Nematic liquid crystal elements spatially structured in the azimuthal direction represent a special type diffraction grating capable of achieving a diffraction efficiency of up to 100%. This paper is devoted to the development and study of the spatial polarization properties of a liquid crystal diffraction polarizing twist element, in which a phase profile equivalent to a diffraction polarizing grating is realized by spatially modulating the topology of the director orientation using the method of polarization holographic recording of surface anisotropy in the azo dye AtA-2 layer. The efficient electrical control possibility of the diffraction and the structured element polarization properties, their ambivalence when changing the side of light input into the liquid crystal element are experimentally demonstrated. The optimal value of the external control voltage corresponding to the maximum diffraction efficiency above 90% for circularly polarized radiation is determined.
This paper presents a continuation of a study of taking into account the quantum-gravity corrections (qgcs) for primordial black holes (pbhs) and of the corresponding inferences in cosmology. The first part is devoted to application of the above-mentioned qgcs in the de Sitter epoch to solution of the Dark Energy Problem. In the process the Dark Energy is understood as the vacuum energy the density of which is identical to the cosmological constant within constant factors. The explicit formula for the transformation of this density due to qgcs for pbhs with the Schwarzschild-de Sitter metric in the de Sitter space has been derived. The case of a dynamic cosmological term in this formalism has been considered. The second part of the work is devoted to studies of qgcs for pbhs during the radiation-dominant era. At first it has been demonstrated that the inclusion of these corrections shifts all the parameters in accordance with the formulae applicable to the de Sitter epoch. Then the accretion and evaporation processes in this pattern are studied for pbhs. It has been shown that, compared with a semiclassical consideration, in this case the accretion rate is lowering, whereas the evaporation rate is growing, i.e., in this pattern the evaporation processes are intensified.
In this article, we propose a modified model for the spread of SIR disease. The model is based on an assumption of the existence of two relatively separate groups, which are called bubbles. In the model we use, it is assumed that the rates of disease spread vary in both groups under consideration and between the groups too. A numerical experiment has been conducted for the proposed model.
In this work, we explore the possibility of chimera states in the locally coupled q-deformed Lozi maps. The deformation is introduced using the Jaganathan-Sinha q-deformation scheme, which modifies the local map while smoothly recovering the standard Lozi map in the limit q -> 1. We observe that the q-deformed map exhibits significant modifications in its dynamical behavior compared to the standard case. Through analytical considerations and numerical simulations, we analyze fixed points, stability, bifurcation structure, and routes to chaos as functions of the deformation parameter. Chimera states are represented by stable coexisting synchronous and asynchronous domains in a lattice of coupled identical oscillators promoted by non-local or global coupling. An analogous 'spatial intermittency' is also observed in coupled map lattices. This work demonstrates that chimera states are not restricted to oscillatory systems or standard maps, but can also emerge in discrete-time systems with rationally deformed dynamics under purely local interactions. This indicates that chimera states are a robust and generic feature of a wide class of non-linear systems.
This article looks at modern cybersecurity methods in educational institutions, using a university with a developed IT infrastructure as an example. The relevance of this work is determined by the growing number of cyberattacks on educational institutions' networks and the need to implement smart information protection systems. A number of typical threats are considered, including unauthorized access, malware, and phishing, as well as the characteristics of the university infrastructure that affect the risk of cybersecurity incidents. The object of this study is deep learning methods applied to the analysis of information security events and the detection of malicious activity. Approaches to normalizing operating system and application event logs are proposed to ensure data unification, as well as a method for vectorizing events by highlighting key features to prepare high-quality input data for deep learning algorithms. The proposed architecture of the information protection system based on explainable deep learning includes modules for collecting and normalizing events, vectorizing them, analyzing them using neural networks, and visualizing the results.
