Dissertations / Theses on the topic 'Self-organisation'

Autonomic computing suggests the need for dynamic systems that adapt their struc-ture in response to environmental changes in a bottom-up fashion. Such systems can be considered to be composed of agents that self-organise in support of more effective operation. While various self-organisation approaches that aim to meet this need for continuous adaptation have been developed, these typically operate on structures that are not constrained to particular patterns (pipelines, hierarchies). Yet the ubiquitous use of such patterns when structuring task workflows, communication protocols, and traditional organisational design, suggests a need for their preservation when reorganis-ing. In cases where specific patterns, or topologies, result from self-organisation, these are artefacts of the self-organisation mechanism, rather than the underlying topology itself being preserved. In contrast, this thesis explicitly tackles such adaptation, while accommodating the need to preserve topology. The thesis introduces techniques for adapting a system’s structure to improve task throughput, and builds on these techniques, to provide a means of preserving particular topologies. A framework for the reorganisation of defined topologies is introduced, and specific solutions are given for the case of pipelines and hierarchies, which reorganise to improve performance based on application-specific metrics, while preserving topology. Importantly, efficacy is only slightly diminished when topology is maintained, but at the cost of diminished autonomy.

Cette thèse porte sur la fonctionnalisation des métalloarchitectures auto-assemblés, les grilles [2 × 2], avec deux modes distincts: chargé "axial" et neutre "latéral". La base de l'auto-assemblage consiste d'entités bien défini de la gamme des ions métalliques paramagnétiques / diamagnétiques [Fe (II), Co (II), Ni (II), Cu (II), Zn (II), etc] en symétrique très manière. La couche organique est formée par ligands bis-hydrazone aromatiques, et les périphéries sont décorées avec les résidus fonctionnels. Ces récepteurs sont multivalents, bien défini (entre 1. 5 nm), solubles dans les solvants organiques et aqueux. Ils peuvent, en principe, reconnaître les molécules hôtes guests par des interactions supramoléculaires coopératives. En raison de leur nature dynamique à cause de la réversibilité de la liason de coordination, une bibliothèque dynamique combinatoire de ces entités pourrait être utilisée pour explorer des cibles biologiques. Dans une première partie, on décrit la décoration axiale d’architectures grilles [2 × 2]. Le fonctionnalisation de ligands bishydrazone a commencé avec l'alkylation par les haloacyl-acetals. Tentatives diverses ont abouti aux ligands fonctionnalisés par bisaldehydes avec les longueurs de chainons différentes. Tous ont donné les grilles [2 × 2] avec plusieurs métaux de transition. Les propriétés magnétiques de ces complexes ont été étudiées. La possibilité d'avoir hydrazides ou des amines comme des groupes fonctionnels dans les grille [2 × 2] a également été étudiée. Les grilles fonctionnalisés avec groupes urées ont été caracterisées. La reconnaissance d'anions n'a pas réussi en raison de l'instabilité des grilles eux-mêmes. La reconnaissance des oligosaccharides a échoué et en raison de la polarité forte des solvants (acétonitrile), qui empêche la formation de liaisons hydrogènes de manière significative. Décoration de l'amide des grilles par des liens avec des acides aminés comme les groupes fonctionnels sont allés plus facilement. La formation des grilles a été démontrée par RMN et structures cristallines. Enfin, nous avons été en mesure de greffer carboxylique unités carboxyliques sur la grille de base, qui devrait être de grands intérêts comme unité de former un nouveau type de polymère supramoléculaire par la formation de liaisons hydrogènes. Dans une deuxième partie, on décrit la décoration latérale d’architectures grilles [2 × 2]. Contrairement à la décoration axial - dont la plupart des produits sont solubles que dans l'acétonitrile et le nitrométhane, nous avons proposé un type de novo de la décoration. Il s'agit de “l'auto-assemblage” à partir de hydrazides de donner une grille [2 × 2] décor latéralement. En raison de la disponibilité de les hydrazides, par conséquent, la fonctionnalisation est devenu beaucoup plus simple. Nous avons montré que les grilles peut être formé par ions métalliques différents [Fe (II), Co (II), Ni (II), Cu (II), Zn (II), etc] et avec groupes alkyls et aromatiques. Pour explorer les caractéristiques multivalentes des grilles, nous avons intégré des acides aminés ou des segments oligopeptides dans le structure des grilles neutres. Il a été constaté que le groupe amino doivent être protégés. Tous les grilles ont été bien caractérisés. En utilisant les Girard réactifs T / P, grilles neutres [2 × 2] portant huit groupes à charge positive a été généré en une seule étape. L'interaction de ces réseaux multicationic avec espèces multianionic telles que le pyrène tetrasulfonate a été étudiée. En modifiant les contre-anions, les grilles peuvent être solubles dans les deux en solution aqueuse et organique. Huit urées ont été introduites dans une grille [2 × 2] d'agir en tant que pierre angulaire de l’ auto-assemblage hiérarchie de polymères supramoecular par liasons hydrogènes urée-urée. Un gel rigide a été créé en toluène. Différences ont été constatées entre les ligands bis-urées et les grilles portant huit groupes urées: alors que le dernier a formé un joli gel, le premier a donné seulement du gel partiel, même si la concentration est très élevé
This dissertation describes the multifunctionalization of a self-assembled, inorganic, grid-like architecture, namely that of [2×2] grids (Figure), with two distinct modes of substitution: "axial" in charged grids and "lateral" in neutral. The core of the self-assembled entities consists well-defined, highly-symmetrical, tetranuclear arrays of paramagnetic and diamagnetic metal ions [Fe(II), Co(II), Ni (II), Cu(II), Zn(II), etc. ]. In the grids which are the subject of this thesis, an organic sheath is formed by aromatic bis-hydrazone ligands, and their peripheries are variously functionalised. These multivalent receptors are well-defined, ranging in size between 1-5 nm, and can be soluble both in organic and aqueous solvents. They may in principle be able to recognize guest molecules via cooperative supramolecular interactions. Due to their constitutionally dynamic nature arising from the reversibility of the coordinate bonds, a dynamic combinatorial library of these entities might be used to explore biological targets. Chapter 2 describes the axial decoration of [2×2] grid architectures. The functionalization of bishydrazone ligands started with alkylation by haloalkyl-acetals. This resulted in the desired bisaldehyde-functionalized ligands with alkyl chains of two different lengths. Both types gave [2×2] grids with several transition metal ions. The magnetic properties of these metal complexes were investigated. The possibility of having hydrazides or amines as functional groups on the [2×2] grid backbones was also investigated. Grids were also prepared bearing urea substituents functionalised with various alkyl chains. The recognition of anions was not successful due to the instability of the grids. The recognition of oligosaccharides was unsuccessful as well, due to the need to use polar solvents which were efficient competitors for H-bonding. Decoration of the grids with amino acids as functional groups went more smoothly and grid formation was demonstrated by NMR spectroscopy and crystal structure determinations. Finally, we were able to graft carboxylic acid units onto the grid backbone, potentially of interest for the formation of a new type of supramolecular polymer via hydrogen bonding. Chapter 2 and Chapter 3 describe the lateral decoration of [2×2] grid architectures. Unlike the axial decoration of the charged grids, this can be achieved very conveniently by a template reaction involving acid hydrazides and a dialdehyde in the presence of the desired metal ion. The ready availability of many different hydrazides made this a very straightforward method of obtaining laterally decorated grids. We demonstrated that grids derived from numerous functionalised alkyl and aromatic acid hydrazides can be formed by various metal ions [Fe(II), Co(II), Ni (II), Cu(II), Zn(II), etc. ]. Using Girard T/P reagents, neutral [2×2] grid cores bearing eight positively charged groups were generated in one step. The interaction of these multicationic grids with multianionic species such as pyrene tetrasulfonate was investigated. By changing the counter anions, the grids can be made soluble both in aqueous and organic solutions. To explore the multivalency of grids, we also incorporated amino acid and oligopeptide segments into the neutral grid framework. Provided the amino group was first protected, well-characterised grids were readily obtained. Eight urea substituents were attached to a [2×2] grid to act as building block for the hierarchical self-assembly of a supramoecular polymer via urea-urea hydrogen bonding. A rigid gel was formed with toluene. Distinct differences were found between the bis-urea bearing ligand and the grid bearing eight urea groups: while the latter readily produced gelation, the former, even at very high concentrations, was almost ineffective. Lateral crown ether substituents were introduced to form tweezer-like receptors for diammonium ions. The eight units arrayed on the grid proved to give effectively a four-armed tweezer

