Communication parameter determination method and apparatus, and electronic device

Abstract

The embodiment of the invention discloses a communication parameter determination method and device and electronic equipment. The method comprises the following steps: inputting a signal parameter of a received signal into a deep neural network model so as to obtain a channel state through the deep neural network model; the channel state is input into a reinforcement learning model, so that an adjustment strategy of communication parameters is obtained through the reinforcement learning model, the communication parameters are used for determining a mode for processing a signal to be sent by the electronic equipment, and the reinforcement learning model is a model obtained through training and used for predicting the adjustment strategy of the communication parameters; and updating the communication parameters of the electronic equipment based on the adjustment strategy of the communication parameters. In this way, the electronic equipment can process the signal to be sent in a mode which is most matched with the current environment (channel state), so that the communication performance of the electronic equipment is improved.

Description

Technical Field

[0001] The present application relates to the field of communication technology, and more specifically, to a method, device and electronic device for determining communication parameters. Background Art

[0002] In the current field of communication technology, electronic devices (such as smart phones, IoT devices, wireless sensors, etc.) generally have the ability to communicate based on preset communication parameters. These communication parameters include but are not limited to frequency, channel, modulation method, transmission power, data rate, etc., which together determine the quality and efficiency of information transmission between devices. In a related manner, these communication parameters are set before the device leaves the factory or when the user configures it for the first time, and remain unchanged during subsequent use, thereby affecting the communication performance of the electronic device. Summary of the invention

[0003] In view of the above problems, the present application proposes a communication parameter determination method, device and electronic device to improve the above problems.

[0004] In a first aspect, the present application provides a method for determining communication parameters, the method comprising: inputting signal parameters of a received signal into a deep neural network model to obtain a channel state through the deep neural network model, the deep neural network model being a trained network model for predicting the channel state; inputting the channel state into a reinforcement learning model to obtain an adjustment strategy for communication parameters through the reinforcement learning model, the communication parameters being a method for determining a processing method for a signal to be sent by an electronic device, the reinforcement learning model being a trained model for predicting an adjustment strategy for communication parameters; and updating the communication parameters of the electronic device based on the adjustment strategy for the communication parameters.

[0005] In a second aspect, the present application provides a communication parameter determination device, the device comprising: a channel state acquisition unit, used to input the signal parameters of a received signal into a deep neural network model to obtain the channel state through the deep neural network model, the deep neural network model is a trained network model for predicting the channel state; a communication parameter acquisition unit, used to input the channel state into a reinforcement learning model to obtain an adjustment strategy for the communication parameters through the reinforcement learning model, the communication parameters are used to determine a method for processing a signal to be sent by an electronic device, and the reinforcement learning model is a trained model for predicting the adjustment strategy for the communication parameters; a communication parameter updating unit, used to update the communication parameters of the electronic device based on the adjustment strategy for the communication parameters.

[0006] In a third aspect, the present application provides an electronic device comprising one or more processors and a memory; one or more programs are stored in the memory and configured to be executed by the one or more processors to implement the above method.

[0007] In a fourth aspect, the present application provides a computer-readable storage medium, in which program code is stored, wherein the above method is executed when the program code is executed by a processor.

[0008] The present application provides a communication parameter determination method, device and electronic device, in which, after obtaining the signal parameters of the received signal, the signal parameters of the received signal can be input into a deep neural network model to obtain a channel state through the deep neural network model, and then the channel state is input into a reinforcement learning model to obtain a communication parameter adjustment strategy through the reinforcement learning model, and then the communication parameters of the electronic device are updated based on the communication parameter adjustment strategy. Thus, through the above-mentioned method, after obtaining the channel state through the signal parameters and the deep neural network model, the communication parameter adjustment strategy can be obtained through the reinforcement learning model and the channel state, so that the communication parameter adjustment strategy is associated with the current channel state, and the communication parameter adjustment strategy can be dynamically updated according to the current channel state, and then, when the communication parameters are used to determine the method for processing the signal to be sent by the electronic device, the electronic device can use the method that best matches the current environment (channel state) to process the signal to be sent, so as to improve the communication performance of the electronic device. BRIEF DESCRIPTION OF THE DRAWINGS

[0009] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For those skilled in the art, other drawings can be obtained based on these drawings without creative work.

[0010] Figure 1 A schematic diagram showing an application scenario proposed in an embodiment of the present application is shown;

[0011] Figure 2 A schematic diagram showing another application scenario proposed in an embodiment of the present application;

[0012] Figure 3 A flow chart of a method for determining communication parameters proposed in an embodiment of the present application is shown;

[0013] Figure 4 A flow chart of a communication parameter determination method proposed in another embodiment of the present application is shown;

[0014] Figure 5 A flow chart of a communication parameter determination method proposed in another embodiment of the present application is shown;

[0015] Figure 6 The overall flow chart of the adjustment strategy from inputting signal parameters to obtaining communication parameters in the embodiment of the present application is shown;

[0016] Figure 7 A structural block diagram of a communication parameter determination device proposed in an embodiment of the present application is shown;

[0017] Figure 8 A structural block diagram of an electronic device used to execute a communication parameter determination method according to an embodiment of the present application is shown in the present application in real time;

[0018] Fig. 9 A storage unit in real time of the present application for storing or carrying program codes for implementing the communication parameter determination method according to an embodiment of the present application is shown. DETAILED DESCRIPTION

[0019] 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 in 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.

