antennas to be processed.

[0025] Specifically, the multi-antenna satellite system collects input data for a period to be processed and sends the input data to a server.

[0026] The input data includes the I / Q components and signal amplitude of the input signal; wherein the I / Q components are the I component (in-phase component) and the Q component (quadrature component).

[0027] Step S220: splice the I / Q components and signal amplitudes of multiple antennas to obtain a spliced ​​vector.

[0028] Step S230: Use the trained pre-distortion model to process the splicing vector to obtain the target I / Q components of each antenna, and convert the obtained target I / Q components into pre-distortion signals so that the power amplifier corresponding to each antenna processes the corresponding pre-distortion signal and then transmits it.

[0029] It can be understood that each predistortion signal is a complex signal after the corresponding target I / Q component is converted.

[0030] Before executing step S230, the training process of the pre-distortion model includes: A. Configure a pre-trained pre-distortion model; the pre-trained pre-distortion model includes a first number of expert networks, a second number of gated networks, and a second number of private networks, wherein the first number is manually determined, and the second number is determined according to the number of antennas in the multi-antenna satellite system.

[0031] That is to say, the pre-trained pre-distortion model of this application is a complex neural network structure, consisting of multiple expert networks and gating networks. Expert network: sharing the underlying structure, each expert network is relatively independent and can capture the unique relevant features in a single task (the processing task of the channel corresponding to any antenna). Gating network: adaptively select the contribution of the expert network for different tasks. Private network: each task has a private network, which is specifically used to learn the private features of each power amplifier.

[0032] B. Construct a training set. Since this scheme only uses a single DPD module to linearize the power amplifiers of all channels (antennas) of the multi-antenna satellite system, it is necessary to consider the nonlinear crosstalk problem between multiple antennas. And since the DPD module simulates the reverse characteristics of the power amplifier, the input and output signals of the power amplifier need to be reversed during the DPD coefficient extraction process. Therefore, the input of the pre-trained pre-distortion model is the I / Q components and signal amplitudes (signal amplitude related items) of the output signals of the multi-antenna power amplifier at the first moment and multiple second moments in the historical period. Among them, the second moment is the historical moment of the first moment.

[0033] Specifically, since the power amplifier has a memory effect, that is, the output signal of the power amplifier at the first moment is not only related to the input signal of the power amplifier at the first moment, but also to the input signal of the power amplifier at the historical moment (second moment) of the first moment. Therefore, when extracting the characteristics of the power amplifier, it is necessary to consider the input signal at the second moment at the same time. Therefore, the training set is the signal collected in the historical period, and the "signal at each moment of the historical period" is processed according to the following expression (here, the input signal at each moment is equivalent to splicing the signal of the second moment with a memory length of P), and finally the corresponding historical splicing vectors at all moments are used as the training set.

[0034] The specific expression can be:

[0035] in, is the history splicing vector, and Respectively The power amplifier in the antenna is at the first moment The I and Q components of the input signal, is the memory length, For the The power amplifier in the antenna is at the first moment The signal amplitude of the input signal, is the nonlinear order.

[0036] The splicing vectors in actual applications are determined in the same way as the historical splicing vectors during training.

[0037] C. In feedforward transmission, the underlying structure shared by multiple gates is divided into multiple expert networks, each of which is relatively independent, so as to capture the unique features of a single task. At the same time, each power amplifier learning task (which can also be understood as the task corresponding to each antenna) is configured with a gating network, which outputs a probability distribution (weight) to regulate the contribution of each expert network to each task. Therefore, the outputs of multiple expert networks are weighted averaged according to the weights generated by the gating network of a specific task, and then input into the private network related to each task to learn and extract the private features of the task.

[0038] During the training process of the pre-trained pre-distortion model, the processing process of each network can be expressed as:

[0039]

[0040] in, is the historical common feature vector output by the i-th expert network, is the first parameter of the i-th expert network, is the history splicing vector, is the second parameter of the i-th expert network, is the weight of the i-th expert network output by the k-th gating network, is the weight matrix of the gating network, For the The input of a private network, For the The power amplifier in the antenna is at the first moment The historical target I / Q component, is the private network function of the kth task, is the third parameter, is the fourth parameter; wherein the third parameter and the fourth parameter are both parameters of the private network function.

