Obstacle detection is very important for people with visual impairment to avoid collision. However, traditional methods usually suffer from extra devices or susceptibility to light. To address these issues, we propose an obstacle detection system, ObstaDetect, which senses echoes in the environment to identify different obstacles in a independent manner on smartphones leveraging acoustic signals. It detects whether there is an obstacle ahead in real time and detects whether we are approaching an obstacle in a dynamic manner. First, the smartphone emits ultrasound, and the built-in microphone records the echo reflected by the obstacle. After that, we use band pass filter to remove noise. Finally, we build a Convolutional Neural Network (CNN) to extract the features of obstacles to distinguish them. In addition, Generate Adversarial Network (GAN) is used to generate large amounts of data similar to the original samples. Ultimately, ObstaDetect is implemented on a smartphone without additional hardware and devices. The experimental results show that ObstaDetect achieves a high accuracy of 96%.
Mobile edge computing (MEC) is growing popular with the development of mobile networks and applications. However, MEC services rely on wireless channels to transmit data and thus are vulnerable to malicious behaviors such as eavesdropping. D2D (Device-to-Device) communication, com-bined with cooperative jamming techniques can efficiently protect MEC customers from such hostiles, and create a physical layer secured MEC environment. Hence in this paper, we propose a D2D-assisted secure MEC mechanism from a physical layer perspective. To tackle the challenge that eavesdroppers' locations are usually unknown, we formulate an eavesdropping risk zone (ERZ) to model the information leakage risk. We introduce a D2D-based cooperative jamming scheme where mobile users (MUs) transmit signals combined with data and artificial noise to reduce ERZ while motivating MUs' participation. Then we propose a contract-based incentive mechanism to promote D2D-assisted cooperative jamming that limits ERZ and improves the secure offloading rate of MEC services. Such incentives can tackle information asymmetry between MUs and incentive designers due to features like individual rationality (IR) and incentive comparability (IC). Further, we jointly optimize incentive costs and transmit power of MEC and D2D users to maximize the operator's MEG utility, where an IC-cost-based method is applied. By simulation, we first evaluate the feasibility of designed incentives, then we demonstrate the performance of the proposed mechanism under different conditions.
With the continuous development of technology in 5G network, different forms of terminal devices are designed to serve the demand of a large consumers and commercial industries. In 5G network, the 5G terminal devices have to store more identity information, certificates and secret keys, and the 5G terminal devices should support more application scenarios, like Internet of Things, smart homes, smart transportation, etc. Besides, abundant terminal devices along with data produces make 5G network more complex, and also bring out many new threats to 5G network. In this paper, we research the terminal security of 5G network under different attack scenes, and study the robustness and connectivity of 5G network.
Recent development of Industrial Internet produces a large number of sensor data that record the production status of equipment in the field of industrial intelligence. This paper focuses on how to detect abnormal event from multivariate time series data. We propose an anomaly detection model DWGAT, that first uses DeepWalk method to generate an embedding vector for each sensor node that can represent sensor's behavior characteristics according to prior information, and then establishes a graph describing the initial relationships between sensor nodes. The graph attention network is used to mine and aggregate the data features to predict the multivariate data at the next moment. Finally, the predicted value is compared with the actual observed value, and a calculation method of anomaly score is designed to judge whether there are anomalies in the system according to whether the anomaly score is larger than the threshold value learned by the model. Experiments are established on two real-world public industrial datasets and compared with two baseline models. The results show that the F1 scores of the proposed anomaly detection model on the two data sets are higher than those of the current optimal model.
Based on the population, urbanization level, per capita GDP, industrial structure, energy technology and industry investment level from 2011 to 2020, this paper established a modified STIRPAT model, and analyzed the relationship between total carbon emissions and various influencing factors by utilizing the ridge regression. Subsequently, by virtue of the scenario analysis method, for different scenarios, Anji County predicted the time to reach the peak of carbon emissions. The research results show that the population, urbanization level, per capita GDP, industrial structure, energy technology and industry investment level are related to the carbon emissions of Anji County, but the effect of each factor is different. By means of the scenario analysis of the STIRPAT model, it can be predicted that about Anji County will reach the target of peak carbon emissions by 2030 under the low carbon scenario. Finally, some positive and effective measures require to be taken, such as adjusting the energy structure, promoting technological innovation, and promoting energy-saving emission reduction measures.