[0020] In today's highly developed field of communication technology, electronic devices have become an indispensable part of people's lives and work. Among them, various electronic devices such as smart phones, Internet of Things devices, and wireless sensors are widely available and frequently used by people. These electronic devices generally have the ability to communicate based on specific preset communication parameters. However, in the current common related methods, these communication parameters are often set before the device is manufactured or when the user configures it for the first time. And, they will remain unchanged during subsequent use. This situation will inevitably have a certain impact on the communication performance of electronic devices, limit the full performance of the equipment, and fail to meet the increasingly complex and diverse communication needs.

[0021] Therefore, the inventor proposes a communication parameter determination method, device and electronic device in the present application, in which, after obtaining the signal parameters of the received signal, the signal parameters of the received signal can be input into a deep neural network model to obtain a channel state through the deep neural network model, and then the channel state is input into a reinforcement learning model to obtain a communication parameter adjustment strategy through the reinforcement learning model, and then the communication parameters of the electronic device are updated based on the communication parameter adjustment strategy. Thus, through the above-mentioned method, after obtaining the channel state through the signal parameters and the deep neural network model, the communication parameter adjustment strategy can be obtained through the reinforcement learning model and the channel state, so that the communication parameter adjustment strategy is associated with the current channel state, and the communication parameter adjustment strategy can be dynamically updated according to the current channel state, and then, when the communication parameters are used to determine the method for processing the signal to be sent by the electronic device, the electronic device can use the method that best matches the current environment (channel state) to process the signal to be sent, so as to improve the communication performance of the electronic device.

[0022] An exemplary application scenario provided by an embodiment of the present application is first introduced below.

[0023] In the embodiment of the present application, the communication parameter determination method provided can be performed by an electronic device. In this way, all steps in the communication parameter determination method provided by the embodiment of the present application can be performed by the electronic device. For example, Figure 1 As shown, all steps in the communication parameter determination method provided in the embodiment of the present application can be executed by the processor of the electronic device 100.

[0024] Alternatively, the communication parameter determination method provided in the embodiment of the present application may also be executed by a server. Correspondingly, in this mode of execution by the server, the server may start executing the steps in the communication parameter determination method provided in the embodiment of the present application in response to a trigger instruction. The trigger instruction may be sent by an electronic device used by a user, or may be triggered locally by the server in response to some automation events.

[0025] In addition, if Figure 2As shown, the communication parameter determination method provided in the embodiment of the present application can also be performed collaboratively by the electronic device and the server. In this manner performed collaboratively by the electronic device and the server, some steps in the communication parameter determination method provided in the embodiment of the present application are performed by the electronic device, while other steps are performed by the server. Exemplarily, the electronic device 100 can perform the communication parameter determination method including: obtaining the signal parameters of the received signal, and then the electronic device 100 can transmit the signal parameters of the received signal to the server 200. After receiving the signal parameters, the server 200 can perform subsequent steps to obtain the adjustment strategy of the communication parameters, and then transmit the adjustment strategy of the communication parameters to the electronic device 100. After receiving the adjustment strategy of the communication parameters, the electronic device 100 can adjust the communication parameters currently used.

[0026] It should be noted that in this method of collaborative execution by the electronic device and the server, the steps respectively executed by the electronic device and the server are not limited to the methods described in the above examples. In practical applications, the steps respectively executed by the electronic device and the server can be dynamically adjusted according to actual conditions. In addition, in addition to being able to collaboratively execute with the server 200, the electronic device 100 can also collaboratively execute the communication parameter determination method with other electronic devices.

[0027] It should be noted that the electronic device 100 is Figure 1 and Figure 2 In addition to the smartphone shown in , it can also be a tablet computer, a smart watch, an intelligent voice assistant and other devices with communication functions. Server 200 can be an independent 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 computing, cloud storage, network services, and artificial intelligence platforms. Among them, in the case where the communication parameter determination method provided in the embodiment of the present application is executed by a server cluster or a distributed system composed of multiple physical servers, different steps in the communication parameter determination method can be executed by different physical servers respectively, or can be executed in a distributed manner by a server built based on a distributed system.

[0028] The embodiments of the present application will be described in detail below with reference to the accompanying drawings.

[0029] See also Figure 3 , a communication parameter determination method provided by an embodiment of the present application, the method comprising:

[0030] S110: Inputting signal parameters of the received signal into a deep neural network model to obtain a channel state through the deep neural network model, wherein the deep neural network model is a trained network model for predicting the channel state.

[0031] Wherein, in the case where the communication parameter determination method is executed by an electronic device, or is executed by the electronic device in collaboration with other devices (for example, the aforementioned server), the received signal can be understood as a signal received by the electronic device. The received signal can be understood as being sent by a device that is communicating with the electronic device. Wherein, in the case where the communication parameter determination method is executed by the electronic device and other devices in collaboration, the device that is communicating with the electronic device may be a device that executes the communication parameter determination method in collaboration with the electronic device, or the device that is communicating with the electronic device may be different from the device that executes the communication parameter determination method in collaboration with the electronic device.

