The Bayesian Approach to Signal Modelling

In this paper, an introduction to Bayesian methods in signal processing will be given. The paper starts by considering the important issues of model selection and parameter estimation and derives analytic expressions for the model probabilities of two simple models. The idea of marginal estimation of certain model parameter is then introduced and expressions are derived for the marginal probabilitiy densities for frequencies in white Gaussian noise and a Bayesian approach to general changepoint analysis is given. Numerical integration methods are introduced based on Markov chain Monte Carlo techniques and the Gibbs sampler in particular.

Ground vibration generated by trains in underground tunnels

A popular method used to reduce vibration transmitted from underground railways into nearby buildings is floating-slab track, whereby a concrete slab supporting the two rails is mounted on rubber bearings or steel springs to isolate it from the tunnel invert. This paper adds a track model to a previously developed three-dimensional tunnel model in order to assess the effectiveness of floating-slab track. A slab beam coupled to the tunnel in the wavenumber domain, with the slab bearings represented by an elastic layer, is examined first. A second beam representing the two rails together is then coupled to the slab, and axle masses representing a train are added to the rail beam. Power-spectral densities and RMS levels of soil vibration due to random roughness-displacement excitation between the masses and the rail beam are calculated. Analytical techniques are used to minimise the computational requirements of the model. The results demonstrate the inadequacy of simple mass-spring and Winkler-beam models with rigid foundations for the assessment of the vibration-isolation performance of railway track. They suggest that the achievable insertion loss is modest and that floating the track slab may in fact cause increased transmission of vibration under certain conditions. © 2006 Elsevier Ltd. All rights reserved.

Physiological modelling for improved reliability in silhouette-driven gradient-based hand tracking

We present a gradient-based motion capture system that robustly tracks a human hand, based on abstracted visual information - silhouettes. Despite the ambiguity in the visual data and despite the vulnerability of gradient-based methods in the face of such ambiguity, we minimise problems related to misfit by using a model of the hand's physiology, which is entirely non-visual, subject-invariant, and assumed to be known a priori. By modelling seven distinct aspects of the hand's physiology we derive prior densities which are incorporated into the tracking system within a Bayesian framework. We demonstrate how the posterior is formed, and how our formulation leads to the extraction of the maximum a posteriori estimate using a gradient-based search. Our results demonstrate an enormous improvement in tracking precision and reliability, while also achieving near real-time performance. © 2009 IEEE.

Methods of patterning magnetic fields using bulk superconductors

High Temperature superconductors are able to carry very high current densities, and thereby sustain very high magnetic fields. There are many projects which use the first property and these have concentrated on power generation, transmission and utilization, however there are relatively few which are currently exploiting the ability to sustain high magnetic fields. There are two main reasons for this: high field wound magnets can and have been made from both BSCCO and YBCO but currently their cost is much higher than the alternative provided by low Tc materials such as Nb3Sn and NbTi. An alternative form of the material is the bulk form which can be magnetized to high fields and using flux pumping this can be done in situ. This paper explores some of the applications of bulk superconductors and describes methods of producing field patterns using the highly uniform magnetic fields required for MRI and accelerator magnets as the frame of reference. The patterns are not limited to uniform fields and it is entirely possible to produce a field varying sinusoidally in space such as would be required for a motor or a generator. The scheme described in this paper describes a dipole magnet such as is found in an accelerator magnet. The tunnel is 30 × 50 × 1000 mm and we achieve a uniformity of better than 200 ppm over the 1000 mm length and better than 1 ppm over the central 500 mm region. The paper presents results for both the overall uniformity and the integrated uniformity which is 302 ppm over the 1000 mm length. © 2010 IEEE.