Short-term traffic flow forecasting is a fundamental and challenging task due to the stochastic dynamics of the traffic flow, which is often imbalanced and noisy. This paper presents a sample-rebalanced and outlier-rejected k-nearest neighbor regression model for short-term traffic flow forecasting. In this model, we adopt a new metric for the evolutionary traffic flow patterns, and reconstruct balanced training sets by relative transformation to tackle the imbalance issue. Then, we design a hybrid model that considers both local and global information to address the limited size of the training samples. We employ four real-world benchmark datasets often used in such tasks to evaluate our model. Experimental results show that our model outperforms state-of-the-art parametric and non-parametric models.

At present, the UAV (Unmanned Aerial Vehicle) has been widely used both in civilian and military fields. Most of the current object detection algorithms used to detect UAVs require more parameters, and it is difficult to achieve real-time performance. In order to solve this problem while ensuring a high accuracy rate, we further lighten the model and reduce the number of parameters of the model. This paper proposes an accurate and lightweight UAV detection model based on YOLOv4. To verify the effectiveness of this model, we made a UAV dataset, which contains four types of UAVs and 20,365 images. Through comparative experiments and optimization of existing deep learning and object detection algorithms, we found a lightweight model to achieve an efficient and accurate rapid detection of UAVs. First, from the comparison of the one-stage method and the two-stage method, it is concluded that the one-stage method has better real-time performance and considerable accuracy in detecting UAVs. Then, we further compared the one-stage methods. In particular, for YOLOv4, we replaced MobileNet with its backbone network, modified the feature extraction network, and replaced standard convolution with depth-wise separable convolution, which greatly reduced the parameters and realized 82 FPS and 93.52% mAP while ensuring high accuracy and taking into account the real-time performance.

Energy balance consumption is an important research topic in the field of wireless sensor networks (WSNs). Clustering protocols are widely used to reduce WSN energy consumption. However, this many-to-one cluster head (CH) model can also easily lead to energy exhaustion of some critical nodes near the base station results in the decline of the network quality and lifetime. This article proposes a clustering election game-based and two-level management clustering protocol (CEGT) for WSNs; the network is divided into layers of equal size, and the layer heads (LHs) are selected within the layers based on the location and energy information. Each layer is further divided into clusters of equal size, and the nodes within the clusters undergo a two-stage screening followed by local clustering and multiple rounds of election games to dynamically select the staged CHs. This protocol effectively prevents low-energy nodes from participating in the CH competitions, and selects high-quality nodes through election games which are based on customized qualification values, and ensures reasonable energy utilization as well as efficient network load balancing. Finally, our simulations show that our CEGT protocol outperforms other clustering protocols in terms of the network lifetime, the number of surviving nodes, and the average number of clusters, and the throughput, which also indicate that the CEGT protocol is more suitable for large-scale networks.

Arrhythmia is a significant cause of death, and it is essential to analyze the electrocardiogram (ECG) signals as this is usually used to diagnose arrhythmia. However, the traditional time series classification methods based on ECG ignore the nonlinearity, temporality, or other characteristics inside these signals. This paper proposes an electrocardiogram classification method that encodes one-dimensional ECG signals into the three-channel images, named ECG classification based on Mix Time-series Imaging (EC-MTSI). Specifically, this hybrid transformation method combines Gramian angular field (GAF), recurrent plot (RP), and tiling, preserving the original ECG time series’ time dependence and correlation. We use a variety of neural networks to extract features and perform feature fusion and classification. This retains sufficient details while emphasizing local information. To demonstrate the effectiveness of the EC-MTSI, we conduct abundant experiments in a commonly-used dataset. In our experiments, the general accuracy reached 93.23%, and the accuracy of identifying high-risk arrhythmias of ventricular beats and supraventricular beats alone are as high as 97.4% and 96.3%, respectively. The results reveal that the proposed method significantly outperforms the existing approaches.

In mobile wireless sensor networks (MWSNs), the mobility of nodes imposes heightened energy management demands on the network. This study proposes a game-based dynamic clustering routing (GDCR) protocol for MWSNs. The GDCR protocol employs multidimensional clustering and constructs a mixed strategy game model among heterogeneous nodes. This model considers various factors for cluster head selection, including residual energy, movement, distance to the sink, and node spacing, ensuring comprehensive decision making. In addition, the protocol incorporates detached node management to ensure network security and stability. Simulation experiments demonstrate that the proposed protocol reduces the average network energy consumption when nodes have unrestricted movements. Furthermore, it effectively curtails fluctuations in energy consumption as the network scales up, thereby prolonging network lifetime and enhancing stability.