Cyber Physical System (CPS) is a combination of physical systems with cyber systems, where there is a tight coupling between the two systems. It is widely used in critical national infrastructure, such as electric power, petroleum and chemical industries. Once an attack against the CPS obtains success, the consequence will be unimaginable. A well-designed risk assessment of CPS will provide an overall view of CPS security status and support efficient allocations of safeguard resources. Though there is much relationship between CPS and IT system, they are still different in various aspects, especially the requirement for real-time. Therefore, traditional risk assessment method for IT system can't be directly applied in CPS. New ideas on CPS risk assessment are in urgent need and one idea about this is addressed in this paper. Firstly, it presents a depict description of a three-level CPS architecture and makes an analysis on the corresponding security features in each level. Secondly, it sums up traditional risk assessment methods analyzes the differences between cyber physical system security and traditional IT system security. Finally, the authors blaze a trail under the new perspective of CPS after breaking the restriction of traditional risk assessment methods and propose a risk assessment idea for CPS.

In recent years, the security of cyberphysical system (CPS) has been focused on increasingly. The most common example of CPS is industrial control system (ICS), which is prevalent in almost every critical infrastructure, such as electricity, oil and gas, water, chemical processing, and healthcare. So ICS security has become a top priority in the security field. Based on a general description of the wireless sensor network (WSN), which is an important element of CPS, this paper first gives a comprehensive and deep understanding of CPS. Secondly, it provides a comprehensive description of the current situation of ICS security in the U.S. and the corresponding approaches the U.S. government and some industries have taken, including management, technology, standards and regulations, and researches of national laboratories. Thirdly, the paper shows the research on ICS in Europe, focusing on the most important report issued by ENISA. Then, compared with developed countries, it presents the grim situation of ICS security and describes the efforts of ICS security management in China.

No abstract

No abstract

As edge computing paradigm achieves great popularity in recent years, there remain some technical challenges that must be addressed to guarantee smart device security in Internet of Things (IoT) environment. Generally, smart devices transmit individual data across the IoT for various purposes nowadays, and it will cause losses and impose a huge threat to users since malware may steal and damage these data. To improve malware detection performance on IoT smart devices, we conduct a malware categorization analysis based on the Kaggle competition of Microsoft Malware Classification Challenge (BIG 2015) dataset in this article. Practically speaking, motivated by temporal convolutional network (TCN) structure, we propose a malware categorization scheme mainly using Word2Vec pre-trained model. Considering that the popular one-hot encoding converts input names from malicious files to high-dimensional vectors since each name is represented as one dimension in one-hot vector space, more compact vectors with fewer dimensions are obtained through the use of Word2Vec pre-training strategy, and then it can lead to fewer parameters and stronger malware feature representation. Moreover, compared with long short-term memory (LSTM), TCN demonstrates better performance with longer effective memory and faster training speed in sequence modeling tasks. The experimental comparisons on this malware dataset reveal better categorization performance with less memory usage and training time. Especially, through the performance comparison between our scheme and the state-of-the-art Word2Vec-based LSTM approach, our scheme shows approximately 1.3% higher predicted accuracy than the latter on this malware categorization task. Additionally, it also demonstrates that our scheme reduces about 90 thousand parameters and more than 1 hour on the model training time in this comparison.