http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
One-dimensional CNN Model of Network Traffic Classification based on Transfer Learning
Lingyun Yang,Yuning Dong,Zaijian Wang,Feifei Gao 한국인터넷정보학회 2024 KSII Transactions on Internet and Information Syst Vol.18 No.2
There are some problems in network traffic classification (NTC), such as complicated statistical features and insufficient training samples, which may cause poor classification effect. A NTC architecture based on one-dimensional Convolutional Neural Network (CNN) and transfer learning is proposed to tackle these problems and improve the fine-grained classification performance. The key points of the proposed architecture include: (1) Model classification--by extracting normalized rate feature set from original data, plus existing statistical features to optimize the CNN NTC model. (2) To apply transfer learning in the classification to improve NTC performance. We collect two typical network flows data from Youku and YouTube, and verify the proposed method through extensive experiments. The results show that compared with existing methods, our method could improve the classification accuracy by around 3-5%for Youku, and by about 7 to 27% for YouTube.
2C-CTP : A Centralized Clustering Data Collection Protocol based on CTP
Lingyun Xu,Tao Li,Baowei Wang,Xingming Sun,Xingang You,Yanhua Zhu 보안공학연구지원센터 2016 International Journal of Smart Home Vol.10 No.9
WSN (wireless sensor network) is widely applied in environmental monitoring, assets tracking, battlefield monitoring and smart building. Since nodes in WSN are constrained by finite battery power and operational capability, it will have the direct effects on the lifetime of whole network. Well-designed routing protocols could make network succinct and efficient. In order to prolong the overall lifetime of WSN, lots of clustering algorithms have been proposed. Most of them are based on LEACH and proved energy-efficient, but few are implemented in real systems. CTP (collection tree protocol) has been extensively used in many systems and provides a reliable protocol for data collection. However, it neither considers energetic balance nor follows clustering feature. In this paper, we propose an efficient routing protocol by improving the CTP, adding clustering and centrally controlled features. Furthermore, we introduce a lightweight method of command distribution and some implementation technologies of clustering based on CTP. The related experiments that carried on TelosB demonstrate that our scheme is outstanding in lifetime and efficiency.
Generative Linguistic Steganography: A Comprehensive Review
Lingyun Xiang,Rong Wang,Zhongliang Yang,Yuling Liu 한국인터넷정보학회 2022 KSII Transactions on Internet and Information Syst Vol.16 No.3
Text steganography is one of the most imminent and promising research interests in the information security field. With the unprecedented success of the neural network and natural language processing (NLP), the last years have seen a surge of research on generative linguistic steganography (GLS). This paper provides a thorough and comprehensive review to summarize the existing key contributions, and creates a novel taxonomy for GLS according to NLP techniques and steganographic encoding algorithm, then summarizes the characteristics of generative linguistic steganographic methods properly to analyze the relationship and difference between each type of them. Meanwhile, this paper also comprehensively introduces and analyzes several evaluation metrics to evaluate the performance of GLS from diverse perspective. Finally, this paper concludes the future research work, which is more conducive to the follow-up research and innovation of researchers.
Improved Particle Swarm Optimization with Dynamic Fractional Order Velocity and Wavelet Mutation
Lingyun Zhou,Lixin Ding 보안공학연구지원센터 2016 International Journal of Hybrid Information Techno Vol.9 No.5
Particle Swarm Optimization (PSO) is one of the most powerful algorithms for optimization. Traditional PSO algorithm tends to suffer from slow convergence and trapping into local optimum. In this paper, an improved PSO algorithm is proposed by combining dynamic fractional order technology and the wavelet mutation strategy. In the proposed method, a dynamic fractional order velocity update equation is designed to control the convergence rate. Furthermore, the wavelet mutation mechanism is employed to improve the swarm diversity and escape from the local optimums. The experimental results show that the proposed algorithm can provide fast convergence speed and high convergence precision based on the ten classic test functions.
On the Existence of Subliminal Channel in Instant Messaging Systems
Lingyun Xiang,Yuhua Xie,Gang Luo,Weizheng Wang 보안공학연구지원센터 2015 International Journal of Security and Its Applicat Vol.9 No.3
Subliminal channel is a covert communication channel that can securely exchange secret information. By analyzing the communication protocols and encryption algorithms of two popular instant messaging systems QQ and Skype, this paper devotes the efforts to find whether they satisfy the condition of establishing subliminal channel or not. The broad-band subliminal channel can be established in Skype and QQ messages. Therefore, two subliminal channel schemes are proposed to embed and extract subliminal information. Finally, performances of the proposed subliminal channels are compared with those of the ones based on ElGamal digital signature and Newton channel. Experimental results and theoretical analysis show that they are easy to be implemented with large bandwidth and execute fast.
Lingyun Tong,Anne Toppinen,Lei Wang 한국식품연구원 2021 Journal of Ethnic Foods Vol.8 No.-
Tea should be perceived more as a cultural product than a fast-moving consumer good with its roots deep in different cultures across nations. Nevertheless, consumer demands on tea have not been explored sufficiently from a cultural perspective. This research aims to examine motives that are driving Chinese tea consumption on two usage occasions, under the influence of the Chinese renqing, mianzi, collectivism, and man-nature unity culture. Linkage was established between Chinese culture and consumer motives. The factor analysis on 280 respondents identified five consumer psychological motives. Further logistic regression analysis suggested consumer motives and usage situations were related. The contribution of the findings is summarized. On one hand, this study adds to the limited understandings on consumer motives of tea in an emerging country from the cultural aspect. On the other hand, deep understandings of brand chasing, sustainability, and pragmatism motives should businesses to better serve consumers.
Fast transient response based on digital single‑cycle charge regulation (SCCR) control
Lingyun Li,Shen Xu,Limin Yu,Weifeng Sun 전력전자학회 2023 JOURNAL OF POWER ELECTRONICS Vol.23 No.9
The Buck converter is a commonly used voltage regulator (VR) structure that supplies power to the load devices and responds to load transients. In most of the applications, the VR is controlled by voltage or current mode control, and some nonlinear approaches are used to improve the transient response which brings more complexities to the closed loop design. To achieve a fast transient response without over-regulation under a load step transient, this paper proposes a digital single-cycle regulation (SCCR) controller, which comprised a fast path, a slow path, and a neutralization path. Unlike the conventional PID control, the neutralization path of the SCCR controller is used to offset the excessive energy change in the first cycle after a load step occurs, which makes the inductor current reach its new static condition in a shorter period of time. The mathematical model of the SCCR control is established in this paper and a scanning algorithm is proposed for the design of the compensator parameters. Experimental results are presented to verify the improvement of the transient response when compared to the conventional and nonlinear PID control, the recovery time is decreased by 64.3% and 43% for load step-up and step-down when compared with nonlinear PID control.