[0032] Among them, the signal parameters can be understood as parameters that characterize channel characteristics. As a way, the signal parameters of the received signal include: at least one of the frequency, bandwidth, noise level, interference information, signal attenuation and signal quality of the signal. Among them, in wireless communication, the frequency of the signal refers to the vibration rate of the electromagnetic wave in the signal, which is usually measured in Hertz (Hz). The frequency determines the basic characteristics of the signal propagation in space, including its propagation distance, penetration ability, etc. The bandwidth of the signal refers to the width of the signal spectrum, that is, the difference between the highest frequency component and the lowest frequency component of the signal. The noise level refers to the useless and randomly changing components in the signal, which can be expressed in power spectrum density or decibels (dB). Interference information refers to the degree of interference of other signals to the signal during transmission, including co-channel interference, adjacent channel interference, intermodulation interference, etc. Signal attenuation refers to the situation in which the signal strength is reduced due to factors such as propagation medium, distance, obstacles, etc. during the transmission process. Signal quality is an important indicator to measure the quality of the signal, including signal stability, clarity, distortion, etc.

[0033] As a way, in order to ensure the consistency and robustness of the model input, the signal parameters of the received signal can be preprocessed to obtain preprocessed signal parameters, and the preprocessing includes at least one of normalization and denoising. Then, the preprocessed signal parameters are input into the deep neural network model to obtain the channel state through the deep neural network model.

[0034] Among them, consistency refers to the fact that signals received from different sources or at different times can maintain similar formats and features when input into the deep neural network model, so that the deep neural network model can stably process these signals. Robustness refers to the tolerance of the deep neural network model to noise, outliers or small changes in the input signal, that is, the deep neural network model can still maintain good performance under these conditions. Among them, normalization can be understood as scaling the value of the signal parameter to a specific range (such as 0 to 1 or -1 to 1) to eliminate the dimensional differences between different parameters, making it easier for the model to learn and understand these parameters. Normalization helps to speed up the training of the model and may improve the accuracy of the model. Denoising can be understood as removing noise from the signal to highlight the true characteristics of the signal. Noise may come from environmental interference, equipment failure, etc., which will affect the model's accurate recognition of signal features.

[0035] In an embodiment of the present application, a deep neural network model (DNN) may be obtained through training and has the ability to output channel states according to input signal parameters. A large amount of historical data may be used to train the deep neural network model to be trained to obtain a deep neural network model for outputting channel states. The historical data may include previously collected wireless communication channel state data, which may specifically include channel impulse response information, signal frequency, bandwidth, noise level, interference information, signal attenuation, and signal quality.

[0036] In the embodiment of the present application, a multilayer perceptron (MLP) or a convolutional neural network (CNN) can be used as the basic architecture of a deep neural network model. Among them, the multilayer perceptron and the convolutional neural network are two different basic architectures of a deep neural network model.

[0037] Among them, the multilayer perceptron is a widely used artificial neural network model, whose structure consists of multiple layers, including an input layer, one or more hidden layers, and an output layer. Each layer contains multiple neurons, which are connected to the neurons of the next layer through weights.

[0038] Among them, the input layer of the multilayer perceptron is used to receive external input data, and its number of neurons usually corresponds to the feature dimension of the input data. The hidden layer of the multilayer perceptron is responsible for extracting the potential features of the input data and performing nonlinear transformations. The multilayer perceptron can contain one or more hidden layers, each of which consists of multiple neurons. The neurons in the hidden layer receive the output from the neurons in the previous layer as input through weighted connections, and generate their own output as input for the neurons in the next layer. The output layer of the multilayer perceptron is responsible for generating the final prediction results. The number of neurons in the output layer usually corresponds to the target number of the task. For example, in a binary classification task, the output layer may contain one neuron, whose output value represents the probability of belonging to a certain category; in a multi-classification task, it may contain neurons equal to the number of categories, and the output value of each neuron represents the probability of belonging to the corresponding category. In a multilayer perceptron, each neuron is a basic processing unit, which receives weighted input from the neurons in the previous layer and generates output through an activation function. Common activation functions include sigmoid, tanh, and ReLU, which are used to introduce nonlinear factors so that neural networks can handle complex nonlinear relationships.

[0039] The training process of a multilayer perceptron usually includes two stages: forward propagation and back propagation. In the forward propagation stage, the input data (for example, the aforementioned historical data) is passed from the input layer to the output layer through the hidden layer layer by layer, and the neurons in each layer calculate the output value based on the current weights and biases. Finally, the output layer generates a prediction result. In the back propagation stage, the weights and biases are updated layer by layer through the back propagation algorithm based on the error between the predicted result and the actual result. The back propagation algorithm calculates the gradient of the error with respect to each weight and bias based on the chain rule, and uses an optimization algorithm (such as stochastic gradient descent SGD, Adam, etc.) to update these parameters.

[0040] Convolutional neural networks include input layers, convolutional layers, pooling layers (optional), fully connected layers, and output layers. By introducing unique operations such as convolutional layers and pooling layers, convolutional neural networks can effectively extract local features and gradually build an understanding of complex data through hierarchical feature abstraction.

