Integrating multi-modal environment sensing and wireless channel prediction provides an innovative unmanned aerial vehicle (UAV) channel modeling solution, which can serve as a foundation for future UAV communication system design and network optimization. This paper proposes a novel UAV channel predictive model using multi-modal environment sensing information and measured channel data. To enable comprehensive understanding of the complex and dynamic UAV communication surroundings, the Global position system (GPS) location data, inertial measurement unit (IMU) data, channel data, and environment information are fused as the model’s input. Within a generative adversarial network (GAN) framework, the model enhances the prediction accuracy of channel data in complex environments through adversarial training of the generator and discriminator, producing channel data as the model’s output that closely approximates real-world measurements. Furthermore, visual perception data from UAVs is incorporated into the prediction process, allowing the model to abstract scatterers by capturing changes in environment obstacles. During channel prediction, an attention mechanism is also incorporated into the multi-modal data fusion process, dynamically adjusting the importance of each modality to ensure the model concentrate on the most crucial features. Results from the experiments indicate that the proposed method facilitates real-time predictions of ground channel data across a range of flight altitudes and communication frequencies. This advancement contributes important knowledge to UAV communication studies and significantly boosts the effectiveness and reliability of intelligent air-to-ground communication networks.

A channel prediction modeling method based on multimodal fusion perception is proposed for complex environments in unmanned aerial vehicle (UAV) air-to-ground (A2G) communications. Two-dimensional (2D) environmental information and three-dimensional (3D) point cloud data are fused to enhance the model’s ability to capture environment blockage, reflection, and multipath effects. The 2D information provides target objects’ planar distribution and texture features, while the 3D point clouds supplement spatial position, size, and height information. These complementary modalities comprehensively describe the geometric structures present in complex environments. A multimodal modeling network is constructed to explore the nonlinear mapping between environmental perception data and channel data. The network takes 2D building distribution, 3D point cloud data, global image features, UAV and receiver positions, and communication parameters as joint inputs. Feature extraction and fusion modules achieve effective joint encoding of heterogeneous multimodal features. A spatial feature decoupling (SFD) module is designed to address interference caused by coupled features. It separates the data distributions corresponding to different channel characteristics, improving the accuracy of channel impulse response (CIR) prediction. Experimental results demonstrate that the proposed method significantly improves the reliability and adaptability of UAV channel modeling in complex urban scenarios.

Unmanned-aerial-vehicle (UAV) communications, as a critical application scenario in the sixth generation (6G) wireless communication field, has garnered widespread attention. During UAV-to-ground communication, channel data plays a pivotal role. Analyzing channel data enables an understanding of communication environments’ diversity and temporal variability, thereby facilitating the construction of more efficient communication systems. This article proposes a novel UAV-to-ground channel prediction method based on multimodal fusion. The method aims to achieve real-time and precise prediction of UAV-to-ground channel data from UAVs in the 3-D airspace by integrating various sources of information, including UAV-captured images, location data of transmitters and receivers, and communication settings. The network uses a fused architecture combining convolutional neural network (CNN) and Transformer architecture to extract and integrate features from diverse information sources. This fusion strategy significantly enhances the accuracy of UAV-to-ground channel prediction. Incorporating image information enables the network better to comprehend the complexity and dynamics of communication environments, thereby assisting in achieving more precise UAV-to-ground channel prediction. Experimental results demonstrate that the proposed method achieves real-time prediction of ground channels across various flight altitudes and communication frequency bands. This provides robust technical support for advancing UAV communication and offers new insights for optimizing and upgrading future wireless communication systems.

