Article
Automation & Control Systems
Xuejun Ding, Tsan-Ming Choi, Yong Tian
Summary: This article introduces a new device-free localization algorithm which utilizes the change in received signal strength of wireless links to locate the target. It achieves high localization accuracy and low computational complexity through techniques such as hierarchic radio imaging, grid normalized model, and square human model.
IEEE TRANSACTIONS ON SYSTEMS MAN CYBERNETICS-SYSTEMS
(2022)
Article
Automation & Control Systems
Lei Shi, Wei Xing Zheng, Qingchen Liu, Yang Liu, Jinliang Shao
Summary: This article proposes a dual privacy-preserving scheme for accurately determining the locations of sensor nodes in wireless sensor networks. The localization algorithm is analyzed in terms of convergence and privacy-preserving performance, and it is demonstrated that accurate localization can be achieved without leaking sensor node privacy information. The performance of the privacy-preserving localization algorithm is verified through experiments with six robots.
IEEE TRANSACTIONS ON INDUSTRIAL ELECTRONICS
(2023)
Article
Computer Science, Information Systems
Jie Zhang, Yanjiao Li, Wendong Xiao
Summary: This article proposes a modified hierarchical framework for WiFi-based device-free localization (DFL), which improves the localization performance in cluttered environments through data separation, feature mapping, and probability distribution embedding.
IEEE INTERNET OF THINGS JOURNAL
(2022)
Article
Engineering, Electrical & Electronic
Martin Schmidhammer, Christian Gentner, Stephan Sand, Uwe-Carsten Fiebig
Summary: State-of-the-art device-free localization systems are enhanced by exploiting multipath propagation between network nodes, particularly indoors where wireless channels have multipath components due to reflection and scattering. Geometrically derived propagation paths of these components, when considered as additional links in the network, improve localization performance and increase coverage area significantly based on theoretical performance bounds on the localization error.
IEEE ANTENNAS AND WIRELESS PROPAGATION LETTERS
(2021)
Article
Engineering, Electrical & Electronic
Naciye Nur Arslan, Durmus Ozdemir, Hasan Temurtas
Summary: Electrocardiography is crucial for early diagnosis and treatment of heart diseases. This study proposes a method that simultaneously trains an autoencoder and a classifier for ECG heartbeat classification. Testing on the MIT-BIH dataset demonstrates that this approach achieves a classification accuracy of 99.99%.
SIGNAL IMAGE AND VIDEO PROCESSING
(2023)
Article
Computer Science, Information Systems
Lingjun Zhao, Huakun Huang, Chunhua Su, Shuxue Ding, Huawei Huang, Zhiyuan Tan, Zhenni Li
Summary: Device-free localization (DFL) is an emerging technology under the Internet-of-Things architecture with applications in intrusion detection, mobile robot localization, and location-based services. Current DFL-related machine learning algorithms face challenges of low localization accuracy and weak dependability. To address these issues, a dependable block-sparse scheme named block-sparse coding with the proximal operator (BSCPO) is proposed in this work, showing improved robustness and accuracy in noisy conditions compared to state-of-the-art DFL methods.
IEEE INTERNET OF THINGS JOURNAL
(2021)
Review
Computer Science, Interdisciplinary Applications
Pranjal Kumar, Siddhartha Chauhan, Lalit Kumar Awasthi
Summary: This paper conducts a comprehensive survey and analysis of the application of deep learning in human activity recognition. The focus is on the key contributions of deep learning and the description of various databases and performance metrics used in HAR methodologies. The paper explores the potential uses of HAR in domains such as healthcare, emotion calculation, assisted living, security, and education.
ARCHIVES OF COMPUTATIONAL METHODS IN ENGINEERING
(2023)
Article
Green & Sustainable Science & Technology
V. Kavitha, Kirupa Ganapathy
Summary: Nodes in a Wireless Sensor Network (WSN) monitor and collect data in different environments, and can send data through the network using routers to determine the most efficient way of transmission. To ensure system security, an energy-efficient routing algorithm in a convolution neural network can be used, along with an optimized convolute network for predicting malicious nodes. This approach improves network security and data transmission efficiency.
SUSTAINABLE ENERGY TECHNOLOGIES AND ASSESSMENTS
(2022)
Article
Engineering, Electrical & Electronic
Yongtao Ma, Wanru Ning, Bobo Wang, Xiuyan Liang
Summary: DFL is essential in device-free applications, where AugRF, a CNN-DAE based architecture, improves localization performance without retraining. The proposed method has been validated through simulations and real-world experiments, showing superiority and meeting real-time localization requirements.
IEEE TRANSACTIONS ON INSTRUMENTATION AND MEASUREMENT
(2021)
Article
Engineering, Electrical & Electronic
Rui Zhou, Huanhuan Hou, Ziyuan Gong, Zuona Chen, Kai Tang, Bao Zhou
Summary: Device-free localization based on WiFi CSI fingerprinting is promising, but the instability of CSI fingerprints due to environmental changes poses a significant challenge in adapting the localization model. The proposed AdapLoc method, utilizing 1D-CNN and DA with SA, achieves effective adaptation of the localization model in dynamic environments with reduced recalibration efforts, outperforming existing work in both localization and adaptation aspects.
