Symmetric Encryption Algorithms in a Polynomial Residue Number System by I. Yakymenko, M. Karpinski, R. Shevchuk, M. Kasianchuk

“…Our studies show that the cryptanalysis of the proposed algorithm requires combinatorial complexity, which leads to an NP-complete problem.…”
Get full text

Using a Solution Construction Algorithm for Cyclic Shift Network Coding under Multicast Network to the Transformation of Musical Performance Styles by Xiuqin Wang, Jun Geng, Zhiyuan Li

“…A two-way recurrent neural network based on the gated recurrent unit is used to extract a sequence of note feature vectors of different styles, and a one-dimensional convolutional neural network is used to predict the intensity of the extracted note feature vector sequence for a specific style, which better learns the intensity variation of different styles of MIDI music.…”
Get full text

A New Scheduling Quantitative Feedback Theory-Based Controller Integrated with Fault Detection for Effective Vibration Control by R. Jeyasenthil, Yang-Sup Lee, Seung-Bok Choi

“…The idea is centered on comparing the closed-loop output between the faulty system and fault-free one to schedule/switch the feedback control once the fault occurs. The problem addressed in this work is the output disturbance rejection. …”
Get full text

Research on Wind Turbine Unbalance Fault Diagnosis Based on Wavelet Transform and Convolutional Neural Network by Zhenling Li, Yukun Gao

“…It consists of two feature extractors of different scales, which are combined in the fully connected layer. …”
Get full text

Surrounding Object Material Detection and Identification Method for Robots Based on Ultrasonic Echo Signals by Bo Zhu, Tao Geng, Gedong Jiang, Zheng Guan, Yang Li, Xialun Yun

“…This method primarily adopts the 16-dimensional feature vector extracted from intrinsic mode functions that we gain from empirical mode decomposition as inputs of machine learning algorithms to recognize different materials, and we use K-nearest neighbor, decision tree, and support vector machine algorithms on the feature vector set to decide the best classifier, and its acoustic theoretical model is established additionally. …”
Get full text

Analysis of Internet Marketing Forecast Model Based on Parallel K-Means Algorithm by Xiaolei Chen, Sikun Ge

“…During the simulation experiment, the weight calculation formula is used to calculate the weight of the feature term. The basic idea is that if a feature word appears more often in this document but less frequently in other nodes, the word will be assigned higher. …”
Get full text

Presenting a Novel Hybrid Approach of Text Mining Sentiment Analysis in Twitter Using CART Decision Tree by Nasir Tayarani, Mehrdad Jalali

“…CART is very effective in processing discrete and continuous data in text mining. The unique CART feature is a complex data structure analysis that can support regression as well as classification operations, according to the input of the problem. …”
Get full text

CNN-GRU Battery SOC Estimation Method Fused with Attention Mechanism for Electric Multiple Units by WANG Shenghui, TIAN Qin, LIU Lihao, FENG Enlai, YU Tianjian

“…The model employs a convolutional neural network (CNN) to extract short-term feature dependencies from long sequences within the battery cycling data. …”
Get full text

Transient Voltage Unit Protection Based on Fault Information Fusion by Zhenwei Guo, Ruiqiang Zhao, Haojie Li, Yongyan Jiang

“…To address these issues, a new method of transient voltage protection is proposed by mining the equipment configuration information of the bus and line, as well as the frequency characteristics of the line boundary. To solve the problem that single-terminal transient line protection cannot distinguish between line fault and bus fault and the fault of adjacent line head end, a new fault feature is constructed by introducing a difference of instantaneous amplitude of the high-frequency signals between line voltage and bus voltage. …”
Get full text

Construction and Simulation of a Strategic HR Decision Model Based on Recurrent Neural Network by Xiaorong Li, Lijun Zhang, Dongchen Li, Dan Guo

“…A strategy for optimizing the training of hybrid convolutional neural networks is proposed, using an exponential linear unit activation function to solve the problem of neuron dead zones, a hybrid pooling strategy is used to improve the problem of feature information loss in maximum pooling processing, and an improved cross-entropy loss function is used to solve the problem of insufficient learning of difficult classification samples. …”
Get full text

Robust Single Sample Per Person Face Recognition with Probabilistic Illumination Enhancement by Muhammad Tariq Siddique, Ibrahim Venkat, Shah Hasan Shah Newaz, Sharul Tajuddin

“…In this paper, we investigate the SSPP face recognition problem in the presence of potential illumination variation and propose a two-fold approach. …”
Get full text

ET-Mamba: A Mamba Model for Encrypted Traffic Classification by Jian Xu, Liangbing Chen, Wenqian Xu, Longxuan Dai, Chenxi Wang, Lei Hu

