Distributed Placement of Service Facilities in Large-Scale Networks

The effectiveness of service provisioning in largescale networks is highly dependent on the number and location of service facilities deployed at various hosts. The classical, centralized approach to determining the latter would amount to formulating and solving the uncapacitated k-median (UKM) problem (if the requested number of facilities is fixed), or the uncapacitated facility location (UFL) problem (if the number of facilities is also to be optimized). Clearly, such centralized approaches require knowledge of global topological and demand information, and thus do not scale and are not practical for large networks. The key question posed and answered in this paper is the following: "How can we determine in a distributed and scalable manner the number and location of service facilities?" We propose an innovative approach in which topology and demand information is limited to neighborhoods, or balls of small radius around selected facilities, whereas demand information is captured implicitly for the remaining (remote) clients outside these neighborhoods, by mapping them to clients on the edge of the neighborhood; the ball radius regulates the trade-off between scalability and performance. We develop a scalable, distributed approach that answers our key question through an iterative reoptimization of the location and the number of facilities within such balls. We show that even for small values of the radius (1 or 2), our distributed approach achieves performance under various synthetic and real Internet topologies that is comparable to that of optimal, centralized approaches requiring full topology and demand information.

Large-Scale Neural Systems for Vision and Cognition

— Consideration of how people respond to the question What is this? has suggested new problem frontiers for pattern recognition and information fusion, as well as neural systems that embody the cognitive transformation of declarative information into relational knowledge. In contrast to traditional classification methods, which aim to find the single correct label for each exemplar (This is a car), the new approach discovers rules that embody coherent relationships among labels which would otherwise appear contradictory to a learning system (This is a car, that is a vehicle, over there is a sedan). This talk will describe how an individual who experiences exemplars in real time, with each exemplar trained on at most one category label, can autonomously discover a hierarchy of cognitive rules, thereby converting local information into global knowledge. Computational examples are based on the observation that sensors working at different times, locations, and spatial scales, and experts with different goals, languages, and situations, may produce apparently inconsistent image labels, which are reconciled by implicit underlying relationships that the network’s learning process discovers. The ARTMAP information fusion system can, moreover, integrate multiple separate knowledge hierarchies, by fusing independent domains into a unified structure. In the process, the system discovers cross-domain rules, inferring multilevel relationships among groups of output classes, without any supervised labeling of these relationships. In order to self-organize its expert system, the ARTMAP information fusion network features distributed code representations which exploit the model’s intrinsic capacity for one-to-many learning (This is a car and a vehicle and a sedan) as well as many-to-one learning (Each of those vehicles is a car). Fusion system software, testbed datasets, and articles are available from http://cns.bu.edu/techlab.

Development of large-scale colloidal crystallisation methods for the production of photonic crystals

Colloidal photonic crystals have potential light manipulation applications including; fabrication of efficient lasers and LEDs, improved optical sensors and interconnects, and improving photovoltaic efficiencies. One road-block of colloidal selfassembly is their inherent defects; however, they can be manufactured cost effectively into large area films compared to micro-fabrication methods. This thesis investigates production of ‘large-area’ colloidal photonic crystals by sonication, under oil co-crystallization and controlled evaporation, with a view to reducing cracking and other defects. A simple monotonic Stöber particle synthesis method was developed producing silica particles in the range of 80 to 600nm in a single step. An analytical method assesses the quality of surface particle ordering in a semiquantitative manner was developed. Using fast Fourier transform (FFT) spot intensities, a grey scale symmetry area method, has been used to quantify the FFT profiles. Adding ultrasonic vibrations during film formation demonstrated large areas could be assembled rapidly, however film ordering suffered as a result. Under oil cocrystallisation results in the particles being bound together during film formation. While having potential to form large areas, it requires further refinement to be established as a production technique. Achieving high quality photonic crystals bonded with low concentrations (<5%) of polymeric adhesives while maintaining refractive index contrast, proved difficult and degraded the film’s uniformity. A controlled evaporation method, using a mixed solvent suspension, represents the most promising method to produce high quality films over large areas, 75mm x 25mm. During this mixed solvent approach, the film is kept in the wet state longer, thus reducing cracks developing during the drying stage. These films are crack-free up to a critical thickness, and show very large domains, which are visible in low magnification SEM images as Moiré fringe patterns. Higher magnification reveals separation between alternate fringe patterns are domain boundaries between individual crystalline growth fronts.