[0041] Since the private network at the top of the pre-trained pre-distortion model is designed to independently learn the private features of each power amplifier, each private network outputs the historical target I / Q components corresponding to the predicted output of each power amplifier, denoted as:

[0042] in, and Respectively The historical target I component and the historical target Q component are output by the pre-trained pre-distortion model in each antenna at the first moment n, thereby determining the historical target I / Q component.

[0043] In some embodiments, during the feedback training process, the pre-trained pre-distortion model is trained using back propagation and gradient descent to optimize the joint loss function of multiple power amplifier learning tasks. Therefore, in order to achieve accurate learning and feature extraction of the multi-antenna power amplifier inverse model, the loss function of the pre-distortion model proposed in this application is:

[0044] in, and is the I / Q component of the actual input signal of the kth antenna power amplifier; the actual input signal is a signal input to the corresponding antenna without any crosstalk; since the pre-distortion model needs to obtain the mapping relationship between the output signal of the power amplifier and the input signal of the power amplifier in order to extract the inverse characteristics including crosstalk and the superposition of the power amplifier itself, the output signal of the pre-distortion model needs to be as close to the input signal of the power amplifier as possible.

[0045] In order to achieve efficient sharing and independent learning among multiple tasks, the common features of the input signals of multiple power amplifiers are learned through the underlying expert network. At the top level of the model, a private network corresponds to each specific task, and based on the output of the underlying expert network, the private network learns and identifies the private features of multiple power amplifiers.

[0046] In a specific embodiment, the input signals of multiple antennas are collected, such as Figure 3 As shown, since the present application only utilizes a single DPD module to perform linearization processing on multiple power amplifiers at the same time, when training the pre-trained pre-distortion model, it is necessary to process the input signals of multiple power amplifiers and combine the historical splicing vectors to form a training set, and train the pre-trained pre-distortion model based on the training set.

[0047] The input and output signals of multiple power amplifiers in the historical period are collected. For each moment in the historical period, the output signal of the power amplifier at each moment is shaped into a historical splicing vector that is a series splicing of the output signal at the first moment and multiple historical I / Q components and signal amplitudes (signal amplitude related items) output at the second moment. Since a single DPD module is used and the nonlinear crosstalk between channels needs to be considered, the historical splicing vectors after the shaping of multiple power amplifiers are used as training samples. Correspondingly, the input signals of multiple power amplifiers are shaped into the I / Q components of the input signals at the first moment, and the splicing combination is used as a training label.

[0048] The input data of the predistortion model in the training set includes the cross-distortion terms of other antennas. Therefore, in the offline training, not only the inverse relationship between the input data and output data of each antenna power amplifier is simulated, but also the inverse relationship between the power amplifier input data and the cross-distortion terms is simulated.

[0049] The pre-distortion model based on multi-task learning combines the inputs of all antennas and combines it with a shared learning mode to extract the crosstalk characteristics of the current antenna to the current antenna. In practical applications, the nonlinear distortion of the power amplifier and the nonlinear crosstalk between multiple antennas can be suppressed at the same time. The pre-compensated signal is sent to the power amplifier of each antenna for amplification, and the influence of its nonlinear distortion and crosstalk is offset by the pre-compensation, so that the output signal of the power amplifier of each antenna is close to a linear relationship with the original input signal.

[0050] Furthermore, crosstalk means that the input signal of the current antenna is affected by the input signal of the adjacent antenna, so that the input signal of the power amplifier of the current antenna contains the signal components of other antennas, and further distortion occurs after passing through the power amplifier. Therefore, in order to more accurately obtain the characteristics of the power amplifier and compensate for the crosstalk, the input signals of other antennas are spliced ​​in the training set. In short, not only the features of the output and input of the current antenna are extracted, but also the features of the output of the current antenna and other antennas (equivalent to extracting the features of the crosstalk part).

[0051] The pre-trained pre-distortion model is trained through training samples and training labels, that is, the nonlinear characteristics of multiple power amplifiers are learned and extracted.

[0052] Combination Figure 4 As shown, the pre-trained pre-distortion model includes multiple expert networks, multiple gating networks, and multiple private networks.