Nowadays, the rapid growth of mobile communications is changing the world towards a fully connected society. Current 4G networks account for almost half of total mobile traffic, and in the forthcoming years, the overall mobile data traffic is expected to dramatically increase. To manage this increase in data traffic, operators adopt network topologies such as Heterogeneous Networks. Thus, operators can de­ ploy hundreds of small cells for each macro cell, allowing them to reduce coverage hales and/or lack of capacity. The advent of this technology is expected to tremendously increase the number of nodes in this new ecosystem, so that traditional network management activities based on, e.g., classic manual and field trial design approaches are just not be viable anymore. As a consequence, the academic J literature has dedicated a significant amount of effort to Self-Organising Network (SON) algorithms. These solutions aim to bring intelligence and autonomous adaptability into cellular networks, thereby reducing capital and operation expenditures (CAPEX/OPEX). Another aspect to take into account is that, these type of networks generate a large amount of data during their normal operation in the form of control, management and data measurements. This data is expected to increase in SG due to different aspects, such as densification, heterogeneity in layers and technologies, additional control and management complexity in Network Functions Virtualisation (NFV) and Software Defined Network (SDN), and the advent of the Internet of Things (loT), among others. In this context, operators face the challenge of de ­ signing efficient technologies, while introducing new services, reaching challenges in terms networks, which are self-aware, self-adaptive, and intelligent. This dissertation provides a contribution to the design, analysis, and evaluation of SON solutions to improve network opera tor performance, expenses, and users' experience, by making the network more self-adaptive and intelligent. It also provides a contribution to the design of a self-aware network planning tool, which allows to predict the Quality of Service (QoS) offered to end-users, based on data al ­ ready available in the network . The main thesis contributions are divided into two parts. The first part presents a novel functional architecture based on an automatic and self-organised Reinforcement Learning (RL) based approach to model SON functionalities, in which the main task is the self-coordination of different actions taken by different SON functions to be automatically executed in a self-organised realistic Long Term Evolution (LTE) network. The proposed approach introduces a new paradigm to deal with the conflicts genera ted by the concurrent execution of multiple SON functions, revealing that the proposed approach is general enough to modelali the SON functions and their derived conflicts. The second part of the thesis is dedicated to the problem of QoS prediction. In particular, we aim at finding patterns of knowledge from physical layer data acquired from heterogeneous LTE networks. We propose an approach that not only is able to verify the QoS level experienced by the users, through physical layer measurements of the UEs, but it is a lso able to predict it based on measurements collected at different time, and from different regions of the heterogeneous network. We propose then to make predictions independently of the physical location, in order to exploit the experience gained in other sectors of the network, to properly dimension and deploy heterogeneous nodes. In this context, we use Machine Learning (ML) as a tool to allow the network to learn from experience, improving performances, and big data analytics to drive the network from reactive to predictive.
Hoy en día, el rápido crecimiento de las comunicaciones móviles está cambiando el mundo hacia una sociedad completamente conectada. Las redes 4G actuales representan casi la mitad del tráfico móvil total, y en los próximos años se espera que el tráfico total de los dispositivos móviles aumente drásticamente. Para gestionar este incremento de tráfico de datos, los operadores adoptan tecnologías de redes como las redes heterogéneas. De esta manera, los operadores pueden desplegar centena res de pequeñas celdas por cada macro celda, permitiendo reducir zonas sin cobertura y/o falta de capacidad. Con la introducción de esta tecnología, se espera que incremente de manera sustancia l el número de nodos en el nuevo ecosistema, de manera que las actividades de gestión de las redes tradicionales, basadas en, por ejemplo, el diseño manual, sean inviables. Como consecuencia, la literatura académica ha dedicado un esfuerzo significativo al diseño de algoritmos de redes auto-organizadas (SON). Estas soluciones tienen como objetivo introducir inteligencia y capacidad autónoma a las redes móviles, reduciendo la capacidad y costes operativos. Otro aspecto a tener en cuenta es que este tipo de redes generan una gran cantidad de datos durante su funcionamiento habitual, en forma de medidas de control y gestión de datos. Se espera que estos datos incrementen con la tecnología SG, debido a diferentes aspectos como los son la densificación de redes heterogéneas, la complejidad adicional en el control y la gestión de la virtualización de las funciones de redes (NFV) y las redes definidas por software (SON), así como la llegada del internet de las cosas (loT), entre otros. En este contexto, los operadores se enfrentan al reto de diseñar tecnologías eficientes, mientras introducen nuevos servicios, consiguiendo objetivos en términos de satisfacción del cliente, en donde el objetivo global del operador es la construcción de redes auto-conscientes, auto-adaptables e inteligentes. Esta tesis ofrece una contribución al diseño y evaluación de soluciones SON para mejorar el rendimiento de las redes, los costes y la experiencia de los usuarios, consiguiendo que la red sea auto-adaptable e inteligente. Así mismo, proporciona una contribución al diseño de una herramienta de planificación de red auto-consciente, que permita predecir la calidad de servicio brindada a los usuarios finales, basada en la explotación de datos disponibles en la red.
Avui en dia, el ràpid creixement de les comunicacions mòbils està canviant el món cap a una societat completament connectada. Les xarxes 4G actuals representen casi la m trànsit mòbil total, i en els propers anys s’espera que el trànsit total de dades mòbils augmenti dràsticament. Per gestionar aquest increment de trànsit de dades, els operadors adopten topologies de xarxa com ara les xarxes heterogènies (HetNets). D’aquesta manera, els operadors poden desplegar centenars de cel·les petites per a cada cella macro, permetent reduir forats en la cobertura i/o la manca de capacitat. Amb l’arribada d’aquesta tecnologia, s’espera que incrementi enormement el nombre de nodes en el nou ecosistema, de manera que les activitats de gestió de xarxa tradicionals, basades en, per exemple, el disseny manual i els assaigs de camp esdevenen simplement inviables. Com a conseqüència, la literatura acadèmica ha dedicat una quantitat significativa d’esforç als algorismes de xarxa auto organitzada (SON). Aquestes solucions tenen com a objectiu portar la intel·ligència i capacitat d’adaptació autònoma a les xarxes mòbils, reduint el capital i les despeses operatives (CAPES/OPEX). Un altre aspecte a tenir en compte és que aquest tipus de xarxes generen una gran quantitat de dades durant el seu funcionament habitual, en forma de mesuraments de control, gestió i dades. S’espera que aquestes dades incrementin amb la tecnologia 5G, degut a diferents aspectes com ara la densificació, l’heterogeneïtat en capes i tecnologies, la complexitat addicional en el control i la gestió de la virtualització de les funcions de xarxa (NFV) i xarxes definides per software (SDN), i l’adveniment de la internet de les coses (IoT), entre d’altres. En aquest context, els operadors s’enfronten al repte de dissenyar tecnologies eficients, mentre introdueixen nous serveis, aconseguint objectius en termes de satisfacció del client, i on l’objectiu global d’un operador és la construcció de xarxes que són autoconscients, auto-adaptables i intel·ligents. Aquesta tesis ofereix una contribució al disseny, l’anàlisi i l’avaluació de les solucions SON per millorar el rendiment de l’operador de xarxa, les xi despeses i l’experiència dels usuaris, fent que la xarxa sigui més auto-adaptable i intel·ligent. També proporciona una contribució al disseny d’una eina de planificació de xarxa autoconscient, el que permet predir la qualitat de servei (QoS) oferta als usuaris finals, basada en dades ja disponibles a la xarxa. Les contribucions principals d’aquesta tesis es divideixen en dues parts. La primera part presenta una nova arquitectura funcional basada en un aprenentatge per reforç (RL) automàtic i auto-organitzat, enfocat en modelar funcionalitats SON, on la tasca principal és l’auto-coordinació de les diferents accions dutes a terme perles diferents funcions SON a ser executades de forma automàtica en una xarxa Long Term Evolution (LTE) auto-organitzada. L’enfocament proposat introdueix un nou paradigma perfer front als conflictes generats per l’execució simultània de múltiples funcions SON, revelant que l’enfocament proposat és prou general per modelar totes les funcions SON i els seus conflictes derivats. La segona part de la tesis està dedicada al problema de la predicció de la qualitat de servei. En particular, el nostre objectiu és trobar patrons de coneixement a partir de dades de la capa física adquirides de xarxes LTE heterogènies. Proposem un enfocament que no només és capaç de verificar el nivell de QoS experimentat pels usuaris, a través de mesuraments de la capa física dels UEs, sinó que també és capaç de predir-ho basant-se en mesuraments adquirits en diferents instants, i de diferents regions de la xarxa heterogènia. Proposem per tant fer prediccions amb independència de la ubicació física, aprofitant l’experiència adquirida en altres sectors de la xarxa, per dimensionar i desplegar nodes heterogenis correctament. En aquest context, utilitzem l’aprenentatge automàtic (ML) com a eina per permetre que la xarxa aprengui de l’experiència, millorant el rendiment, i l’anàlisi de grans volums de dades per a conduir la xarxa de reactiva a predictiva. Durant l’elaboració d’aquesta tesis, s’han extret dues conclusions principals clau. En primer lloc, destaquem la importància de dissenyar algorismes SON eficients per fer front eficaçment a diversos reptes, com ara la ubicació més adequada de funcions SON i algorismes per resoldre adequadament el problema d’implementació distribuïda o centralitzada, o la solució de conflictes entre funcions SON executades a diferents nodes o xarxes. En segon lloc, en termes d’eines de planificació de xarxes, es poden trobar diferents eines cobrint una àmplia gamma de sistemes i aplicacions orientades a la indústria, així com per a fins d’investigació. En aquest context, les solucions investigades són sotmeses contínuament a canvis importants, on un del principals impulsors és presentar solucions més rentables

