Article
Engineering, Electrical & Electronic
Ronghui Zhang, Chunxiao Jiang, Sheng Wu, Quan Zhou, Xiaojun Jing, Junsheng Mu
Summary: Gesture recognition is crucial for human-computer interaction, and gesture-based user identification can enhance system security. Recently, Wi-Fi-integrated sensing and communication technology has shown potential in gesture recognition. In this study, we propose a system called WiGesID, using Wi-Fi sensing to achieve joint gesture recognition and human identification.
IEEE JOURNAL ON SELECTED AREAS IN COMMUNICATIONS
(2022)
Article
Engineering, Electrical & Electronic
Yingxiang Sun, Haoqiu Xiong, Danny Kai Pin Tan, Tony Xiao Han, Rui Du, Xun Yang, Terry Tao Ye
Summary: This study introduces a dual-frequency continuous wave radar that can achieve both localization and activity/gesture recognition simultaneously. Features of different movements are classified by the lightweight network AGRNet, and data corresponding to walking are used for moving target localization by comparing phase differences in the Doppler domain between dual frequencies. Additionally, a segmentation method is proposed to effectively extract individual time-periods corresponding to different motions from continuous signals. Experimental results demonstrate classification accuracy over 91% and centimeter-level localization accuracy.
IEEE SENSORS JOURNAL
(2021)
Article
Engineering, Aerospace
Hongbo Sun, Lek Guan Chia, Sirajudeen Gulam Razul
Summary: This article demonstrates the effectiveness of passive through-wall human sensing technique using opportunistic WiFi signals, showing better detection sensitivity and capability to sense various human motions inside a room, expanding the potential of WiFi passive radar for through-wall sensing applications.
IEEE TRANSACTIONS ON AEROSPACE AND ELECTRONIC SYSTEMS
(2021)
Article
Computer Science, Information Systems
Anum Ali, Priyabrata Parida, Vutha Va, Saifeng Ni, Khuong Nhat Nguyen, Boon Loong Ng, Jianzhong Charlie Zhang
Summary: This paper presents a study on gesture recognition using millimeter-wave radar sensors. The authors propose a set of 6 micro-gestures and an end-to-end solution based on machine learning, including an activity detection module and a gesture classifier. Evaluation on data collected from 11 users shows that the proposed solution achieves an end-to-end accuracy of 95%.
Article
Computer Science, Information Systems
Kang Ling, Haipeng Dai, Yuntang Liu, Alex X. Liu, Wei Wang, Qing Gu
Summary: With the rise of AR/VR technology and miniaturization of mobile devices, gesture recognition is becoming popular in the research area of human-computer interaction. This paper presents UltraGesture, an ultrasonic finger motion perception and recognition system that can run on mobile devices without any hardware modification. Experimental results demonstrate that UltraGesture achieves high accuracy in recognizing various gestures.
IEEE TRANSACTIONS ON MOBILE COMPUTING
(2022)
Article
Engineering, Electrical & Electronic
Xin Tong, Zhaoyang Zhang, Zhaohui Yang
Summary: Integrated sensing and communication (ISAC) has emerged as a research hotspot for future wireless communication systems to achieve environment sensing within the wireless communication framework. This article proposes both centralized and distributed architectures for multi-view sensing in wireless communications, and identifies key performance indicators for each architecture. Various signal processing and machine learning methods are discussed for realizing multi-view sensing, with detailed classifications and comparisons. Simulation results confirm the effectiveness of the proposed architectures and schemes.
IEEE COMMUNICATIONS MAGAZINE
(2023)
Article
Engineering, Electrical & Electronic
Ali Safa, Andre Bourdoux, Ilja Ocket, Francky Catthoor, Georges G. E. Gielen
Summary: Radar processing using spiking neural networks (SNNs) is a solution for ultralow-power wireless human-computer interactions. SNNs are significantly more energy-efficient compared to traditional deep learning methods, with successful application in radar gesture recognition, and providing evaluation code for future research.
