Deep Learning Models for Continuous Authentication on Mobile Devices

Häftad, Engelska, 2027

1 970 kr

Kommande

Beskrivning

Sensor-based continuous authentication has emerged as a critical approach for strengthening mobile security, enabling persistent user verification without disrupting device usage. However, the field faces significant hurdles, including limited training data, complex feature representation, environmental noise, and the strict resource constraints of mobile hardware.

Deep Learning Models for Continuous Authentication on Mobile Devices provides a unified and structured treatment of data-driven continuous authentication, presenting a systematic study of sensor-based continuous authentication on mobile devices, focusing on modern machine learning and deep learning techniques. It guides readers in designing, analyzing, and deploying reliable systems that effectively balance security, robustness, and computational efficiency. Featuring data augmentation strategies for data scarcity, multi-sensor feature fusion, discriminative feature learning via two-stream CNNs, data synthesis using conditional Wasserstein GANs, lightweight networks for efficient deployment, neural architecture search for automated optimization, and neuromorphic computing with spiking neural networks,

Deep Learning Models for Continuous Authentication on Mobile Devices balances methodological rigor with practical system design, offering robust solutions for real-world mobile security.

Introduces representative sensor-based continuous authentication methods on mobile devices, spanning data augmentation, feature fusion, convolutional and generative models, automated architecture search, and neuromorphic learning, offering comprehensive guidance for students and researchers
Presents practical strategies to address critical challenges in the field, including limited training data, inter-user behavioral variability, robustness to environmental noise and mimic behaviors, and the requirements for efficient deployment on mobile platforms
Includes systematic experimental analysis and implementation insights derived from both public and real-world datasets, helping practitioners understand the performance of continuous authentication methods in practical scenarios and design their own effective security solutions

Produktinformation

Utgivningsdatum:2027-01-01
Mått:191 x 235 x undefined mm
Format:Häftad
Språk:Engelska
Antal sidor:400
Förlag:Elsevier Science
ISBN:9780443494154

Utforska kategorier

Grafik och bildbehandling inom Data och IT
Systemvetenskap och AI inom Data och IT
Artificiell intelligens inom Data och IT

Mer om författaren

Yantao Li received the Ph.D. degree in computer science and technology from Chongqing University, Chongqing, China, in December 2012. He is currently a tenure-track Assistant Professor with the College of Computer Science, Chongqing University, Chongqing, China. He received the Best Paper Award from IEEE Internet Computing in 2022. He was a recipient of the Outstanding Ph.D. Thesis Award, Chongqing, in 2014, and the Outstanding Master's Thesis Award, in 2011. His research interests include mobile computing and security, the Internet of Things, sensor networks, and ubiquitous computing. Prof. Li currently serves as an Associate Editor for the IEEE Internet of Things Journal (IoT-J). His main research interests include machine learning, networked control systems, and decentralized algorithm. He has published more than 40 research papers.Qingguo Lü is a Graduate Research Assistant at Southwest University, Chongqing, China, where he is currently pursuing his Ph.D. degree in Computational Intelligence and Information Processing. His research interests include privacy protection of networked systems, Distributed Optimization, Neurodynamics, and Smart Grids. Huafeng Qin received the B.Sc. degree from the School of Mathematics and Physics and the M.Eng. degree from the College of Electronic and Automation, Chongqing University of Technology, China, and the Ph.D. degree from the College of Opto-Electronic Engineering, Chongqing University, China. He was a visiting student at Nanyang Technological University, Singapore, for 12 months, and subsequently a postdoctoral researcher for two years at Université Paris-Saclay, France. He is currently a Professor with the National Research Base of Intelligent Manufacturing Service, Chongqing Technology and Business University, China. His research interests include biometrics (e.g., vein, face, and gait recognition) and machine learning. Hailong Hu received his Ph.D. degree from the University of Luxembourg, and both his Master and Bachelor degrees from Southwest University. He is currently an Assistant Professor at Chongqing Technology and Business University. His research interests include trustworthy AI and biometrics. His work has been published in leading international journals and conferences such as TIFS, TAI, TOSN, and IoT-J. He has received two Best Paper Honorable Mention Awards (ACSAC 2021 and IEEE NAS 2018). He also serves as a reviewer for journals and conferences such as TIFS, TDSC, TOPS, PR, and CCBR.

Innehållsförteckning

1. SensorAuth: Data Augmentation for Smartphone Authentication2. FusionAuth: Feature Fusion Strategies for Mobile Authentication3. SCANet: Two-Stream CNNs for Multimodal Behavioral Biometrics4. CAGANet: GAN-Enhanced CNN Models for Robust Authentication5. DeFFusion: Deep Feature Fusion with Convolutional Networks6. SearchAuth: Neural Architecture Search for Authentication Model7. ADFFDA: Adaptive Deep Feature Fusion with Augmented Data8. SNNAuth: Spiking Neural Networks for Efficient Authentication