Wireless channel prediction method based on machine learning and geographical environment information

Abstract

This invention discloses a wireless channel prediction method based on machine learning and geographic environment information, belonging to the field of intelligent channel prediction. This method uses a multi-layered convolutional neural network structure to extract terrain feature information and channel state information values ​​from terrain images from low-level to high-level layers, and encodes these into high-dimensional state vectors as input to a fully connected neural network. The fully connected neural network analyzes and learns the complex relationship between the high-dimensional terrain feature state vectors and the channel state information (CSI) values ​​of the corresponding area to achieve outdoor wireless channel prediction. This invention improves the accuracy of deep neural channel prediction networks, facilitating more efficient deployment of drone base stations in remote areas or disaster sites lacking infrastructure.

Description

Technical Field

[0001] The present invention relates to a wireless channel prediction method based on machine learning and geographical environment information, belonging to the technical field of intelligent channel prediction. Background Art

[0002] In modern wireless communication systems, achieving widespread and efficient network coverage has always been a challenge, especially in remote areas or disaster sites where infrastructure is lacking. Traditional base station layouts rely on existing infrastructure, such as communication towers, which have high deployment costs and poor flexibility, making it difficult to quickly adapt to the communication needs of emergencies or special environments. With the rapid development of Unmanned Aerial Vehicle (UAV) technology, drone-assisted communication has become an emerging solution that can be quickly deployed in environments without infrastructure support to provide temporary wireless communication services. However, the effective deployment of drone base stations requires accurate wireless channel information to optimize coverage and communication quality, especially in outdoor scenarios with complex and unknown geographical environments.

[0003] Existing channel prediction methods are primarily categorized as those based on deterministic models and intelligent channel prediction based on machine learning. While deterministic modeling methods accurately describe channel propagation characteristics in specific environments, facilitating precise communication system design and performance evaluation in known environments, they are computationally complex, require extensive environmental information, and are difficult to apply to situations where environmental information is incomplete or changing. To address the limitations of deterministic modeling channel prediction methods in unknown and complex environments, channel prediction research has gradually evolved from deterministic channel modeling to predictive channel modeling driven by large amounts of data. Machine learning enables computers to discover patterns in data and process complex, nonlinear relationships. By identifying the relationship between terrain and environmental factors and channel state information from large amounts of data, and providing a more accurate description of these relationships, UAV-assisted communication systems can achieve more reliable and efficient channel information.

[0004] However, when deploying drone base stations in remote areas lacking infrastructure or at disaster sites, the complexity of the unknown terrain (slope, aspect, and terrain height) and the height of the drone base station can significantly impact the performance of the communication system. Therefore, achieving more accurate channel prediction in remote areas lacking infrastructure or at disaster sites with unknown terrain has become a key issue for the rapid deployment of drone base stations. Summary of the Invention

[0005] The technical problem to be solved by the present invention is to provide a wireless channel prediction method based on machine learning and geographic environment information, aiming to solve the technical problem that in unknown complex environments, deterministic modeling channel prediction has limitations, which makes it difficult to quickly deploy drone base stations.

[0006] To solve the above technical problems, the technical solution of the present invention is to provide a wireless channel prediction method based on machine learning and geographical environment information, comprising the following steps:

[0007] Step S1: Acquire a scene image of the drone base station to be deployed;

[0008] Step S2: inputting the scene image into a trained convolutional neural network module, and outputting a high-dimensional state vector of terrain features;

[0009] Step S3: inputting the high-dimensional state vector into a trained fully connected neural network channel prediction model to obtain a channel prediction result of the scene image;

[0010] Step S4: Based on the channel prediction results, the evaluation of the above model is completed to achieve rapid deployment of drone base stations in the scene.

[0011] The step S1 is specifically as follows:

[0012] Digital terrain images of different areas are obtained through digital elevation model (DEM), and the contour density IL, slope S and aspect A of each terrain image are extracted from DEM using geographic information system (GIS).

[0013] Select multiple locations in different terrain images to deploy ground receiving points;

[0014] The terrain feature information of each ground point, including contour line density IL, slope S, and slope aspect A, is combined with the CSI information at the corresponding drone altitude, including channel capacity C, ground node received power P, and delay D, and the drone altitude H to form a seven-dimensional data set (IL, S, A, C, P, D, H).

