Dissertations / Theses on the topic 'Network behavior mechanisms'

In recent years, the network concept has become a central component of administrative scholarship. One cannot ignore the increased use of networks as both an explicit policy choice and a condition of public funding. This trend suggests that the network concept now represents an approach to governance. Regardless, active participation in these multi-organizational, multi-governmental, and multi-sectoral relationships has outpaced empirical description and theoretical explanation. Making the important case that network management has become a critical activity in public administration, researchers have neglected the relationship between front-line participatory behavior and the use of the network approach. As a result, the vocabulary and the imagery needed to describe and theorize about the specific front-line participatory behaviors that accompany the use of the network approach does not exist. Due to the limitations of past research, there is little understanding of the front-line participatory behaviors that could help make this happen. This study refers to these types of behaviors as network behaviors. Relying on surveys and elite interviews with participants from thirteen social welfare networks throughout the state of Virginia, this study addresses two primary research questions: What are the front-line participatory behaviors that accompany the use of the network approach? And, how do these behaviors differ along with variations in the network approach? Through examining 14 hypotheses, the study relates a framework of four degrees (variations) of the network approach (cooperation, coordination, consolidation, and collaboration) to three categories of behavior (knowledge management, communicative behavior, and commitment/identity). The findings support the underlying rationale for this study that variations in the network approach may shape front-line participatory behavior differently, and vice versa. Ultimately, by exploring this relationship and integrating the literature on networks with the literature on front-line work, this study may serve as the foundation for future efforts to establish a theory or rationale for developing and choosing among variations in the network approach.
Ph. D.

Many older youth in foster care lack adequate resources and ongoing support in their social networks as they transition into young adulthood, while other youth in these circumstances experience stable social networks providing comprehensive support. Systematically measuring the supportive personal and service-oriented relationships in youth networks expands the scope of inquiry in this area by identifying patterns of social network structure, member composition, and relational qualities that are associated with more or less support provision through formal and informal relationships. These can also be measured over time to observe changes in network form and content and assess network stability. This exploratory study (1) describes the support networks for a small sample of youth with foster care experience who are enrolled in post-secondary education and training programs, (2) assesses changes in these networks over time, and (3) demonstrates the reliability and validity of this methodology for broader use with populations of transition-age foster youth. Findings show that family (biological and foster) and friends are the most prevalent informal supports, relationship ties to parent figures are strongest and provide the most stable and multi-dimensional support, and ties with formal service providers are not as strong, but provide more informational support. The stability of a network ties over time is associated with the breadth of support provided, and network-based social support is associated with post-secondary enrollment at follow-up. Support network profiles are described and interpreted in terms of bonding and bridging social capital. Discussion includes implications for future support network research and guidelines for pre-transition assessment of youth networks in practice.

