Article
Computer Science, Information Systems
Pu Huang, Qiaolin Ye, Fanlong Zhang, Guowei Yang, Wei Zhu, Zhangjing Yang
Summary: A novel robust PCA formulation called Double L-2, L-p-norm based PCA (DLPCA) is proposed in this paper for feature extraction, which simultaneously considers the minimization of reconstruction error and the maximization of variance to ensure insensitivity to outliers. Experimental results demonstrate the effectiveness of this method in feature extraction.
INFORMATION SCIENCES
(2021)
Article
Computer Science, Artificial Intelligence
Sisi Wang, Feiping Nie, Zheng Wang, Rong Wang, Xuelong Li
Summary: PCA is a widely used unsupervised dimensionality reduction algorithm, but it is sensitive to outliers. In order to solve this problem, we propose a robust principal component analysis method, RPCA-RP, which combines reconstruction error and projection variance to improve the performance and has anomaly detection capabilities.
IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS
(2022)
Article
Engineering, Multidisciplinary
Zhenfeng Huang, Kuangchi Sun, Dahuan Wei, Hanling Mao, Xinxin Li, Xun Qian
Summary: The study focuses on fault diagnosis of rotating machinery, using time-domain synchronous averaging to reduce random noise, and employing moving interpolation and Kurtosis searching criterion for accurate signal extraction. An improved compensation algorithm is proposed to overcome amplitude attenuation in low SNR signals.
MEASUREMENT SCIENCE AND TECHNOLOGY
(2021)
Article
Automation & Control Systems
Lei Wang, Xiurui Geng
Summary: This article introduces an algorithm based on semidefinite programming to more accurately determine independent components, addressing previous issues with approximate solutions and manual setting of IC numbers. A new modified version called MSDP is presented to exclude redundant ICs and select true ICs. Experimental results demonstrate its superiority in improving the accuracy of ICs and automatically determining the number of ICs.
IEEE TRANSACTIONS ON CYBERNETICS
(2022)
Article
Geochemistry & Geophysics
Deise Santana Maia, Minh-Tan Pham, Erchan Aptoula, Florent Guiotte, Sebastien Lefevre
Summary: Morphological attribute profiles (APs) are a key method for spatial-spectral pixel analysis in remote sensing images. Over the past decade, research has focused on their application to a wider range of tasks, performance improvement, and extension to more complex Earth observation data.
IEEE GEOSCIENCE AND REMOTE SENSING MAGAZINE
(2021)
Article
Medicine, General & Internal
Farheen Siddiqui, Awwab Mohammad, M. Afshar Alam, Sameena Naaz, Parul Agarwal, Shahab Saquib Sohail, Dag Oivind Madsen
Summary: This paper presents a deep learning model for mental task classification using EEG signals. The model is able to classify mental tasks for patients with limited motor movements. By applying spatial filtering and principal component analysis to the EEG signals, task-specific features were extracted and achieved an accuracy of 77.62%.
Article
Engineering, Electrical & Electronic
Daniela Lupu, Ion Necoara, Joseph L. L. Garrett, Tor Arne Johansen
Summary: This paper presents a new stochastic algorithm called SHOICA, which approximates the non-convex loss functions of ICA using higher-order Taylor minorizers. It shows that SHOICA is faster and provides better feature extraction compared to the commonly used FastICA. Experimental results on real hyperspectral images validate the advantages of SHOICA in dimensionality reduction and classification tasks.
IEEE TRANSACTIONS ON COMPUTATIONAL IMAGING
(2022)
Article
Computer Science, Artificial Intelligence
Xiaofeng Wang, Leyan Shi, Jun Liu, Minglu Zhang
Summary: This paper introduces a novel cosine 2DPCA method to tackle the issues in two-dimensional principal component analysis. By applying the cosine objective function and a greedy iterative algorithm, this method demonstrates significant improvements in reconstruction, correlation, complexity, and classification.
IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS
(2023)
Article
Chemistry, Analytical
Nadia Abu Farha, Fares Al-Shargie, Usman Tariq, Hasan Al-Nashash
Summary: Vigilance level assessment is crucial in critical working environments, and the wavelet ICA (wICA) method has shown improvement in artifact reduction.
Article
Computer Science, Artificial Intelligence
Bo Peng, Ying Bi, Bing Xue, Mengjie Zhang, Shuting Wan
Summary: A new diagnosis approach named MFCGPE is proposed to automatically construct high-level features and build an effective ensemble for identifying different fault types. The approach achieves higher diagnostic accuracy than all compared methods on three bearing datasets with a small number of training samples.