Active matter theory studies the collective behaviour of self-propelled organisms or objects. Although the field has made great progress in the past decade, little is known of the role played by confinement and surfaces. This thesis analyses the self-organisation of dense bacterial suspensions in three different microchambers: flattened drops, racetracks and lattices of cavities. Suspensions of swimming bacteria are well-known to spontaneously form macroscopic quasi-turbulent patterns such as jets and swirls. Confinement inside flattened drops and racetracks stabilises their motion into a spiral vortex and wavy streams, respectively. We have quantitatively measured and analysed bacterial circulation and discovered cells at the interfaces to move against the bulk. To understand this phenomenon, we developed a method able to measure simultaneously the directions of swimming and of motion. Experiments in drops reveal that cells align in a helical pattern, facing outward and against the main bulk circulation. Likewise, bacteria in racetracks share a biased orientation against the overall stream. Particle-based simulations confirm these results and identify hydrodynamic interactions as the main driving force: bacteria generate long-range fluid flows which advect the suspension in the bulk against its swimming direction, resulting in the double-circulation pattern. We have finally injected dense suspensions of bacteria into lattices of cavities. They form a single vortex in each cavity, initially spinning clockwise or counterclockwise with equal probabilities. Changing the topology of the lattice and the geometry of connections between cavities allows us to control the lattice state (random, ferromagnetic, antiferromagnetic, or unstable). Edge currents along interfaces and connections appear to determine the lattice organisation. We finally propose an Ising model to understand experimental results and estimate Hamiltonian and interactions parameters. This work opens new perspectives for the study of active matter and, we hope, will have a great impact on the field.

Self-organisation and natural selection are fundamental in shaping the natural world. Substantial progress in understanding how these two forces interact as biological systems evolve has been made through the study of abstract models, for example by evolving boolean networks on computers. Further progress may be made by identifying a model system in which the interaction between self-organisation and selection can be investigated empirically. To this end, we investigate how the self-organising thermoregulatory huddling behaviours displayed by rodents might influence natural selection of the genetic components of metabolism. By applying a simple evolutionary algorithm to a simplistic description of self-organising thermoregulation huddling, we arrive at a clear albeit counterintuitive prediction: Animals able to huddle together in cold environments should evolve an increased thermal conductance at a faster rate than animals reared in isolation. According to the model, within-lifetime adaptation (self organising huddling) is able to guide the evolution of complementary between-lifetime adaptation (natural selection of thermoregulatory genes). Confirmation of these predictions in future experiments would constitute strong evidence of a mechanism by which self-organisation can guide natural selection.

Self-organisation describes emergence of global order from local interactions between components of a system without supervision by external directing forces.This decentralised mode of decision making is central to social phenomena such as swarm behaviour in bacteria, flocking of birds, and schooling of fish. Biological self-organisation also governs morphogenic dynamics during development of multi-cellular organisms. By self-organisation, dichotomies such as proliferation/differentiation are resolved based on simple interactions of cells. During neural morphogenesis, for example, self-organisation cues instruct temporal commitment to differentiation of neural progenitor cells and sub-lineage differentiation outcomes. In this thesis, the identity of molecular switches that orchestrate human neural self-organisation is investigated. In the first data chapter, a primary cellular model of human adult neurogenesis is developed that is not confounded by noise inherent to cell lines. The model is based on direct trans-differentiation of human microvascular pericytes to functional interneurons without exogenous interference. Due to lack of transitional cellular forms, such as transient amplifying cells, the model system demonstrated minimal temporal noise and high fidelity. In the second data chapter of the thesis, the signalling activity of a novel morphogenic microRNA is described that can access and override the self-organisation program of human neural progenitors. The microRNA interacts with an ancient cascade involved in detection of metabolic stressors. By downregulating a key component of the cascade, cdk-18,the miRNA invokes a faux stress response that impacts upon availability of free-cytoplasmic catenin-β1 and synchronises the cycling neural progenitors and also instructs their differentiation fate. In the third data chapter, evolutionary origin of the microRNA is investigated in order to gain insight into the interface of the microRNA and Wnt/catenin-β1 signalling cascades during emergence of self-organisation. Finally, in the last data chapter it is demonstrated that catenin-β1 encodes the self-organisation lexicon of human neural progenitors by coupling the cell cycle of individual cells. By manipulating the subcellular localisation of catenin-β1 and hence the strength of coupling, spatial organisation and differentiation propensity of neural progenitors is reprogrammed.

The rapid growth in mobile communications due to the exponential demand for wireless access is causing the distribution and maintenance of cellular networks to become more complex, expensive and time consuming. Lately, extensive research and standardisation work has been focused on the novel paradigm of self-organising network (SON). SON is an automated technology that allows the planning, deployment, operation, optimisation and healing of the network to become faster and easier by reducing the human involvement in network operational tasks, while optimising the network coverage, capacity and quality of service. However, these SON autonomous features cannot be achieved with the current drive test coverage assessment approach due to its lack of automaticity which results in huge delays and cost. Minimization of drive test (MDT) has recently been standardized by 3GPP as a key self- organising network (SON) feature. MDT allows coverage to be estimated at the base station using user equipment (UE) measurement reports with the objective to eliminate the need for drive tests. However, most MDT based coverage estimation methods recently proposed in literature assume that UE position is known at the base station with 100% accuracy, an assumption that does not hold in reality. In this work, we develop a novel and accurate analytical model that allows the quantification of error in MDT based autonomous coverage estimation (ACE) as a function of error in UE as well as base station (user deployed cell) positioning. We first consider a circular cell with an omnidirectional antenna and then we use a three-sectored cell and see how the system is going to be affected by the UE and the base station (user deployed cell) geographical location information errors. Our model also allows characterization of error in ACE as function of standard deviation of shadowing in addition to the path-loss.

Since the first realization of a laser source in 1960, tremendous progresses have been made in the theoretical understanding and experimental control of interacting atomic-optical systems. Optical fields can nowadays be used to engineer long-range interactions in cold atomic gases, manipulating the external degrees of freedom of the atoms via optical forces. This opens the possibility for the study of highly controllable and tunable long-range interacting systems, in which a complex dynamics for the motional properties of the gas can arise due to the effective atom-atom coupling induced by the field. In this thesis the spontaneous emergence of spatial structures in non-equilibrium atom-optical systems is theoretically and numerically investigated, for different geometries and physical configurations. Extending previous research in hot atomic gases, self-organising instabilities involving the external degrees of freedom are studied, and in contrast to other cold-atom spatial instabilities the spontaneous breaking of continuous symmetries is predicted. The main focus of the work presented in this thesis is on dynamical instabilities in cold gases. However, connections are found with other fields of nonlinear physics, such as synchronisation of coupled oscillators and phase transitions in many-body systems. Part of the research presented here has been conducted in the context of a collaboration with the Photonics group at Strathclyde and the Institut non Linéaire de Nice, in which experimental observations of self-organisation and continuous symmetry breaking were obtained.

Self-organising processes occurring in isotropic turbulence and homogeneous magnetohydrodynamic (MHD) turbulence are investigated in relation to the stability of helical flow structures. A stability analysis of helical triad interactions shows that compared to hydrodynamics, equilibria of the triadic evolution equations have more instabilities with respect to perturbations on scales larger than the characteristic scale of the system. Some of these instabilities can be mapped to Stretch-Twist-Fold dynamo action and others to the inverse cascade of magnetic helicity. High levels of cross-helicity are found to constrain small-scale instabilities more than large scale instabilities and are thus expected to have an asymmetric damping effect on forward and inverse energy transfer. Results from a numerical investigation into the influence of helicity on energy transfer and dissipation are consistent with this observation. The numerical work also confirms the predictions of an approximate method describing the Reynolds number dependence of the dimensionless dissipation coefficient for MHD turbulence. These predictions are complemented by the derivation of mathematically rigorous upper bounds on the dissipation rates of total energy and cross-helicity in terms of applied external forces. Large-scale helical flows are also found to emerge in relaminarisation events in direct numerical simulations of isotropic hydrodynamic turbulence at low Reynolds number, where the turbulent fluctuations suddenly collapse in favour of a large-scale helical flow, which was identified as a phase-shifted ABC-flow. A statistical investigation shows similarities to relaminarisation of localised turbulence in wall-bounded parallel shear flows. The turbulent states have an exponential survival probability indicating a memoryless process with a characteristic lifetime, which is found to depend super-exponentially on Reynolds number akin to well-established results for pipe and plane Couette flow. These and further similarites suggest that the phase space dynamics of isotropic turbulence and wall-bounded shear flows are qualitatively similar and that the relaminarisation of isotropic turbulence can also be explained by the escape from a chaotic saddle.