L’émission acoustique (EA) est reconnue pour être une technique efficace de surveillance de la santé des structures permettant de détecter la création et la propagation de microfissures dans les matériaux structuraux tels que le béton ou les composites lorsqu'ils sont soumis à des contraintes quasi-statiques. Sur la base de méthodes de traitement de signaux adéquates, différentes études ont établi des liens entre les salves d’EA et les micro-dommages créés. D'autres travaux ont montré qu'il est possible de corréler le temps de relaxation des composites et l'énergie des mécanismes d'endommagement mesurée durant la charge quasi-statique en utilisant les salves d’EA enregistrées. Cette thèse propose d'utiliser un protocole expérimental original pour détecter la relaxation non-linéaire d'échantillons de bétons à l'état intact et endommagés. Ce protocole est basé sur l'utilisation de l’EA pour capter passivement la relaxation non-linéaire d'échantillons de bétons au lieu du signal de faible amplitude habituellement utilisé dans les expériences de dynamique lente. Les résultats montrent que les méthodes de détection passives et actives conduisent à des temps de relaxation équivalents. De plus, le capteur d’EA révèle l’existence d’une ‘période de silence’ pendant les premières minutes de la relaxation non-linéaire après laquelle les salves d’EA commencent à être détectées. De plus, les caractéristiques des salves d’EA enregistrées pendant la relaxation passive ont montré une nette ressemblance avec celles obtenues lors de l'endommagement des mêmes échantillons, où des mécanismes de cisaillement et de compression sont impliqués. Enfin, nous notons qu’en plus de l'utilisation d'une approche de reconnaissance des formes non- supervisée pour la classification des salves d’EA, ce travail propose une nouvelle approche de classification des signaux d’EA basé sur l’image de la représentation en ondelettes continue (CWT) et le réseau de neurones convolutifs (CNN). Les résultats liés aux données d’EA dynamiques non-linéaires et quasi-statiques montrent que les deux approches de traitement du signal ont une grande précision de classification, ce qui représente un intérêt certain pour le développement de méthodes d’EA dynamiques en présence de microfissures
Acoustic emission (AE) is well known to be an efficient structural health monitoring technique to detect the creation and propagation of micro-cracks within structural materials such as concrete or composites when submitted to quasi-static stresses. Based on adequate signal processing methods, different research studies have established links between the detected AE hits and the created micro-damages. Other works have shown that it is possible to correlate the relaxation time in composites and the energy of the damage mechanisms measured during the quasi-static loading using the recorded AE hits. This thesis proposes to use an original experimental protocol to probe the nonlinear relaxation of concrete samples at the intact and damaged states. This protocol is based on the use of AE to passively probe the nonlinear relaxation of concrete samples instead of the weak amplitude signal usually used in slow dynamics experiments. Results show that passive and active probing methods lead to equivalent relaxation times. Furthermore, AE probing reveals the existence of a ‘silence period’ during the first minutes of the nonlinear relaxation after which AE hits start to be detected. In addition, the characteristics of AE hits recorded during the passive relaxation showed a clear resemblance to those obtained during the damaging of the same samples, where shear and compression mechanisms are involved. For the clustering of the AE hits, in addition to use of an unsupervised pattern recognition approach to cluster the detected AE hits, this work proposes a novel ‘image- based AE classification’ approach based on continuous wavelet transform (CWT) and convolutional neural network (CNN). Results related to the nonlinear dynamic and quasi-static AE data show that both signal processing approaches have high classification accuracy, which represents a great interest in the development of dynamic AE methods in the presence of micro-cracks

Port-knocking is a security mechanism used in computer systems to hide available network services. Its operation relies on a drop policy firewall setting in order to make impossible for port-scanning attacks to occur. This project researches the impact of implementing such a software solution. Furthermore, it looks into the behavior of three chosen implementations and make conclusions on the benefits and disadvantages that they bring. In addition, the surrounding implications related to both user and administrator are explored. This thesis includes tests on the resource consumption of the implementations as well as records of the added delay of using the mechanism when initiating a SSH session. There has not been such research performed in this field and the results of it could be beneficial to those who are involved in computer science and network security in particular. Finally, the product of this study state that port-knocking is overlooked and has great benefits in preventing zero-day exploits and hacker tools relying on exposed network services.

The main goal of this thesis is to provide a sociologically informed analysis of tax avoidance and tax evasion in contemporary Mexico and Sweden, focusing particular attention on the explanatory role of social networks, social interactions, and positive feedback mechanisms. Two major data sources are used: (1) A panel dataset that includes all persons, 16 years or older, who resided in Stockholm County during at least one of the years 1990 to 2003 (N=1,967,993). The dataset includes detailed information on the socio-demographic characteristics, kinship networks, and criminal offences of these individuals; (2) A random sample of 36,949 firms that appeared in the Mexican Federal Register of Taxpayers for the year 2002. The records of the Mexican Federal Administrative Fiscal Tribunal provided data on all types of tax claims appealed before them during the 2002-2008 period. A variety of approaches and techniques are used such as agent-based simulation models, discrete time event history models, random effect logit models, and hierarchical linear models. These models are used to test different hypotheses related to the role of social networks, social interactions, and positive feedback mechanisms in explaining tax behaviour. There are five major empirical findings. (1) Networks seem to matter for individuals' tax behaviour because exposure to tax crimes of family members appears to increase a person’s likelihood of committing a tax crime. (2) Positive feedback mechanisms appear relevant because if a person commits a tax crime, it seems to increase the likelihood that the person will commit more tax crimes in the future. (3) Positive feedback mechanisms are also important for explaining corporate tax behaviour because a firm that has engaged in legal tax avoidance in the past appears to be more likely to engage in tax avoidance in the future. (4) Network effects are important in the corporate world because exposure to the tax avoidance of other firms increase the propensity of a firm to engage in tax avoidance. (5) Substitution effects between tax evasion and tax avoidance are likely to exist because when tax evasion becomes more prevalent in a firm’s environment, their likelihood of engaging in legal tax avoidance is lowered. The results underscore the importance of a sociological perspective on tax behaviour that takes into account social interactions and positive feedback mechanisms. In order to understand microscopic as well as macroscopic tax evasion patterns, the results presented in this thesis suggest that much more attention must be given to mechanisms through which taxation crimes breed more taxation crimes.

