Generalized loopy 2U: A new algorithm for approximate inference in credal networks

Credal networks generalize Bayesian networks by relaxing the requirement of precision of probabilities. Credal networks are considerably more expressive than Bayesian networks, but this makes belief updating NP-hard even on polytrees. We develop a new efficient algorithm for approximate belief updating in credal networks. The algorithm is based on an important representation result we prove for general credal networks: that any credal network can be equivalently reformulated as a credal network with binary variables; moreover, the transformation, which is considerably more complex than in the Bayesian case, can be implemented in polynomial time. The equivalent binary credal network is then updated by L2U, a loopy approximate algorithm for binary credal networks. Overall, we generalize L2U to non-binary credal networks, obtaining a scalable algorithm for the general case, which is approximate only because of its loopy nature. The accuracy of the inferences with respect to other state-of-the-art algorithms is evaluated by extensive numerical tests.     

Achieving Brouwer’s law with implicit Runge–Kutta methods

In high accuracy long-time integration of differential equations, round-off errors may dominate truncation errors. This article studies the influence of round-off on the conservation of first integrals such as the total energy in Hamiltonian systems. For implicit Runge–Kutta methods, a standard implementation shows an unexpected propagation. We propose a modification that reduces the effect of round-off and shows a qualitative and quantitative improvement for an accurate integration over long times. AMS subject classification (2000)  65L06, 65G50, 65P10     

Propagation of a vectorial Laguerre–Gaussian beam beyond the paraxial approximation

Based on the vectorial Rayleigh–Sommerfeld integrals, the analytical propagation expression of a vectorial Laguerre–Gaussian beam beyond paraxial approximation is presented. The far field expression and the scalar paraxial result are obtained as special cases of the general formulae. According to the analytical representation, the light intensity distribution of the vectorial Laguerre–Gaussian beam is depicted in the reference plane. The light intensity distribution of a vectorial Laguerre–Gaussian beam with cos m is also compared with that of a vectorial Laguerre–Gaussian beam with sin m.     

Velocity of propagation in diffusional quantum theory

An equation of diffusional quantum theory which takes into account the finite velocity of propagation is derived from Kelvin's telegraph equation and Fürth's relation. The equation is then used to derive the ground state of quantum systems and to derive the Sommerfeld-Dirac expression for the ionization potential of hydrogen-like ions.     

Femtosecond second-order solitons in optical fiber transmission

A propagation of the femtosecond second-order solitons in an optical fiber is studied. We show that a generalized nonlinear Schrödinger equation well describes the propagation of the second-order soliton even containing only a few optical cycles. The propagations of a 50 fs and a 10 fs second-order soliton in an optical fiber are numerically simulated. It is found that, for the case of 10 fs second-order soliton, the soliton decay is dominated by the third-order dispersion, in contrast to the case of 50 fs second-order solitons, where the soliton decay is dominated by the delayed Raman response. It is also found that the exact delayed Raman response form must be used for the propagation of the 50 fs or less than 50 fs second-order soliton.     

An improved theory of microwave open resonators

The analytic theory of microwave open resonators is improved here by means of solving the complex Airy equation, that the Q-value as well as the argument of field profile may be given accurately. The resonance conditions for open cavities with high and moderate Q-values are derived analytically. Theoreticl Q-values are well agreed with the measured ones from experiments.     

Wind effects on noise propagation for complicated geographical and road configurations

Outdoor propagation from roads is influenced by wind, temperature and humidity. It is necessary to predict wind effects to simulate long-term environmental noise accurately. Noise levels have been measured in conjunction with wind speed and wind direction measurements. These noise levels showed noticeable daily changes. Excess attenuations of noise levels were found to depend on the vector wind (U_vec). By relating the vector wind and observed sound attenuations under all wind conditions, the influence of wind can be more accurately predicted for purposes of simulating noise propagation.     

Long-Distance Transmission at Zero Dispersion: Exact Expressions for One-Time Statistical Properties

We study the transmission of a signal through a dispersionless fiber with in-line amplifiers. The nonlinearities of the fiber and the noise generated at each amplifier are taken into account. Exact analytical expressions are obtained for the field averages. From these results we obtain the distributed amplification approximation, and study the conditions under which this approximation is valid.     

Frequency– and Time-domain BEM Analysis of Rigid Track on a Half-Space with Vibration Barriers

The rapid development of high-speed trains like the TGV or the ICE in recent years results in high dynamic loads causing vibrations which propagate from the train-track structure into the ground and further into nearby buildings. In this context it is important to develop rigid tracks with improved dynamic behaviour and to investigate possible means of vibration reduction. The boundary element method in frequency and time domain is used to simulate train-track structures subjected to dynamic loading and the reduction of vibrations which for instance can be achieved via a trench running parallel to the rigid track. In this context the non-causality error, which arises when the time-domain BEM algorithm is applied to mathematically concave domains, is discussed and the substructure method is proposed as a solution to this problem. A two-layered cylindrical elastic structure on a half-space with a trench is added as an example of a possible application.