Search

Profiled side-channel analysis (SCA) leverages leakage from cryptographic implementations to extract the secret key. When combined with advanced methods in neural networks (NNs), profiled SCA can successfully attack even those cryptocores assumed to be protected against SCA. Despite the rise in the number of studies devoted to NN-based SCA, a range of questions has remained unanswered, namely: how to choose an NN with an adequate configuration, how to tune the NN’s hyperparameters, when to stop the training, etc. Our proposed approach, “InfoNEAT,” tackles these issues in a natural way. InfoNEAT relies on the concept of neural structure search, enhanced by information-theoretic metrics to guide the evolution, halt it with novel stopping criteria, and improve time-complexity and memory footprint. The performance of InfoNEAT is evaluated by applying it to publicly available datasets composed of real side-channel measurements. In addition to the considerable advantages regarding the automated configuration of NNs, InfoNEAT demonstrates significant improvements over other approaches for effective key recovery in terms of the number of epochs (e.g.,x6 faster) and the number of attack traces compared to both MLPs and CNNs (e.g., up to 1000s fewer traces to break a device) as well as a reduction in the number of trainable parameters compared to MLPs (e.g., by the factor of up to 32). Furthermore, through experiments, it is demonstrated that InfoNEAT’s models are robust against noise and desynchronization in traces.

A study was conducted on thermal stresses in an inwardly solidifying sphere. A solution was presented which accounts for thermal expansion associated with temperature gradients and volume changes linked with phase transition.The frozen medium was modeled as an elastic-perfectly plastic. Parametric studies using appropriate material properties and cryopreservation protocols indicated that strains associated with phase transition led to higher stresses than those associated with thermal expansion.

Thermal expansion data are essential for analyses of cryodestruction associated with thermal stresses during cryopreservation protocols as well as during cryosurgery. The present study tests a commonly used hypothesis that the thermal expansion of frozen tissues is similar to that of pure water ice crystals. This study further provides insight into the potential effect of the presence of cryoprotectants on thermal expansion. A new apparatus for thermal strain measurements of frozen biological tissues within a cryogenic temperature range is presented. Results are presented for fresh tissue samples taken from beef muscle, chicken muscle, rabbit muscle, rabbit bone, and pig liver. Pilot studies of the effect of cryoprotectants on thermal expansion are further presented for rabbit muscle immersed in dimethyl sulphoxide (2 mols/l) and glycerol (2 mols/l), and for pig liver perfused with dimethyl sulphoxide (2 mols/l). Thermal expansion of frozen soft biological tissues was found to be similar to that of water ice crystals in the absence of cryoprotectant. Thermal expansion of the rabbit bone was found to be about one half of that of frozen soft tissues. A significant reduction in the thermal expansion at higher temperatures was observed in the presence of cryoprotectants. A rapid change of thermal strain near -100 [degrees] C was also observed, which is likely to be associated with the glass transition process of the cryoprotectant solutions.

A multidimensional, finite difference numerical scheme for the freezing process of biological tissues during cryosurgery is presented, which is a modification of an earlier numerical solution for inanimate materials. The tissues are treated as nonideal materials, freezing over a temperature range and possessing temperature-dependent thermophysical properties, blood perfusion, and metabolic heat generation. The numerical scheme is based on the application of an effective specific heat, substituting the intrinsic property, to include the latent heat effect within the phase transition temperature range. Results of the numerical solution were verified against an existing exact solution of a one-dimensional inverse Stefan problem in Cartesian coordinates. Results were further validated against experimental data available from the literature. The utility of the numerical solution for the design and application of cryodevices is demonstrated by parametric studies of the freezing processes around spherical and cylindrical cryoprobes. The parameters studied are the cryoprobe cooling power and the dimensions of the frozen region. Results are calculated for typical thermophysical properties of soft biological tissues, for angioma and for water.

A new, combined solution of the one-dimensional inverse Stefan problem in biological tissues is presented. The tissue is assumed to be a nonideal material in which phase transition occurs over a temperature range. The solution includes the thermal effects of blood perfusion and metabolic heat generation. The analysis combines a heat balance integral solution in the frozen region and a numerical enthalpy-based solution approach in the unfrozen region. The subregion of phase transition is included in the unfrozen region. Thermal effects of blood perfusion and metabolic heat generation are assumed to be temperature dependent and present in the unfrozen region only. An arbitrary initial condition is assumed that renders the solution useful for cryosurgical applications employing repeated freezing/thawing cycles. Very good agreement is obtained between the combined and an exact solution of a similar problem with constant thermophysical properties and a uniform initial condition. The solution indicated that blood perfusion does not appreciably affect either the shape of the temperature forcing function on the cryoprobe or the location and depth of penetration of the freezing front in peripheral tissues. It does, however, have a major influence on the freezing/thawing cycle duration, which is most pronounced during the thawing stage. The cooling rate imposed at the freezing front also has a major inverse effect on the duration of the freezing/thawing.

An investigation of a ground thermal energy storage system, which includes storage units containing phase-change materials (PCM), is presented. This study is related to a large-diameter helical heat exchanger, which is placed vertically in the ground. The PCM storage units under consideration have a cylindrical shell shape and are located inside and/or outside the helix. A modified numerical scheme for the solution of heat transfer in the ground, in the PCM units, and within the heat exchanger pipe, is presented. The theoretical results show that the thermal diffusivity of the PCM dominates the thermal performance of the system. Incorporation of PCM storage units containing paraffin wax results in a reduction of the thermal efficiency in comparison with a system not containing these units. However, incorporation of PCM having the same thermal diffusivity as of the soil results in a significant improvement of the thermal performance.