[0041] Among them, the input layer of the convolutional neural network is used to receive signal parameters as input data. The convolution layer of the convolutional neural network is used to extract local features in the input data through convolution operations. The convolution operation slides on the input data through a small filter (i.e., convolution kernel) to calculate the weighted sum of the local area. Each convolution kernel can learn a specific feature pattern. The convolution layer usually uses multiple convolution kernels so that different feature representations can be learned at the same time. The pooling layer of the convolutional neural network is used to downsample the features output by the convolution layer to reduce the size of the features, reduce the computational complexity, and enhance the robustness of the model. Commonly used pooling operations include maximum pooling and average pooling. The fully connected layer of the convolutional neural network is used to map the features extracted by the convolution layer and the pooling layer to the final output space (such as category labels). The fully connected layer is similar to a traditional neural network by flattening the features and connecting them to all neurons. The output layer of the convolutional neural network is used to produce the final prediction results.

[0042] The training process of convolutional neural networks can use optimization algorithms such as back propagation and gradient descent. During the training process, the weights of the convolution kernel and the fully connected layer are gradually updated through back propagation, and the model continuously optimizes its ability to represent the input data.

[0043] S120: Input the channel state into the reinforcement learning model to obtain an adjustment strategy for communication parameters through the reinforcement learning model, wherein the communication parameters are used to determine a method for processing a signal to be sent by an electronic device, and the reinforcement learning model is a trained model for predicting an adjustment strategy for communication parameters.

[0044] It should be noted that there are many factors that can cause changes in channel status (for example, signal-to-noise ratio, bandwidth occupancy). Specifically, these factors include dynamic characteristics that may appear in wireless communication channels, such as multipath effects, signal attenuation, and noise interference. In addition, channel status will also change if users move, obstacles appear or disappear, and weather conditions change. Therefore, by predicting the channel status, electronic devices can better configure communication parameters.

[0045] After the channel state is obtained through the above steps, the communication parameter adjustment strategy can be further obtained by combining the reinforcement learning model. The reinforcement learning model can be an adjustment strategy for the communication parameters that takes the channel state as input and outputs the corresponding output.

[0046] Optionally, in an embodiment of the present application, a policy gradient-based reinforcement learning model, such as an Actor-Critic model, may be used. When a reinforcement learning model is introduced, the communication strategy can be adjusted according to changes in the environment (changes in channel states), and the decision-making process can be continuously optimized through reinforcement learning. Among them, the Actor-Critic model enables the communication parameters to be dynamically adjusted in actual communication based on past experience and current feedback, thereby achieving optimal performance.

[0047] For the Actor-Critic model (AC model), the Actor network is responsible for generating strategies, and the Critic network evaluates the quality of the strategies generated by the Actor network. The two are combined to optimize the output strategy. The input of the Actor network includes the current channel state, and the output is the adjustment strategy of the communication parameters. Among them, through the policy gradient method (Policy Gradient), the Critic network calculates the value function of the strategy and feeds it back to the Actor network for updating.

[0048] Among them, for the reinforcement learning model, an adaptive reward mechanism can also be designed to encourage the reinforcement learning model to make correct decisions in actual communication. When designing the reward mechanism, factors such as communication success rate, throughput, and delay can be considered as reward indicators. If the electronic device successfully transmits the signal with a low bit error rate, a high reward will be given; if the signal transmission fails or the delay is long, points will be deducted, thereby guiding the strategy to optimize in a better direction.

[0049] It should be noted that the purpose of processing the signal to be sent by the electronic device is to enable the signal to be transmitted better. Among them, better transmission can be understood as a wider transmission range, or preventing the signal from being interfered with during transmission, etc., so as to improve the stability and reliability of signal transmission. Optionally, in an embodiment of the present application, processing the signal may include modulating and / or encoding the signal. In this case, the communication parameters can be understood as parameters used to determine the method of modulating and / or encoding the signal. In other words, through the adjustment strategy of the communication parameters output by the reinforcement learning model, the electronic device can determine the modulation method currently used for signal modulation, and determine the signal encoding method currently used for encoding the signal.

[0050] S130: updating the communication parameters of the electronic device based on the communication parameter adjustment strategy.

[0051] After the communication parameters of the electronic device are adjusted based on the adjustment strategy output by the reinforcement learning model, the electronic device will determine a processing method for the signal to be sent based on the adjusted communication parameters.

[0052] As a method, the communication parameters include at least one of a modulation strategy parameter and a coding strategy parameter, wherein the modulation strategy parameter is used to determine a strategy for modulation processing of a signal to be sent by the electronic device, and the coding strategy parameter is used to determine a strategy for coding processing of a signal to be sent by the electronic device.

[0053] Optionally, the optional modulation modes include binary phase shift keying (BPSK), quadrature phase shift keying (QPSK) or 16-order quadrature amplitude modulation (16-QAM), etc. In this case, the electronic device can select one from the multiple modulation modes based on the modulation strategy of the output communication parameters. For example, when the signal-to-noise ratio of the channel state representation input to the reinforcement learning model is relatively high, the final output adjustment strategy may indicate the selection of a modulation mode such as 16-QAM to increase the data transmission rate. When the signal-to-noise ratio of the channel state representation input to the reinforcement learning model is relatively low, the final output adjustment strategy may indicate the selection of a modulation mode such as BPSK to ensure communication reliability.

[0054] Optionally, the optional coding methods include low-density parity check (LDPC) codes or Turbo codes. In this case, the electronic device can select one from the multiple coding methods based on the modulation strategy of the output communication parameters. For example, when the channel state characterization input to the reinforcement learning model has a low signal-to-noise ratio, the final output adjustment strategy may indicate the selection of a coding method with a lower coding rate to improve noise resistance. When the channel state characterization input to the reinforcement learning model has a high signal-to-noise ratio, the final output adjustment strategy may indicate the selection of a higher coding rate to improve transmission efficiency.

