Search for: All records

Biometric authentication systems are increasingly needed across a broad range of applications including in smart city environments (e.g., entering hotels), and in smart home environments (e.g., controlling smart devices). Traditional methods, such as face-based and fingerprint-based authentication, usually incur high costs to be installed in all this kind of environments. In this paper, we develop a ubiquitous low-effort user authentication approach, mmPalm, based on palm recognition using millimeter wave (mmWave) signals. mmWave technology has been adopted by WiGig and 5G, making mmPalm a low-cost solution that can be widely adopted in public places. In addition, the high resolution of mmWave signals allows mmPalm to extract detailed palm characteristics (e.g., palm geometry, skin thickness, and texture) that can assemble distinctive palmprints for user authentication. Our innovative virtual antennas design further increases the spatial resolution of a commercial mmWave device, enabling it to fully capture the comprehensive palmprint features. Moreover, to address the challenge of small-scale environmental changes (e.g., variations in palm-device distances and palm orientations), we design a novel palm profile augmentation method, utilizing a Conditional Generative Adversarial Network (cGAN) to generate synthetic palm profiles for mitigating palm instability. Furthermore, we design a cross-environment adaptation framework based on transfer learning to address the challenge of large-scale environmental changes, including multipath variations introduced by human bodies and nearby furniture. Extensive experiments with 30 participants through 6 months demonstrate that mmPalm achieves 99% authentication accuracy with resilience against different types of attacks, including random, impersonation, and counterfeit. 

Abstract Multivariate failure time data are frequently analyzed using the marginal proportional hazards models and the frailty models. When the sample size is extraordinarily large, using either approach could face computational challenges. In this paper, we focus on the marginal model approach and propose a divide‐and‐combine method to analyze large‐scale multivariate failure time data. Our method is motivated by the Myocardial Infarction Data Acquisition System (MIDAS), a New Jersey statewide database that includes 73,725,160 admissions to nonfederal hospitals and emergency rooms (ERs) from 1995 to 2017. We propose to randomly divide the full data into multiple subsets and propose a weighted method to combine these estimators obtained from individual subsets using three weights. Under mild conditions, we show that the combined estimator is asymptotically equivalent to the estimator obtained from the full data as if the data were analyzed all at once. In addition, to screen out risk factors with weak signals, we propose to perform the regularized estimation on the combined estimator using its combined confidence distribution. Theoretical properties, such as consistency, oracle properties, and asymptotic equivalence between the divide‐and‐combine approach and the full data approach are studied. Performance of the proposed method is investigated using simulation studies. Our method is applied to the MIDAS data to identify risk factors related to multivariate cardiovascular‐related health outcomes.