A new integrated collector storage (ICS) concept for low-temperature solar heating of water is described. The solar energy is stored in a salt-hydrate phase-change material (PCM) held in the collector and is discharged to cold water flowing through a surface heat exchanger located in a layer of stationary heat transfer liquid (SHTL), floating over an immiscible layer of PCM. A theoretical model for the charging process of the proposed integrated collector is presented. The model assumes one-dimensional transient heat conduction in the PCM and SHTL layers and neglects the effect of convection heat transfer in these regions. The model was solved numerically by an enthalpy-based finite differences method and validated against experimental data. The results of parametric studies on the effect of the transition temperature and of the thickness layer of the salt-hydrate PCM on the thermal performance of the charging process are also presented.

We study the effects of asymmetric bending and twist rigidities and of spontaneous curvature and twist, on the statistical mechanics of fluctuating ribbons. Using a combination of Monte Carlo and differential geometry methods we perform computer simulations and calculate the probability density of the end-to-end distance of a ribbon. We find that for rectilinear ribbons of asymmetric cross section and for spontaneously curved rods with circular cross section, the distribution of end-to-end distance (but not its mean square) is affected by twist rigidity and by spontaneous twist. Possible relevance of these effects to the physics of DNA is discussed. [ABSTRACT FROM AUTHOR]

We derive the nonlinear equations governing vacancy-controlled interdiffusion in a two-component noninteracting lattice gas. We observe that when the diffusion constants of the two species are not equal, the problem cannot be reduced to a simple linear diffusion. In order to treat the strongly nonlinear limit of large kinetic asymmetry and small concentration of vacancies, we introduce an adiabatic approximation in which the ‘‘fast’’ species is in an equilibrium state defined by the instantaneous distribution of the ‘‘slow’’ species, which obeys a linear diffusion equation with a renormalized diffusion coefficient. Comparison with numerical solutions of the nonlinear equations shows that the adiabatic approximation captures the essential physics of the diffusion process. [ABSTRACT FROM AUTHOR]

We investigate the velocity disturbance produced by point-like sources of force in viscous fluids subjected to unbounded, simple shear, and elongational flows of arbitrary strength. The response to the body force is given by the dynamic Oseen tensor which is constructed using the propagator for the linearized Navier–Stokes equation. Application of the method of characteristics leads to ‘‘Kelvin mode’’-type solutions for the velocity disturbances and explicit analytical expressions are obtained for the propagators and Oseen tensors in the above mentioned flows. [ABSTRACT FROM AUTHOR]

The elongational-flow-induced deformation of a polymer segment of arbitrary length and location along the contour of a macromolecule is calculated neglecting excluded-volume and nondraining effects. The resulting expression is valid for arbitrary flow strain rates in both steady-state and transient regimes of extension. For strain rates approaching a critical value at which a steady-state solution ceases to exist, the center of the macromolecule becomes much more stretched than its ends and it is argued that as a result of this nonuniform elongation, the stretching cannot be characterized by a single time scale corresponding to the uniform deformation of the polymer. The emerging picture of the process agrees with earlier conjectures made by the author and with Ryskin’s ‘‘yo-yo’’ model of polymer stretching in elongational flow. [ABSTRACT FROM AUTHOR]

We study solvent dynamics in a solution consisting of polymers and solvent. Our starting point is a system of Langevin equations where the solvent velocity field and solvent velocity fluctuations are explicitly included. We calculate the Fourier transform of the solvent velocity autocorrelation function in the dilute regime. From this correlation function we calculate the dynamic viscosity of the solution and obtain an effective equation of motion of the solvent which turns out to be the generalized Maxwell model of a linear viscoelastic fluid. [ABSTRACT FROM AUTHOR]

A simple rotational isomeric-type model of polymers in elongational flows is developed and used to analyze the coil stretching and chain retraction as a function of polymer and flow parameters. The results are in agreement with available experimental data on dilute polymer solutions in strong elongational flows. [ABSTRACT FROM AUTHOR]

We have developed an apparatus which enables us to perform accurate measurements of the shear viscosity of low ionic strength, dilute polyelectrolyte solutions, down to polymer concentrations below one part per million. We have shown that a theoretical expression for the viscosity of such solutions can be derived using the mode–mode coupling approximation to the hydrodynamics of charged Brownian spheres. Very good agreement between the predicted and observed polymer and salt concentration and molecular weight dependence of the viscosity is observed in the low-added salt, dilute solution range. Furthermore, it appears that the theory gives a qualitatively correct description of the viscosity of semidilute solutions, indicating that independent of polyion concentration, the hydrodynamics of low ionic strength polyelectrolyte solutions is dominated by electrostatic repulsion between polyions. [ABSTRACT FROM AUTHOR]

We review experimental results on the degradation of stretched polymer molecules in strong extensional flow fields. Atactic polystyrene and polyethylene oxide both show closely central scission along the backbone of the chain. We present a theoretical approach based upon a modified thermally activated barrier to scission model. This successfully describes the scission process as a function of molecular weight and stiffness and strain rate. [ABSTRACT FROM AUTHOR]