The latent variable prior of the variational autoencoder (VAE) often utilizes a standard Gaussian distribution because of the convenience in calculation, but has an underfitting problem. This paper proposes a variational autoencoder with optimizing Gaussian mixture model priors. This method utilizes a Gaussian mixture model to construct prior distribution, and utilizes the Kullback-Leibler (KL) distance between posterior and prior distribution to implement an iterative optimization of the prior distribution based on the data. The greedy algorithm is used to solve the KL distance for defining the approximate variational lower bound solution of the loss function, and for realizing the VAE with optimizing Gaussian mixture model priors. Compared with the standard VAE method, the proposed method obtains state-of-the-art results on MNIST, Omniglot, and Frey Face datasets, which shows that the VAE with optimizing Gaussian mixture model priors can learn a better model.

Driving can take up a substantial part of daily life and frequently trigger negative emotions like anger or anxiety, which have a significant adverse impact on driving safety as well as long-term human health. To identify driver emotions, thereby improving the safety and humanization of intelligent driving, we explore how to model the discriminative emotion features from both speech and facial expressions in this work. More specifically, an effective attention-based network for facial expression and a lightweight speech emotion network are proposed, separately. Then, audio and video features are combined at the feature level to construct our multimodal driver emotion recognition model. This paper proposes a new audio feature extractor that uses a multi-scale residual structure to extract spectrogram features. In terms of video, a set of frame sequences using Local Binary Pattern Histograms (LBPH) is obtained through preprocessing, which generates a fixed-dimensional feature representation. These features are then input into a fine-tuned ResNet18 model to analyze spatial information. This model is further augmented by integrating both a temporal attention module and a Gated Recurrent Unit (GRU), enhancing its capability to create a highly discriminative video representation. Additionally, we propose an Internet of Vehicles (IoV) platform, specifically designed for driver emotion recognition. The IoV platform consists of sensor layer, data acquisition and transport layer, server layer and data application layer. The IoV platform uses sensors to collect multimodal data from drivers, which can provide data support for the proposed multimodal driver emotion recognition algorithm. The performance of this proposed algorithm is evaluated on two multimodal emotional datasets, Ryerson Audio-Visual Dataset of Emotional Speech and Song (RAVDESS) and Surrey Audio-Visual Expressed Emotion (SAVEE), using a variety of performance indicators. Compared to other baseline methods, this proposed multimodal model achieves state-of-the-art results on the RAVDESS and SAVEE datasets, demonstrating superior recognition accuracy with rates of 0.93 and 0.99, respectively. Additionally, it exhibits precision scores of 0.93 on RAVDESS and 0.99 on SAVEE, along with exceptional specificity scores of 0.99 and 1.00, respectively.

Intelligent Reflecting Surface exhibits significant potential in 6G Vehicle-to-Everything (V2X) communications, helping to alleviate communication link blockage issues caused by the high dynamics of vehicle networks. This paper investigates the fairness resource allocation problem in IRS-assisted multi-user Multiple-Input Single-Output V2X system. By jointly optimizing the transmit beamforming at the Roadside Unit and the phase-shift matrix at the IRS to maximize the minimum vehicle users (V-UEs) rate. To tackle the non-convexity of the joint optimization, we model it as a reinforcement learning Markov decision process and propose a Deep Deterministic Policy Gradient based IRS resource allocation algorithm. The algorithm learns the optimal policy through interaction with the model. Simulation results demonstrate that the proposed algorithm enhances the minimum rate of V-UEs and ensures fairness among them.

Sleep apnea-hypopnea syndrome (SAHS) is a disease that seriously affects human sleep. Due to its strong concealment and significant harm, it has gradually attracted people's attention. Therefore, it is necessary to use the SAHS detection model to screen the disease as early as possible. However, most of the current SAHS detection models use convolutional neural network (CNN) architectures such as Resnet, which are ineffective in capturing images' details. Moreover, the Vision Transformer (ViT) model relies heavily on the number of SaO2 signal image data during training. Otherwise, it isn't easy to achieve better performance. In addition, due to the existence of fuzzy and uncertain information in the SaO2 signal image, the detection accuracy of previous models is low. Therefore, in this work, we propose a vision transformer enabled deep fuzzy neural network (ViTDFNN) to detect SAHS in an Internet of Medical Things platform. The ViTDFNN model can analyze the patient's vital signs and sleep status on time and identify potential SAHS conditions. The ViTDFNN model first uses ViT to learn the features of the SaO2 signal image so that global information and long-range dependencies can be learned. Subsequently, the ViTDFNN model uses VGG to extract the features of the SaO2 signal image to learn the local features in the image, which is convenient for distinguishing the texture, edge, and other details. Finally, the ViTDFNN model linearly fuses the two image features and inputs them into the deep fuzzy neural network and fully connected network to detect the patient's symptoms. The deep fuzzy neural network can convert each deterministic pixel into an uncertain pixel, thereby solving the fuzzy and uncertain information in the SaO2 signal image. In addition, the ViTDFNN model also uses the masked strategy to form a task of image reconstruction, which is used to reduce the dependence of the ViT model on the labeled SaO2 signal image. We verified the superiority of ViTDFNN by conducting extensive experiments on a real sleep dataset. The scores of the ViTDFNN model under the F1-Score, Recall and Accuracy metrics are 0.994, 0.992 and 0.994, respectively.

This study aims to investigate the information-geometric optimization method for ship detection in synthetic aperture radar (SAR) images. The detection strategy relies on an optimization-based representation of images, which considers the relationship between local and global statistical information and is more robust to sea clutter. The algorithm has three main stages. First, a constant false alarm rate (CFAR) detector is used to accomplish the initial detection. It is designed to acquire the initial distribution parameters of pure sea clutter and ship targets. Second, a Riemannian metric is constructed according to the local statistical information. Its purpose is to design the energy function of the optimization model. Third, the detection problem is formulated as a maximum decision rule. The experimental results show that the model can effectively detect ships in SAR images, is not sensitive to clutter, and can eliminate the effect of inhomogeneous sea state conditions. Compared with similar algorithms, the proposed algorithm has higher detection precision and good stability in ship detection in SAR images.

Time synchronization is essential for node localization, target tracking, data fusion, and various other Wireless Sensor Network (WSN) applications. To improve the estimation accuracy of continuous clock offset and skew of mobile nodes in WSNs, we propose a novel time synchronization algorithm, the Rao-Blackwellised (RB) particle filter time synchronization algorithm based on the Dirichlet process mixture (DPM) model. In a state-space equation with a linear substructure, state variables are divided into linear and non-linear variables by the RB particle filter algorithm. These two variables can be estimated using Kalman filter and particle filter, respectively, which improves the computational efficiency more so than if only the particle filter was used. In addition, the DPM model is used to describe the distribution of non-deterministic delays and to automatically adjust the number of Gaussian mixture model components based on the observational data. This improves the estimation accuracy of clock offset and skew, which allows achieving the time synchronization. The time synchronization performance of this algorithm is also validated by computer simulations and experimental measurements. The results show that the proposed algorithm has a higher time synchronization precision than traditional time synchronization algorithms.