[0015] By repeating the above operations at multiple different points in different scenes, n sample data are obtained and a dataset covering different drone base station heights and terrain features is constructed.

[0016] The step S2 is specifically as follows:

[0017] The input terrain image and its corresponding initial channel state information C, P, D and the UAV deployment height H are input into the convolutional neural network as joint features;

[0018] In the second convolutional layer, the convolution kernel performs mid-level abstraction on the low-level features extracted in the first layer, gradually learning the mid-scale relationship between terrain characteristics and channel delay D, as well as the distribution trend of channel capacity C at different drone altitudes H. After the second convolutional layer, the mid-level features are spatially compressed through the maximum pooling layer to extract key channel-terrain correlation characteristics.

[0019] The pooled feature map is input into the third convolutional layer, which performs high-level abstraction and fusion of the global relationship between terrain features IL, S, A and channel features C, P, D and drone height H. The spatial size of the feature map is further compressed through the maximum pooling layer, and the feature representation is concentrated in a smaller spatial range to reduce data redundancy and retain the global information of high-level features, which is represented by a high-dimensional state vector.

[0020] The step S3 is specifically as follows:

[0021] The high-dimensional state vector output by the convolutional neural network is input into the input layer of the fully connected neural network. The fully connected network gradually learns the potential characteristic patterns of the high-dimensional state vector through layer-by-layer processing and predicts the channel states C, P, D and the optimal UAV deployment height H.

[0022] Perform preliminary feature combination on the input high-dimensional state vector, and use nonlinear activation function to capture the complex nonlinear relationship between terrain features IL, S, A and channel states C, P, D and drone height H, to generate a preliminary comprehensive high-dimensional representation;

[0023] The initially extracted comprehensive features are passed to the fully connected layer to further abstract high-dimensional features and extract the deep correlation between terrain features and channel performance indicators;

[0024] The features are further compressed and optimized, integrating the terrain characteristics with the global pattern of channel performance, and combining the impact of the UAV altitude H on the channel state to generate a feature representation with predictive capabilities for the channel state and the optimal UAV deployment altitude.

[0025] The comprehensive features extracted from the hidden layer are mapped to specific predicted values ​​of the channel states C, P, D and the drone deployment height H, providing channel performance prediction for the deployment of drone base stations and the optimal deployment height of drones.

[0026] The step S4 is specifically as follows:

[0027] During the model training phase, the back propagation algorithm is used to dynamically adjust the weights and biases of the convolutional network and the fully connected network to calculate the channel state values ​​predicted by the model. and drone altitude The error with its true value (C, P, D, H), the loss function Loss uses the mean square error, the formula is:

[0028]

[0029] Where n is the total number of samples, (C i ,P i ,D i ,H i) represents the actual channel state and drone height of the i-th sample, The error signal is propagated back to the hidden layer and convolution layer through the chain rule to dynamically adjust the weights and biases of each layer.

[0030] In the model optimization stage, the gradient descent method is used to iteratively update the network parameters. The update formula is:

[0031]

[0032] Among them, w t is the current weight, η is the learning rate, is the gradient of the loss function with respect to the weight;

[0033] The absolute error AE and mean absolute error MAE are used for evaluation:

[0034] The absolute error AE is calculated as follows:

[0035]

[0036] Among them, y is the actual value, is the predicted value;

[0037] The mean absolute error (MAE) is calculated as follows:

[0038]

[0039] Among them, n is the total number of samples, y i and are the actual value and predicted value of the i-th sample respectively;

[0040] Based on the channel prediction results, the model generates the optimal deployment strategy for different areas.

[0041] The beneficial effects of the present invention are: compared with the existing technology, the present invention extracts terrain environment information from terrain images through convolutional neural networks, thereby improving the accuracy of deep neural channel prediction networks, which helps to more efficiently deploy drone base stations in remote areas or disaster sites that lack infrastructure. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] In order to make the implementation process of the embodiment of the present invention clearer, the technical solutions in the embodiment of the present invention will be clearly and completely described below in combination with the drawings in the embodiment of the present invention. Obviously, the described embodiment is only a partial embodiment of the present invention, not all embodiments.

