Search

To prevent implicit privacy disclosure in sharing gradients among data owners (DOs) under federated learning (FL), differential privacy (DP) and its variants have become a common practice to offer formal privacy guarantees with low overheads. However, individual DOs generally tend to inject larger DP noises for stronger privacy provisions (which entails severe degradation of model utility), while the curator (i.e., aggregation server) aims to minimize the overall effect of added random noises for satisfactory model performance. To address this conflicting goal, we propose a novel dynamic privacy pricing (DyPP) game which allows DOs to sell individual privacy (by lowering the scale of locally added DP noise) for differentiated economic compensations (offered by the curator), thereby enhancing FL model utility. Considering multi-dimensional information asymmetry among players (e.g., DO's data distribution and privacy preference, and curator's maximum affordable payment) as well as their varying private information in distinct FL tasks, it is hard to directly attain the Nash equilibrium of the mixed-strategy DyPP game. Alternatively, we devise a fast reinforcement learning algorithm with two layers to quickly learn the optimal mixed noise-saving strategy of DOs and the optimal mixed pricing strategy of the curator without prior knowledge of players' private information. Experiments on real datasets validate the feasibility and effectiveness of the proposed scheme in terms of faster convergence speed and enhanced FL model utility with lower payment costs., Comment: Accepted by IEEE ICC2023

In this article, physical layer security (PLS) in an intelligent reflecting surface (IRS) assisted multiple-input multiple-output multiple antenna eavesdropper (MIMOME) system is studied. In particular, we consider a practical scenario without instantaneous channel state information (CSI) of the eavesdropper and assume that the eavesdropping channel is a Rayleigh channel. To reduce the complexity of currently available IRS-assisted PLS schemes, we propose a low-complexity deep learning (DL) based approach to design transmitter beamforming and IRS jointly, where the precoding vector and phase shift matrix are designed to minimize the secrecy outage probability. Simulation results demonstrate that the proposed DL-based approach can achieve a similar performance of that with conventional alternating optimization (AO) algorithms for a significant reduction in the computational complexity.

This paper investigates physical layer security (PLS) in the intelligent reflecting surface (IRS)-assisted multiple-user uplink channel. Since the instantaneous eavesdropper's channel state information (CSI) is unavailable, the secrecy rate can not be measured. In this case, existing investigations usually focus on the maximization of the minimum (max-min) of signal to interference plus noise power ratio (SINRs) among multiple users, and do not consider secrecy outage probability caused by eavesdroppers. In this paper, we first formulate the minimization of the maximum (min-max) secrecy outage probability among multiple users. The formulated problem is solved by alternately optimizing receiving matrix and phase shift matrix. Simulations demonstrate that the maximum secrecy outage probability is significantly reduced with the proposed algorithm compared to max-min SINR strategies, meaning our scheme has a higher security performance.

This paper investigates an endogenous security architecture for computation offloading in the Internet of Things (IoT), where the blockchain technology enables the traceability of malicious behaviors, and the task data uploading link from sensors to small base station (SBS) is protected by intelligent reflecting surface (IRS)-assisted physical layer security (PLS). After receiving task data, the SBS allocates computational resources to help sensors perform the task. The existing computation offloading schemes usually focus on network performance improvement, such as energy consumption minimization, and neglect the Gas fee paid by sensors, resulting in the discontent of high Gas payers. Here, we design a Gas-oriented computation offloading scheme that guarantees the degree of satisfaction of sensors, while aiming to reduce energy consumption. Also, we deduce the ergodic secrecy rate of IRS-assisted PLS transmission that can represent the global secrecy performance to allocate computational resources. The simulations show that the proposed scheme ensures that the node paying higher Gas gets stronger computational resources, and just raises $4\%$ energy consumption in comparison with energy consumption minimization schemes.

Measurements of sub-3 nm particles are technically challenging and characterized with significant uncertainties. This study selected two types of widely used commercial instruments: an Airmodus® Particle Size Magnifier (PSM) and a TSI® diethylene glycol-based Scanning Mobility Particle Sizer (DEG-SMPS), and examined factors that govern their abilities on measurements of sub-3 nm particles. Monodisperse standard particles used in the experiments include positively and negatively charged tungsten oxide particles and nickel-chromium (Ni/Cr) oxide particles, positive and negative ions produced in a soft X-ray neutralizer, and positively charged organic particles including tetraheptylammonium bromide (THABr), tetrabutylammonium bromide (TBABr), and polyethylene glycol (PEG) particles. Our results show that both PSM and DEG-SMPS are subject to the influence of particle composition. Negatively charged tungsten oxide particles were detected at higher efficiencies than positively charged ones, whereas opposite results were observed for Ni/Cr particles and background ions produced in the soft X-ray neutralizer. In addition, the concentration accuracy of PSM and DEG-SMPS are explored after calibration using Ni/Cr particles. The inverted total concentrations by DEG-SMPS are usually underestimated because of the overestimation of penetration efficiency through different modules, whereas the inverted total concentrations by PSM can be either overestimated or underestimated, when the properties of analyte particles are significantly different from the calibration particles. A better understanding on the effects of particle chemical composition and charging state, and their diffusion losses in the instruments will help to diminish the uncertainties in the measurements.