Animals face frequent threats from predators and must generate appropriate behavioural responses to ensure their survival. To achieve this, they process sensory cues to correctly identify the presence and imminence of a predatory threat, and transform this information into defensive actions. However, despite much research in identifying the circuits that may be responsible for such transformations, little is known about how this occurs mechanistically. We focus on how escape behaviour in the mouse is generated from visual predatory threats, and use a combination of behavioural, neurophysiological and anatomical methods to identify the relevant neurons and understand how they perform this computation. In this work, we developed an innate decision making paradigm in which a mouse detects and assesses sensory stimuli of varying threat evidence during exploration, choosing whether to escape to a shelter, or not. The performance data in this task were best formalised with a drift-diffusion model of decision making, providing a framework to understand innate behavioural tasks in terms of evidence accumulation and boundaries. Next, we performed calcium imaging in freely-moving mice to probe for neural correlates of decision elements and flight behaviour in brain areas that we show to be necessary for the flight responses: we found that VGluT2 neurons in the deeper medial superior colliculus (dmSC) increase their activity during a repeated threatening stimulus, while VGluT2 neurons of the dorsolateral periaqueductal gray (dPAG) are silent until just before the initiation of escape, and are maximally active during escape. These results suggest that the dmSC accumulates evidence of threat which dPAG neurons threshold. This interpretation is supported by optogenetic activation of mSC-VGluT2 neurons in vivo, which recapitulates the statistics of escape probability evoked with a visual stimulus, while activation of VGluT2 neurons in the dPAG evokes an all-or-nothing escape response. Finally, using channelrhodopsin-2-assisted circuit mapping and monosynaptic viral tracing, we reveal that over half of dPAG-VGluT2 neurons receive monosynaptic connections from mSC-VGluT2 neurons with a low probability of release, allowing this synapse to act as a high-pass filter and providing a mechanism for the computation of an escape decision. These findings advance our understanding of how defensive behaviours are generated at circuit and single-cell level, and of how neurons process information in a circuit critical for implementing basic behaviours.

Attention-deficit/hyperactivity disorder (ADHD) is one of the most common child-onset neurodevelopment disorders, affecting 5% of children in the United States (American Psychiatric Association, 2013). Treatment matching in ADHD is difficult and unsatisfactory; the same general treatment algorithm is recommended for all children. It is therefore important to consider the development of specialized treatment programs based on a variety of behavioral and neurological biomarkers. Unfortunately, due to its multi-faceted classification, the heterogeneity of this behavioral disorder is under-investigated (Costa Dias et al., 2015). Scientific research in this area is especially limited as the severity of ADHD goes undiagnosed, children tend to have difficulties remaining still in MRI scanners, and the hyperactivity-impulsivity that is associated with ADHD may cause further challenges when trying to remain motionless in the scanner. Furthermore, tasks such as Facial Emotion Perception Task (FEPT) and Theory of Mind (ToM) have not been used to analyze social and behavioral deficits in children with ADHD. More research needs to be allocated to helping uncover the neural substrates underlying the inattention and hyperactivity traits of this disorder. For this reason, we acquired functional magnetic resonance imaging (fMRI) data from five children with ADHD performing the FEPT and ToM tasks. The results showed the children have an easier and quick time correctly identifying happy emotional states, as compared to the fearful, angry, and neutral conditions. Results from the FEPT task also revealed that the participants were thinking and reasoning more (i.e., taking longer to deduce an ending) when identifying emotions than identifying animals. The ToM task showed that the default mode network (DMN) may not be fully suppressed when the children are choosing the correct cartoon ending, and therefore the children may be having lapses in attention. These findings may assist the current hypothesis that the default mode network has reduced network homogeneity in people with ADHD. Overall, the findings presented in this thesis provide a good diving board into discovering the reason(s) for the social cognition and emotion recognition impairments associated with ADHD, but further research is needed in order to one day pinpoint and ultimately correct the regions(s) of dysfunction.