[0053] Expert network: The shared underlying structure in the pre-trained pre-distortion model is divided into multiple expert networks, each of which is relatively independent. Multiple expert networks all use a fully connected network structure, which is divided into an expert input layer, an expert hidden layer, and an expert output layer. The activation functions of the expert hidden layer and the expert output layer both use the Tanh function. The input data of multiple expert networks is the same, and the structures of the expert networks are the same, but the network parameters of each expert network are randomly initialized during training, so that the results cannot approach a single result. Different data views of the input data can be learned, solving the problem that a single network of the same size cannot effectively capture the common features shared in all tasks. By dividing the structure into multiple expert networks, each expert network can learn the representation of a specific task, thereby capturing the unique relevant features in a single task.

[0054] Gated network: Each power amplifier learning task is equipped with a gating network, so the number of gating networks is equal to the number of tasks. Among them, the gating network is usually lightweight, containing only one network layer, and its activation function is the Softmax function. The input data is the same as the input data of the expert network, and the input data is the constructed training set. The output data is a probability distribution, which represents the weight of each expert network in a power amplifier learning task.

[0055] Private network: The output data of the expert network and the output data of the gated network are weighted and used as the input of the private network. Since the private network is used to extract the private features of each power amplifier learning task, the number of private networks is equal to the number of tasks. All of them adopt a fully connected architecture, which is divided into a private input layer, a private hidden layer and a private output layer. The activation functions of the private hidden layer and the private output layer both use the Leaky ReLU function. The private network combines the common features of multiple tasks and the private features obtained for specific tasks to obtain the model output, and its output corresponds to the historical target I / Q components of the predicted power amplifier input.

[0056] Step S230 specifically includes: integrating the trained pre-distortion model into the multi-antenna system, and connecting the output of the pre-distortion model to the power amplifiers of the multiple antennas.

[0057] The trained pre-distortion model is combined with the power amplifier in a multi-antenna satellite communication system to pre-compensate the nonlinearity of the multi-channel power amplifier.

[0058] Afterwards, when the input signal of multiple antennas flows to the pre-distortion model, the input signal at each moment in the current processing period is stored, the memory length is filled, and it is shaped into a format of I / Q components and signal amplitudes at each moment in the processing period corresponding to the memory length, which is a splicing vector, to be suitable for the pre-distortion model input; the splicing vector is transmitted to the pre-distortion model for processing to obtain the target I / Q components of each antenna. Specifically, it can be: using the first number of expert networks in the pre-distortion model to process the splicing vectors respectively to obtain the common feature vectors corresponding to each expert network; and using the second number of gated networks in the pre-distortion model to process the splicing vectors to obtain the weights of each expert network; using the second number of private networks in the pre-distortion model to process the weights of each expert network and the common feature vectors corresponding to each expert network to obtain the target I / Q components of each antenna.

[0059] In this method, the pre-distortion model pre-compensates the received input signal, outputs the target I / Q components corresponding to the input signal of the multi-antenna power amplifier, and converts the target I / Q components of each antenna into a complex signal, which is transmitted to the power amplifier of each antenna for corresponding processing.

[0060] The linearization compensation of multi-channel power amplifiers is regarded as a multi-task problem. A pre-distortion model is constructed using a multi-gated hybrid expert network. The linearization of multi-channel power amplifiers is performed simultaneously through only one DPD module to ensure the linearity of the omnidirectional beam while limiting the significant increase in hardware complexity. Through efficient sharing and independent learning between multiple tasks, the linearization performance of multi-channel power amplifiers can be effectively guaranteed and the nonlinear crosstalk problem between multiple antennas can be alleviated.

[0061] The present application provides a digital predistortion method for a multi-antenna satellite system, which is based on a multi-feedback loop architecture and uses a multi-gated hybrid expert network to model the predistortion model of the multi-antenna system. Among them, the multi-gated hybrid expert network uses an expert network and a private network to learn common features and private features, and uses a gated network to adaptively select features for different tasks. By automatically adjusting the parameterization between shared information and task-specific information, the predistortion model can effectively learn complex data distributions, improve linearization performance, and reduce nonlinear crosstalk across channels. Compared with the existing neural network-based predistortion model, the predistortion model of the present application can obtain a more accurate power amplifier inverse model under the same number of iterations of training to effectively compensate for multi-channel power amplifier distortion and crosstalk. Compared with the existing polynomial machine learning method, the predistortion model has better performance in both nonlinear distortion compensation and crosstalk suppression.