Introduction: Fatal familial insomnia (FFI) is a rare clinical case. The study was mainly to report the clinical symptoms and imaging and genetic characteristics of a FFI case with depression, with relevant literature summarized. Patient concerns: A male, aged 57 years old, with mental disorders and progressive memory decline one year before admission. Diagnosis: Clinical manifestations: he had obvious abnormal mental behavior, rapidly progressing dementia symptoms, stubborn insomnia, abnormal movements and laryngeal stridor after falling asleep at night. Imaging and genetic test results: the cranial magnetic resonance imaging showed frontal temporal lobe atrophy; the polysomnography results showed no effective sleep; the 14-3-3 test result of cerebrospinal fluid was negative; the prion protein (PRNP) test showed that the D178N gene locus had mutations. And the patient was finally diagnosed as FFI. Interventions: There were no obvious effects in the treatment using medicines such as Risperidone, Olanzapine, Alprazolam, Clonazepam, and Deanxit. Outcomes: Mobility dysfunction of the patient was further aggravated. He was no longer able to move around on his own, and there were serious mental disorders. Conclusion: PRNP examination is of guiding significance for the diagnosis of the FFI of depression. Hence, it is very necessary to perform PRNP examination in clinical diagnosis of FFI of depression.

This paper mainly studies a multi-input single-output (MISO) system assisted by artificial noise (AN), and provides a system level implementation based on software-defined radio (SDR) for the AN-assisted MISO communications. Firstly, the paper studies the power allocation between the AN and the useful signals in a null-space-based AN scheme, and deduces the secrecy rate of this AN scheme. Then, we establish the system model of the AN-assisted MISO system under different modulation modes. Finally, a verification platform of the AN-assisted MISO system is built using nanoBEE software radio platform. The experimental results are in good agreement with the theoretical analysis and MATLAB simulations. The results also show that under the optimal power allocation strategy, the AN can effectively improve the secrecy rate of the system, but the secrecy rate is not affected by modulation modes. They just affect the bit error rate (BER) performance of the system.

In our study, solid alkali were produced and characterized by a series of methods. Dissolution of corn stalk was achieved in solvent systems at 170 °C for 5 h by cooking with solid alkali. The components of ionic liquid and water in solvent systems showed synergistic effects on lignocellulosic dissolution and degradation. Alkaline molecular sieves were proven to be the most effective solid alkali for lignin degradation, which promoted the dissolution efficiency of the corn stalk in solvent systems. Under the combined action of solid alkali and the nucleophilic component in solvent systems, lignin and hemicellulose in corn stalk were degraded. It effectively enhanced cellulose solubilization in solvent systems. The regenerated cellulose was demonstrated to be similar to microcrystalline cellulose according to Fourier Transform infrared spectroscopy and Nuclear Magnetic Resonance analyses. X-ray diffraction curves showed that the crystallinity of the regenerated cellulose is lower than that of microcrystalline cellulose.

Telecentric and cryogenic optical system is a good solution to space-based IR sensor with wide field, sensitivity and determination of accurate orientation. A refractive telecentric optical system was designed with squared field of view in 8 degrees, entrance pupil of 265 millimeters and operating at 200 K. The aperture of the lens was 280 millimeters. By multilevel spring holders,the opto-mechanism got a available tradeoff between enough stiffness at room temperature and stress-free at 200 K, accordingly prevents thermal stress on lens and heat leakage from room temperature supports to cryogenic lens. After lenses alignment, vibration test and image test at operating temperature were performed. The ensquared energy reached 75% on axis and 72% off-axis respectively.

We deal with anti-periodic problems for differential inclusions with nonmonotone perturbations. The main tools in our study are the maximal monotone property of the derivative operator with anti-periodic conditions and the theory of pseudomonotone perturbations of maximal monotone mappings. We then apply our results to evolution hemivariational inequalities and parabolic equations with nonmonotone discontinuities, which generalize and extend previously known theorems.

This paper deals with the initial-value problem of nonlinear evolution inclusions of the form dB(u)/dt+A(u) ∈ f, v0 ∈ B(u)(0), where the operator B is induced by a subgradient and A is pseudomonotone. Existence theorem is established via the time discretization technique and the regularization method. In contrast to the previous results, here we impose a weaker coerciveness condition on A and remove the strong monotonicity from B.

In order to investigate the influence of plant roots on the silty clays slope with low liquid limit in North China area, a case study of a slope of The Beijing-Taipei Expressway in Langfang City, screened two kinds of plants, Caragana microphylla Lam. and Astragalus adsurgens Pall., planted in the silty clays slope with low liquid limit, they are endurable to drought and barren resistance. Made tests of the mechanical strength of root-soil composites an pure soil, atter enalyzed and evaluated the influence of two kinds of plants on slope by ANSYS, a finite element software. The results showed that the the roots of Caragana microphylla Lam. and Astragalus adsurgens Pall. could improve the stability of slope, and Caragana microphylla Lam. boasts more advantage than Astragalus adsurgens Pall. in terms of slope protection., and the toe of the slopes were danger zone, therefore, dealing with the toe of slope in projects. The results of this test have great significance to the ecological protection technology of the silty clays slope with low liquid limit slope and ecological management of mining area, such as, plant selection and configuration.

Keep composting matrix in continuous collision and friction under a relatively high-temperature can significantly accelerate the progress of composting. A bioreactor was designed according to the novel process. Using this technology, organic fertilizer could be produced within 96h. The electric conductivity (EC) and pH value reached to a stable value of 2.35mS/cm and 7.7 after 96h of fermentation. The total carbon/total nitrogen (TC/TN) and dissolved carbon/dissolved nitrogen (DC/DN) ratio was decrease from 27.3 and 36.2 to 17.4 and 7.6 respectively. In contrast, it needed 24days to achieve the similar result in traditional static composting (TSC). Compost particles with different size were analyzed to explore the rapid degradation mechanism of food waste. The evidence of anaerobic fermentation was firstly discovered in aerobic composting.