IEEE SENSORS JOURNAL
(2021)
Article
Automation & Control Systems
Kawsar Ali, Daniel J. Rogers
Summary: Deploying energy harvesting-based wireless sensor nodes in challenging environments often means lack of control over node placement and orientation. This article introduces a smart WSN that operates independently of placement and orientation, sharing energy and information among cube faces to maximize energy harvesting and signal transmission.
IEEE TRANSACTIONS ON INDUSTRIAL ELECTRONICS
(2021)
Article
Engineering, Electrical & Electronic
Rivo Randriatsiferana, Frederic Alicalapa, Richard Lorion, Lala Rajaoarisoa, Blaise Ravelo, Christophe Moy
Summary: This article presents an accurate energy model for wireless sensor devices based on power consumption measurement, and validates its effectiveness through experiments, highlighting the importance of data transfer in WSD energy consumption.
IEEE SENSORS JOURNAL
(2022)
Article
Environmental Sciences
Qiaoqiao Sun, Xuefeng Liu, Salah Bourennane
Summary: A novel unsupervised multi-level feature extraction framework based on a three-dimensional convolutional autoencoder is proposed in this paper to improve hyperspectral classification. The framework allows for spectral-spatial information to be mined simultaneously and can be trained without labeled samples, providing more efficient feature extraction compared to using multiple networks.
Article
Chemistry, Analytical
Jonas Ninnemann, Paul Schwarzbach, Andrea Jung, Oliver Michler
Summary: This paper discusses passive localization using wireless communication signals, determining object positions by measuring signal delays. A CIR environmental mapping method is proposed, allowing for more accurate object localization without the need for device participation in the process.
Article
Automation & Control Systems
Sang Su Lee, Dhong Hun Lee, Dong Kyu Lee, Choon Ki Ahn
Summary: This article presents a new mobile robot localization algorithm that overcomes the performance degradation problem caused by linearization errors. The algorithm, based on an objective function minimization approach, shows superior accurate, robust, real-time performance in real mobile robot localization experiments.
IEEE-ASME TRANSACTIONS ON MECHATRONICS
(2022)
Article
Engineering, Electrical & Electronic
Huakun Huang, Lingjun Zhao, Huawei Huang, Song Guo
IEEE COMMUNICATIONS MAGAZINE
(2020)
Article
Computer Science, Hardware & Architecture
Lingjun Zhao, Chunhua Su, Zeyang Dai, Huakun Huang, Shuxue Ding, Xinyi Huang, Zhaoyang Han
JOURNAL OF SUPERCOMPUTING
(2020)
Article
Computer Science, Information Systems
Huakun Huang, Shuxue Ding, Lingjun Zhao, Huawei Huang, Liang Chen, Honghao Gao, Syed Hassan Ahmed
IEEE INTERNET OF THINGS JOURNAL
(2020)
Article
Computer Science, Information Systems
Lingjun Zhao, Huakun Huang, Chunhua Su, Shuxue Ding, Huawei Huang, Zhiyuan Tan, Zhenni Li
Summary: Device-free localization (DFL) is an emerging technology under the Internet-of-Things architecture with applications in intrusion detection, mobile robot localization, and location-based services. Current DFL-related machine learning algorithms face challenges of low localization accuracy and weak dependability. To address these issues, a dependable block-sparse scheme named block-sparse coding with the proximal operator (BSCPO) is proposed in this work, showing improved robustness and accuracy in noisy conditions compared to state-of-the-art DFL methods.
IEEE INTERNET OF THINGS JOURNAL
(2021)
Article
Computer Science, Artificial Intelligence
Haoli Zhao, Peng Zhong, Haiqin Chen, Zhenni Li, Wuhui Chen, Zibin Zheng
Summary: This paper proposes an efficient group non-convex sparsity regularized partially shared dictionary learning method for multi-view learning. The method utilizes the partially shared dictionary learning model to extract both consistency and complementarity from multi-view data and employs generalized group non-convex sparsity for discriminative and sparse representations. Experimental results validate the effectiveness of both group information and non-convexity, and demonstrate that appropriate coefficient sharing ratios can improve clustering performances. The proposed algorithm outperforms compared algorithms in terms of convergence performance and has reasonable running time costs.
KNOWLEDGE-BASED SYSTEMS
(2022)
Article
Automation & Control Systems
Zhenni Li, Haoli Zhao, Yongcheng Guo, Zuyuan Yang, Shengli Xie
Summary: The article introduces a new CTL framework with a log regularizer for accurate representations and strong sparsity. By employing PDCA algorithm and extrapolation technology, the algorithm is accelerated for fast and efficient CTL learning. The experimental results show stable convergence with lower approximation error and faster speed compared to existing CTL algorithms.