Mobile telecommunications networks based on Long Term Evolution (LTE) technology promise faster throughput to their users. LTE networks are however susceptible to a phenomenon known as inter-cell interference which can greatly reduce the throughput of the network causing unacceptable degradation of performance for cell edge users. A number of approaches to mitigating or minimising inter-cell interference have been presented in the literature such as randomisation, cancellation and coordination. The possibility of coordination between network nodes in an LTE network is made possible through the introduction of the X2 network link. This thesis explores approaches to reducing the effect of inter-cell interference on the throughput of LTE networks by using the X2 link to coordinate the scheduling of radio resources. Three approaches to the reduction of inter-cell interference were developed. Localised organisation is a centralised scheme in which a scheduler is optimised by a Genetic Algorithm (GA) to reduce interference. Networked organisation makes use of the X2 communications link to enable the network nodes to exchange scheduling information in a way that lowers the level of interference across the whole network. Finally a more distributed and de-centralised approach is taken in which each of the network nodes optimises its own scheduling in coordination with its neighbours. An LTE network simulator was built to allow for experimental comparison between these techniques and a number of existing approaches and to serve as a test bed for future algorithm development. These approaches were found to significantly improve the throughput of the cell edge users who were most affected by intereference. In particular the networked aspect of these approaches yielded the best initial results showing clear improvement over the existing state of the art. The distributed approach shows significant promise given further development.

The article reviews water-soluble polymers characterized by surfactant side chains, and related amphiphilic polymers. Various synthetic approaches are presented, and rules for useful molecular architectures are given. Models for the self-organization of such polymers in water are presented comparing them with the micellization of low molecular weight surfactants. Highlighting key properties of aqueous polysoap solutions such as viscosity, surface tension and solubilization power, some structure-property relationships are established. Further, the formation of mesophases and of superstructures in bulk is addressed. Finally, the functionalization of polysoaps, and potential applications are discussed.

The increasing market fluctuations and customized products demand have dramatically changed the focus of industry towards organizational sustainability and supply chain agility. Such critical changes in the strategic vision of the companies inevitably have a direct impact on the shop-floor operational requirements. In this sense, traditional shop-floor approaches are becoming increasingly inadequate leading to the adoption of more pluggable and reusable solutions. The emergence o modern manufacturing paradigms translates the effort undertaken by the academia in order to provide the required background to support the implementation of such distributed mechatronic systems. Biological systems, due to their similar distributed network-like structure, represent naturally a common analogy and source of inspiration for such distributed modular approaches. Hence, modern manufacturing paradigms usually rely on complexity science biologically inspired concepts to attain distributed control, adaptability, evolution, flexibility and robustness as core concepts. This originated the implementation of a number of different multi-agent based architectures. Nevertheless, with time the majority of the these implementation efforts left behind most of the bio-inspired concepts resulting in simple distributed approaches with considerable limitations regarding scalability, reconfigurability and distributed problem resolution. Particularly under the scope of Evolvable Production System (EPS) the implementation of self-organising mechanisms based on negotiation interaction protocols and dynamic coalition-based hierarchical complexity, have considerable hindered the system performance and limited the full exploitation of the paradigm potential. In this context, this licentiate thesis is focused on the development of a self-organising manufacturing systems that holistically mimics the main structural and regulatory principles followed by natural systems. For this purpose, the present approach was designed as opposed to the current tendency followed by modern productions approaches, in which the product holds the production knowledge and is responsible for the management of its own production. Instead, the production knowledge was reduced to the minimum and distributed over the manufacturing components. Self-organising principles heavily inspired on the regulatory mechanisms of biological systems, were then devised to regulate the critical control mechanisms of the manufacturing system. Hence, similarly to the natural world the characteristics and the system overall production emerge as consequence of the micro-dynamics of the systems. In this way, it becomes therefore possible to attain a system that is not only highly reconfigurable and scalable but also able to distributively tackle the manufacturing processes. Although the present work has been developed under the Evolvable Production System context, the introduced approach can be easily adapted to a wider range of modular networked-based systems.

QC 20131001

Internal Models (IMs) play a significant role in autonomous robotics. They are mechanisms able to represent the input-output characteristics of the sensorimotor loop. In developmental robotics, open-ended learning of skills and knowledge serves the purpose of reaction to unexpected inputs, to explore the environment and to acquire new behaviours. The development of the robot includes self-exploration of the state-action space and learning of the environmental dynamics. In this dissertation, we explore the properties and benefits of the self-organisation of robot behaviour based on the homeokinetic learning paradigm. A homeokinetic robot explores the environment in a coherent way without prior knowledge of its configuration or the environment itself. First, we propose a novel approach to self-organisation of behaviour by artificial curiosity in the sensorimotor loop. Second, we study how different forward models settings alter the behaviour of both exploratory and goal-oriented robots. Diverse complexity, size and learning rules are compared to assess the importance in the robot’s exploratory behaviour. We define the self-organised behaviour performance in terms of simultaneous environment coverage and best prediction of future sensori inputs. Among the findings, we have encountered that models with a fast response and a minimisation of the prediction error by local gradients achieve the best performance. Third, we study how self-organisation of behaviour can be exploited to learn IMs for goal-oriented tasks. An IM acquires coherent self-organised behaviours that are then used to achieve high-level goals by reinforcement learning (RL). Our results demonstrate that learning of an inverse model in this context yields faster reward maximisation and a higher final reward. We show that an initial exploration of the environment in a goal-less yet coherent way improves learning. In the same context, we analyse the self-organisation of central pattern generators (CPG) by reward maximisation. Our results show that CPGs can learn favourable reward behaviour on high-dimensional robots using the self-organised interaction between degrees of freedom. Finally, we examine an on-line dual control architecture where we combine an Actor-Critic RL and the homeokinetic controller. With this configuration, the probing signal is generated by the exertion of the embodied robot experience with the environment. This set-up solves the problem of designing task-dependant probing signals by the emergence of intrinsically motivated comprehensible behaviour. Faster improvement of the reward signal compared to classic RL is achievable with this configuration.

The phytohormone auxin plays a key role in many plant developmental processes. Its polar cell-to-cell transport is linked to and dependent on auxin efflux transporters and their polar localisation in cell membranes. This relies on feedback loops between auxin and its transport on many levels. Hypotheses brought forward in auxin biology, trying to elucidate the nature of these feedbacks, such as the canalisation hypothesis, depend on mechanisms by which auxin transport is established and maintained on specific routes through tissues. Auxin transport canalisation is based on a proposed feedback between auxin flux and auxin transport polarisation, with the result that auxin transport is directed by the strength of auxin fluxes. Despite this phenomenon being well described in biology, its underlying mechanisms are largely unknown. Many of them are occurring at the cell level, justifying a focus on cells in elucidating the nature of this feedback. In this thesis, computational modelling of self-organising mechanisms potentially leading to such phenomena at a cell level has been accomplished. While many auxin transport models are already available at tissue and whole plant scales, such a single cell model is a novel contribution. With the main focus on auxin/proton interactions grounded on the results of biological experiments, a feedback by which auxin influences its own transport by the activation of plasma membrane-bound proton pumps is described. In simulations, it is shown to lead to increased allocation of auxin in cells as well as to enhancement of all auxin transport fluxes over the membrane, and in due course to the establishment of canalisation-type polarisation patterns, without polarised transporter localisation. The results point towards a functional redundancy of polarisation in auxin transport and lead to hypotheses on differential energisation of auxin transporters, which may play a role in auxin transport polarisation events.