[0055] A communication parameter determination method provided in this embodiment enables, through the above-mentioned method, after obtaining the channel state through signal parameters and a deep neural network model, to obtain an adjustment strategy for the communication parameters through a reinforcement learning model and the channel state, so that the adjustment strategy for the communication parameters is associated with the current channel state, and the adjustment strategy for the communication parameters can be dynamically updated according to the current channel state. Furthermore, when the communication parameters are used to determine a method for processing a signal to be sent by an electronic device, the electronic device can process the signal to be sent in a method that best matches the current environment (channel state) to improve the communication performance of the electronic device.

[0056] See also Figure 4 , a communication parameter determination method provided by an embodiment of the present application, the method comprising:

[0057] S210: Inputting signal parameters of the received signal into a deep neural network model to obtain a channel state through the deep neural network model, wherein the deep neural network model is a trained network model for predicting the channel state.

[0058] S220: Input the channel state into the reinforcement learning model to obtain an adjustment strategy for communication parameters through the reinforcement learning model, wherein the communication parameters are used to determine a method for processing a signal to be sent by an electronic device, and the reinforcement learning model is a trained model for predicting an adjustment strategy for communication parameters.

[0059] S230: updating the communication parameters of the electronic device based on the communication parameter adjustment strategy.

[0060] S240: Update the network parameters of the deep neural network model based on the environmental data and the online learning algorithm.

[0061] It should be noted that in the embodiments of the present application, in the process of actually applying the deep neural network model, the deep neural network model can still be fine-tuned to update the network parameters of the deep neural network model. Among them, fine-tuning the deep neural network model can be understood as using incremental learning technology to make small adjustments to the deep neural network model that is already in the application process, avoiding the high computational overhead caused by retraining the deep neural network model in full.

[0062] The deep neural network model may be fine-tuned based on an online learning algorithm. The online learning algorithm may fine-tune the deep neural network model based on environmental data. The environmental data may include: signal parameters of a received signal, or signal parameters of a received signal and user location data.

[0063] As a method, the learning parameters of the online learning algorithm are determined according to the communication performance of the electronic device during the communication process, and the learning parameters include a learning rate. If the model performs poorly, the learning rate may need to be increased to speed up the adjustment of the model; if the model is already stable, the learning rate may need to be reduced to avoid over-adjustment to ensure the effectiveness and robustness of the model in a dynamic environment.

[0064] The present embodiment provides a method for determining communication parameters, so that through the above-mentioned method, after obtaining the channel state through the signal parameters and the deep neural network model, the adjustment strategy of the communication parameters can be obtained through the reinforcement learning model and the channel state, so that the adjustment strategy of the communication parameters is associated with the current channel state, and the adjustment strategy of the communication parameters can be dynamically updated according to the current channel state. Then, when the communication parameters are used to determine the method for processing the signal to be sent by the electronic device, the electronic device can use the method that best matches the current environment (channel state) to process the signal to be sent, so as to improve the communication performance of the electronic device. In addition, in the present embodiment, the network parameters of the deep neural network model and / or the reinforcement learning model can also be dynamically updated according to the current actual environment conditions, so that the model (deep neural network model and / or reinforcement learning model) can be more adapted to the current communication environment.

[0065] See also Figure 5 , a communication parameter determination method provided by an embodiment of the present application, the method comprising:

[0066] S310: Detect whether the communication parameter dynamic update mode is started.

[0067] It should be noted that in the embodiment of the present application, the electronic device may execute the communication parameter determination method provided in the embodiment of the present application only when the communication parameter dynamic update mode is started. Correspondingly, if the communication parameter dynamic update mode is not started, the electronic device may keep the currently used communication parameters until the user manually updates them, or until the communication parameter dynamic update mode is started, and then adaptively updates them.

[0068] The communication parameter dynamic update mode is manually started by the user, or the communication parameter dynamic update mode is started when the electronic device is detected to be in a busy communication state. It should be noted that when the electronic device is in a busy communication state, the electronic device needs to have a greater probability of ensuring stable and reliable communication. Furthermore, the communication parameter dynamic update mode can be started when the communication is busy.

[0069] In the embodiment of the present application, there are multiple ways to determine whether the current communication state is busy.

[0070] As a method, the number of programs that use the communication function among the currently running programs can be detected. If the number of programs that use the communication function is greater than a threshold number, it is determined that the current state is in a busy communication state. Correspondingly, if the number of programs that use the communication function is less than or equal to a threshold number, it is determined that the current state is in an idle communication state. Among them, the communication function can be understood as a function of communicating with devices other than electronic devices.

[0071] As another way, optionally, in the embodiment of the present application, time can be divided into a communication idle time period and a communication busy time period based on the operation of the electronic device. The communication idle time period can be understood as a time period when there are relatively few programs that use the communication function among the programs running on the electronic device, and the communication busy time period can be understood as a time period when there are relatively more programs that use the communication function among the programs running on the electronic device.

[0072] In the embodiment of the present application, the communication idle time period and the communication busy time period can be divided according to the historical operation status of the electronic device.

[0073] As one approach, the communication idle time period and the communication busy time period may be divided according to the number of programs that use the communication function and are run by the electronic device.