Sensor networks are being used in several emerging applications not even imagined some years ago due to advances in sensing, computing, and communication techniques. However, these advances also pose various challenges that must be faced. One important challenge is related to the autonomous capability needed to setup and adapt the networks, which decentralizes the control of the network, saving communication and energy resources. Middleware technology helps in addressing this kind of problem, but there is still a need for additional solutions, particularly considering dynamic changes in users' requirements and operation conditions. This paper presents an agent-based framework acting as an integral part of a middleware to support autonomous setup and adaptation of sensor networks. It adds interoperability among heterogeneous nodes in the network, by means of autonomous behavior and reasoning. These features also address the needs for system setup and adaptations in the network, reducing the communication overhead and decentralizing the decision making mechanism. Additionally, preliminary results are also presented.

Les principaux objectifs de la thèse sont : d'explorer les motivations sociales et stratégiques et leurs conséquences, de modéliser et de décortiquer les mécanismes relationnels et instrumentaux des stratégies des organisations sur le marché du carbone. Pour cela, trois études empiriques ont été menées sur le terrain du marché du carbone régit par le Mécanisme de Développement Propre (MDP) élaboré dans le cadre du protocole de Kyoto. La première recherche exploratoire qui se base sur une étude de cas inductive a été réalisée dans un pays en voie de développement. Les résultats suggèrent que les préoccupations institutionnelles et concurrentielles influencent la stratégie des organisations du marché du carbone. Traitant de l'influence sociale comme opérant au sein d’une structure relationnelle, les deux études suivantes analysent la dynamique du comportement de l'organisation et son réseau relationnel. Les résultats suggèrent que, dans le cadre du marché MDP, l'ancrage et la centralité du réseau des entreprises affectent leurs comportements. Cependant, tout en contrôlant les contingences structurelles d’ordre relationnel à l’échelle des acteurs, les dynamiques co-évolutionnaires liées au réseau et au comportement suggèrent que les motivations instrumentales d’une entreprise sont plus promptes à influencer son comportement. Cette thèse contribue à la littérature sur la stratégie climatique par l'explication des dynamiques comportementales sociales et stratégiques, ainsi que par l’analyse des caractéristiques structurelles du réseau d’acteurs sur le marché carbone
The main objectives of the dissertation are to explore social and strategic motivations and consequences, and to model and disentangle relational and rational-instrumental mechanisms of carbon market strategies of organizations. To achieve these objectives, three empirical studies were conducted in the context of carbon market that operates under the Kyoto Protocol’s Clean Development Mechanism (CDM). The first inductive case study based exploratory research was carried out in a developing country. Findings suggest that institutional and competitive concerns motivate carbon market strategy of organizations. Treating social influence as operating through relational structure, the next two papers then analyze dynamics of organization’s behavior and its relational network. Results suggest that in the CDM market, firm’s network embeddedness, and centrality affect firms’ behaviors. However, while controlling for certain actor level and relational-structural contingencies, network and behavior co-evolutionary dynamics suggest that firm’s instrumental motivations are more pronounced in influencing the behavior of firms. This dissertation contributes to the climate strategy literature by explicating the social and strategic dynamics of behavior, and network structural characteristics, of the carbon market actors