Une rivière alluviale s'écoule sur une épaisse couche de sédiments. Lorsqu'elle construit son lit, elle entraîne, transporte et dépose des sédiments, façonnant ainsi sa propre forme. Ainsi, le couplage entre l'écoulement et le transport sédimentaire régit la taille et la forme de la rivière. Dans cette thèse, nous étudions l'influence du transport sédimentaire sur la forme et la stabilité d'une rivière alluviale. Pour ce faire, nous reproduisons des rivières en laboratoire en laissant s'écouler un liquide visqueux sur un lit granulaire. L'aspect du chenal ainsi formé dépend des débits de liquide et de sédiment injectés en entrée. A l'aide de ces expériences, nous mettons en évidence les deux mécanismes qui contrôlent l'équilibre d'une rivière. D'abord, la gravité entraîne les grains vers le centre du chenal. Ce mécanisme érode continuellement les berges de la rivière, et tend donc à l'élargir. Cependant, les collisions d'un grain avec le lit dévient sa trajectoire dans la direction transverse à l'écoulement. Les grains se comportent ainsi comme des marcheurs aléatoires, qui, collectivement, diffusent vers les berges de la rivière. A l'équilibre, cette diffusion compense la gravité, et fixe ainsi la forme de la rivière. Lorsque la diffusion prend le dessus sur la gravité, elle peut induire une instabilité. En effet, si on perturbe un lit sédimentaire avec des stries longitudinales, le cisaillement fluide est plus faible là où l'écoulement est moins profond. Par conséquent, les grains diffusent depuis les creux de la perturbation vers ses crêtes. Cette rétroaction déstabilisante pourrait générer de nouveaux chenaux, et expliquer la formation des rivières en tresses
An alluvial river builds its own bed with the sediment it transports. The channel bounds the flow, which in turns deforms the channel through erosion and deposition. This coupling between flow and sediment transport selects the shape and the size of the river. In this manuscript, we investigate it using laboratory experiments. The first ingredient of this coupling is gravity, which pulls the moving grains towards the center of the channel, thus continually eroding the banks. However, due to the roughness of the bed, the trajectory of a moving grain fluctuates across the stream. The bedload layer is therefore a collection of random walkers which diffuse towards the less active areas of the bed. In a river at equilibrium, this diffusion counteracts gravity to maintain the banks. When gravity and diffusion are out of balance, their interaction causes an instability. Indeed, if an initially flat bed of sediment is perturbed with longitudinal streaks, the flow-induced shear stress is weaker where the flow is shallower. Therefore, bedload diffusion induces a sediment flux towards the crests of the perturbation. This positive feedback induces an instability which can generate new channels. We suggest that this mechanism could initiate the braiding of alluvial rivers

In this thesis we develop computational techniques for modelling molecular selforganisation. After a short review of the current nanotechnological applications of molecular self-assembly and the main problems encountered in modelling the selforganised behaviour of chemical systems, we introduce a set of methods, from both chemistry and complexity science, for the prediction of self-assembled structures, with particular focus on Monte Carlo (MC) based methods. We apply the MC method to two systems of experimental interest. First we model the silica nanoparticles on the surface of spherical polystyrene latex droplets, synthesised by the S. Bon Group at the University of Warwick, as a set of soft spheres on a spherical surface, to study their packing patterns as a function of the broadening of the nanoparticle size distribution. Then we develop a hexagonal lattice model for the study of the two-dimensional self-organisation of planar molecules capable of complementary interactions, to study their phase diagrams as a function of the strength of their complementary interactions and bonding motif. In both cases, the phases are characterised using a number of order parameters. We show that these simplified models are able to reproduce the experimental observations. We then develop an Agent Based (AB) algorithm, traditionally used for the study of complex systems, for the modelling of molecular self-organisation. In this algorithm, an agent is identified with a stable portion of the system under investigation. The agents can then evolve following a set of rules which include elements of adaptation (new configurations induce new types of moves) and learning (past successful choices are repeated), in order to drive the system towards its lowest energy configuration. We first apply the method to the study of the packing of a set of idealised shapes, then we extend it to the study of a realistic system. The latter is achieved by linking the AB algorithm to an available molecular mechanics code, in order to calculate the interaction energies of atomistic models. In both cases we compare the AB result with that of MC based methods, showing that for all the systems studied, the AB method consistently finds significantly lower energy minima than the MC algorithms in less computing time. Finally, we show how the AB algorithm can be used as a part of the protocol to calculate the phase diagram of a rigid organic molecule (1,4-benzene-dicarboxylic acid or TPA) with less computational effort than standard techniques.

Convection is a well-observed atmospheric phenomenon, which has a fundamental role in global weather and climate. The transport of heat, moisture and momentum that result from convection are significant at a range of temporal and spatial scales. Convective clouds have a wide range of non-linear interactions with other atmospheric processes which make them difficult to understand and model numerically. Due to resolution constraints in climate models, sub-grid convection is represented by a parameterisation. Many parameterisation schemes are based on an assumed temporal and spatial separation between convection and its forcing. The implied equilibrium relates current convection directly to the large-scale forcing. The validity of the temporal scale separation is directly tested in this thesis by examining the convective response to a time-varying forcing.

Many of the great challenges of our age are ultimately challenges of complexity. Anthropogenic climate change threatens to destabilise the ecological connections that span the biosphere as well as the delicate geopolitical functioning of our globalised civilisation. It seems a push towards developing an understanding of the seemingly unpredictable behaviour of such highly connected complex systems could hardly come at a more critical time. It is the spontaneous emergence of macroscopic phenomena from these complex systems that forms the core of this thesis. Complexity generally results from the multiplicity of system components and is ubiquitous in natural systems. Developing an understanding of how macroscopic structures emerge from microscopic dynamics is central to furthering our understanding of complex systems. However, there exists a gulf between our capacity to formulate complex models and our ability to predict their behaviour. Additionally, large and complicated models require vast computational resources and will remain out of reach for many years, even when considering Moore's law, the observation that the increase of computing power is approximately exponential with time. This thesis contributes to building the bridge between model formulation and prediction with multiple directions of attack. We begin by analysing Watson and Lovelock's (1983) Daisyworld model by making explicit the relationship between the various model time scales, laying the foundation for our own conceptual ecosystem model with an arbitrarily diverse biota and a multidimensional environment. We make the important step of providing a framework to translate the model formulation and determine the collective behaviour of the biota. Cellular automata and networks are methods for modelling discrete complex systems and are widely considered a staple of the complex systems simulator's toolbox. We go on to develop the analytical tools available to these models with a real space renormalisation of two dimensional non-equilibrium cellular automata along with demonstrating the utility of a master equation approach by analysing the properties of a new algorithm for growing complex networks. Lastly we investigate the utility of thermodynamic principles in predicting self-organisation in dissipative systems.

Spikes are an important part of information transmission between neurons in the brain. There is biological evidence (such as rapid information processing in the electro sensory system of electric fish, the auditory system of echo-locating bats, and the visual system of flies) to prove the use of the precise timing of spikes for information coding, rather than only the firing rate. Recent research has shown the potential capability of spiking neural networks to model complex information processing in the brain. However, the exact learning mechanism in which the neuron is trained to fire at precise times remains an open problem. The aim of this project is to design novel biologically plausible learning algorithms for spiking neuron models at the neuron and the network levels. The first learning algorithm proposed is called BPSL (Biologically Plausible Supervised Learning algorithm), and it uses Spike Timing Dependent Plasticity (STOP) to adjust the weights. Additionally, the method sets strong teacher inputs to drive the neuron response and to prevent the silent neuron problem. Furthermore, local dendritic depolarization impacts the STOP in BPSL. A second approach called DL-ReSuMe, Delay Learning based Remote Supervised Method, is then proposed to merge weight adjustment and a delay shift approach to enhance learning performance and its biological plausibility. The weights are adjusted by LTP, Long-Term Potentiation, characteristic of BPSL and LTD, Long-Term Depression, of anti-STOP. The delay learning property helps DL-ReSuMe to solve the silent window problem. Then a third approach called EDL, Extended DL-ReSuMe, is proposed. Delays of appropriate groups of excitatory and inhibitory inputs are adjusted multiple times to find more accurate values for the delays. The delays and weights are adjusted cooperatively to construct a stable learning method. Subsequently, an approach termed Multi-DL-ReSuMe is proposed to extend DL-ReSuMe to a learning algorithm for a layer of spiking neurons for classification of spatiotemporal input patterns. The number of neurons in Multi-DL-ReSuMe is also increased to the number of the classes to improve its accuracy compared to a single neuron. Finally, a supervised learning algorithm for multilayer SNNs to train precisely the timing of multiple spikes is proposed. Weights of hidden and output neurons are adjusted in parallel to train the output neurons to fire at desired times. The delays of the output neurons are also adjusted in cooperation with the weight adjustments. EDL is used in the learning procedure of the multilayer network. Interactions between different layers of the network are governed through a biofeedback signal sent back by the output neurons. It uses another training procedure to prevent misclassification. During a misclassification, the learning algorithm adjusts the network parameters to train it to not fire close to the spikes that causes the misclassifications. The classification learning enables the proposed method to overcome the difficulty of learning of real world data. Simulation results are presented for all the proposed approaches to validate the propositions and benchmark against the current state of the art methods.

