arXiv subjects

Yongsheng Zhu

Publications and source records attributed to Yongsheng Zhu.

6 recordsLinked to original sources

arXiv2025

A two-dimensional terahertz smart wristband for integrated sensing and communication

The development of wearable devices for terahertz (THz) integrated sensing and communication (ISAC) is pivotal for forthcoming 6G Internet of Things (IoT) and wearable optoelectronics. However, existing THz system suffers from bulkiness, narrow spectral response and limited flexibility constrained by their dependence on external antennas, complex coupling architectures and rigid components. Here, we present a 2D THz smart wristband based on a graphene plasmon polariton atomic cavity (PPAC) array, which integrates sensing and communication within a monolithic microdetector. Operating without any external antenna, the compact and flexible device enables self-powered, polarization-sensitive and frequency-selective THz detection across a broad response spectrum from 0.25 to 4.24 THz, with a responsivity of 6 V/W, a response time of 62 ms, and mechanical robustness maintained over 2000 bending cycles. Notably, we further exploit its multi-parameter THz responses for dual-purpose ISAC functionality. For sensing, the polarization- and strain-dependent THz responses are utilized as high-dimensional features for a convolutional neural network (CNN), enabling circuit fault diagnosis with 97% accuracy. For communication, the device implements secure encrypted communication under simulated on-body wearing condition through dual-channel encoding of THz polarization and on-off signals. This 2D ISAC platform paves the way for miniaturized, intelligent wearable systems for advanced human-machine interaction.

physics.optics↗

arXiv2025

Coherent Absorption Synergizes with Plasmon-Enhanced Graphene Terahertz Photo-thermoelectric Response

Terahertz (THz) technology shows great potential in 6G communications and imaging, but faces challenges related to detector sensitivity, noise, and cryogenic operation. Here, we integrate interferometric enhancement of absorption (IEA) from a metal reflection layer with a graphene plasmon polariton atomic cavity (PPAC)-based photodetector. The hybrid configuration enhances the in-plane electric field and improves the plasmon-induced thermal gradient. Numerical simulations and photoresponse measurements were employed to systematically investigate the influence of a metal reflective layer on the photothermoelectric behavior of the device, which reveals the IEA design significantly boosts the THz absorption rate in graphene nanostructures and promotes asymmetry in the lateral diffusion of hot carriers. Compared with the bare device, the responsivity of the device is enhanced by approximately 30-folds, while maintaining a response time below 130 microseconds. We further demonstrate the potential of the device to distinguish concealed liquids, advancing high-responsivity, room-temperature, and compact terahertz imaging technology.

cond-mat.mes-hall↗

arXiv2025

Deep Learning Empowered Sub-Diffraction Terahertz Backpropagation Single-Pixel Imaging

Terahertz single-pixel imaging (THz SPI) has garnered widespread attention for its potential to overcome challenges associated with THz focal plane arrays. However, the inherently long wavelength of THz waves limits imaging resolution, while achieving subwavelength resolution requires harsh experimental conditions and time-consuming processes. Here, we propose a sub-diffraction THz backpropagation SPI technique. We illuminate the object with continuous-wave 0.36-THz radiation (λ0 = 833.3 μm). The transmitted THz wave is modulated by prearranged patterns generated on a 500-μm-thick silicon wafer and subsequently recorded by a far-field single-pixel detector. An untrained neural network constrained with the physical SPI process iteratively reconstructs the THz images with an ultralow sampling ratio of 1.5625%, significantly reducing the long sampling times. To further suppress the THz diffraction-field effects, a backpropagation SPI from near field to far field is implemented by integrating with a THz physical propagation model into the output layer of the network. Notably, using the thick wafer where THz evanescent field cannot be fully recorded, we achieve a spatial resolution of 118 μm (~λ0/7) through backpropagation SPI, thus eliminating the need for ultrathin photomodulators. This approach provides an efficient solution for advancing THz microscopic imaging and addressing other inverse imaging challenges.

eess.IV↗

arXiv2024

Lurking in the shadows: Unveiling Stealthy Backdoor Attacks against Personalized Federated Learning

Federated Learning (FL) is a collaborative machine learning technique where multiple clients work together with a central server to train a global model without sharing their private data. However, the distribution shift across non-IID datasets of clients poses a challenge to this one-model-fits-all method hindering the ability of the global model to effectively adapt to each client's unique local data. To echo this challenge, personalized FL (PFL) is designed to allow each client to create personalized local models tailored to their private data. While extensive research has scrutinized backdoor risks in FL, it has remained underexplored in PFL applications. In this study, we delve deep into the vulnerabilities of PFL to backdoor attacks. Our analysis showcases a tale of two cities. On the one hand, the personalization process in PFL can dilute the backdoor poisoning effects injected into the personalized local models. Furthermore, PFL systems can also deploy both server-end and client-end defense mechanisms to strengthen the barrier against backdoor attacks. On the other hand, our study shows that PFL fortified with these defense methods may offer a false sense of security. We propose \textit{PFedBA}, a stealthy and effective backdoor attack strategy applicable to PFL systems. \textit{PFedBA} ingeniously aligns the backdoor learning task with the main learning task of PFL by optimizing the trigger generation process. Our comprehensive experiments demonstrate the effectiveness of \textit{PFedBA} in seamlessly embedding triggers into personalized local models. \textit{PFedBA} yields outstanding attack performance across 10 state-of-the-art PFL algorithms, defeating the existing 6 defense mechanisms. Our study sheds light on the subtle yet potent backdoor threats to PFL systems, urging the community to bolster defenses against emerging backdoor challenges.

cs.LG↗

arXiv2015

Combining upper limits with a Bayesian approach

We discuss how to determine and combine upper limits based on observed events and estimated backgrounds with a Bayesian method, when insignificant signals are observed in independent measurements. In addition to some general features deduced from the analytical formulae, systematic numerical results are obtained by a C$++$ program (CULBA) for low-count experiments, which can be used as a reference to combine two upper limits.

physics.data-an↗

arXiv2006

On the Statistical Significance

A definition for the statistical significance by constructing a correlation between the normal distribution integral probability and the p-value observed in an experiment is proposed, which is suitable for both counting experiment and continuous test statistics.

physics.data-an↗