The study of learning has been a central focus of psychology and neuroscience since their inception. Cognitive neuroscience’s traditional approach to understanding learn-ing has been to decompose it into discrete cognitive processes with separable and localized underlying neural systems. While this focus on modular cognitive functions for individual brain areas has led to considerable progress, there is increasing evidence that much of learn-ing behavior relies on overlapping cognitive and neural systems, which may be harder to disentangle than previously envisioned. This is not surprising, as the processes underlying learning must involve widespread integration of information from sensory, affective, and motor sources. The standard tools of cognitive neuroscience limit our ability to describe processes that rely on widespread coordination of brain activity. To understand learning, it will be necessary to characterize dynamic co-activation at the circuit level. In this dissertation, I present three studies that seek to describe the roles of distrib-uted brain networks in learning. I begin by giving an overview of our current understand-ing of multiple forms of learning, describing the neural and computational mechanisms thought to underlie incremental feedback-based learning and flexible episodic memory. I will focus in particular on the difficulties in separating these processes at the cognitive level and in localizing them to individual regions at the neural level. I will then describe recent findings that have begun to characterize the brain’s large-scale network structure, emphasiz-ing the potential roles that distributed networks could play in understanding learning and cognition more generally. I will end the introduction by reviewing current attempts to char-acterize the dynamics of large-scale brain networks, which will be essential for providing a mechanistic link to learning behavior. Chapter 2 is a study demonstrating that intrinsic connectivity between the hippo-campus and the ventromedial prefrontal cortex, as well as between these regions and dis-tributed brain networks, is related to individual differences in the transfer of learning on a sensory preconditioning task. The hippocampus and ventromedial prefrontal cortex have both been shown to be involved in this type of learning, and this study represents an early attempt to link connectivity between individual regions and broader networks to learning processes. Chapter 3 is a study that takes advantage of recent developments in mathematical modeling of temporal networks to demonstrate a relationship between large-scale network dynamics and reinforcement learning within individuals. This study shows that the flexibil-ity of network connectivity in the striatum is related to learning performance over time, as well as to individual differences in parameters estimated from computational models of re-inforcement learning. Notably, connectivity between the striatum and visual as well as or-bitofrontal regions increased over the course of the task, which is consistent with an inte-grative role for the region in learning value-based associations. Network flexibility in a dis-tinct set of regions is associated with episodic memory for object images presented during the learning task. Chapter 4 examines the role of dopamine, a neurotransmitter strongly linked to val-ue updating in reinforcement learning, in the dynamic network changes occurring during learning. Patients with Parkinson’s disease, who experience a loss of dopaminergic neu-rons in the substantia nigra, performed a reversal-learning task while undergoing functional magnetic resonance imaging. Patients were scanned on and off of a dopamine precursor medication (levodopa) in a within-subject design in order to examine the impact of dopa-mine on brain network dynamics during learning. The reversal provided an experimental manipulation of dynamic connectivity, and patients on medication showed greater modula-tion of striatal-cortical connectivity. Similar results were found in a number of regions re-ceiving midbrain projections including the prefrontal cortex and medial temporal lobe. This study indicates that dopamine inputs from the midbrain modulate large-scale network dy-namics during learning, providing a direct link between reinforcement learning theories of value updating and network neuroscience accounts of dynamic connectivity. Together, these results indicate that large-scale networks play a critical role in multi-ple forms of learning behavior. Each highlights the potential importance of understanding dynamic routing and integration of information across large-scale circuits for our concep-tion of learning and other cognitive processes. Understanding the when, where, and how of this information flow in the brain may provide an alternative or compliment to traditional theories of distinct learning systems. These studies also illustrate challenges in integrating this perspective with established theories in cognitive neuroscience. Chapter 5 will situate the studies in a broader discussion of how brain activity relates to cognition in general, while pointing out current roadblocks and potential ways forward for a cognitive network neuroscience of learning.