[0043] Figure 1 Schematic diagram of the implementation process of the present invention;

[0044] Figure 2An image of a scene where a drone base station needs to be deployed in an embodiment of the present invention;

[0045] Figure 3 is the terrain feature information of the scene image in the embodiment of the present invention, Figure 3 (a) is a DEM digitized terrain image of the scene image, Figure 3 (b) Figure 3 (a) The scene image containing terrain height information is produced. Figure 3 (c) Figure 3 (a) The image containing the slope information is produced. Figure 3 (d) Figure 3 (a) The resulting image containing terrain direction information;

[0046] Figure 4 It is a channel state information diagram of the different height distributions of communication users and drone base stations in a certain ground area in the scene image;

[0047] Figure 5 This is a structural diagram of the neural network model of the present invention. DETAILED DESCRIPTION

[0048] The present invention will be further described below with reference to the accompanying drawings and specific embodiments.

[0049] Example 1: Figure 1 As shown in FIG, the wireless channel prediction method based on machine learning and geographical environment information has the following specific steps:

[0050] Step S1: Obtain a scene image of the drone base station to be deployed.

[0051] Obtain different scene images and their corresponding terrain feature information and channel state information values, and perform labeling operations on the scene images. Specifically, the data style in the constructed dataset is (IL, S, A, C, P, D, H). The first three columns represent the terrain feature information of the contour line density IL, slope S and slope direction A of the ground point; the seventh column data H represents the deployment height of the drone. In order to reduce the amount of data, this patent considers three different heights of 800 meters, 1000 meters and 2000 meters; the fourth, fifth and sixth columns respectively represent the three CSI information of the channel capacity C, the ground node receiving power P and the delay D between the drone and the ground node at a certain height.

[0052] Acquire terrain images: obtain digital terrain images of different areas through digital elevation model (DEM), and use geographic information system (GIS) to extract contour line density (IL), slope (S), and aspect (A) information of each terrain image from DEM.

[0053] Channel State Information (CSI) Acquisition: Using Wireless Insite software, we selected multiple ground receiving points within various terrain images and deployed them. At each ground point, we deployed a drone base station at altitudes of 800, 1000, and 2000 meters. For each ground point and drone altitude combination, we used the Wireless Insite simulation tool to measure and record the three CSI values: channel capacity (C), received power (P), and latency (D).

[0054] Dataset Construction: The terrain feature information (IL, S, A) of each ground point is combined with the CSI information (C, P, D) at the corresponding drone altitude and the drone height (H) to form a seven-dimensional data point (IL, S, A, C, P, D, H). By repeating this process at multiple different points in different scenarios, n sample data are obtained, and a dataset covering different drone base station altitudes and terrain features is constructed.

[0055] like Figure 2 Shown is an image of a scene where a drone base station needs to be deployed. The present invention aims to deploy drones in remote areas or disaster sites that lack infrastructure to achieve extensive and efficient network coverage.

[0056] like Figure 3 Shown is the terrain feature information of the scene image;

[0057] in Figure 3 (a) is a DEM digitized terrain image of the scene image, Figure 3 (b) Figure 3 (a) The scene image containing terrain height information is produced. Figure 3 (c) Figure 3 (a) The image containing the slope information is produced. Figure 3 (d) Figure 3 (a) The resulting image containing terrain direction information;

[0058] Figure 4 This is a channel state information diagram of the different height distributions of communication users and drone base stations in a certain ground area in the scene image. It can be seen that when the height distribution of drones is different, there will be line-of-sight transmission channels and non-line-of-sight transmission channels between mobile users and drone base stations;

[0059] in accordance with Figure 2 Obtained Figure 3 and Figure 4To account for the varying distribution heights of mobile users and drone base stations in different areas of the same scene, datasets (training, evaluation, and test datasets) were constructed containing different channel state information at different drone base station heights and different scene terrain features. The scene images were labeled based on the regional channel state information corresponding to different terrain features and drone base station heights, generating the datasets. The constructed convolutional neural network was trained using the datasets to obtain a convolutional neural network model capable of extracting features from terrain feature images.

[0060] In an embodiment of the present invention, there are multiple methods for obtaining the scene image of the drone base station to be deployed, the terrain feature information of the scene image, and the channel state information of the communication users and the drone base station distributed at different heights in a certain ground area in the scene image, and the specific acquisition method is not limited.

[0061] Step S2: Input the scene image into a trained convolutional neural network module, and output a high-dimensional state vector of terrain features.

