Tracking the states of a nonlinear system in the weight-space of a feed-forward neural network

Nonlinear, non-stationary signals are commonly found in a variety of disciplines such as biology, medicine, geology and financial modeling. The complexity (e.g. nonlinearity and non-stationarity) of such signals and their low signal to noise ratios often make it a challenging task to use them in critical applications. In this paper we propose a new neural network based technique to address those problems. We show that a feed forward, multi-layered neural network can conveniently capture the states of a nonlinear system in its connection weight-space, after a process of supervised training. The performance of the proposed method is investigated via computer simulations.

Globalisation and connectedness: A network approach to international business

There have been recent calls for the field of International Business to retool its routines by becoming genuinely interdisciplinary. This paper takes such an approach by using recent advances in the fields of evolutionary economics and applying them to IB. Evolutionary economists are now viewing the economy as an actual network. Consequently, one the key analytical tools in this approach is network analysis. Some of the basic methods in network analysis are reviewed. The paper then looks at how using these tools might be of use in IB studies. In particular, it outlines fruitful research paths in the areas of globalisation and regionalisation, and the measurement of performance in multi-national firms and alliances. In each case, propositions are put forward which can be analytically tested with the use of network analysis. The paper concludes with a brief outline of a research agenda which utilises this approach in International Business studies.

Performance analysis of multi-radio AODV in hybrid wireless mesh networks

Wireless Mesh Networks (WMNs), based on commodity hardware, present a promising technology for a wide range of applications due to their self-configuring and self-healing capabilities, as well as their low equipment and deployment costs. One of the key challenges that WMN technology faces is the limited capacity and scalability due to co-channel interference, which is typical for multi-hop wireless networks. A simple and relatively low-cost approach to address this problem is the use of multiple wireless network interfaces (radios) per node. Operating the radios on distinct orthogonal channels permits effective use of the frequency spectrum, thereby, reducing interference and contention. In this paper, we evaluate the performance of the multi-radio Ad-hoc On-demand Distance Vector (AODV) routing protocol with a specific focus on hybrid WMNs. Our simulation results show that under high mobility and traffic load conditions, multi-radio AODV offers superior performance as compared to its single-radio counterpart. We believe that multi-radio AODV is a promising candidate for WMNs, which need to service a large number of mobile clients with low latency and high bandwidth requirements.

Improved multi-beam neural network scatterometer forward models

Current methods for retrieving near surface winds from scatterometer observations over the ocean surface require a foward sensor model which maps the wind vector to the measured backscatter. This paper develops a hybrid neural network forward model, which retains the physical understanding embodied in ¸mod, but incorporates greater flexibility, allowing a better fit to the observations. By introducing a separate model for the mid-beam and using a common model for the fore- and aft-beams, we show a significant improvement in local wind vector retrieval. The hybrid model also fits the scatterometer observations more closely. The model is trained in a Bayesian framework, accounting for the noise on the wind vector inputs. We show that adding more high wind speed observations in the training set improves wind vector retrieval at high wind speeds without compromising performance at medium or low wind speeds.

Attentional changes in pre-stimulus oscillatory activity within early visual cortex are predictive of human visual performance

Physiological and neuroimaging studies provide evidence to suggest that attentional mechanisms operating within the fronto-parietal network may exert top–down control on early visual areas, priming them for forthcoming sensory events. The believed consequence of such priming is enhanced task performance. Using the technique of magnetoencephalography (MEG), we investigated this possibility by examining whether attention-driven changes in cortical activity are correlated with performance on a line-orientation judgment task. We observed that, approximately 200 ms after a covert attentional shift towards the impending visual stimulus, the level of phase-resetting (transient neural coherence) within the calcarine significantly increased for 2–10 Hz activity. This was followed by a suppression of alpha activity (near 10 Hz) which persisted until the onset of the stimulus. The levels of phase-resetting, alpha suppression and subsequent behavioral performance varied between subjects in a systematic fashion. The magnitudes of phase-resetting and alpha-band power were negatively correlated, with high levels of coherence associated with high levels of performance. We propose that top–down attentional control mechanisms exert their initial effects within the calcarine through a phase-resetting within the 2–10 Hz band, which in turn triggers a suppression of alpha activity, priming early visual areas for incoming information and enhancing behavioral performance.