[0074] Optionally, multiple time periods may be pre-divided, and then the number of programs that use the communication function and are running in each time period is counted, and then each time period is divided into a communication idle time period or a communication busy time period according to the number of programs that use the communication function and are running in each time period. For example, a time period in which the number of programs that use the communication function and are running is less than a first quantity threshold may be determined as a communication idle time period, and a time period in which the number of programs that use the communication function and are running is not less than (greater than or equal to) the first quantity threshold may be determined as a communication busy time period.

[0075] The length of each pre-divided time period may be of various types, for example, the length of each time period may be 1 hour, 2 hours, or 3 hours.

[0076] Optionally, the time period can be divided by counting the running status of the programs in the electronic device, and the divided time period can be synchronously determined to be a communication idle time period or a communication busy time period. In this way, the moment when the number of running programs that use the communication function changes compared to the first number threshold can be used as the moment when the communication idle time period and the communication busy time period intersect. After the electronic device is started, the number of currently running programs that use the communication function can be detected, and the moment of startup is used as the starting moment of the communication idle time period. If it is detected that the number of running programs that use the communication function is greater than the first number threshold, the moment when the number of programs that use the communication function is detected to be greater than the first number threshold can be used as the starting moment of the communication busy time period, and the starting moment is also the end moment of a communication idle time period before the communication busy time period. If after detecting that the number of running programs that use the communication function is greater than the first number threshold, it is detected that the number of programs that use the communication function is not greater than the first number threshold, the moment when the number of programs that use the communication function is detected to be less than the first number threshold will be used as the end moment of the current communication busy time period, and the end moment is also the start moment of a new communication idle time period entered. In this way, the electronic device does not need to divide the time periods into multiple time periods in advance, thus having better flexibility.

[0077] As a determination method, the user of the electronic device may determine the specific method of dividing the communication idle time period and the communication busy time period. As another method, the electronic device may determine the specific method of dividing the communication idle time period and the communication busy time period according to actual conditions.

[0078] The communication idle time period and the communication busy time period obtained by different division methods may be different.

[0079] S320: If the communication parameter dynamic update mode has been started, the signal parameters of the received signal are input into the deep neural network model to obtain the channel state through the deep neural network model.

[0080] S330: Input the channel state into the reinforcement learning model to obtain an adjustment strategy for communication parameters through the reinforcement learning model, wherein the communication parameters are used to determine a method for processing a signal to be sent by an electronic device, and the reinforcement learning model is a trained model for predicting an adjustment strategy for communication parameters.

[0081] S340: updating the communication parameters of the electronic device based on the communication parameter adjustment strategy.

[0082] S350: If the communication parameter dynamic update mode is not activated, the electronic device may keep the currently used communication parameters until the user manually updates them, or until the communication parameter dynamic update mode is activated and adaptively updated.

[0083] This embodiment provides a communication parameter determination method, so that, in the above manner, when the communication parameter is used to determine the method for processing the signal to be sent by the electronic device, the electronic device can use the method that best matches the current environment (channel state) to process the signal to be sent, so as to improve the communication performance of the electronic device. In addition, in this embodiment, the electronic device can dynamically turn on or off the communication parameter dynamic update mode, thereby improving the flexibility of the method.

[0084] like Figure 6 As shown, an overall process from the input of signal parameters to the adjustment strategy of communication parameters is shown. It also includes fine-tuning the model (deep neural network model and reinforcement learning model) by incremental learning. The obtained communication parameter adjustment strategy may include a modulation strategy and a coding strategy, wherein the modulation strategy can be understood as a method for determining the modulation method of the signal to be transmitted. The coding strategy is used to determine the method for encoding the signal to be transmitted.

[0085] See also Figure 7 , an embodiment of the present application provides a communication parameter determination device 400, the device 400 comprising:

[0086] The channel state acquisition unit 410 is used to input the signal parameters of the received signal into the deep neural network model to obtain the channel state through the deep neural network model. The deep neural network model is a trained network model for predicting the channel state.

[0087] The communication parameter acquisition unit 420 is used to input the channel state into the reinforcement learning model to obtain the adjustment strategy of the communication parameters through the reinforcement learning model, wherein the communication parameters are used to determine the method for processing the signal to be sent by the electronic device, and the reinforcement learning model is a trained model for predicting the adjustment strategy of the communication parameters.

[0088] The communication parameter updating unit 430 is configured to update the communication parameters of the electronic device based on the communication parameter adjustment strategy.

[0089] As a method, the communication parameter updating unit 430 is also used to update the network parameters of the deep neural network model based on environmental data and an online learning algorithm, wherein the environmental data includes: signal parameters of the received signal, or signal parameters of the received signal and user location data.

[0090] Optionally, the communication parameter updating unit 430 is further specifically used to determine the learning parameters of the online learning algorithm according to the communication performance of the electronic device during the communication process, and the learning parameters include a learning rate.

[0091] As a method, the channel state acquisition unit 410 is specifically used to preprocess the signal parameters of the received signal to obtain the preprocessed signal parameters, wherein the preprocessing includes at least one of normalization processing and denoising processing; the preprocessed signal parameters are input into the deep neural network model to obtain the channel state through the deep neural network model.

