Common spatial pattern (CSP) is an efficient algorithm widely used in feature extraction of EEG-based motor imagery classification. Traditional CSP depends only on spatial filtering, that aims to maximize or minimize the ratio of variances of filtered EEG signals in different classes. Recent advances of CSP approaches show that temporal filtering is also preferable to extract discriminative features. In view of this perspective, a novel spatio-temporal filtering strategy is proposed in this paper. To improve computational efficiency and alleviate the overfitting issue frequently encountered in the case of small sample size, the same temporal filter is designed by EEG signals of the same class and shared by all the spatial channels. Spatial and temporal filters can be updated alternatively in practice. Furthermore, each of the resulting designs can still be cast as a CSP problem and tackled efficiently by the eigenvalue decomposition. To alleviate the adverse effects of outliers or noisy EEG channels, sparse spatial or temporal filters can also be achieved by incorporating an ℓ1-norm-based regularization term in our CSP problem. The regularized spatial or temporal filter design is iteratively reformulated as a CSP problem via the reweighting technique. Two sets of motor imagery EEG data of BCI competitions are used in our experiments to verify the effectiveness of the proposed algorithm.

Unmanned Surface Vehicle (USV) has a broad application prospect and autonomous path planning as its crucial technology has developed into a hot research direction in the field of USV research. This paper proposes an Improved Dueling Deep Double-Q Network Based on Prioritized Experience Replay (IPD3QN) to address the slow and unstable convergence of traditional Deep Q Network (DQN) algorithms in autonomous path planning of USV. Firstly, we use the deep double Q-Network to decouple the selection and calculation of the target Q value action to eliminate overestimation. The prioritized experience replay method is adopted to extract experience samples from the experience replay unit, increase the utilization rate of actual samples, and accelerate the training speed of the neural network. Then, the neural network is optimized by introducing a dueling network structure. Finally, the soft update method is used to improve the stability of the algorithm, and the dynamic $ϵ$-greedy method is used to find the optimal strategy. The experiments are first conducted in the Open AI Gym test platform to pre-validate the algorithm for two classical control problems: the Cart pole and Mountain Car problems. The impact of algorithm hyperparameters on the model performance is analyzed in detail. The algorithm is then validated in the Maze environment. The comparative analysis of simulation experiments shows that IPD3QN has a significant improvement in learning performance regarding convergence speed and convergence stability compared with DQN, D3QN, PD2QN, PDQN, PD3QN. Also, USV can plan the optimal path according to the actual navigation environment with the IPD3QN algorithm.

Image degradation technique has been the focus of current research in image super-resolution( SR). The High-order Deterioration Model(HDM) of Real-ESRGAN is more effective than the conventional bicubic kernel interpolation in simulating the degradation of real-world images. However, the images reconstructed by Real-ESRGAN suffer from two notable weaknesses. Firstly, the rebuilt image is overly smooth and suffers from substantial texture information loss, making it worse than classical models like SRGAN and ESRGAN. Moreover, despite the reconstructed images having a better visualization effect, they are entirely different from the original image, violating the principle of image reconstruction. In order to address these two issues, an improved model built on Real-ESRGAN (IRE) is presented in this work. We remove the first-order degradation modeling of HDM and keep its second-order degradation modeling to lessen the degree of visual deterioration. Then we utilized PatchGAN as the fundamental structure of the discriminator, added a channel attention mechanism to the dense block of the generator, and increased the texture detail in the reconstructed images. Finally, we replaced the L1 loss function with the SmoothL1 loss function to improve the convergence speed with better model performance. The ablation study shows that the average training time was decreased by about 28% when employing PatchGAN rather than U-net as the discriminator. In addition, the ablation study presents that the RankIQA measures assessed by IRE on Set5, Set14, and DIV2K100 were decreased by 15.20%, 20.46%, and 10.27%, respectively, and the PI measures considered by IRE+ on BSD100, Urban100 were decreased by 15.76%, 8.67%, respectively, compared to Real-ESRGAN. Also, the comparison experiment illustrates that the NIQE measures assessed by our suggested model on Set5, RealSR-Canon, and RealSR-Nikon were decreased by 12.31%, 3%, and 2.41%, respectively, compared to Real-ESRGAN.

Based on various on-road sensor observations, dynamic modeling and analysis of urban road networks becomes an important task of an intelligent transportation system. The major difficulty of this task is that traffic states vary dynamically in both spatial and temporal domains. In this paper, we propose a novel urban road network modeling approach. A road network is described by a weighted undirected graph composed of vertices and edges denoting, respectively, traffic intersections and their pairwise connections. Given the topology of the network, an effective weight estimation algorithm is proposed to extract spatial correlations among adjacent traffic intersections from physical sensor observations. Graph weights can be regularly updated to capture the dynamic essence of traffic states over time. On the basis of weight estimation, we further develop a dynamic spatio-temporal traffic prediction model by using the spatio-temporal autoregressive integrated moving average (STARIMA) model. The effectiveness of the proposed graph-based STARIMA is validated by a series of numerical experiments.

In this paper, we present an efficient localization algorithm for sensor networks using range information. Since each sensor can only communicate with its neighbors, the Euclidean distance matrix (EDM), composed by squared distances between each pair of sensors, is incomplete. The first step of the proposed algorithm is to fulfill the EDM completion and relative maps of sensor networks are then achieved by the multidimensional scaling technique. Besides the EDM, the centralized Gram matrix of sensors' coordinates is also used to model the localization problem. Compared with other EDM-based localization algorithms, mathematical properties of both the EDM and the Gram matrix are appropriately exploited so as to improve the estimation accuracy. The resulting localization model is formulated as a semidefinite programming problem. An alternating direction method of multipliers is further developed to enhance the scalability of the proposed algorithm. The numerical experiments demonstrate that the proposed algorithm can effectively improve the estimation accuracy of both the EDM completion and the final localization.

Accurate estimation of time difference of arrival (TDOA) between sensors is a fundamental problem in various applications of sensor networks. In this letter, a novel algorithm is proposed to estimate TDOAs based on low-rank essence of a TDOA matrix, whose entry (i, j) denotes TDOA value between sensors i and j with respect to a common reference source. To enhance the robustness of the proposed algorithm, we further introduce the Gram matrix of the TDOA matrix, resulting in a semidefinite programming estimation problem. An efficient alternating direction method of multipliers is developed to improve the scalability of the proposed algorithm. Simulation results validate the performance of the proposed algorithm.