[0062] Specifically, the convolutional neural network model extracts the multi-level relationship between terrain features and CSI layer by layer by inputting the constructed seven-dimensional data set and generates a high-dimensional state vector;

[0063] First, the input terrain image, its corresponding initial channel state information C, P, D, and the drone deployment altitude H are fed as joint features into a convolutional neural network. The first convolutional layer (Conv2D-32x3x3, ReLU) extracts low-level visual features of the terrain (such as contour boundaries, local variations in slope, and the directionality of slope aspect) using 32 3×3 filters. It also initially learns the local relationships between these terrain features and the channel state information. For example, the network is able to capture the distribution trend of weak signals between dense contour areas and the channel capacity C, as well as the impact of the drone altitude H on the channel received power P. This convolutional layer initially fuses the local features of the terrain image with the channel state information. After processing with the ReLU activation function, the network enhances its ability to model local nonlinear relationships. Subsequently, a maximum pooling layer (MaxPool2D-2x2, stride 2) reduces the dimensionality of the extracted features, highlighting key feature areas and reducing computational complexity.

[0064] Next, in the second convolutional layer (Conv2D-64x3x3, ReLU), the convolution kernel performs intermediate-level abstraction on the low-level features extracted in the first layer. At this point, the network can gradually learn the medium-scale relationship between terrain features and channel delay D, as well as the distribution trend of channel capacity C at different drone heights H. For example, in areas with steeper slopes A, the network can capture the significant fluctuations in channel received power P that may occur, and further encode these intermediate-level features into feature maps output by the convolution. After the second convolutional layer, the intermediate-level features are again spatially compressed through the maximum pooling layer to extract more critical channel-terrain correlation characteristics.

[0065] The pooled feature map is then fed into the third convolutional layer (Conv2D-128x3x3, ReLU). At this layer, the network abstracts and fuses the global relationship between terrain features IL, S, A, channel features C, P, D, and drone altitude H at a high level. For example, the network can integrate the complex shapes of dense contour areas, the spatial fluctuations of slope, and the overall impact of drone altitude H on channel capacity C and latency D, forming a high-level representation of the coupled relationship between terrain and channel characteristics. The feature extraction results of this layer can more comprehensively reflect the impact of terrain information on the distribution of channel states. The maximum pooling layer (MaxPool2D-2x2, stride 2) further compresses the spatial size of the feature map, concentrating the feature representation within a smaller spatial range. This reduces data redundancy, preserves the global information of high-level features, and represents it as a high-dimensional state vector.

[0066] Step S3: inputting the high-dimensional state vector into the trained fully connected neural network channel prediction model to obtain the channel prediction result of the scene image.

[0067] The fully connected neural network channel prediction model uses the high-dimensional state vector generated in step S2 as input to complete the prediction of channel state information and optimal UAV deployment height. The specific implementation is as follows:

[0068] First, the high-dimensional state vector output by the convolutional neural network is input into the input layer of a fully connected neural network. This high-dimensional state vector is a comprehensive representation of terrain features (contour density IL, slope S, aspect A), channel state information (channel capacity C, received power P, delay D), and drone altitude H, encapsulating the multi-level relationships between these features. The fully connected network gradually learns the underlying characteristic patterns of the high-dimensional state vector through layer-by-layer processing and predicts the channel states C, P, D and the optimal drone deployment altitude H.

[0069] In the fully connected layer (Dense-1024, ReLU), the network performs a preliminary feature combination on the input high-dimensional state vector, using a nonlinear activation function (ReLU) to capture the complex nonlinear relationship between the terrain features IL, S, A and the channel states C, P, D and the drone's altitude H. This layer integrates the raw terrain features and channel features in the input vector to generate a preliminary, comprehensive high-dimensional representation.

[0070] Next, the initially extracted comprehensive features are passed to a fully connected layer (Dense-512, ReLU). At this layer, the network further abstracts high-dimensional features and extracts deep correlations between terrain features and channel performance metrics. For example, this layer can learn the local impact of high-density contours on channel capacity, as well as the nonlinear relationship between slope variation and received power and latency.

[0071] In the fully connected layer (Dense-256, ReLU), the network further compresses and optimizes the features to generate a highly abstract low-dimensional feature representation that integrates terrain characteristics and global patterns of channel performance, incorporating the impact of drone altitude H on channel state. For example, this layer can learn the global impact of terrain structure on channel state (such as latency D) from high-dimensional features, thereby generating a feature representation that has predictive power for channel state and optimal drone deployment altitude.

