Bluetooth Low Energy (BLE) is emerging as an appealing solution for several Industrial Internet of Things (IIoT) applications that require low cost and low power consumption. In particular, in this context, some appli...
Bluetooth Low Energy (BLE) is emerging as an appealing solution for several Industrial Internet of Things (IIoT) applications that require low cost and low power consumption. In particular, in this context, some applications need to collect data from sensors distributed over a wide area and require predictable behavior and real-time guarantees for processing such data with bounded delays. For this reason, some works in the literature proposed approaches that enable BLE-based real-time mesh networks. However, these approaches require offline configurations, thus limiting the flexibility and adaptability of the network. This work introduces a solution to build real-time mesh networks by enabling BLE devices to find a suitable configuration at runtime.
The integration of machine learning (ML) into mobile applications presents unique challenges, particularly in resource-constrained environments such as iOS devices. Skin lesion classification is a critical task in der...
详细信息
Federated learning (FL) is an efficient, scalable, and privacy-preserving technology in which clients collaborate on machine learning or deep learning model training. However, malicious clients can send poisoned model...
Federated learning (FL) is an efficient, scalable, and privacy-preserving technology in which clients collaborate on machine learning or deep learning model training. However, malicious clients can send poisoned model updates to the central server without being identified, which makes FL vulnerable to backdoor attacks. In this work, we propose a novel defence approach, FLSec, to mitigate backdoor attacks caused by adversarial local model updates. FLSec utilizes an original measurement, GradScore, computed from the loss gradient norm of the final layer of the local models for backdoor defence. We show that GradScore is efficient and robust in identifying malicious model updates through analysis and experiments. Our extensive evaluation also demonstrates FLSec is highly effective in mitigating three state-of-the-art backdoor attacks on well-known datasets, MNIST, LOAN, and CIFAR-10. The accuracy on a benign dataset with the proposed defence approach is nearly unchanged, with the accuracy on the backdoor dataset being reduced to 0%. In addition, our experiments show that FLSec significantly outperforms existing backdoor defences in multi-round backdoor attacks.
Considering that optical intelligent reflecting surface (OIRS) can change the optical channel, the enhancement performance of OIRS on the energy efficiency (EE) of the non-orthogonal multiple access (NOMA)-based visib...
详细信息
In this paper, we introduce a novel energy harvester (EH) using rolling magnets, which can supply power to a wireless sensor system (WSS) for monitoring the status of rotating ship shafts. The EH consists of twelve co...
详细信息
We propose an augmented reality near-eye-display which can present varifocal virtual images with corresponding masks. The varifocal mask occludes the real scene behind the virtual images, enhancing their visibility. &...
详细信息
The oxide thin-film transistors (TFTs) with low processing temperature and high-performance metrics exhibit great potentials in advanced displays and flexible electronics. The demands of high integration density and h...
The oxide thin-film transistors (TFTs) with low processing temperature and high-performance metrics exhibit great potentials in advanced displays and flexible electronics. The demands of high integration density and high flexibility drive the investigations on the downscaling behavior and mechanical stability of oxide TFTs. This work investigates the channel length (L) downscaling of self-aligned top-gate (SATG) oxide TFTs and explores the mechanical stress instabilities of flexible double-gate (DG) TFTs. By optimizing 4-nm AlO x gate insulator (GI) and modifying the channel/GI interface, the high performance is successfully maintained on sub-500 nm oxide TFTs. Furthermore, the study reveals the impacts of the laser lift-off (LLO) process and mechanical stress on the flexible oxide TFT. The DG structure and fluorine plasma treatment can noticeably enhance the mechanical robustness of flexible oxide TFTs.
In this paper,considering the cost of base station,coverage,call quality,and other practical factors,a multi-objective optimal site planning scheme is ***,based on practical needs,mathematical modeling methods were us...
详细信息
In this paper,considering the cost of base station,coverage,call quality,and other practical factors,a multi-objective optimal site planning scheme is ***,based on practical needs,mathematical modeling methods were used to establish mathematical expressions for the three sub-objectives of cost objectives,coverage objectives,and quality ***,a multi-objective optimization model was established by combining threshold and traffic volume *** order to reduce the time complexity of optimization,a non-dominated sorting genetic algorithm(NSGA)is used to solve the multi-objective optimization problem of site ***,a strategy for clustering and optimizing weak coverage areas was *** order to avoid redundant neighborhood retrieval during cluster expansion,the Fast Density-Based Spatial Clustering of Applications with Noise(FDBSCAN)clustering method was *** different sub-objectives as the main objectives,this paper obtained the distribution map of weak coverage areas before and after the establishment of new base stations,as well as relevant site planning maps,and provided three planning schemes for different main *** simulation results show that the traffic coverage of the three station planning schemes is above 90%.The change in the main optimization objective will result in a significant difference between the cost of the three solutions and the coverage of weak coverage points.
Federated learning (FL) is widely used in edge-cloud collaborative training due to its distributed architecture and privacy-preserving properties without sharing local data. FLTrust, the most state-of-the-art FL defen...
详细信息
The rapid advancement of computer-generated holography has bridged deep learning with traditional optical principles in recent ***,a critical challenge in this evolution is the efficient and accurate conversion from t...
详细信息
The rapid advancement of computer-generated holography has bridged deep learning with traditional optical principles in recent ***,a critical challenge in this evolution is the efficient and accurate conversion from the amplitude to phase domain for high-quality phase-only hologram(POH)*** computational models often struggle to address the inherent complexities of optical phenomena,compromising the conversion *** this study,we present the cross-domain fusion network(CDFN),an architecture designed to tackle the complexities involved in POH *** CDFN employs a multi-stage(MS)mechanism to progressively learn the translation from amplitude to phase domain,complemented by the deep supervision(DS)strategy of middle features to enhance task-relevant feature learning from the initial ***,we propose an infinite phase mapper(IPM),a phase-mapping function that circumvents the limitations of conventional activation functions and encapsulates the physical essence of *** simulations,our proposed method successfully reconstructs high-quality 2K color images from the DIV2K dataset,achieving an average PSNR of 31.68 dB and SSIM of ***,we realize high-quality color image reconstruction in optical *** experimental results highlight the computational intelligence and optical fidelity achieved by our proposed physics-aware cross-domain fusion.
暂无评论