Projective Limit Random Probabilities on Polish Spaces

A pivotal problem in Bayesian nonparametrics is the construction of prior distributions on the space M(V) of probability measures on a given domain V. In principle, such distributions on the infinite-dimensional space M(V) can be constructed from their finite-dimensional marginals---the most prominent example being the construction of the Dirichlet process from finite-dimensional Dirichlet distributions. This approach is both intuitive and applicable to the construction of arbitrary distributions on M(V), but also hamstrung by a number of technical difficulties. We show how these difficulties can be resolved if the domain V is a Polish topological space, and give a representation theorem directly applicable to the construction of any probability distribution on M(V) whose first moment measure is well-defined. The proof draws on a projective limit theorem of Bochner, and on properties of set functions on Polish spaces to establish countable additivity of the resulting random probabilities.

Micro-vortex generator flow control for supersonic engine inlets

To investigate the flow control potential of micro-vortex generators for supersonic mixed-compression inlets, a basic model experiment has been designed which combines a normal shock wave with a subsonic diffuser. The diffuser is formed by a simple expansion corner, with a divergence angle of 6 degrees. The diffuser entry Mach numbers were M=1.3 and M=1.5 and a number of shock locations relative to the corner position were tested. Flow control was applied in the form of counter-rotating micro-vanes with heights of approximately 20% of boundary layer thickness. Furthermore, corner fences where employed to reduce sidewall effects. It was found that micro-vortex generators were able to significantly reduce the extent of flow separation under all conditions, but could not eliminate it altogether. Corner fences also demonstrated potential for improving the flow in rectangular cross section channels and the combination of corner fences with micro-vortex generators was found to give the greatest benefits. At M=1.3 the combination of corner fences and micro-vanes placed close to the diffuser entry could prevent separation for a wide range of conditions. At the higher diffuser entry Mach number the benefits of flow control were less significant although a reduction of separation size and an improved pressure recovery was observed. It is thought that micro-vortex generators can have significant flow control potential if they are placed close to the expected separation onset and when the adverse pressure gradient is not too far above the incipient separation level. The significant beneficial effects of corner fences warrant a more comprehensive further investigation. It is thought that the control methods suggested here are capable of reducing the bleed requirement on an inlet, which could provide significant performance advantages.

Bayesian signal processing

The application of Bayes' Theorem to signal processing provides a consistent framework for proceeding from prior knowledge to a posterior inference conditioned on both the prior knowledge and the observed signal data. The first part of the lecture will illustrate how the Bayesian methodology can be applied to a variety of signal processing problems. The second part of the lecture will introduce the concept of Markov Chain Monte-Carlo (MCMC) methods which is an effective approach to overcoming many of the analytical and computational problems inherent in statistical inference. Such techniques are at the centre of the rapidly developing area of Bayesian signal processing which, with the continual increase in available computational power, is likely to provide the underlying framework for most signal processing applications.

Eccentricity effects on acoustic radiation from a spherical source suspended within a thermoviscous fluid sphere.

Acoustic radiation from a spherical source undergoing angularly periodic axisymmetric harmonic surface vibrations while eccentrically suspended within a thermoviscous fluid sphere, which is immersed in a viscous thermally conducting unbounded fluid medium, is analyzed in an exact fashion. The formulation uses the appropriate wave-harmonic field expansions along with the translational addition theorem for spherical wave functions and the relevant boundary conditions to develop a closed-form solution in form of infinite series. The analytical results are illustrated with a numerical example in which the vibrating source is eccentrically positioned within a chemical fluid sphere submerged in water. The modal acoustic radiation impedance load on the source and the radiated far-field pressure are evaluated and discussed for representative values of the parameters characterizing the system. The proposed model can lead to a better understanding of dynamic response of an underwater acoustic lens. It is equally applicable in miniature transducer analysis and design with applications in medical ultrasonics.

On the diffraction of acoustic waves by a quarter-plane

This paper follows the work of A.V. Shanin on diffraction by an ideal quarter-plane. Shanin's theory, based on embedding formulae, the acoustic uniqueness theorem and spherical edge Green's functions, leads to three modified Smyshlyaev formulae, which partially solve the far-field problem of scattering of an incident plane wave by a quarter-plane in the Dirichlet case. In this paper, we present similar formulae in the Neumann case, and describe a numerical method allowing a fast computation of the diffraction coefficient using Shanin's third modified Smyshlyaev formula. The method requires knowledge of the eigenvalues of the Laplace-Beltrami operator on the unit sphere with a cut, and we also describe a way of computing these eigenvalues. Numerical results are given for different directions of incident plane wave in the Dirichlet and the Neumann cases, emphasising the superiority of the third modified Smyshlyaev formula over the other two. © 2011 Elsevier B.V.

The bacterial cytoskeleton modulates motility, type 3 secretion, and colonization in Salmonella.

Although there have been great advances in our understanding of the bacterial cytoskeleton, major gaps remain in our knowledge of its importance to virulence. In this study we have explored the contribution of the bacterial cytoskeleton to the ability of Salmonella to express and assemble virulence factors and cause disease. The bacterial actin-like protein MreB polymerises into helical filaments and interacts with other cytoskeletal elements including MreC to control cell-shape. As mreB appears to be an essential gene, we have constructed a viable ΔmreC depletion mutant in Salmonella. Using a broad range of independent biochemical, fluorescence and phenotypic screens we provide evidence that the Salmonella pathogenicity island-1 type three secretion system (SPI1-T3SS) and flagella systems are down-regulated in the absence of MreC. In contrast the SPI-2 T3SS appears to remain functional. The phenotypes have been further validated using a chemical genetic approach to disrupt the functionality of MreB. Although the fitness of ΔmreC is reduced in vivo, we observed that this defect does not completely abrogate the ability of Salmonella to cause disease systemically. By forcing on expression of flagella and SPI-1 T3SS in trans with the master regulators FlhDC and HilA, it is clear that the cytoskeleton is dispensable for the assembly of these structures but essential for their expression. As two-component systems are involved in sensing and adapting to environmental and cell surface signals, we have constructed and screened a panel of such mutants and identified the sensor kinase RcsC as a key phenotypic regulator in ΔmreC. Further genetic analysis revealed the importance of the Rcs two-component system in modulating the expression of these virulence factors. Collectively, these results suggest that expression of virulence genes might be directly coordinated with cytoskeletal integrity, and this regulation is mediated by the two-component system sensor kinase RcsC.

Construction of nonparametric Bayesian models from parametric Bayes equations

We consider the general problem of constructing nonparametric Bayesian models on infinite-dimensional random objects, such as functions, infinite graphs or infinite permutations. The problem has generated much interest in machine learning, where it is treated heuristically, but has not been studied in full generality in non-parametric Bayesian statistics, which tends to focus on models over probability distributions. Our approach applies a standard tool of stochastic process theory, the construction of stochastic processes from their finite-dimensional marginal distributions. The main contribution of the paper is a generalization of the classic Kolmogorov extension theorem to conditional probabilities. This extension allows a rigorous construction of nonparametric Bayesian models from systems of finite-dimensional, parametric Bayes equations. Using this approach, we show (i) how existence of a conjugate posterior for the nonparametric model can be guaranteed by choosing conjugate finite-dimensional models in the construction, (ii) how the mapping to the posterior parameters of the nonparametric model can be explicitly determined, and (iii) that the construction of conjugate models in essence requires the finite-dimensional models to be in the exponential family. As an application of our constructive framework, we derive a model on infinite permutations, the nonparametric Bayesian analogue of a model recently proposed for the analysis of rank data.