IEEE TRANSACTIONS ON CYBERNETICS
(2022)
Article
Automation & Control Systems
Haoli Zhao, Jiqiang Wu, Zhenni Li, Wuhui Chen, Zibin Zheng
Summary: This article proposes a double sparse DRL algorithm using multilayer sparse coding and nonconvex regularized pruning to obtain deep sparse representations for control in reinforcement learning. Experimental results demonstrate its superior performance in control compared to existing sparse-coding-based DRL methods, with significant rewards increase and parameter reduction.
IEEE TRANSACTIONS ON CYBERNETICS
(2023)
Article
Computer Science, Artificial Intelligence
Haoli Zhao, Zhenni Li, Wensheng Su, Shengli Xie
Summary: This paper proposes a Value Estimation Network (VEN) model based on Dynamic Sparse Coding (DSC) to address the issues of interference and redundant parameters in Deep Reinforcement Learning (DRL). The proposed algorithm achieves higher control performances in both discrete-action and continuous-action environments compared to existing benchmark DRL algorithms.
Article
Computer Science, Artificial Intelligence
Haoli Zhao, Zhenni Li, Wuhui Chen, Zibin Zheng, Shengli Xie
Summary: This article proposes an efficient multiview dictionary learning algorithm for multiview clustering. The algorithm uses a partially shared DL model to excavate both consistency and complementarity in the multiview data. Experimental results demonstrate that the proposed algorithm can effectively recover the synthetic dictionary and perform well in multiview clustering.
IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS
(2023)
Article
Computer Science, Artificial Intelligence
Zhenni Li, Zuyuan Yang, Haoli Zhao, Shengli Xie
Summary: In this article, a novel direct-optimization-based dictionary learning algorithm is proposed using the minimax concave penalty as a sparsity regularizer. The algorithm can enforce strong sparsity and obtain accurate estimation. The nonconvex MCP is decomposed into two convex components and processed using convex functions algorithm and nonconvex proximal-splitting algorithm. The proposed algorithm can be applied to a broader class of dictionary learning problems and has proven convergence guarantee.
IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS
(2023)
Article
Green & Sustainable Science & Technology
Huakun Huang, Dingrong Dai, Longtao Guo, Sihui Xue, Huijun Wu
Summary: Reducing carbon emissions from buildings is crucial, but the building sector faces challenges such as low accuracy in forecasting energy consumption and lacking effective measurement methods. However, studies show that AI and big data technologies can significantly improve the accuracy of building energy consumption prediction, enabling effective building operation management for emission reduction goals.
Article
Computer Science, Interdisciplinary Applications
Huakun Huang, Sihui Xue, Lingjun Zhao, Dingrong Dai, Weijia Wang, Huijun Wu, Zhou Cao
Summary: This article proposes a smart energy management system that utilizes intelligent edge clients and distributed electric vehicles to improve energy utilization and maximize the use of renewable energy. The system treats a virtual power plant as an energy storage facility and efficiently manages battery energy.
ENGINEERING REPORTS
(2023)
Article
Computer Science, Information Systems
Haiping Huang, Lingjun Zhao, Yisheng Wu
Summary: This paper introduces a novel DHM simulation approach that combines the Kinect motion capture system, IoT devices, and advanced machine learning techniques. The approach captures real-time human motion data and utilizes deep learning for time-series modeling. Virtual reality technology is also employed for motion simulation and optimization. Experimental evaluations demonstrate that the approach outperforms conventional methods in terms of real-time simulation capabilities. This research pioneers an innovative technological paradigm for DHM simulation, opening up new avenues for research and applications in related domains.
COMPUTER COMMUNICATIONS
(2023)
Proceedings Paper
Computer Science, Artificial Intelligence
Ying Xie, Zhenni Li, Haoli Zhao
Summary: This paper introduces a gradient-free neural network training method by using deep dictionary learning and logarithm function as sparse regularizer for feature extraction in network training. Proximal block coordinate descent method and log-thresholding operator are employed for optimizing non-convex and nonsmooth subproblems.
PATTERN RECOGNITION AND COMPUTER VISION, PT IV
(2021)
Proceedings Paper
Computer Science, Artificial Intelligence
Ling-jun Zhao, Man-Wai Mak
Summary: In this work, the challenges of recognizing speakers from a distance using far-field microphones are addressed by enhancing the frame-level processing and feature aggregation of x-vector networks through the use of Res2Net blocks, squeeze-and-excitation units, and a channel-dependent attention mechanism. The proposed CE-Res2Net architecture shows a relative improvement of about 16% in EER and 17% in minDCF on the VOiCES 2019 Challenge's evaluation set, demonstrating the effectiveness of the approach.
2021 12TH INTERNATIONAL SYMPOSIUM ON CHINESE SPOKEN LANGUAGE PROCESSING (ISCSLP)
(2021)