An interpolated flux scheme for cellular automaton device simulation

We present BTE-Barna (Boltzmann Transport Equation - Beyond the RTA for NAnosystems), a software package that extends the Monte Carlo (MC) module of the almaBTE solver of the Peierls-Boltzmann transport equation for phonons (PBTE) to work with nanosystems based on 2D materials with complex geometries. To properly capture how the phonon occupations evolve in momentum space as a result of scattering, we have supplemented the relaxation-time approximation with an implementation of the propagator for the full linearized version of the PBTE. The code can now find solutions for finite and extended devices under the effect of a thermal gradient, with isothermal reservoirs or with an arbitrary initial temperature distribution in space and time, writing out the temperature and heat flux distributions as well as their spectral decompositions. Besides the full deviational MC solver, a number of useful approximations for highly symmetric devices are also included. Program summaryProgram Title::BTE-BarnaCPC Library link to program files:https://doi.org/10.17632/t58z289tpc.1Developer's repository link:https://github.com/sousaw/BTE-BarnaCode Ocean capsule:https://codeocean.com/capsule/4171806Licensing provisions: Apache-2.0Programming language: C++Nature of problem: Calculation of temperature profiles, thermal flux and/or effective thermal conductivities for nanosystems based on 2D materials.Solution method: For highly symmetric systems confined along some direction(s), the linearized phonon Boltzmann transport equation is solved iteratively by partially suppressing the phonon lifetimes due to boundaries. For more complex geometries, an energy-based deviational Monte Carlo method including off-diagonal terms in the linearized scattering operator arising from the phonon-phonon interaction is used.Additional comments including restrictions and unusual features: Depends on the following external libraries: boost, Eigen, HDF5, MessagePack, MPI, oneTBB, spglib, and almaBTE.

An improved cellular automaton(CA) method is proposed in which an interpolated flux concept is introduced to suppress a well-known artifitial diffusion problem. The new method allows larger mesh sizes both in real and momentum space without losing numerical accuracy. Consequently, it becomes a practical modeling tool of non-linear carrier transport in semiconductor devices. In its application to Si MOSFETs, obtained results are extremely stable beyond the advanced weighted Monte Carlo method. Furthermore, making the best use of the stability, gate currents are studied in detail. 2 peaks of gate currents on gate bias, so-called drain avalanche hot carrier(DAHC) and channel hot electron(CHE), are well explained by thermionic emission and Fowler Nordheim(FN) tunneling of hot carriers respectively.

In typical particle simulations applied to device problems, it is desirable to simulate regions having widely different carrier concentration with similar resolution of the energy distribution function. Standard Monte Carlo (MC) algorithms use variance reduction techniques to limit the statistical error in regions of low carrier density. Here, we describe an alternative approach, based on a local iterative MC algorithm, where the carrier distribution is calculated by an iterative application of short MC steps for test particles that represent the local density. The information obtained from the local MC procedure is used to compute directly the evolution of the energy distribution function from an initial distribution until a steady state is reached. This local iterative procedure treats the charge density, represented by the MC, with arbitrary weight, independently of the actual number of physical particles corresponding to a given density. In this way, the computation time is the same for low- or high-density regions, leading to a more efficient use of computational resources and uniform statistical error. The computation time can be reduced also by tabulating and using more than once the results from the local MC steps in the overall iteration process. While the local iterative MC approach does not have the generality of the standard MC technique, much faster calculations and greatly reduced statistical noise are possible when applied to steady-state problems. We present a number of simulation examples for silicon devices, which illustrate the capabilities of the approach.

Cellular automata modelling of a surface catalytic reaction with Eley-Rideal step: the case of CO oxidation

Renormalization-group (RG) methods developed previously for the study of the growth of order in unstable systems are extended to treat the spinodal decomposition of the two-dimensional spin-exchange kinetic Ising model. The conservation of the order parameter and fixed-length sum rule are properly preserved in the theory. Various correlation functions in both coordinate and momentum space are calculated as functions of time. The scaling function for the structure factor is extracted. We compare our results with direct Monte Carlo (MC) simulations and find them in good agreement. The time rescaling parameter entering the RG analysis is temperature dependent, as was determined in previous work through a RG analysis of MC simulations. The results exhibit a long-time logarithmic growth law for the typical domain size, both analytically and numerically. In the time region where MC simulations have previously been performed, the logarithmic growth law can be fitted to a power law with an effective exponent. This exponent is found to be in excellent agreement with the result of MC simulations. The logarithmic growth law agrees with a physical model of interfacial motion which involves an interplay between the local curvature and an activated jump across the interface.