Self-organisation is an inherent mechanism in all laser systems although it is often overlooked. The self formation of spatial and spectral modes, competition between modes and much spatio-temporal dynamics are driven by the intrinsic non-linearity of saturable gain in the laser amplifier medium coupled with feedback from a resonator structure. It is highly insightful to consider the growth, extinction and competition of modes in a laser system as an evolving ecosystem with modes as species growing by stimulated emission in the gain medium, decaying by intracavity and output coupling losses, and having to compete for the common, but finite, resource of gain which is supplied by an external excitation source. The outcome of species competition is a survival-of-the-fittest that determines the final steady-state output or dynamical set of modes that can continue to persist. This thesis presents investigations into the design solid-state laser systems which utilise the inherent dynamics of optical fields and gain media in order to self-organise the system to operate in a desirable manner. The Nd:YVO4 bounce geometry laser amplifier is employed throughout this thesis. A numerical investigation of the thermally induced lensing within the laser crystal is reported. Optimisation of the geometry parameters is explored as well as investigation into future developments, such as the utilisation of an additional sapphire crystal to directly cool the laser crystal pump face. This is shown to theoretically reduce the horizontal thermally induced lens strengths by a factor of 4. Single longitudinal mode single longitudinal mode (SLM) ring lasers where the unidirectionality is imposed either by an extra-cavity ‘parasitic’ pass of the gain media or by retro-reflection of one of the two outputs are investigated. SLM TEM00 output powers of up to 20W are demonstrated without the need for a Faraday isolator. A self-adaptive sensor which allows the measurement of remote surface vibrations is demonstrated. The two-wave mixing interaction within a saturable gain media is shown to allow measurement of high frequency phase modulations (>10kHz) whilst adapting to cancel out low frequency perturbations. This sensor system is shown to have potential as a remote ultrasound detector as the holographic nature allows high frequency measurement of the vibrations of rough remote surfaces. Self-starting self-adaptive lasers, where a four wave mixing interaction within the saturable gain medium is utilised to generate phase conjugate and aberration corrective laser systems are experimentally investigated. This work is extended to show that the gain hologram is capable of adapting to low frequency phase modulations in order to maintain a high quality output. A demonstration of self-organised coherent beam combination of two bounce geometry laser oscillators into a single output beam is reported. A combined output beam of 35.7W was demonstrated from 94W of pump power. This coherent beam combination is extended into the technique of phase conjugate self-organised coherent beam combination (PCSOCBC) where a first demonstration of the combination of two self-starting self-adaptive modules is reported. It is shown that the adaptive modules allow efficient beam combination (94%) with a combined output of 27W. As the self-starting self-adaptive modules do not have predefined spatial or spectral modes it is believed that this system could be scaled to much higher numbers of modules than is possible with conventional self-organised coherent beam combination.

The microtubule cytoskeleton determines the internal architecture of cells, crucial for their proper functioning. In many vertebrate cells the interphase microtubule cytoskeleton has an astral organisation with stable microtubule ends focused near the cell centre and dynamic microtubule ends emanating outwards towards the plasma membrane. This arrangement is determined by a complex self-organisation process involving the cell boundary and many protein activities within and at the periphery of a cell. This project aimed at the elucidation of the minimal set of activities required for this process to understand the rules governing the self-organisation of microtubule arrays within a boundary. I reconstituted self-organising microtubule arrays inside lipid monolayer surrounded droplets in oil from purified components and hence established a well-controlled system to systematically study the interplay of organising activities by fluorescence microscopy. My studies showed that restriction in space as such has no influence on de novo polymerisation of microtubule polymerisation, however on their arrangement. In presence of a surrounding lipid monolayer microtubule nucleation and polymerisation could be achieved using a lipid composition close to that of a natural plasma membrane and microtubule nucleation enhancing buffer conditions. Without any further organising activity, the arrangement of microtubules depends on the droplet diameter and microtubule nucleation efficiency. The microtubule-crosslinking and organising kinesin 14 forms single microtubules asters in larger droplets, in smaller droplets it bundles microtubules indicating a microtubule length – droplet diameter interdependency influencing the aster formation inside a confined volume. Therefore I investigated whether microtubule shortening by a kinesin-13 or severing by Spastin, either homogenously distributed in the droplet or located to the droplet boundary, allows motor-driven aster formation in small droplets. In order to transfer that basic model system into artificial liposomes and investigate the influence of a pliable boundary, I designed a flow chamber to appropriately image vesicles.

Peer-to-peer systems are a highly decentralised form of distributed computing, which has advantages of robustness and redundancy over more centralised systems. When the peer-to-peer system has a stable and static population of nodes, variations and bursts in traffic levels cause momentary levels of congestion in the system, which have to be dealt with by routing policies implemented within the peer-to-peer system in order to maintain efficient and effective routes. Peer-to-peer systems, however, are dynamic in nature, as they exhibit churn, i.e. nodes enter and leave the system during their use. This dynamic nature makes it difficult to identify consistent routing policies that ensure a reasonable proportion of traffic in the system is routed successfully to its destination. Studies have shown that chum in peer-to-peer systems is difficult to model and characterise, and further, is difficult to manage. The task of creating and maintaining efficient routes and network topologies in dynamic environments, such as those described above, is one of dynamic optimisation. Complex adaptive systems such as ant colony optimisation and genetic algorithms have been shown to display adaptive properties in dynamic environments. Although complex adaptive systems have been applied to a small number of dynamic optimisation problems, their application to dynamic optimisation problems is new in general and also application to routing in dynamic environments is new. Further, the problem characteristics and conditions under which these algorithms perform well, and the reasons for doing so, are not yet fully understood. The assessment of how good the complex adaptive systems are at creating solutions to the dynamic routing optimisation problem detailed above is dependent on the metrics used to make the measurements. A contribution of this thesis is the development of a theoretical framework within which we can analyse the behaviours and responses of any peer-to-peer system. We do this by considering a peer-to-peer system to be a graph generating algorithm, which has input parameters and has outputs which can be measured using topological metrics and statistics that characterise the traffic through the network. Specifically, we consider the behaviour of an ant-based peer-topeer system and we have designed and implemented an ant-based peer-to-peer simulator to enable this. Recently methods for characterising graphs by their scaling properties have been developed and a small number of distinct categories of graphs have been identified (such as random graphs, lattices, small world graphs, and scale-free graphs). These graph characterisation methods have also enabled the creation of new metrics to enable measurements of properties of the graphs belonging to different categories. We use these new graph characterisation techniques mentioned above and the associated metrics to implement a systematic approach to the analysis of the behaviour of our ant peer-topeer system. We present the results of a number of simulation runs of our system initiated with a range of values of key parameters. The resulting networks are then analysed from both the point of view of traffic statistics, and also topological metrics. Three sets of experiments have been designed and conducted using the simulator created during this project. The first set, equilibrium experiments, consider the behaviour of the system when the number of operational nodes in the system is constant and also the demand placed on the system is constant. The second set of experiments considers the changes that occur when there are bursts in traffic levels or the demand placed on the system. The final set considers the effect of churn in the system, where nodes enter and leave the system during its operation. In crafting the experiments we have been able to identify many of the major control parameters of the ant-based peer-to-peer system. A further contribution of this thesis is the results of the experiments which show that under conditions of network congestion the ant peer-to-peer system becomes very brittle. This is characterised by small average path lengths, a low proportion of ants successfully getting through to their destination node, and also a low average degree of the nodes in the network. This brittleness is made worse when nodes fail and also when the demand applied to the system changes abruptly. A further contribution of this thesis is the creation of a method of ranking the topology of a network with respect to a target topology. This method can be used as the basis for topological control (i.e. the distributed self-assembly of network topologies within a peer-to-peer system that have desired topological properties) and assessing how best to modify a topology in order to move it closer to the desired (or reference) topology. We use this method when measuring the outcome of our experiments to determine how far the resulting graph is from a random graph. In principle this method could be used to measure the distance of the graph of the peer-to-peer network from any reference topology (e.g. a lattice or a tree). A final contribution of this thesis is the definition of a distributed routing policy which uses a measure of confidence that nodes in the system are in an operational state when making calculations regarding onward routing. The method of implementing the routing algorithm within the ant peer-to-peer system has been specified, although this has not been implemented within this thesis. It is conjectured that this algorithm would improve the performance of the ant peer-to-peer system under conditions of churn. The main question this thesis is concerned with is how the behaviour of the ant-based peer-to-peer system can best be measured using a simulation-based approach, and how these measurables can be used to control and optimise the performance of the ant-based peer-to-peer system in conditions of equilibrium, and also non-equilibrium (specifically varying levels of bursts in traffic demand, and also varying rates of nodes entering and leaving the peer-to-peer system).