For the development of the sixth-generation (6G) wireless communication technologies and the realization of integrated space–air–ground–sea networks, a novel nonstationary unmanned aerial vehicle (UAV) to multi-unmanned surface vehicle (multiUSV) channel model for maritime communications is proposed. The impacts of evaporation duct and sea surface scattering paths on the channel are considered in the model. To describe the related cooperative, noncooperative, and original characteristics among different UAV-to-unmanned surface vehicle (USV) channels, a cooperative degree parameter based on power distribution is introduced as an influencing factor. For classifying cooperative clusters, the KPowerMeans (KPM) algorithm and birth–death (B–D) process are applied to extract and cluster the cooperative and noncooperative paths, followed by their evolution. Furthermore, the channel impulse responses (CIRs) of the cooperative, noncooperative, and original channels are represented. The distribution of cooperative clusters is then analyzed and fitted by the normal distribution. Several typical statistical properties of the proposed channel model, including auto-correlation function (ACF), cross-correlation function (CCF), root mean square delay spread (RMS DS), stationary intervalss (SIs), cooperative degree, and channel correlation across different USVs, are examined. The effects of speed and altitude on specific channel characteristics are also analyzed. Finally, the model’s accuracy is verified through a comparison of available measured and simulated data. This model provides theoretical guidance for the design and evaluation of future 6G maritime multiuser cooperative communication systems.

With the development of sixth-generation (6G) wireless communication technology, establishing an accurate and effective channel model has become an indispensable technical foundation for the exploitation of novel systems. However, traditional wireless channel modeling methods display obvious deficiencies in the face of more complex scenarios and massive data requirements in 6G communication. In view of this, a novel generative adversarial network (GAN)-based channel generation and modeling method for 6G intelligent industrial internet of things (IIoT) communications is proposed, which is validated with the measurement data obtaining from multi-frequency and multi-scenes IIoT channels. In this methodological framework, the Wasserstein generative adversarial network with gradient penalty (WGAN-GP) is employed to make predictions on real measurement data. It enables the reconstruction of lost channel data and data augmentation, which effectively address the problem of shortage of the real measurement data. Using the generated channel data, the birth-death (B-D) process of clusters is tracked and then a cluster-based channel model is constructed for 6G intelligent IIoT communicaitons. The characteristics and the survival distance of clusters with different scenes in different frequency bands are further investigated and analyzed. The relevant research provides an innovative approach for channel generation and modeling of 6G IIoT scenarios.

With the development of the sixth-generation (6G) wireless communications and Industrial Internet of Things (IIoT), numerous sensors, smart devices, and mobile terminals will be deployed in factories. Accurate modeling of IIoT channels facilitates the design and evaluation of the communication system, thereby ensuring the stability and efficiency of IIoT production systems. However, the huge data volume, the high dynamism, and complexity of IIoT communication environments pose significant challenges for traditional channel models in accurately capturing their characteristic variations. To address these issues, a deep-learning (DL)-based multimodal fusion predictive channel model is proposed in this article. The model is designed with the convolutional neural network (CNN) and multilayer perceptron (MLP) to extract feature information from input dual-camera images and basic physical parameters, respectively, and utilizes the fused features to predict received power and root-mean-square delay spread (RMS DS) in factory environments. The model integrates multiple modal information, capable of fully capturing the complex mapping relationship between the environment and channel characteristics. Comprehensive experiments demonstrate that the model achieves optimal prediction performance when integrating image information from both the transmitter (Tx) and receiver (Rx) simultaneously. Compared to several existing methods, our proposed predictive model exhibits superior performance. It presents a practical and feasible solution for the design and optimization of advanced IIoT systems in the future.

Many existing advanced deep learning-based autonomous systems have recently been used for autonomous vehicles. In general, a deep learning-based visual perception system heavily relies on visual perception to recognize and localize dynamic interest objects (e.g., pedestrians and cars) and indicative traffic signs and lights to assist autonomous vehicles in maneuvering safely. However, the performance of existing object recognition algorithms could degrade significantly under some adverse and challenging scenarios including rainy, foggy, and rainy night conditions. The raindrops, light reflection, and low illumination pose a great challenge to robust object recognition. Thus, A robust and accurate autonomous driving system has attracted growing attention from the computer vision community. To achieve robust and accurate visual perception, we target to build effective and efficient augmentation and fusion techniques based on visual perception under various adverse conditions. The unpaired image-to-image (I2I) synthesis is integrated for visual perception enhancement and effective synthesis-based augmentation. Besides, we design a two-branch architecture to utilize the information from both the original image and the enhanced image synthesized by I2I. We comprehensively and hierarchically investigate the performance improvement and limitation of the proposed system based on visual recognition tasks and network backbones. An extensive experimental analysis of various adverse weather conditions is also included. The experimental results have demonstrated the proposed system could promote the ability of autonomous vehicles for robust and accurate perception under adverse weather conditions.