Purpose: To perform a systematic comparison of the Monte Carlo (MC), convolution/superposition (CS), and equivalent path length (EPL)‐based dose calculation algorithms for the purposes of outcomes modeling in lung cancer treatment planning. Methods: Several treatment plans (originally planned using EPL) from a large database of patients treated on a lung dose escalation protocol were retrospectively recalculated using MC and CS. Doses were computed in the homogeneous (unit‐density) and heterogeneous geometries; homogeneous calculations were used to elicit differences in the beam models To evaluate algorithmic differences due to heterogeneity effects, beam model differences were minimized by adjusting beam weights in the homogeneous plans to achieve the same prescribed dose with each algorithm. These beam weights were then applied to the heterogeneous geometries. Absolute dose distributions were compared using: color‐wash dose difference displays, isodose lines, EUD (for the target) and mean lung dose (MLD) and NTCP (for the normal lungs). Results: For the target, MC and CS‐computed EUDs were in good agreement for both homogeneous and heterogeneous cases, with maximum dose differences of 1.2 Gy noted. Differences between EPL and MC (or CS) were generally much larger, in the heterogeneous plans extending up to 6 Gy. Differences in MLD computed with MC and CS ranged between 2% and 15% in the heterogeneous plans. These differences were similar in the corresponding homogeneous geometries, illustrating the importance of beam model disparities. For EPL, differences in the MLD and NTCP (relative to MC or CS) were much larger in the heterogeneous plans indicating systematic differences in the normal lung dose prediction. Conclusion: Evidence thus far is suggestive that discrepancies in dose computed with EPL and MC (or CS) will lead to differences in correlations of dose with outcome with respect to the target as well as normal tissue complications (radiation induced pneumonitis) and calculated NTCP.

This paper presents a groove reconstruction method based on the local rule of Cellular Automata (CA). A local rule for groove reconstruction is proposed to design and optimize the groove of milling insert. The insert is firstly divided into a large number of cells that typically form a regular grid over the domain. The domain consists of regular square cells with discrete variables and the discrete CA model is built. As the cells in the CA domain only interact with their neighboring cells when performing local computations, the governing equation for the whole domain is not necessary. Then the states of cells are modified according to the local rule applied. The state of the entire system is updated based on the state of the cell and its neighboring cells. Collectively, these cells’ states define the state of the entire domain, and the groove can be reconstructed according to the state of the domain. The reconstructed groove is tested by a FEM simulation. The simulation results show that the reconstructed groove has a satisfied performance on the stress field.

In today’s world, advancement of Information Technology has been simultaneously followed by cyber crimes resulting in offensive and distressful digital contents. Threat to the digital content has initiated the need for application of forensic activities in digital field seeking evidence against any type of cyber crimes for the sake of reinforcement of the law and order. Digital Forensics is an interdisciplinary branch of computer science and forensic sciences, rapidly utilizing the recovery and/or investigation works on digital data explored in electronic memory based devices with reference to any cyber based unethical, illegal, and unauthorized activities. A typical digital forensic investigation work follows three steps to collect evidence(s): content acquisition, content analysis and report generation. In digital content analysis higher amount of data volumes and human resource(s) exposure to distressing and offensive materials are of major concerns. Lack of technological support for processing large amount of offensive data makes the analytical procedure quite time consuming and expensive. Thus, it results in a degradation of mental health of concerned investigators. Backlog in processing time by law enforcement department and financial limitations initiate huge demand for digital forensic investigators turning out trustworthy results within reasonable time. Forensic analysis is performed on randomly populated sample, instead of entire population size, for faster and reliable analysis procedure of digital contents. Present work reports about an efficient design methodology to facilitate random sampling procedure to be used in digital forensic investigations. Cellular Automata (CA) based approach has been used in our random sampling method. Equal Length Cellular Automata (ELCA) based pseudo-random pattern generator (PRPG) has been proposed in a cost effective manner utilizing the concept of random pattern generator. Exhibition of high degree randomness has been demonstrated in the field of randomness quality testing. Concerned cost effectiveness refers to time complexity, space complexity, design complexity and searching complexity. This research includes the comparative study for some well known random number generators, e.g., recursive pseudo-random number generator (RPRNG), atmospheric noise based true-random number generator (TRNG), Monte-Carlo (M-C) pseudo-random number generator, Maximum Length Cellular Automata (MaxCA) random number generator and proposed Equal Length Cellular Automata (ELCA) random number generator.Resulting sequences for all those above mentioned pattern generatorshave significant improvement in terms of randomness quality. Associated fault coverage is being improved using iterative methods. Emphasis on cost effectiveness has been initiated for proposed random sampling in forensic analysis.