Kinesin is a molecular motor which walks on microtubule tracks in the eukaryotic cytoskeleton. It transports cargo but is also involved in cytoskeletal organisation. This thesis demonstrates fusing kinesin and DNA to construct a molecular transport system using self-organised tracks and to study the mechanics of the minimal motor unit of kinesin. The programmability of DNA allows for the formation of nanostructures with controllable interactions. Kinesin is conjugated to various DNA nanostructures to accomplish different tasks. Instructions encoded into DNA sequences are used to direct the assembly of a polar array of microtubules, to control the loading, active concentration and unloading of cargo on this track network and to trigger the disassembly of the network. Fluorescence microscopy was used to observe these microtubule arrays and the movement of cargo. It was found that the DNA signals used to control the unloading of cargo and the disassembly of the network had to be actively transported, rather than relying on diffusion, for effective delivery of the signal. This work lead to a first author publication, Wollman et al. (2013). DNA was also used to study kinesin by linking defined numbers of minimal functional motor units, single kinesin heads, into teams of 4-12 heads and observing their movement along microtubules via fluorescent labelling. A minimum of 5 heads were required for sustained movement, in agreement with the predictions of Hancock and Howard (1998). The velocity of teams increased with more heads, up to 8, and then a decrease was observed in teams with more heads.

Rip channels are frequently observed morphological pattern on the sandy beach, which are critical to the erosion of beaches as well as the safety of beach visitors. Field observations indicate the wave conditions and sediment characteristics exert the controlling effects on the states of beaches and determine the development of rip channel system. Although the role of wave conditions on the rip channel dynamics has been extensively studied, little attention has been paid to the effects of sediment characteristics. In this thesis, the impacts of grain size of both uniform and non-uniform sediments on the formation and nonlinear evolution of rip channels have been investigated using a nonlinear morphodynamic model that is adapted from the open source program XBeach. To quantitatively describe the impacts of sediment grain size, a number of indicators that represent the dynamical and geometrical characteristics of rip channels have been used. Specifically, the indicators that related to the dynamical properties of rip channels include growth rate, migration speed, response time and saturation time. The indicators that related to the geometrical characteristics of rip channels include predominant spacing and rip channel three-dimensionality. For beaches that consist of uniform sediment, the grain size can affect number aspects of rip channel dynamics. It is found that when the grain size increases, both alongshore migration speed and growth rate decreases significantly, while the response and saturation time increases. On the other hand, the influence of grain size on the geometrical properties of the rip channel is much less pronounced, as the predominant wavelength of rip channels hardly changes and the rip channel three-dimensionality only vary slightly, depending on either the variation of grow rate or saturation time. The “global analysis” method that takes the variables over the whole modelling domain into account is then applied to investigate the mechanism underlying the variations of growth rate and migration speed. The results of calculation using “global analysis” method reveal that the variations of growth rate and migration speed are mainly caused by the amount of sediments being stirred up into the water column which is clearly grain-size-dependent. Previously, all modeling studies assume that the heterogeneous sediment on sandy beach can be represented by single homogeneous sediment, which implicitly suggest that the actual dynamics of rip channels for heterogeneous sediment are either similar to those within an environment comprised of uniform sediment or equivalent to the linear summation of dynamics determined for individual grains. However, our results of simulations that using heterogeneous sediment show that in some occasion this assumption is not valid, as the values of indicators that concern to the temporal evolution properties of rip channels for heterogeneous sediment can locate outside of the range that restricted by prediction using uniform sediment. This is because the gradient of sediment concentration depends on the spatial distribution of sediment size and is not always the linear summation of the gradient of sediment concentration calculated using uniform sediment grains. Therefore, our results suggest that special care for the spatial distribution of sediment size should be taken when predicting the development of rip channels.

Energy released from the Earth’s crust in the form of earthquakes commonly follows a powerlaw gamma type probability distribution. This spontaneous organisation is in apparent contradiction to the second law of thermodynamics that states that a system should naturally evolve to a state of maximum disorder or entropy. However, developments in the field of modern thermodynamics suggest that some systems can undergo organisation locally, at the expense of increasing disorganisation (or entropy) globally through a process of entropy production. The primary aim of this thesis is to investigate self-organisation in the Earth’s seismogenic lithosphere as a driven, far-from-equilibrium, self-organising ‘dissipative structure’ in a very near critical steady-state and the underlying general mechanisms involved. The secondary aim is to test in more detail the applicability of the Bak, Tang and Wiesenfeld (BTW) model of Self-Organised Criticality (SOC) in describing Earth’s seismicity. This is done by: 1. Mathematical derivation of analytical solutions for system energy and entropy using the tools of equilibrium statistical mechanics; 2. The study of conservative and non-conservative versions of the BTW numerical model and 3. Analysis of temporal and spatial properties of earthquake data from the Harvard Centroid Moment Tensor catalogue and the Global Heat Flow Database. The modified gamma distribution predicts analytically that entropy S is related to the energy probability distribution scaling exponent B and the expectation of the logarithm of seismic energy hlnEi in the form of the gamma entropy equation S » BhlnEi. This solution is con- firmed for both numerical model results and real earthquake data. Phase diagrams of B vs. hlnEi suggest that the universality in B need not be maintained for a system to remain critical provided there is a corresponding change in hlnEi and S. The power-law systems examined are different from equilibrium systems since the critical points do not occur at global maximum entropy. For the dissipative BTW model at a steady-state, the externally radiated energy follows out-of-equilibrium power-law gamma type statistics, but, the internal energy has two icharacteristics that are indicative of equilibrium systems; a Gaussian type energy probability distribution and a Brownian noise power-spectrum for the internal energy fluctuations. This suggests an observer dependency in assessing criticality. The internal and external entropies calculated for the model are negatively correlated suggesting that driven systems self-organise at the expense of increasing entropy globally through a process of dissipation. A power-law dependency of mean radiated energy hEi on dissipation 1¡® is confirmed for a locally driven dissipative system in the form hEi » (1¡®)¡0:975. The BTW model shows spatial heterogeneity whilst maintaining universality in contradiction to previous assumptions. The quantitative analysis of real data reveals that earthquakes are more predictable spatially then temporally. Regionalisation using the Flinn-Engdahl classification shows that mid-ocean ridges are more organised (lower entropy) than subduction zones. A regional study of three different scaling exponents suggests that universality in earthquake scaling is violated, in contradiction to the original model of SOC. A model of self-organised sub-criticality (SOSC) is proposed as an alternative model for Earth seismicity. Overall, the results suggest that the tools of equilibrium thermodynamics can be applied to a steady-state far-from-equilibrium system such as the Earth’s seismogenic lithosphere, and that the resulting self-organisation occurs at the expense of maximising dissipation and hence entropy production.

This research locates my work in a theoretical and practical context, exemplified by a five-day intensive fieldwork project with a group of eight participants at my home and consulting rooms. My practice is based on two main ways of working with the body: Ortho-Bionomy (SOBI, 2006); and Bowen Technique (Navratil, 2003), especially the former, which is the focal modality (or method) of my work and this research. Ortho-Bionomy is a respectful way to attend to aspects of the body’s self-organising ability, also known as emergence. The work basically involves taking the body into positions of comfort, allowing time and space for the easing of pain and tension. In the research I have also explored two other modalities: Open Space Technology, designed by Harrison Owen (1994) for group meetings, and Process Work (sometimes known as Process Oriented Psychology), developed by Arnold Mindell and colleagues in Switzerland and USA (Mindell, 1982). Like Ortho-Bionomy (Kain and Berns, 1992), these two ways of working also provide time, space and structure for issues of concern to be resolved. All are described in the Modalities chapter. This research follows on from my Masters research (Baensch, 2002b) about the effects of my learning, life experiences and work, on my body. This time my focus is on researching my practices using two meanings of the word ‘practice’. One meaning relates to the hands-on work I carry out and the principles and concepts behind that, which I will generally call my ‘work’. As well, I touch on aspects of my practice as an enterprise in the community.

Thesis (Ph.D.)--University of Western Sydney, 2009.
Accompanied by DVD entitled: Body of knowledge. DVD can be viewed at UWS Library. A thesis presented to the University of Western Sydney, College of Arts, Social Justice and Social Change Research Group, in fulfilment of the requirements for the degree of Doctor of Philosophy. Includes bibliographies.

It is considered that our memories of experience constrain what the self can be because this tethers the self in reality. Likewise, what we can retrieve about our past probably also influences how we describe and think about our self. Furthermore, one way of thinking about the self is as a store of personally relevant information meaning that it is linked to the very notion of memory. Using a novel fluency paradigm, this thesis explores the idea that the accessibility and organisation of self statements is intimately associated with autobiographical memory. Experimental work in Study 1a revealed that it is possible to use memory in order to ‘boost’ access to the self. Recall of an autobiographical memory increased the retrieval of self statements in a novel fluency task, and psychological selves in particular (e.g., I am kind). Study 2 showed that the accessibility of self statements varies across the lifespan and that selves emerge around key autobiographical milestones (e.g., during the period of the reminiscence bump). In Study 3, a neuropsychological approach was used to explore the impact of amnesia on the accessibility of temporally extended aspects of the self (i.e. past, present and hoped for self) in a single case study on patient SA. Findings showed that retrograde amnesia had no impact on the accessibility of self statements, yet SA failed to generate a single psychological description of her current identity. Study 4 indicated that autobiographical memories cluster around times when a new self emerges, reflecting the importance of memory in maintaining a coherent sense of self in spite of change. Finally, Study 5 aimed to extend the results from Study 3 to a group of people with severe memory impairment. Unlike SA, these participants were impaired in their ability to access self statements, but no relationship between episodic impairment and the accessibility of psychological selves was found. Across all chapters, this thesis presents the I Am Fluency Task as a valuable tool for meaningfully ‘measuring’ the self, and the clinical implications of this will be considered throughout. The experiments in this thesis help to delineate more clearly the role of memory processes in the retrieval of self statements. On the whole, it is found that whilst episodic retrieval processes seem to be implicated in the retrieval of self statements, such as in the first experiment, the other studies show that even when autobiographical memory is impaired, access to certain aspects of the self - probably maintained by semantic memory -persist. Thus, the self is a complex dynamic structure which draws upon both autobiographical memory and also personal semantics, and may be updated and maintained by a complex co-ordination of information from multiple memory systems.