[0062] Corresponding to the above method, the embodiment of the present application also provides a digital predistortion device for a multi-antenna satellite system, such as Figure 5 As shown, the device comprises: An acquisition unit 510 is used to acquire input data of multiple antennas to be processed; the input data includes I / Q components and signal amplitude of the input signal; A splicing unit 520, configured to splice I / Q components and signal amplitudes of multiple antennas to obtain a splicing vector; The processing unit 530 is used to use the trained pre-distortion model to process the splicing vector to obtain the target I / Q components of each antenna, and convert each target I / Q component into a pre-distortion signal so that the power amplifier corresponding to each antenna processes the corresponding pre-distortion signal and then transmits it; wherein the pre-distortion model includes a first number of expert networks, a second number of gated networks, and a second number of private networks.

[0063] The functions of the functional units of the digital predistortion device for a multi-antenna satellite system provided in the above embodiment of the present application can be implemented through the above method steps. Therefore, the specific working process and beneficial effects of each unit in the digital predistortion device for a multi-antenna satellite system provided in the embodiment of the present application are not repeated here.

[0064] The present application also provides an electronic device, such as Figure 6 As shown, it includes a processor 610 , a communication interface 620 , a memory 630 and a communication bus 640 , wherein the processor 610 , the communication interface 620 , and the memory 630 communicate with each other via the communication bus 640 .

[0065] Memory 630, for storing computer programs; The processor 610 is used to implement the following steps when executing the program stored in the memory 630: Acquire input data of multiple antennas to be processed; the input data includes I / Q components and signal amplitude of the input signal; The I / Q components and signal amplitudes of multiple antennas are spliced ​​to obtain a spliced ​​vector; The splicing vector is processed by using a trained pre-distortion model to obtain a target I / Q component of each antenna, and each target I / Q component is converted into a pre-distortion signal, so that a power amplifier corresponding to each antenna processes the corresponding pre-distortion signal and then transmits it; wherein the pre-distortion model includes a first number of expert networks, a second number of gated networks, and a second number of private networks.

[0066] The communication bus mentioned above can be a Peripheral Component Interconnect (PCI) bus or an Extended Industry Standard Architecture (EISA) bus, etc. The communication bus can be divided into an address bus, a data bus, a control bus, etc. For ease of representation, only one thick line is used in the figure, but it does not mean that there is only one bus or one type of bus.

[0067] The communication interface is used for communication between the above electronic device and other devices.

[0068] The memory may include a random access memory (RAM) or a non-volatile memory (NVM), such as at least one disk memory. Optionally, the memory may also be at least one storage device located away from the aforementioned processor.

[0069] The above-mentioned processor can be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; it can also be a digital signal processor (DSP), an application specific integrated circuit (ASIC), a field programmable gate array (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components.

[0070] The implementation methods and beneficial effects of the components of the electronic device in the above embodiments to solve the problems can be seen in Figure 2 The various steps in the illustrated embodiment are implemented, therefore, the specific working process and beneficial effects of the electronic device provided by the embodiment of the present application are not repeated here.

[0071] In another embodiment provided in the present application, a computer-readable storage medium is also provided, in which instructions are stored. When the computer-readable storage medium is executed on a computer, the computer executes a digital predistortion method for a multi-antenna satellite system as described in any one of the above embodiments.

[0072] In another embodiment provided by the present application, a computer program product including instructions is also provided. When the computer program product is executed on a computer, the computer executes a digital predistortion method for a multi-antenna satellite system as described in any one of the above embodiments.

[0073] Those skilled in the art will appreciate that the embodiments in the present application can be provided as methods, systems, or computer program products. Therefore, the embodiments in the present application can be in the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the embodiments in the present application can be in the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program codes.

