International Journal of Modern Physics C

This paper discusses the performance studies of a coarse grained parallel neural network training code for control of nonlinear dynamical systems, implemented in the shared memory and message passing parallel programming environments OpenMP and MPI, respectively. In addition, these codes are compared to an implementation utilizing SHMEM the native data passing SGI/Cray environment for parallel programming. The multiprocessor platform used in the study is a SGI/Cray Origin 2000 with up to 32 processors, which supports all these programming models efficiently. The dynamical system used in this study is a nonlinear 0D model of a thermonuclear fusion reactor with the EDA-ITER design parameters. The results show that OpenMP outperforms the other two environments when large number of processors are involved, while yielding a similar or a slightly poorer behavior for small number of processors. As expected the native SGI/Cray environment outperforms MPI for the entire range of processors used. Reasons for the observed performance are given. The parallel efficiency of the code is always greater than 60% regardless of the parallel environment for the range of processors used in this study.

An impact of the spin–orbit interaction on the electron quantum confinement is considered theoretically for narrow gap semiconductor cylindrical quantum dots. To study the phenomena for InAs quantum dot embedded into GaAs semiconductor matrix, the effective one electronic band Hamiltonian, the energy position dependent electron effective mass approximation, and the spin-dependent Ben Daniel–Duke boundary conditions are considered, formulated and solved numerically. To solve the nonlinear Schrödinger equation, we propose a nonlinear iterative algorithm. This calculation algorithm not only converges for all simulation cases but also has a good convergent rate. With the developed quantum dot simulator, we study the effect of the spin–orbit interaction for narrow gap InAs/GaAs semiconductor cylindrical quantum dots. From the numerical calculations, it has been observed that the spin–orbit interaction leads to a sizeable spin-splitting of the electron energy states with nonzero angular momentum. Numerical evidence is presented to show the splitting result is strongly dependent on the quantum dot size.

EGS is a very popular Monte Carlo code, used in the simulation of Nuclear Medicine devices. Simulation techniques are particularly effective to optimize collimator configuration and camera design in Single Photon Emission studies. With the EGS code, users must define the geometry where particles are transported. This can be both a very hard task and a source of inefficiency, especially in the case of complex geometries as, for instance, hexagonal hole collimators or pixellated detectors. In this paper we present a modular description of such geometries. Our method allows the computation of the region a point belongs to in a few steps; thus we are able to calculate this region in a reduced number of operations, independently of the collimator and detector dimensions. With a modular description we can reduce the computational time by 30%, with respect to a "traditional" description of the geometry. We validated the modular description in the simulation of a Nuclear Medicine apparatus for scintimammography. Two different collimators have been considered: one with square holes and one with hexagonal holes. We accomplished their characterization and tested their performance in a torso–breast phantom. Outcomes of the two collimators are comparable, even if it seems that the hexagonal hole collimator, thanks to its greater septal penetration, could give slightly better results for small tumors located near the collimator.

We present a method for estimating the density of states of a classical statistical model. The algorithm successfully combines the Wang–Landau flat histogram method with the N-fold way in order to improve the efficiency of the original single-spin-flip version. We test our implementation of the Wang–Landau method with the two-dimensional nearest neighbor Ising model for which we determine the tunneling time and the density of states on lattices with sizes up to 50 × 50. Furthermore, we show that our new algorithm performs correctly at right edges of an energy interval over which the density of states is computed. This removes a disadvantage of the original single-spin-flip Wang–Landau method where results showed systematically higher errors in the density of states at right boundaries. In order to demonstrate the improvements made, we compare our data with the detailed numerical tests presented in a study by Wang and Swendsen where the original Wang–Landau method was tested against various other methods, especially the transition matrix Monte Carlo method (TMMC). Finally, we apply our method to a thin Ising film of size 32 × 32 × 6 with antiparallel surface fields. With the density of states obtained from the simulations we calculate canonical averages related to the energy such as internal energy, Gibbs free energy and entropy, but we also sample microcanonical averages during simulations in order to determine canonical averages of the susceptibility, the order parameter and its fourth order cumulant. We compare our results with simulational data obtained from a conventional Monte Carlo algorithm.