Dans cette thèse, nous étudions les organoïdes, mini-organes auto-assemblés issus de quelques cellules souches, qui offrent une perspective unique pour étudier l'organogenèse. Notre recherche relie les formes et les mouvements collectifs des organoïdes à la dynamique hors équilibre du cortex d’acto-myosine. À l'interface entre la physique et la biologie, nous concevons des expériences pour quantifier les propriétés cellulaires et tissulaires et nous intégrons ces mesures dans des modèles physiques révélant les règles d’auto-organisation des organoïdes. En utilisant des cystes MDCK, un modèle organotypique, nous explorons (i) le rôle des asymétries corticales sur la forme des cellules et du cyste, (ii) comment les protéines de jonction serrée influent sur la morphologie et la mécanique du cyste, et (iii) l'émergence de la rotation collective spontanée en 3D de doublets cellulaires due à la rupture de symétrie de la dynamique d’acto-myosine. Notre travail établit un lien entre l'auto-organisation des organoïdes et la dynamique d’acto-myosine, révélant comment les propriétés hors équilibre dirigent la morphogenèse
In this PhD study, we investigate organoids—self-assembled mini-organs derived from a few stem cells, offering a unique perspective on organogenesis. Our research links organoid shapes and collective motions to the out-of-equilibrium dynamics of the acto-myosin cortex. At the interface between Physics and Biology, we design experiments to quantify cellular and tissue properties and use theoretical physics to integrate measurements into models revealing the self-organization of organoids. Using MDCK cysts, an organotypic model, we explore (i) the role of cortical asymmetries on cell shape and cyst structure, (ii) how tight junction proteins influence cyst morphology and mechanics, and (iii) the emergence of spontaneous 3D collective rotation of cell doublets due to symmetry breaking of acto-myosin dynamics. Our work highlights the intricate link between organoid self-organisation and acto-myosin dynamics further revealing how out-of-equilibrium properties drive morphogenesis

This study is set in the broad context of the changing world of work that is characterised by the dissolution of full-time stable employment and the emergence of precarious, insecure forms of work (see e.g. Gorz, 1999, Hardt and Negri, 2005, Huws, 2016, Lorey, 2015, Ross, 2003, Ross, 2009, Smith, 2001, Standing, 2011). As a response to these labour market uncertainties a growing number of individuals are managing multiple areas of the self as part of their work or occupation. This trend has been termed 'the new worker-subjectivity' or 'the entrepreneurial self' that is formed through practices of self-management (Bührmann, 2005, Lorey, 2009). Despite increasing awareness of the emergence of the entrepreneurial worker-subjectivity, research into practices of self-management has only focused on occupational groups in formal work. Knowledge about the trend in the context of virtual workers who operate outside of conventional working relations and have no publicly recognised work identity is largely missing. In order to address this gap, this study explores how entrepreneurial worker-subjectivities manifest in professional online poker players as an emerging online occupation. It investigates how these workers manage themselves in the absence of formal organisational control and socially recognised occupational norms, and asks what are the effects of this self-management on the quality of their working lives? The study is based on 39 in-depth interviews with people involved in online poker or other similar activities such as online gaming or trading. The interviews were conducted either face-to-face in Estonia, Bulgaria, Romania and UK or over Skype between December 2012 and May 2014. The study develops an analytical framework for researching entrepreneurial worker-subjectivities in the context of an emerging occupation and a three-stage-model of the trajectory that provides a basis for exploring the career paths of professional online poker players. Using these framework, the study finds that professional online poker players manage various areas of the self by following informal occupational rules and that their sense of professionalism is largely derived from various practices of self-management that help them distinguish from recreational players. The study also discovers conflicting relations of autonomy and control among the workers and a range of negative effects that self-management practices have on professional online poker players. It concludes that professional online poker is not a sustainable long-term career option. These findings contribute to a better understanding of virtual work, the emergence of online poker playing as a form of work and the development of the entrepreneurial worker-subjectivity.

Many human languages in the world exhibit phrase structure. Phrase structure combines words, phrases, and both, into phrases, and it may empower language systems to exploit recursion. This thesis pushes forward the hypothesis that phrase structure is not an accidental structural property of language, but rather an adaptation of language systems to enable the computation of language. I propose a minimal operational model of communication as a specific language game, which together with concrete learning operators shows how a population of artificial agents is able to self-organise a system exhibiting phrase structure. After demonstrating that phrase structure reduces the complexity of language computation, I propose concrete mechanisms in the form of learning operators whose application introduces variation in the language of the agents and selection on the reduction of the computational cost. The mechanisms are implemented and tested in computer simulations as evolutionary explanation for the emergence of phrase structure, including cases exploiting recursion.
Els llenguatge naturals es basen en molts casos en gramàtiques sintagmàtiques. Les gramàtiques sintagmàtiques combinen paraules i sintagmes en altres sintagmes, i poden capacitar els sistemes lingüístics a fer ús de la recursió. Aquesta tesi enforteix la hipòtesi que les gramàtiques sintagmàtiques no són una propietat estructural accidental del llenguatge, sinó que són una adaptació dels sistemes lingüístics que permet que el llenguatge pugui ser processat adequadament. Proposo un model minimal de comunicació basat en un joc del llenguatge en concret que defineixo, i que juntament amb operadors d'aprenentatge específics mostra com una població d'agents artificials és capaç d'autoorganitzar un sistema que exhibeix l'ús de gramàtiques sintagmàtiques. Un cop demostrat que les gramàtiques sintagmàtiques redueixen la complexitat del processament del llenguatge, proposo mecanismes concrets en forma d'operadors d'aprenentatge l'aplicació dels quals introdueix variació en les gramàtiques dels agents i selecció en la reducció del cost de processament. Els mecanismes són implementats i avaluats en simulacions com a una explicació evolutiva de l'emergència de les gramàtiques sintagmàtiques, incluent casos en què es fa ús de la recursió.

A disaster always means a huge death toll, displacement and inconceivable destruction for a poor country such as Bangladesh. Recently, Bangladesh has taken a holistic approach to prioritising interrelated activities and the involvement of various organisations in disaster management. A number of disaster management committees (DMCs) have been formed to coordinate and implement risk reduction measures. But the levels of success of these organisations have varied in different regions. Improper consideration of local knowledge, corruption of actors, lack of coordination and capacity of actors, etc., are perceived as major causes of this. Primarily, this mini-thesis aims to measure the impact of self-organisation in disaster risk management.

The aim of this work is to introduce modular processing mechanisms for cortical functions implemented in networks of spiking neurons. Neural maps are a feature of cortical processing found to be generic throughout sensory cortical areas, and self-organisation to the fundamental properties of input spike trains has been shown to be an important property of cortical organisation. Additionally, oscillatory behaviour, temporal coding of information, and learning through spike timing dependent plasticity are all frequently observed in the cortex. The traditional self-organising map (SOM) algorithm attempts to capture the computational properties of this cortical self-organisation in a neural network. As such, a cognitive module for a spiking SOM using oscillations, phasic coding and STDP has been implemented. This model is capable of mapping to distributions of input data in a manner consistent with the traditional SOM algorithm, and of categorising generic input data sets. Higher-level cortical processing areas appear to feature a hierarchical category structure that is founded on a feature-based object representation. The spiking SOM model is therefore extended to facilitate input patterns in the form of sets of binary feature-object relations, such as those seen in the field of formal concept analysis. It is demonstrated that this extended model is capable of learning to represent the hierarchical conceptual structure of an input data set using the existing learning scheme. Furthermore, manipulations of network parameters allow the level of hierarchy used for either learning or recall to be adjusted, and the network is capable of learning comparable representations when trained with incomplete input patterns. Together these two modules provide related approaches to the generation of both topographic mapping and hierarchical representation of input spaces that can be potentially combined and used as the basis for advanced spiking neuron models of the learning of complex representations.