[0072] Finally, in the output layer (Dense-4, Linear), the network maps the comprehensive features extracted by the hidden layer into specific predictions for the channel state (C, P, D) and the drone deployment altitude (H). A linear activation function ensures that the range of the predicted values ​​aligns with real-world physical meaning, such as positive channel capacity and received power, and that the drone altitude is within a reasonable range (e.g., 800 to 2000 meters). Through the output of this layer, the model can provide channel performance predictions and optimal drone deployment altitudes for subsequent drone base station deployments.

[0073] Step S4: Based on the channel prediction results, the evaluation of the above model is completed to achieve rapid deployment of drone base stations in the scene.

[0074] First, during the model training phase, the backpropagation algorithm (BP) is used to dynamically adjust the weights and biases of the convolutional network and the fully connected network to minimize the error between the predicted value and the actual channel state value. Specifically, the channel state value predicted by the model is calculated. and drone altitude The error between the actual value (C, P, D, H) and the loss function is the mean square error (MSE), the formula is:

[0075]

[0076] Where n is the total number of samples, (C i ,P i ,D i ,H i ) represents the actual channel state and drone height of the i-th sample, is the corresponding model prediction value. The error signal is back-propagated to the hidden layer and convolutional layer through the chain rule, dynamically adjusting the weights and biases of each layer.

[0077] Secondly, in the model optimization stage, the gradient descent method (such as Adam optimizer) is used to iteratively update the network parameters, gradually reducing the error function value and optimizing the model performance. The parameter update formula is:

[0078]

[0079] Among them, w t is the current weight, η is the learning rate, is the gradient of the loss function with respect to the weights. Through multiple iterative updates, the model can effectively learn the complex relationship between terrain features, channel status, and drone altitude, improving channel prediction accuracy and deployment efficiency.

[0080] To verify the accuracy and reliability of the model, absolute error AE and mean absolute error MAE are used for evaluation; the absolute error AE is calculated as follows:

[0081]

[0082] Among them, y is the actual value, is the predicted value;

[0083] The mean absolute error (MAE) is calculated as follows:

[0084]

[0085] Among them, n is the total number of samples, y i and are the actual value and predicted value of the i-th sample, respectively. By calculating AE and MAE, the prediction accuracy of the model for channel status and drone altitude is comprehensively evaluated.

[0086] Finally, based on the channel prediction results and recommended drone deployment heights, the model generates optimal deployment strategies for different regions. For example, in areas with low channel capacity C and complex terrain, it recommends deploying low-altitude drone base stations to improve received power P. In areas with steep slopes and significant channel delay D, it recommends deploying higher-altitude drone base stations to improve latency performance. In areas with flat terrain and high coverage requirements, medium-altitude drone base stations are recommended to balance coverage and performance. By comprehensively analyzing the distribution relationship between terrain characteristics, channel conditions, and drone altitude, the model achieves precise deployment, making it particularly suitable for improving communication quality in remote areas or disaster relief scenarios.

[0087] The above are the implementation steps of the present invention, which include the preparation of the data set used for training and evaluation, the model structure and the functions of each module, as well as the implementation of model training and channel result prediction;

[0088] A complete channel prediction process is as follows Figure 1 As shown in the figure, the scene image of the UAV base station to be deployed is obtained, and the image containing the terrain feature information and channel state information of the image is input into the pre-trained model. The model structure is as follows Figure 5 As shown, the channel state prediction results in different areas in the scene are obtained.

[0089] The wireless channel prediction method based on machine learning and geographic environment information in an embodiment of the present invention can obtain terrain feature images in unknown scenes to complete real-time channel prediction, which is used for the deployment of drone base stations; the above is a detailed description of the specific implementation methods of the present invention in conjunction with the accompanying drawings, but the present invention is not limited to the above implementation methods. Various changes can be made within the scope of knowledge possessed by ordinary technicians in this field without departing from the purpose of the present invention.