IEEE MICROWAVE AND WIRELESS COMPONENTS LETTERS
(2022)
Article
Engineering, Electrical & Electronic
Xianglong Zeng, Chaoyang Wu, Wen-Bin Ye
Summary: In recent years, radar-based dynamic hand gesture recognition systems have been widely used, researchers typically use machine learning algorithms, especially deep learning algorithms for high precision gesture recognition. However, most deep learning models only accept predefined gestures as input, not allowing user-defined gestures. This study proposes a neural network model trained with meta-learning method to handle few-shot classification tasks and enable user-definable DHGR.
IEEE SENSORS JOURNAL
(2021)
Article
Computer Science, Artificial Intelligence
Zhiwen Yu, Dong Zhang, Zhu Wang, Qi Han, Bin Guo, Qi Wang
Summary: In recent years, intelligent activity recognition systems based on radio frequency signals have been developed rapidly. This study proposes a multitarget gesture recognition system based on Doppler radar and demonstrates its high accuracy through experiments.
IEEE TRANSACTIONS ON HUMAN-MACHINE SYSTEMS
(2022)
Article
Computer Science, Information Systems
Jie Zhang, Yang Li, Haoyi Xiong, Dejing Dou, Chunyan Miao, Daqing Zhang
Summary: Recent advances in wireless sensing techniques have enabled gesture recognition using WiFi devices. Existing methods fail to work well when dealing with changes in locations, orientations, and sizes. This study proposes a new approach to accurately recognize handwriting gestures using WiFi devices and achieves superior performance compared to state-of-the-art methods.
IEEE INTERNET OF THINGS JOURNAL
(2022)
Article
Geochemistry & Geophysics
Yong Wang, Yuhong Shu, Xiuqian Jia, Mu Zhou, Liangbo Xie, Lei Guo
Summary: This letter presents a method using FMCW radar for short-range hand gesture sensing and recognition. The range, Doppler, and angle parameters of hand gestures are measured using FFT and MUSIC algorithms. The proposed CMFF-HGR method extracts features and achieves hand gesture recognition.
IEEE GEOSCIENCE AND REMOTE SENSING LETTERS
(2022)
Article
Computer Science, Information Systems
Zhixiong Yang, Xu Liu, Zijian Li, Bo Yuan, Yajun Zhang
Summary: Gesture recognition through wireless signals is a novel aspect of human perception. While RFID has gained popularity for its advantages, there are unresolved complexities in RFID sensing research.
Article
Chemistry, Analytical
Michael C. Brown, Changzhi Li
Summary: This paper discusses the challenges and potential solutions of incorporating digital modulation into radar systems. It proposes a reconfigurable phase-modulated continuous wave (PMCW) radar system for vital sign detection and gesture recognition. Experimental results show that this radar system has high flexibility and accuracy under different modulation schemes.
Article
Engineering, Electrical & Electronic
Sevgi Z. Gurbuz, Ali Cafer Gurbuz, Evie A. Malaia, Darrin J. Griffin, Chris S. Crawford, Mohammad Mahbubur Rahman, Emre Kurtoglu, Ridvan Aksu, Trevor Macks, Robiulhossain Mdrafi
Summary: The article proposes the use of RF sensors for remote recognition of ASL, benefiting the Deaf community in human-computer interactions. By studying the linguistic properties of RF ASL data, it is possible to differentiate between sign language and daily activities, as well as design smart environments. Feature-level fusion of RF sensor network data can achieve high accuracy in ASL sign classification.
IEEE SENSORS JOURNAL
(2021)
Article
Computer Science, Information Systems
Rui Min, Xing Wang, Jie Zou, Jing Gao, Liying Wang, Zongjie Cao
Summary: This article presents an early gesture recognition method that achieves almost the same accuracy as complete sequence recognition while improving recognition speed. Experimental results show high recognition rates on the Google Project Soli dataset and potential application to other radar sensors.