Three-state von Neumann cellular automata and pattern generation

We report on the 3D modeling of ultra-small MOS structures using a newly developed full-band device simulator. The simulation tool is based on a novel approach, featuring a hybrid Monte Carlo (MC)-Cellular Automata (CA) simulation engine self-consistently coupled with a 3D multi-grid Poisson solver.

A cellular automaton-based technique for modeling mesoscale damage evolution

This thesis addresses the computational notion of aesthetics in the framework of multistate two-dimensional cellular automata (2D CA). The measure of complexity is a core concept in computational approaches to aesthetics. Shannon's information theory provided an objective measure of complexity, which led to the emergence of various informational theories of aesthetics. However, entropy fails to take into account the spatial characteristics of 2D patterns; these characteristics are fundamental in addressing the aesthetic problem, in general, and of CA-generated patterns, in particular. This thesis proposes two empirically evaluated alternative measures of complexity, taking into account the spatial characteristics of 2D patterns and experimental studies on human aesthetic perception in the visual domain. The measures are extended to robustly quantify the complexity of multi-state 2D CA-generated patterns. The first model, spatial complexity, is based on the probabilistic spatial distribution of homogeneous/heterogeneous neighbouring cells over the lattice of a multi-state 2D cellular automaton. The second model is based on algorithmic information theory (Kolmogorov complexity) which is extended to estimate the complexity of 2D patterns. The spatial complexity measure presents performance advantage over information-theoretic models, specifically in discriminating symmetries and the orientation in CA-generated patterns, enabling more accurate measurement of complexity in relation to aesthetic evaluations of 2D patterns. A series of experimental stimuli with various structural characteristics and levels of complexity were generated by seeding 3-state 2D CA with different initial configurations for psychological experiments. The results of experimentation demonstrate the presence of correlation between spatial complexity measures and aesthetic judgements of experimental stimuli. The same results were obtained for the estimations of Kolmogorov complexity of experimental stimuli.

To investigate self-heating effects in double gate MOSFETs, a simulator solving self-consistently the Boltzmann transport equations (BTE) for both electrons and phonons has been developed. A Monte Carlo (MC) solver for electrons is coupled with a direct solver for the phonon transport. This method is particularly efficient to evaluate accurately the phonon emission and absorption spectra in both real and energy spaces. The resulting degradation of the I-V characteristics is estimated for a 20 nm-long Double-Gate MOSFET.

The immense size and complex geometry of the ITER experimental fusion reactor require the development of special techniques that can accurately and efficiently perform neutronics simulations with minimal human effort. This paper shows the effect of the hybrid Monte Carlo (MC)/deterministic techniques—Consistent Adjoint Driven Importance Sampling (CADIS) and Forward-Weighted CADIS (FW-CADIS)—in enhancing the efficiency of the neutronics modeling of ITER and demonstrates the applicability of coupling these methods with computer-aided-design-based MC. Three quantities were calculated in this analysis: the total nuclear heating in the inboard leg of the toroidal field coils (TFCs), the prompt dose outside the biological shield, and the total neutron and gamma fluxes over a mesh tally covering the entire reactor. The use of FW-CADIS in estimating the nuclear heating in the inboard TFCs resulted in a factor of ˜275 increase in the MC figure of merit (FOM) compared with analog MC and a factor of ˜9 compared with the traditional methods of variance reduction. By providing a factor of ˜21 000 increase in the MC FOM, the radiation dose calculation showed how the CADIS method can be effectively used in the simulation of problems that are practically impossible using analog MC. The total flux calculation demonstrated the ability of FW-CADIS to simultaneously enhance the MC statistical precision throughout the entire ITER geometry. Collectively, these calculations demonstrate the ability of the hybrid techniques to accurately model very challenging shielding problems in reasonable execution times.

The paper presents the spectrum of methods and results for design of structures, which are frequently applied in numerical simulations of properties and processes taking place in cellular materials. The methods described here are universal for many applications at various length-scales. They can be efficiently applied for complex cellular structures such as polycrystals or foams, where the elements (grains or pores) are distributed and shaped in a controlled way. The digital material representations created by these methods can be used for a number of numerical techniques such as: Molecular dynamics (MD), Monte Carlo (MC), Cellular Automaton (CA), Finite Element Method (FEM) or Finite Volume Method (FVM). The examples of structures consisting of atoms, fields or finite elements are presented in this paper. The applications of such structures are demonstrated by the properties and processes relevant to the specific length scales.