Nucleation in the two-dimensional q-state Potts model has been studied by means of Monte Carlo simulations using the heat-bath dynamics. The initial metastable state has been prepared by magnetic quench of the ordered low-temperature phase. The magnetic field dependence of the nucleation time has been measured as the function of the magnetic field for different q and lattice sizes at T = 0.5T_c. A size-dependent crossover from the coalescence to nucleation region is observed at all q. The magnetic field dependence of the nucleation time is roughly described by the classical nucleation theory. Our data show increase of the anisotropy in the shape of the critical droplets with increase of q.

A model of Poincaré gauge self-dual theory over the Minkowski space–time, endowed with spherical symmetry, is considered. The self-duality (S-D) conditions are imposed and two sets of the self-duality equations for the gauge fields are obtained: one for the torsion tensor and another for the curvature tensor. An analytical solution of the gauge field equations is also obtained. The Yang–Mills (Y–M) equations of the gauge fields are also derived. It is shown that these Y–M equations can be obtained from the S-D equations, a result which is generally true for any S-D gauge theory. Most of all calculations are performed using the GRTensorII package, running on the MapleV platform.

Possible schemes for sending N-bit secret message with established N EPR pairs and a classical channel between two parties, are considered. The corresponding N-particle quantum key-distribution for N-bit secret message communication is outlined.

In this paper the kinetics of Muon Catalyzed Fusion (μCF) in H/D/T mixture, considering the muon transfer from hydrogen isotopes (p,d,t) to helium isotopes (³He, ⁴He) in the range of temperatures of 300 K < T < 1300 K at the density of Φ = 1 LHD, is presented. Calculation of cycling rate in different branches of fusion such as dtμ, ptμ, pdμ, ddμ, ttμ and energy gain showed that in certain physical conditions, the H/D/T mixture is comparable to D/T one. Finally the results of this research are compared with the experimental results reported by other researchers; and good agreements were found.

We empirically investigate distributions of individual consumption expenditure for four commodity categories conditional on fixed income levels. The data stems from the Family Expenditure Survey carried out annually in the United Kingdom. We use graphical techniques to test for normality and lognormality of these distributions. While prevailing economic theory does not predict any structure for these distributions, we find that in at least three commodity categories individual consumption expenditure conditional on a fixed income level is lognormally distributed.

A nonlinear differential equation of Sornette–Ide type with noise, for a complex variable, yields endogenous crashes, preceded by roughly log-periodic oscillations in the real part, and a strong increase in the imaginary part. The latter is interpreted as the trader expectation.

We study the question whether matrix models obtained in the zero volume limit of 4d Yang–Mills theories can describe large N QCD strings. The matrix model we use is a variant of the Eguchi–Kawai model in terms of Hermitian matrices, but without any twists or quenching. This model was originally proposed as a toy model of the IIB matrix model. In contrast to common expectations, we do observe the area law for Wilson loops in a significant range of scale of the loop area. Numerical simulations show that this range is stable as N increases up to 768, which strongly suggests that it persists in the large N limit. Hence the equivalence to QCD strings may hold for length scales inside a finite regime.

A robust and reliable implicit method is proposed for application in data interpolation. The algorithm is based on a recently developed analytic approximation method, namely the distributed approximating functionals (DAFs), which is known to have the "well-tempered" property of UNIFORMLY approximating a function and its derivatives. In comparison with the conventionally used local explicit interpolation algorithms, the implicit method achieves much more accurate interpolation results because it couples all sample values (both known and unknown) in the domain of interest using a set of simultaneous linear algebraic equations. Due to the fact that the well-tempered DAFs also are very good low-pass filters, the performance of the DAF-based implicit method is not affected very much by the high frequency noise in the input signal. As an application, the proposed algorithm is applied to signals that are corrupted with impulse noise.