IEEE INTERNET OF THINGS JOURNAL
(2021)
Article
Computer Science, Information Systems
En Xu, Zhiwen Yu, Bin Guo, Helei Cui
Summary: CTR is a crucial index in online advertising systems, and accurately predicting it can enhance ad recommendation accuracy. Extracting users' interests is vital for CTR prediction tasks, and the proposed CIN model outperforms existing solutions in handling long sequences. The CIN model first extracts core interests of users and utilizes refined data for subsequent learning tasks.
ACM TRANSACTIONS ON KNOWLEDGE DISCOVERY FROM DATA
(2021)
Article
Computer Science, Information Systems
Yunji Liang, Huihui Li, Bin Guo, Zhiwen Yu, Xiaolong Zheng, Sagar Samtani, Daniel D. Zeng
Summary: Traditional text classification methods rely on labor-intensive feature engineering and are not suitable for rapidly evolving domains. In contrast, the novel SVA-CNN method leverages multi-view representation learning and attention mechanisms to automatically extract and weight multiple granularities and fine-grained text representations, outperforming existing deep learning text classification methods in both performance and training time.
INFORMATION SCIENCES
(2021)
Article
Computer Science, Artificial Intelligence
Bin Guo, Hao Wang, Yasan Ding, Wei Wu, Shaoyang Hao, Yueqi Sun, Zhiwen Yu
Summary: With the development of deep learning technology, text-generation technology has made great progress and provided a variety of services for human beings. Researchers have recently focused on more anthropomorphic text-generation technology, especially conditional text generation, including emotional text generation and personalized text generation. Therefore, conditional text generation has become a research hotspot and attracted widespread attention.
ACM TRANSACTIONS ON INTELLIGENT SYSTEMS AND TECHNOLOGY
(2021)
Article
Computer Science, Hardware & Architecture
Zumin Wang, Jiyu Tian, Hui Fang, Liming Chen, Jing Qin
Summary: This paper proposes a lightweight log anomaly detection algorithm, LightLog, which achieves real-time processing and accurate anomaly detection on edge devices. Experimental results demonstrate that LightLog outperforms several benchmarking methods.
Article
Computer Science, Artificial Intelligence
Wenxi Wang, Huansheng Ning, Feifei Shi, Sahraoui Dhelim, Weishan Zhang, Liming Chen
Summary: This article introduces the new developments in theoretical research and practical applications of social computing, particularly under the influence of artificial intelligence. By combining human intelligence and AI, H-AI shows significant advantages in dealing with social problems.
IEEE TRANSACTIONS ON HUMAN-MACHINE SYSTEMS
(2022)
Article
Engineering, Electrical & Electronic
Jinqiang Wang, Tao Zhu, Jingyuan Gan, Liming Luke Chen, Huansheng Ning, Yaping Wan
Summary: In this study, a novel sensor data augmentation method is proposed by resampling, which introduces variable domain information and simulates realistic activity data, improving the performance of contrastive learning for sensor-based human activity recognition (HAR).
IEEE SENSORS JOURNAL
(2022)
Article
Computer Science, Information Systems
Sahraoui Dhelim, Nyothiri Aung, Mohand Tahar Kechadi, Huansheng Ning, Liming Chen, Abderrahmane Lakas
Summary: Trust Management System (TMS) is crucial in IoT networks to ensure network security, data integrity, and promote legitimate devices while punishing malicious activities. Trust scores assigned by TMSs reflect devices' reputations, which help predict future behaviors and assess reliability in IoT networks. This article proposes Trust2Vec, a TMS for large-scale IoT systems that leverages a random-walk network exploration algorithm and network embeddings community detection algorithm to manage trust relationships and mitigate large-scale trust attacks by malicious devices.
IEEE INTERNET OF THINGS JOURNAL
(2023)
Article
Computer Science, Information Systems
Jinqiang Wang, Tao Zhu, Liming Luke Chen, Huansheng Ning, Yaping Wan
Summary: Contrastive learning is a promising self-supervised learning paradigm for sensor-based human activity recognition. However, it may lead to overclustering. To address this, we propose ClusterCLHAR, a new contrastive learning framework that selects negatives through clustering in HAR. Experimental results demonstrate that ClusterCLHAR outperforms state-of-the-art methods in self-supervised and semi-supervised learning for HAR.