Traffic based energy consumption optimisation to improve the lifetime and performance of ad hoc wireless sensor networks

Ad hoc wireless sensor networks (WSNs) are formed from self-organising configurations of distributed, energy constrained, autonomous sensor nodes. The service lifetime of such sensor nodes depends on the power supply and the energy consumption, which is typically dominated by the communication subsystem. One of the key challenges in unlocking the potential of such data gathering sensor networks is conserving energy so as to maximize their post deployment active lifetime. This thesis described the research carried on the continual development of the novel energy efficient Optimised grids algorithm that increases the WSNs lifetime and improves on the QoS parameters yielding higher throughput, lower latency and jitter for next generation of WSNs. Based on the range and traffic relationship the novel Optimised grids algorithm provides a robust traffic dependent energy efficient grid size that minimises the cluster head energy consumption in each grid and balances the energy use throughout the network. Efficient spatial reusability allows the novel Optimised grids algorithm improves on network QoS parameters. The most important advantage of this model is that it can be applied to all one and two dimensional traffic scenarios where the traffic load may fluctuate due to sensor activities. During traffic fluctuations the novel Optimised grids algorithm can be used to re-optimise the wireless sensor network to bring further benefits in energy reduction and improvement in QoS parameters. As the idle energy becomes dominant at lower traffic loads, the new Sleep Optimised grids model incorporates the sleep energy and idle energy duty cycles that can be implemented to achieve further network lifetime gains in all wireless sensor network models. Another key advantage of the novel Optimised grids algorithm is that it can be implemented with existing energy saving protocols like GAF, LEACH, SMAC and TMAC to further enhance the network lifetimes and improve on QoS parameters. The novel Optimised grids algorithm does not interfere with these protocols, but creates an overlay to optimise the grids sizes and hence transmission range of wireless sensor nodes.

Automatic control of attenuation in a digital communication network

This thesis describes the investigation of an adaptive method of attenuation control for digital speech signals in an analogue-digital environment and its effects on the transmission performance of a national telecommunication network. The first part gives the design of a digital automatic gain control, able to operate upon a P.C.M. signal in its companded form and whose operation is based upon the counting of peaks of the digital speech signal above certain threshold levels. A study was ma.de of a digital automatic gain control (d.a.g.c.) in open-loop configuration and closed-loop configuration. The former was adopted as the means for carrying out the automatic control of attenuation. It was simulated and tested, both objectively and subjectively. The final part is the assessment of the effects on telephone connections of a d.a.g.c. that introduces gains of 6 dB or 12 dB. This work used a Telephone Connection Assessment Model developed at The University of Aston in Birmingham. The subjective tests showed that the d.a.g.c. gives advantage for listeners when the speech level is very low. The benefit is not great when speech is only a little quieter than preferred. The assessment showed that, when a standard British Telecom earphone is used, insertion of gain is desirable if speech voltage across the earphone terminals is below an upper limit of -38 dBV. People commented upon the presence of an adaptive-like effect during the tests. This could be the reason why they voted against the insertion of gain at level only little quieter than preferred, when they may otherwise have judged it to be desirable. A telephone connection with a d.a.g.c. in has a degree of difficulty less than half of that without it. The score Excellent plus Good is 10-30% greater.

Spatially orthogonal chemical functionalization of a hierarchical pore network for catalytic cascade reactions

The chemical functionality within porous architectures dictates their performance as heterogeneous catalysts; however, synthetic routes to control the spatial distribution of individual functions within porous solids are limited. Here we report the fabrication of spatially orthogonal bifunctional porous catalysts, through the stepwise template removal and chemical functionalization of an interconnected silica framework. Selective removal of polystyrene nanosphere templates from a lyotropic liquid crystal-templated silica sol–gel matrix, followed by extraction of the liquid crystal template, affords a hierarchical macroporous–mesoporous architecture. Decoupling of the individual template extractions allows independent functionalization of macropore and mesopore networks on the basis of chemical and/or size specificity. Spatial compartmentalization of, and directed molecular transport between, chemical functionalities affords control over the reaction sequence in catalytic cascades; herein illustrated by the Pd/Pt-catalysed oxidation of cinnamyl alcohol to cinnamic acid. We anticipate that our methodology will prompt further design of multifunctional materials comprising spatially compartmentalized functions.