[0092] As a method, the channel state acquisition unit 410 is also used to detect whether the communication parameter dynamic update mode is activated. Furthermore, if the communication parameter dynamic update mode is activated, the signal parameters of the received signal are input into the deep neural network model to obtain the channel state through the deep neural network model.

[0093] The following will be combined Figure 8 An electronic device provided by the present application is described.

[0094] See also Figure 8 Based on the above communication parameter determination method and device, the embodiment of the present application also provides another electronic device 2000 that can execute the above communication parameter determination method. The electronic device 2000 includes one or more (only one is shown in the figure) processors 202, a memory 204, a network module 206, a sensor module 208 and an audio acquisition device 210 that are coupled to each other. Among them, the memory 204 stores a program that can execute the content of the above embodiment, and the processor 202 can execute the program stored in the memory 204.

[0095] Among them, the processor 202 may include one or more processing cores. The processor 202 uses various interfaces and lines to connect various parts within the entire electronic device 2000, and executes various functions and processes data of the electronic device 2000 by running or executing instructions, programs, code sets or instruction sets stored in the memory 204, and calling data stored in the memory 204. Optionally, the processor 202 can be implemented in at least one hardware form of digital signal processing (Digital Signal Processing, DSP), field programmable gate array (Field-Programmable Gate Array, FPGA), and programmable logic array (Programmable Logic Array, PLA). The processor 202 can integrate one or a combination of a central processing unit (Central Processing Unit, CPU), a graphics processing unit (Graphics Processing Unit, GPU) and a modem. Among them, the CPU mainly processes the operating system, user interface and application programs; the GPU is responsible for rendering and drawing display content; and the modem is used to process wireless communications. It can be understood that the above-mentioned modem may not be integrated into the processor 202, but may be implemented separately through a communication chip.

[0096] The memory 204 may include a random access memory (RAM) or a read-only memory (ROM). The memory 204 may be used to store instructions, programs, codes, code sets or instruction sets. The memory 204 may include a program storage area and a data storage area, wherein the program storage area may store instructions for implementing an operating system, instructions for implementing at least one function (such as a touch function, a sound playback function, an image playback function, etc.), instructions for implementing the following various method embodiments, etc.

[0097] The network module 206 is used to implement information interaction between the electronic device 2000 and other devices, for example, transmission of device control instructions, operation request instructions, and status information acquisition instructions, etc. When the electronic device 2000 is specifically a different device, the corresponding network module 206 may be different.

[0098] The sensor module 208 may include at least one sensor. Specifically, the sensor module 208 may include but is not limited to: a light sensor, a motion sensor, a pressure sensor, an infrared heat sensor, a distance sensor, an acceleration sensor, and other sensors.

[0099] Among them, the pressure sensor can be a sensor that detects the pressure generated by pressing on the electronic device 2000. That is, the pressure sensor detects the pressure generated by the contact or pressing between the user and the electronic device, such as the pressure generated by the contact or pressing between the user's ear and the mobile terminal. Therefore, the pressure sensor can be used to determine whether contact or pressing occurs between the user and the electronic device 2000, and the magnitude of the pressure.

[0100] Among them, the acceleration sensor can detect the magnitude of acceleration in all directions (generally three axes), and can detect the magnitude and direction of gravity when stationary, and can be used for applications that recognize the posture of the electronic device 2000 (such as horizontal and vertical screen switching, related games, magnetometer posture calibration), vibration recognition related functions (such as pedometer, tapping), etc. In addition, the electronic device 2000 can also be equipped with other sensors such as gyroscopes, barometers, hygrometers, thermometers, etc., which will not be repeated here.

[0101] The audio acquisition device 210 is used to collect audio signals. Optionally, the audio acquisition device 210 includes a plurality of audio acquisition devices. The audio acquisition device may be a microphone. For example, in one embodiment, the audio acquisition device 210 may include two microphones, in which one of the microphones may correspond to an analog-to-digital converter, and the other microphone may correspond to an analog-to-digital converter with two different analog gains. In another embodiment, the audio acquisition device 210 may include three microphones. In this embodiment, two of the microphones (e.g., a main microphone and a secondary microphone) may each correspond to an analog-to-digital converter, and the other microphone (e.g., a camera microphone) may correspond to an analog-to-digital converter with two different analog gains.

[0102] As a method, the network module of the electronic device 2000 is a radio frequency module, which is used to receive and send electromagnetic waves, realize the mutual conversion between electromagnetic waves and electrical signals, and thus communicate with a communication network or other devices. The radio frequency module may include various existing circuit elements for performing these functions, such as antennas, radio frequency transceivers, digital signal processors, encryption / decryption chips, user identity modules (SIM) cards, memories, etc. For example, the radio frequency module can exchange information with an external device through the electromagnetic waves sent or received, and then receive audio signals sent by the external device.

[0103] Furthermore, the electronic device 2000 may also include an image acquisition device for image acquisition, for example, a video, a static picture or a dynamic picture may be taken by the image acquisition device.

[0104] Please refer to Fig. 9, which shows a structural block diagram of a computer-readable storage medium provided in an embodiment of the present application. The computer-readable medium 800 stores program codes, which can be called by a processor to execute the method described in the above method embodiment.

[0105] The computer readable storage medium 800 may be an electronic memory such as a flash memory, an EEPROM (electrically erasable programmable read-only memory), an EPROM, a hard disk, or a ROM. Optionally, the computer readable storage medium 800 includes a non-transitory computer-readable storage medium. The computer readable storage medium 800 has storage space for program code 810 that performs any method step in the above method. These program codes can be read from or written to one or more computer program products. The program code 810 can be compressed, for example, in an appropriate form.