[0074] The embodiments of the present application are described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the process. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0075] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0076] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0077] Unless otherwise defined, the technical terms or scientific terms used in this application should be understood by people with ordinary skills in the field to which the present invention belongs. "First", "second" and similar words used in this application do not indicate any order, quantity or importance, but are only used to distinguish different components. "Include" or "comprise" and similar words mean that the elements or objects appearing before the word cover the elements or objects listed after the word and their equivalents, without excluding other elements or objects. "Connect", "couple" or "connected" and similar words are not limited to physical or mechanical connections, but may include electrical connections, whether direct or indirect. "Up", "down", "left", "right" and the like are only used to indicate relative positional relationships. When the absolute position of the described object changes, the relative positional relationship may also change accordingly.

[0078] Although the preferred embodiments of the present application have been described, those skilled in the art may make additional changes and modifications to these embodiments once they are aware of the basic creative concepts. Therefore, the present application embodiments are intended to be interpreted as including the preferred embodiments and all changes and modifications that fall within the scope of the present application embodiments.

[0079] Obviously, those skilled in the art can make various changes and modifications to the embodiments in the present application without departing from the spirit and scope of the embodiments in the present application. Thus, if these modifications and variations of the embodiments in the present application are within the scope of the embodiments in the present application and their equivalents, the embodiments in the present application are also intended to include these modifications and variations.

Claims

1. A digital predistortion method for a multi-antenna satellite system, characterized in that: The method comprises: Acquire input data of multiple antennas to be processed; the input data includes I / Q components and signal amplitude of the input signal; The I / Q components and signal amplitudes of multiple antennas are spliced ​​to obtain a spliced ​​vector; The splicing vector is processed by using a trained pre-distortion model to obtain a target I / Q component of each antenna, and each target I / Q component is converted into a pre-distortion signal, so that a power amplifier corresponding to each antenna processes the corresponding pre-distortion signal and then transmits it; wherein the pre-distortion model includes a first number of expert networks, a second number of gated networks, and a second number of private networks.

2. The method according to claim 1, characterized in that For any expert network, the expert network is a fully connected network structure, including an expert input layer, an expert hidden layer and an expert output layer; The activation functions of the expert hidden layer and the expert output layer are both Tanh functions.

3. The method according to claim 1, characterized in that For any gated network, the gated network only includes a network layer; The activation function of the network layer is the Softmax function.

4. The method according to claim 1, characterized in that The private network is a fully connected network structure, including a private input layer, a private hidden layer and a private output layer; The activation functions of the private hidden layer and the private output layer are both Leaky ReLU functions.

5. The method according to claim 1, characterized in that The spliced ​​vector is processed using a trained predistortion model to obtain target I / Q components of each antenna, including: Using a first number of expert networks in the predistortion model to process the splicing vectors respectively, to obtain common feature vectors corresponding to the expert networks; and, using a second number of gated networks in the predistortion model to process the concatenated vector to obtain weights of each expert network; The second number of private networks in the predistortion model are used to process the weights of the expert networks and the common feature vectors corresponding to the expert networks to obtain target I / Q components of each antenna.

6. A digital predistortion device for a multi-antenna satellite system, characterized in that: The device comprises: An acquisition unit, configured to acquire input data of multiple antennas to be processed; the input data includes I / Q components and signal amplitude of the input signal; A splicing unit, used for splicing I / Q components and signal amplitudes of multiple antennas to obtain a splicing vector; A processing unit is used to use a trained pre-distortion model to process the splicing vector to obtain a target I / Q component of each antenna, and convert each target I / Q component into a pre-distortion signal so that a power amplifier corresponding to each antenna processes the corresponding pre-distortion signal and then transmits it; wherein the pre-distortion model includes a first number of expert networks, a second number of gated networks, and a second number of private networks.

7. The device according to claim 6, characterized in that For any expert network, the expert network is a fully connected network structure, including an expert input layer, an expert hidden layer and an expert output layer; The activation functions of the expert hidden layer and the expert output layer are both Tanh functions.

8. The device according to claim 6, characterized in that For any gated network, the gated network only includes a network layer; The activation function of the network layer is the Softmax function.

9. An electronic device, characterized in that: The electronic device comprises a processor, a communication interface, a memory and a communication bus, wherein the processor, the communication interface and the memory communicate with each other via the communication bus; Memory, used to store computer programs; A processor, for implementing the method steps described in any one of claims 1 to 5 when executing a program stored in a memory.

Images