Claims

1. A wireless channel prediction method based on machine learning and geographical environment information, characterized in that: The following steps are involved: Step S1: Obtain the scene image of the UAV base station to be deployed, as well as the contour line density IL, slope S, and slope aspect A information, and combine it with the CSI information at the corresponding UAV altitude, including channel capacity C, ground node received power P, and delay D, and the UAV altitude H; Step S2: inputting the scene image into a trained convolutional neural network module, and outputting a high-dimensional state vector of terrain features; Step S3: inputting the high-dimensional state vector into a trained fully connected neural network channel prediction model to obtain a channel prediction result of the scene image; Step S4: Based on the channel prediction results, the fully connected neural network channel prediction model is evaluated to achieve rapid deployment of the drone base station in the scenario; The step S2 is specifically as follows: The input terrain image and its corresponding initial channel state information C, P, D and the UAV deployment height H are input into the convolutional neural network as joint features; In the second convolutional layer, the convolution kernel performs mid-level abstraction on the low-level features extracted in the first layer, gradually learning the mid-scale relationship between terrain characteristics and channel delay D, as well as the distribution trend of channel capacity C at different drone altitudes H. After the second convolutional layer, the mid-level features are spatially compressed through the maximum pooling layer to extract key channel-terrain correlation characteristics. The pooled feature map is input into the third convolutional layer, which performs high-level abstraction and fusion of the global relationship between terrain features IL, S, A and channel features C, P, D and drone height H. The maximum pooling layer further compresses the spatial size of the feature map, concentrating the feature representation into a smaller spatial range, reducing data redundancy, and retaining the global information of high-level features, which is represented by a high-dimensional state vector. The step S3 is specifically as follows: The high-dimensional state vector output by the convolutional neural network is input into the input layer of the fully connected neural network. The fully connected network gradually learns the potential characteristic patterns of the high-dimensional state vector through layer-by-layer processing and predicts the channel states C, P, D and the optimal UAV deployment height H. Perform preliminary feature combination on the input high-dimensional state vector, and use nonlinear activation function to capture the complex nonlinear relationship between terrain features IL, S, A and channel states C, P, D and drone height H, to generate a preliminary comprehensive high-dimensional representation; The initially extracted comprehensive features are passed to the fully connected layer to further abstract high-dimensional features and extract the deep correlation between terrain features and channel performance indicators; The features are further compressed and optimized, integrating the terrain characteristics with the global pattern of channel performance, and combining the impact of the UAV altitude H on the channel state to generate a feature representation with predictive capabilities for the channel state and the optimal UAV deployment altitude. The comprehensive features extracted from the hidden layer are mapped to specific predicted values ​​of the channel states C, P, D and the drone deployment height H, providing channel performance prediction for the deployment of drone base stations and the optimal deployment height of drones.

2. The wireless channel prediction method based on machine learning and geographical environment information according to claim 1, characterized in that: The step S1 is specifically as follows: Digital terrain images of different areas are obtained through digital elevation model (DEM), and the contour density IL, slope S and aspect A of each terrain image are extracted from DEM using geographic information system (GIS). Select multiple locations in different terrain images to deploy ground receiving points; The terrain feature information of each ground point, including contour line density IL, slope S, and slope aspect A, is combined with the CSI information at the corresponding drone altitude, including channel capacity C, ground node received power P, and delay D, and the drone altitude H to form a seven-dimensional data set (IL, S, A, C, P, D, H). By repeating the operation at multiple different points in different scenes, n sample data are obtained to construct a dataset covering different drone base station heights and terrain features.

3. The wireless channel prediction method based on machine learning and geographical environment information according to claim 1, characterized in that: The step S4 is specifically as follows: During the model training phase, the back propagation algorithm is used to dynamically adjust the weights and biases of the convolutional network and the fully connected network to calculate the channel state values ​​predicted by the model. and drone altitude The error between the actual value (C, P, D, H) and the loss function Loss uses the mean square error, and the formula is: Where n is the total number of samples, (C i ,P i ,D i ,H i ) represents the actual channel state and drone height of the i-th sample, The error signal is propagated back to the hidden layer and convolution layer through the chain rule to dynamically adjust the weights and biases of each layer. In the model optimization stage, the gradient descent method is used to iteratively update the network parameters. The update formula is: Among them, w t is the current weight, η is the learning rate, is the gradient of the loss function with respect to the weight; The absolute error AE and mean absolute error MAE are used for evaluation: The absolute error AE is calculated as follows: Among them, y is the actual value, is the predicted value; The mean absolute error (MAE) is calculated as follows: Among them, n is the total number of samples, y i and are the actual value and predicted value of the i-th sample respectively; Based on the channel prediction results, the model generates the optimal deployment strategy for different areas.

Images