“…During the fine-tuning phase, the agent attention mechanism is adopted in the feature extraction phase to achieve global information modeling at a low computational cost, and the SmoothLoss function is designed to solve the problem of the insufficient generalization ability of cross-entropy loss function during training. …”
Get full text

Joint power control and user grouping for uplink power domain non-orthogonal multiple access by Bilal Ur Rehman, Mohammad Inayatullah Babar, Arbab Waheed Ahmad, Hesham Alhumyani, Gamil Abdel Azim, Rashid A. Saeed, Sayed Abdel Khalek

“…Owing to the combinatorial nature of the underlying optimization problem involving user pairing/grouping scheme, power control and decoding order, the computational complexity in determining optimal and sub-optimal solutions remains considerably high. …”
Get full text

Deep learning based model predictive control of active filter inverter as interface for photovoltaic generation by Amin Rasoulian, Hadi Saghafi, Mohammadali Abbasian, Majid Delshad

“…Although MPC is an easily implemented control system, it needs a high computational complexity due to the dependency on solving an iterative optimization problem. To overcome this problem, this study develops an artificial intelligence‐based on one‐dimensional convolutional neural network (1D‐CNN) based MPCs. …”
Get full text

Indoor dynamic environment lo calization algorithm based on transfer learning by Can LIU, Junna SHANG, Ruijiang LI, Keqiang YUE

“…The accuracy of the traditional indoor fingerprint lo calization system is limited bymany factors,such as the density of the reference location node in the fingerprint database and the characteristics of the indoor environment.When the indoor environment changes dynami cally,the RSS fluctuates,and usually does notmeet the assumption of the same distribution.Therefore,it was difficult to obtain high-precision requirements for conventional fingerprint positioningmethod.Aiming at the problem that the traditional algorithm couldn’t locate accurately,an algorithm based on the indoor fingerprint database was designed and implemented.The algorithm adopted the idea ofmigration learning to embed different data sets into the latent feature space,and the adverse effects of environmental changes on the system weremitigated.The simulation results show that the average positioning error of this algorithm is 1.23m.…”
Get full text

Kernel Neighborhood Rough Sets Model and Its Application by Kai Zeng, Siyuan Jing

“…Moreover, the model is used in the problem of feature selection. The proposed method is tested on the UCI datasets. …”
Get full text

Research on Personalized Recommender System Adapting to Different Business by Caixia Tao, Hai Yuan, Kang Chen, Anhua Ma

“…The rapid development of internet technology, especially the developed of Web 2.0 which has the main feature of personality, making a lot of information in front of people at the same time. …”
Get full text

Block cipher mode research based on identifier of text files by Tong-ming ZHANG, Jian-feng GUAN

“…The electronic code book(ECB)mode cannot conceal the data model of the plaintext and has only one private key to the whole plaintext.A block cipher mode based on identity for text files was proposed，which ab-sorbed the advantages of block cipher mode and the feature of plaintext string.A double Hash mapping structure was designed and one layer of the mapping structure maps the plaintext into discontinuous groups,another one which was based on the concept of identify library produced corresponding private key to the plaintext group.The model overcame the discontinuous group and single private key problem and effectively improved the security of the cipher text.The experiment results show that the new model has better security,flexibility and extensibility.…”
Get full text

基于卡尔曼滤波和数学形态学分形分析的齿轮箱故障诊断 by 刘志川, 唐力伟, 曹立军

“…Aiming at the problem that traditional signal time-frequency analysis methods are not suitable for non-stationary signal,by using gearbox time domain vibration signals to build AR model,least squares method to calculate model parameters,and AIC formula to ascertain model order,status space model is also built for Kalman filter.Mathematical morphology fractal dimension method is used for calculating signal fractal dimension after filtering.This method is used for signal analysis of normal gears,tooth surface abrasion,gear tooth crack and tooth fracture.The analysis results of measured signal show that the method based on Kalman filter and mathematical morphology can recognize gear fault types effectively,and it offers a new feature extract method for gearbox fault diagnosis.…”
Get full text

Blockchain-based service publishing scheme with recommendation for cold start users by Xuewen DONG, Haozhe LIU, Hui QIAO, Jiawei ZHENG

“…In order to solve the problem that the existing publishing systems were vulnerable to tampering, and it was difficult to recommend service information to “cold start” users, a blockchain-based service publishing scheme that supported cold-start user recommendations was proposed.The latent feature mapping model of a single auxiliary domain was extended to two auxiliary domains to make more accurate recommendations for cold start users in the target domain.In addition, the microservice architecture and block technology were combined to ensure system scalability, reliability and security.The experimental results of three real data sets extracted from Amazon transaction data show that the proposed recommendation model is better than most other recommendation methods, and the service release information can be safely stored in the blockchain to ensure that it is not changed.…”
Get full text