IEEE INTERNET OF THINGS JOURNAL
(2023)
Article
Computer Science, Information Systems
Zhu Wang, Zilong Wang, Xiaona Li, Zhiwen Yu, Bin Guo, Liming Chen, Xingshe Zhou
Summary: It is important for a recommender system to utilize both explicit and implicit information. The knowledge graph has been widely used to capture auxiliary information, but current models have limitations in information propagation. To address this, we propose a Multi-dimension Interaction based attentional Knowledge Graph Neural Network (MI-KGNN) for enhanced KG-aware recommendation. MI-KGNN optimizes node representation updates by exploring multi-dimension interactions among nodes and introduces a dual attention mechanism to capture and represent structural and semantic information in the knowledge graph. Experimental results show significant improvements compared to existing methods.
IEEE TRANSACTIONS ON BIG DATA
(2023)
Article
Computer Science, Theory & Methods
Sahraoui Dhelim, Liming Chen, Sajal K. Das, Huansheng Ning, Chris Nugent, Gerard Leavey, Dirk Pesch, Eleanor Bantry-White, Devin Burns
Summary: This article surveys the literature on social media analysis for detecting mental distress, with a focus on studies published since the COVID-19 outbreak. The authors propose new approaches to organizing and classifying the large amount of research in this emerging field, providing fresh insights and knowledge for interested communities. The article also discusses future research directions and niche areas in detecting mental health problems using social media data, as well as the technical, privacy, and ethical challenges in this rapidly growing field.
ACM COMPUTING SURVEYS
(2023)
Article
Computer Science, Information Systems
Zhiwen Yu, Minling Dang, Qilong Wu, Liming Chen, Yujin Xie, Yu Wang, Bin Guo
Summary: This research focuses on the predictability of human mobility, proposes a model and method for quantifying predictability, and extends the method to consider external factors. Experimental results show that the method has a tighter upper bound on predictability compared to previous work and predictability slightly increases when considering external factors.
ACM TRANSACTIONS ON KNOWLEDGE DISCOVERY FROM DATA
(2023)
Article
Engineering, Civil
Nyothiri Aung, Sahraoui Dhelim, Liming Chen, Abderrahmane Lakas, Wenyin Zhang, Huansheng Ning, Souleyman Chaib, Mohand Tahar Kechadi
Summary: This paper proposes a social-aware vehicular edge computing architecture that utilizes vehicles as edge servers to deliver popular content to nearby users, addressing the challenges of content delivery in vehicular networks.
IEEE TRANSACTIONS ON INTELLIGENT TRANSPORTATION SYSTEMS
(2023)
Article
Computer Science, Information Systems
Samson Kahsay Gebresilassie, Joseph Rafferty, Liming Chen, Zhan Cui, Mamun Abu-Tair
Summary: This article discusses the definition and importance of the Internet of Things (IoT), as well as the security challenges faced by IoT networks and the limitations of traditional intrusion detection systems. The author designed an intrusion detection system based on transfer learning (TL) and convolutional neural network (CNN), generating their own dataset through real-time data collection and parsing, and evaluated its performance using multiple metrics, with results showing higher accuracy and lower false positive rate.
Article
Engineering, Electrical & Electronic
Furong Duan, Tao Zhu, Jinqiang Wang, Liming Chen, Huansheng Ning, Yaping Wan
Summary: This article introduces deep learning methods for sensor-based human activity recognition (HAR). The segmentation of sensor data stream is a critical task in HAR. However, the current independent preprocessing methods with fixed-size windows have led to two problems: the multiclass window problem and fluctuation of prediction results. To address these issues, a novel multitask DL approach is proposed in this article, including a multiscale window method, a boundary offset prediction algorithm, and a multitask framework. Experimental results show that this approach outperforms current state-of-the-art HAR methods.
IEEE TRANSACTIONS ON INSTRUMENTATION AND MEASUREMENT
(2023)