Improved neural network scatterometer forward models

Current methods for retrieving near-surface winds from scatterometer observations over the ocean surface require a forward sensor model which maps the wind vector to the measured backscatter. This paper develops a hybrid neural network forward model, which retains the physical understanding embodied in CMOD4, but incorporates greater flexibility, allowing a better fit to the observations. By introducing a separate model for the midbeam and using a common model for the fore and aft beams, we show a significant improvement in local wind vector retrieval. The hybrid model also fits the scatterometer observations more closely. The model is trained in a Bayesian framework, accounting for the noise on the wind vector inputs. We show that adding more high wind speed observations in the training set improves wind vector retrieval at high wind speeds without compromising performance at medium or low wind speeds. Copyright 2001 by the American Geophysical Union.

The impact of architecture on the performance of artificial neural networks

A number of researchers have investigated the impact of network architecture on the performance of artificial neural networks. Particular attention has been paid to the impact on the performance of the multi-layer perceptron of architectural issues, and the use of various strategies to attain an optimal network structure. However, there are still perceived limitations with the multi-layer perceptron and networks that employ a different architecture to the multi-layer perceptron have gained in popularity in recent years, particularly, networks that implement a more localised solution, where the solution in one area of the problem space does not impact, or has a minimal impact, on other areas of the space. In this study, we discuss the major architectural issues affecting the performance of a multi-layer perceptron, before moving on to examine in detail the performance of a new localised network, namely the bumptree. The work presented here examines the impact on the performance of artificial neural networks of employing alternative networks to the long established multi-layer perceptron. In particular, networks that impose a solution where the impact of each parameter in the final network architecture has a localised impact on the problem space being modelled are examined. The alternatives examined are the radial basis function and bumptree neural networks, and the impact of architectural issues on the performance of these networks is examined. Particular attention is paid to the bumptree, with new techniques for both developing the bumptree structure and employing this structure to classify patterns being examined.

The relative performance of scalable load balancing algorithms in loosely-coupled distributed systems. Available in 2 volumes.

The computer systems of today are characterised by data and program control that are distributed functionally and geographically across a network. A major issue of concern in this environment is the operating system activity of resource management for different processors in the network. To ensure equity in load distribution and improved system performance, load balancing is often undertaken. The research conducted in this field so far, has been primarily concerned with a small set of algorithms operating on tightly-coupled distributed systems. More recent studies have investigated the performance of such algorithms in loosely-coupled architectures but using a small set of processors. This thesis describes a simulation model developed to study the behaviour and general performance characteristics of a range of dynamic load balancing algorithms. Further, the scalability of these algorithms are discussed and a range of regionalised load balancing algorithms developed. In particular, we examine the impact of network diameter and delay on the performance of such algorithms across a range of system workloads. The results produced seem to suggest that the performance of simple dynamic policies are scalable but lack the load stability of more complex global average algorithms.

Parallel implementation of image segmentation algorithms on a network of transputers

Image segmentation is one of the most computationally intensive operations in image processing and computer vision. This is because a large volume of data is involved and many different features have to be extracted from the image data. This thesis is concerned with the investigation of practical issues related to the implementation of several classes of image segmentation algorithms on parallel architectures. The Transputer is used as the basic building block of hardware architectures and Occam is used as the programming language. The segmentation methods chosen for implementation are convolution, for edge-based segmentation; the Split and Merge algorithm for segmenting non-textured regions; and the Granlund method for segmentation of textured images. Three different convolution methods have been implemented. The direct method of convolution, carried out in the spatial domain, uses the array architecture. The other two methods, based on convolution in the frequency domain, require the use of the two-dimensional Fourier transform. Parallel implementations of two different Fast Fourier Transform algorithms have been developed, incorporating original solutions. For the Row-Column method the array architecture has been adopted, and for the Vector-Radix method, the pyramid architecture. The texture segmentation algorithm, for which a system-level design is given, demonstrates a further application of the Vector-Radix Fourier transform. A novel concurrent version of the quad-tree based Split and Merge algorithm has been implemented on the pyramid architecture. The performance of the developed parallel implementations is analysed. Many of the obtained speed-up and efficiency measures show values close to their respective theoretical maxima. Where appropriate comparisons are drawn between different implementations. The thesis concludes with comments on general issues related to the use of the Transputer system as a development tool for image processing applications; and on the issues related to the engineering of concurrent image processing applications.