[0106] The present application provides a communication parameter determination method, device and electronic device, in which, after obtaining the signal parameters of the received signal, the signal parameters of the received signal can be input into a deep neural network model to obtain a channel state through the deep neural network model, and then the channel state is input into a reinforcement learning model to obtain a communication parameter adjustment strategy through the reinforcement learning model, and then the communication parameters of the electronic device are updated based on the communication parameter adjustment strategy. Thus, through the above-mentioned method, after obtaining the channel state through the signal parameters and the deep neural network model, the communication parameter adjustment strategy can be obtained through the reinforcement learning model and the channel state, so that the communication parameter adjustment strategy is associated with the current channel state, and the communication parameter adjustment strategy can be dynamically updated according to the current channel state, and then, when the communication parameters are used to determine the method for processing the signal to be sent by the electronic device, the electronic device can use the method that best matches the current environment (channel state) to process the signal to be sent, so as to improve the communication performance of the electronic device.

[0107] Among them, the communication parameter determination method provided by the embodiment of the present application can improve the adaptability of the communication system. Among them, through the introduction of AI technology (for example, the aforementioned deep neural network model and reinforcement learning model), the system can adapt to changes in the channel environment in real time, significantly improving the robustness and stability of the communication system. Specifically, deep learning and online learning technology are used to dynamically adjust communication parameters to ensure that the system can achieve optimal communication performance in different environments. In addition, the system can autonomously adjust the communication strategy, reducing dependence on manual operations, thereby reducing maintenance costs.

[0108] In addition, the communication parameter determination method provided in the embodiment of the present application can be implemented in multiple application scenarios such as 5G, Internet of Things, and Internet of Vehicles, effectively improving the overall performance of various communication systems.

[0109] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit it. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A method for determining communication parameters, characterized in that: The method comprises: Inputting signal parameters of the received signal into a deep neural network model to obtain a channel state through the deep neural network model, wherein the deep neural network model is a trained network model for predicting the channel state; Inputting the channel state into a reinforcement learning model to obtain an adjustment strategy for a communication parameter through the reinforcement learning model, wherein the communication parameter is used to determine a method for processing a signal to be sent by an electronic device, and the reinforcement learning model is a trained model for predicting an adjustment strategy for the communication parameter; The communication parameters of the electronic device are updated based on the communication parameter adjustment strategy.

2. The method according to claim 1, characterized in that The signal parameter of the received signal includes at least one of frequency, bandwidth, noise level, interference situation information, signal attenuation and signal quality.

3. The method according to claim 1, characterized in that The method further comprises: The network parameters of the deep neural network model are updated based on environmental data and an online learning algorithm, wherein the environmental data includes: signal parameters of a received signal, or signal parameters of a received signal and user location data.

4. The method according to claim 3, characterized in that The method further comprises: According to the communication performance of the electronic device during the communication process, the learning parameters of the online learning algorithm are determined, and the learning parameters include a learning rate.

5. The method according to claim 1, characterized in that The step of inputting the signal parameters of the received signal into the deep neural network model to obtain the channel state through the deep neural network model includes: Preprocessing the signal parameters of the received signal to obtain preprocessed signal parameters, wherein the preprocessing includes at least one of normalization processing and denoising processing; The preprocessed signal parameters are input into a deep neural network model to obtain a channel state through the deep neural network model.

6. The method according to any one of claims 1 to 5, characterized in that: The communication parameters include at least one of a modulation strategy parameter and a coding strategy parameter, wherein the modulation strategy parameter is used to determine a strategy for modulation processing of a signal to be sent by the electronic device, and the coding strategy parameter is used to determine a strategy for coding processing of a signal to be sent by the electronic device.

7. The method according to claim 1, characterized in that The method further includes: inputting the signal parameters of the received signal into the deep neural network model to obtain the channel state through the deep neural network model; Check whether the communication parameter dynamic update mode is started; If the communication parameter dynamic update mode has been started, inputting the signal parameters of the received signal into the deep neural network model to obtain the channel state through the deep neural network model; The communication parameter dynamic update mode is manually started by a user, or the communication parameter dynamic update mode is started when it is detected that the electronic device is in a busy communication state.

8. A communication parameter determination device, characterized in that: The device comprises: A channel state acquisition unit, used to input the signal parameters of the received signal into a deep neural network model to obtain the channel state through the deep neural network model, wherein the deep neural network model is a trained network model for predicting the channel state; a communication parameter acquisition unit, configured to input the channel state into a reinforcement learning model, so as to obtain an adjustment strategy for a communication parameter through the reinforcement learning model, wherein the communication parameter is used to determine a method for processing a signal to be sent by the electronic device, and the reinforcement learning model is a model obtained through training for predicting an adjustment strategy for the communication parameter; A communication parameter updating unit is used to update the communication parameters of the electronic device based on the communication parameter adjustment strategy.

9. An electronic device, characterized in that: The method comprises one or more processors and a memory; one or more programs are stored in the memory and are configured so that the one or more processors execute the method according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores program code, wherein when the program code is executed by a processor, the method according to any one of claims 1 to 7 is executed.

Images