IEEE COMPUTATIONAL INTELLIGENCE MAGAZINE
(2021)
Article
Engineering, Environmental
Hongqiu Zhu, Qiling Wang, Fengxue Zhang, Chunhua Yang, Yonggang Li
Summary: In this study, a prediction method of the removal rate of the electrocoagulation reactor based on deep learning LSTM network and ARIMA model is proposed. The method effectively predicts the removal rate by analyzing historical data and important factors, and its practicality is confirmed through industrial data verification.
PROCESS SAFETY AND ENVIRONMENTAL PROTECTION
(2021)
Article
Chemistry, Analytical
Teresa M. C. Pereira, Raquel C. Conceicao, Raquel Sebastiao
Summary: This study investigated the potential of electrocardiogram (ECG) as a biometric trait for identification and authentication. Using data from a public database, templates were generated and different algorithms were used for identification and authentication. The results suggest that ECG can be successfully used for personal recognition and authentication.
Article
Engineering, Biomedical
Yue Wang, Guanci Yang, Shaobo Li, Yang Li, Ling He, Dan Liu
Summary: This paper proposes an arrhythmia classification algorithm based on multi-head self-attention mechanism. It preprocesses the ECG signals using wavelet transform and designs a linear projection layer and position encoding-based spatiotemporal characterization method. Experimental results show that this algorithm achieves high classification accuracy.
BIOMEDICAL SIGNAL PROCESSING AND CONTROL
(2023)
Article
Computer Science, Artificial Intelligence
Zahoor Uddin, Muhammad Altaf, Ayaz Ahmad, Aamir Qamar, Farooq Alam Orakzai
Summary: This article presents a technique using independent vector analysis (IVA) for eliminating various artifacts in electrocardiogram (ECG) signals. The technique combines canonical correlation analysis (CCA) and independent component analysis (ICA) to effectively separate mixed data. The proposed technique is practical and outperforms CCA and ICA by minimizing changes to the ECG signals while removing artifacts.
PEERJ COMPUTER SCIENCE
(2023)
Article
Computer Science, Information Systems
Hao Zheng, Liyong Fu, Qiaolin Ye
Summary: Robust principal component analysis (PCA) has been proven effective in data reconstruction and recognition tasks. However, existing methods often suffer from performance and robustness issues. To address this, we propose a new method called flexible capped PCA (FCPCA) that uses capped L2,p-norm distance metric to minimize reconstruction errors. Experimental results demonstrate that FCPCA outperforms existing methods in terms of power and flexibility.
INFORMATION SCIENCES
(2022)
Article
Engineering, Electrical & Electronic
Vinod Kumar, Musheer Ahmad, Adesh Kumari, Saru Kumari, M. K. Khan
Summary: Vehicular cloud computing is a new area of research that equips vehicles with networking and sensing capacities to improve vehicle-to-vehicle or vehicle-to-infrastructure communication, which is beneficial for traffic management and road safety. This technology presents a smart structure for vehicular communication, maintaining autonomous traffic, improved vehicle control, and enhanced system security.
INTERNATIONAL JOURNAL OF COMMUNICATION SYSTEMS
(2021)
Article
Telecommunications
Vinod Kumar, Musheer Ahmad, Dheerendra Mishra, Saru Kumari, Muhammad Khurram Khan
VEHICULAR COMMUNICATIONS
(2020)
Article
Computer Science, Information Systems
Adesh Kumari, Srinivas Jangirala, M. Yahya Abbasi, Vinod Kumar, Mansaf Alam
JOURNAL OF INFORMATION SECURITY AND APPLICATIONS
(2020)
Article
Computer Science, Information Systems
Vinod Kumar, Rajendra Kumar, S. K. Pandey
Summary: The paper proposed a key distribution protocol based on the Chinese Remainder Theorem that significantly reduced computational complexity, increased security, improved efficiency in handling large groups, and effectively prevented illegal access.
MULTIMEDIA TOOLS AND APPLICATIONS
(2021)
Article
Engineering, Electrical & Electronic
Akber Ali Khan, Vinod Kumar, Musheer Ahmad, Saurabh Rana, Dheerendra Mishra
INTERNATIONAL JOURNAL OF ELECTRICAL POWER & ENERGY SYSTEMS
(2020)
Article
Computer Science, Information Systems
Vinod Kumar, Rajendra Kumar, S. K. Pandey
Summary: Smart Grid is a modernized power grid system that requires a secure and robust key management scheme to ensure confidentiality during communication in the Advanced Metering Infrastructure (AMI). This paper proposes a tailored key management scheme for Home Area Networks (HAN) in AMI, which significantly reduces rekeying overhead and enhances privacy protection.
PEER-TO-PEER NETWORKING AND APPLICATIONS
(2021)
Article
Computer Science, Hardware & Architecture
Akber Ali Khan, Vinod Kumar, Musheer Ahmad, Saurabh Rana
Summary: The development of IoT technology presents new challenges for smart grid systems. To enhance security and privacy in smart grids, lightweight authentication and key agreement protocols are proposed. Analyzing and verifying these new protocols can improve the efficiency of smart grid systems.