Anatomical identification of extracellularly recorded cells in large-scale multielectrode recordings.

This study combines for the first time two major approaches to understanding the function and structure of neural circuits: large-scale multielectrode recordings, and confocal imaging of labeled neurons. To achieve this end, we develop a novel approach to the central problem of anatomically identifying recorded cells, based on the electrical image: the spatiotemporal pattern of voltage deflections induced by spikes on a large-scale, high-density multielectrode array. Recordings were performed from identified ganglion cell types in the macaque retina. Anatomical images of cells in the same preparation were obtained using virally transfected fluorescent labeling or by immunolabeling after fixation. The electrical image was then used to locate recorded cell somas, axon initial segments, and axon trajectories, and these signatures were used to identify recorded cells. Comparison of anatomical and physiological measurements permitted visualization and physiological characterization of numerically dominant ganglion cell types with high efficiency in a single preparation.

Utilising computational fluid dynamics (CFD) for the modelling of granular material in large-scale engineering processes

In this paper, the framework is described for the modelling of granular material by employing Computational Fluid Dynamics (CFD). This is achieved through the use and implementation in the continuum theory of constitutive relations, which are derived in a granular dynamics framework and parametrise particle interactions that occur at the micro-scale level. The simulation of a process often met in bulk solids handling industrial plants involving granular matter, (i.e. filling of a flat-bottomed bin with a binary material mixture through pneumatic conveying-emptying of the bin in core flow mode-pneumatic conveying of the material coming out of a the bin) is presented. The results of the presented simulation demonstrate the capability of the numerical model to represent successfully key granular processes (i.e. segregation/degradation), the prediction of which is of great importance in the process engineering industry.

The development of an advanced ship evacuation simulation software product and associated large scale testing facility for the collection of human shipboard behaviour data

When designing a new passenger ship or modifying an existing design, how do we ensure that the proposed design and crew emergency procedures are safe from an evacuation point of view? In the wake of major maritime disasters such as the Herald of Free Enterprise and the Estonia and in light of the growth in the numbers of high density, high-speed ferries and large capacity cruise ships, issues concerned with the evacuation of passengers and crew at sea are receiving renewed interest. In the maritime industry, ship evacuation models offer the promise to quickly and efficiently bring evacuation considerations into the design phase, while the ship is "on the drawing board". maritimeEXODUS-winner of the BCS, CITIS and RINA awards - is such a model. Features such as the ability to realistically simulate human response to fire, the capability to model human performance in heeled orientations, a virtual reality environment that produces realistic visualisations of the modelled scenarios and with an integrated abandonment model, make maritimeEXODUS a truly unique tool for assessing the evacuation capabilities of all types of vessels under a variety of conditions. This paper describes the maritimeEXODUS model, the SHEBA facility from which data concerning passenger/crew performance in conditions of heel is derived and an example application demonstrating the models use in performing an evacuation analysis for a large passenger ship partially based on the requirements of MSC circular 1033.

Application of regularization methods to damage detection in large scale plane frame structures using incomplete noisy modal data

This paper presents an approach for detecting local damage in large scale frame structures by utilizing regularization methods for ill-posed problems. A direct relationship between the change in stiffness caused by local damage and the measured modal data for the damaged structure is developed, based on the perturbation method for structural dynamic systems. Thus, the measured incomplete modal data can be directly adopted in damage identification without requiring model reduction techniques, and common regularization methods could be effectively employed to solve the developed equations. Damage indicators are appropriately chosen to reflect both the location and severity of local damage in individual components of frame structures such as in brace members and at beam-column joints. The Truncated Singular Value Decomposition solution incorporating the Generalized Cross Validation method is introduced to evaluate the damage indicators for the cases when realistic errors exist in modal data measurements. Results for a 16-story building model structure show that structural damage can be correctly identified at detailed level using only limited information on the measured noisy modal data for the damaged structure.