JOURNAL OF SYSTEMS ARCHITECTURE
(2021)
Article
Telecommunications
Akber Ali Khan, Vinod Kumar, Musheer Ahmad, B. B. Gupta, Musheer Ahmad, Ahmed A. Abd El-Latif
Summary: This article introduces the two-way communication of the Internet of Energy to address the security, efficiency, and reliability in energy reform. A mutual authenticated key agreement framework is proposed in the vehicle-grid system using elliptic curve cryptography and hash function to maintain secure communication with reliable computation and communication costs. The proposed protocol is shown to be secure through formal and informal methods, and is more efficient in terms of computation and communication costs compared to related protocols in the same environment.
TELECOMMUNICATION SYSTEMS
(2021)
Article
Computer Science, Hardware & Architecture
Vinod Kumar, Mahmoud Shuker Mahmoud, Ahmed Alkhayyat, Jangirala Srinivas, Musheer Ahmad, Adesh Kumari
Summary: This article presents a secure and lightweight authentication scheme (RAPCHI) for Internet of medical Things (IoMT) in cloud-healthcare infrastructure (CHI) during a pandemic. The proposed framework provides resistance against various security threats and is compared to existing frameworks, demonstrating its superiority in terms of security, computation, and communication. The formal security analysis and simulation tests validate RAPCHI's security against man-in-the-middle and replay attacks.
JOURNAL OF SUPERCOMPUTING
(2022)
Article
Chemistry, Analytical
Vikas Kumar, Rahul Kumar, Akber Ali Khan, Vinod Kumar, Yu-Chi Chen, Chin-Chieh Chang
Summary: The Internet of Things (IoT) is a future trend that connects physical things with the cyber world through the Internet. While IoT technology brings convenience to our daily lives, it also poses security and privacy risks, particularly in RFID-tag-connected devices. To address these concerns, we propose a robust authentication framework for IoT-based RFID infrastructure, which satisfies all security requirements and improves communication performance.
Article
Computer Science, Information Systems
Akber Ali Khan, Vinod Kumar, Musheer Ahmad
Summary: Smart grid adjusts power generation by supervising consumer behavior and promoting cultural heritage. Ensuring environmental security and privacy are key concerns in smart systems, with authentication protocols enabling secure communication.
JOURNAL OF KING SAUD UNIVERSITY-COMPUTER AND INFORMATION SCIENCES
(2022)
Article
Telecommunications
Akber Ali Khan, Vinod Kumar, Jangirala Srinivas, Saru Kumari, Mridul Kumar Gupta
Summary: This article presents a secure and efficient architecture for satellite network systems using elliptic curve cryptography and a hash function. The proposed protocol ensures secure communication and key agreement, and is resistant to various security threats. It also includes different security features and capabilities, and allows easy password updates. The random oracle model is used to demonstrate its security, and security verification is performed against man in the middle and replay attacks using AVISPA software. The proposed protocol is shown to have lower computation and transmission overhead compared to competing methods.
TELECOMMUNICATION SYSTEMS
(2022)
Article
Computer Science, Information Systems
Akber Ali Khan, Vinod Kumar, Musheer Ahmad, Srinivas Jangirala
Summary: This paper proposes a key agreement and authentication framework for vehicular to grid networks based on elliptic curve cryptography and the hash function, aiming to achieve secure and efficient communication in the Energy Internet.
JOURNAL OF INFORMATION SECURITY AND APPLICATIONS
(2022)
Article
Computer Science, Information Systems
Vinod Kumar, Ammar Mohammed Ali Al-Tameemi, Adesh Kumari, Musheer Ahmad, Mayadah Waheed Falah, Ahmed A. Abd El-Latif
Summary: The Vehicular Cloud Environment (VCE) is a new research field that combines cloud computing and vehicular networks to enable communication between vehicles and roadside infrastructure. A hybrid technical solution utilizing vehicle resources, cloud infrastructure, and IoT is necessary for effective vehicular communication networking. Security and privacy are significant challenges in VCE due to the integration of unknown vehicles and infrastructure via the public network. In this regard, we propose a provably secure authentication system for VCE employing smartphones. The proposed system demonstrates both security and efficiency in terms of communication and processing overhead.
Article
Computer Science, Information Systems
Samiulla Itoo, Akber Ali Khan, Vinod Kumar, Ahmed Alkhayyat, Musheer Ahmad, Jangirala Srinivas
Summary: In the traditional medical healthcare system, sharing patient data is challenging. Blockchain technology and cloud computing can address this issue and provide better privacy and security.