博士
國立交通大學
生物資訊及系統生物研究所
102
As an increasing number of genomic, proteomic, signaling and metabolomic data become available, the construction of networks is valuable for understanding the molecular mechanisms and disease mechanism. Many large-scale experimental methods have been proposed to identify the protein-protein interactions (PPIs) for constructing PPI networks, but these experimental PPIs were often studied on some species with high false-positive rates. In addition, dynamic cooperation between proteins for assembly of protein modules in time and space is essential to execute biological processes in networks of cells. Therefore, the construction of dynamic networks and dynamic modules are urgently required to reveal cell behaviors and disease mechanisms. To address these issues, we first proposed PPI and module family to construct (enlarge) the PPI networks through homologous mapping (across multiple species). For example, we derived homologous mapping human network, 17,343 proteins and 100,788 PPIs, which is much more complete than the previous works, ~12,121 proteins and 66,057 PPIs. For analysis of complex PPI networks, we proposed the organizational principles of modules for identifying the modules, homologous modules (called module family), and module-module interaction (MMI) networks. Moreover, we also developed a server (MoNetFamily) to provide the network annotations; the profiles of proteins and PPIs in module families; GO annotations of neighboring modules and module families. Our MMI network inferred two modules, JAK2-PAFR-TYK2 and IL-6/IL6Rα/gp130, to explore the mechanism of HIV-associated psoriasis. Next, we used these networks integrated with human gene expression data across 21 diverse cancer types to explore nicotine-induced cancer mechanisms regulated by nicotinic acetylcholine receptor subunit alpha-9 (α9-nAChR). Currently, all databases lack the records of interaction partners of α9-nAChR. For the α9-nAChR, our strategies successfully identified 9 novel interacting proteins, including EGFR, ERBB2, CSNK1D, COPS6, and NCLN. Our network shows that EGFR, ERBB2, CSNK1D, and COPS6 are involved in the MAPK signaling pathway to regulate cell proliferation and apoptosis. Based on the dynamic networks combined with gene expression profiling of 58 breast tumor tissues and 147 corresponding normal tissues, we observed the activations of these interacting proteins may increase when they disassociated from the interactions with α9-nAChR during tumorigenesis. Interestingly, ERBB2 seems to associate with α9-nAChR both in normal tissues and tumor tissues of basal-like subtype. Therefore, patients with active and passive smoking would induce disassociation between ERBB2 and α9-nAChR to increase activation of ERBB2 signalling, and result in promotion and progression of breast cancer cells, especially in basal-like cancers. In addition, we collected over 7,000 human gene expression data sets from Gene Expression Omnibus and integrated them into modules/networks to construct dynamic modules/networks for reflecting the cellular processes, driving phenotypic variations, in different cells and states; for example, exploring tumor heterogeneity in 21 diverse cancer types among 983 clinical samples. We believe that the dynamic modules in homologous mapping networks are useful for discovering cell behaviors and disease mechanisms.

碩士
大同大學
事業經營學系(所)
100
Nowadays, Botnet has become one of the greatest threats to the network security. Network attackers can use Botnet to launch the distributed denial of service (DDoS) to paralyze the important websites or to steal the confidential data from infected computer, use fishing attacking to steal the sensitive information such as the account and password, send bulk email advertising or conduct click fraud. Even though the detection technology has got improved and some solutions to Internet security have been proposed, the threat of Botnet still exists. Most previous studies used the packet contents or the features of network flows to analyze to detect Botnet. However, there are still some problems with packet encryption and privacy, i.e., Botnet can easily change the packet contents and flow features to avoid the detection system. This study proposes a solution to those problems and develops a mechanism of Botnet detection step-by-step. First of all, three important network behaviors including long communication behavior (ActBehavior), connection failure behavior (FailBehavior), and network scanning behavior (ScanBehavior) are defined in this study by reviewing the related literatures and analyzing the network activities of infected computer. Secondly, the features of network behaviors are extracted from the flow records of Network-Layer and Transport-layer in the network equipment. Finally, Particle Swarm Optimization (PSO) and K-means algorithm are used to detect the members of Botnet in the organization's network. This study uses campus network as a case study. The experimental results show that this mechanism can find out the Botnet members on the network of an organization even the packet encryption or changed features of flows, find out the Botnet members prior to the detection of the other information security systems. Moreover, the mechanism in this study is and simple to implement and can be used in student dormitory network, home network and mobile 3G network as well.

碩士
國立交通大學
資訊管理研究所
99
In cellular networks, when the Base Station (BS) has no the extra available channels, the handover procedure of communicating Mobile Station (MS) will be failed. As a result, it caused to call blocking which will decrease the customer satisfaction and result in financial loss. Therefore, the channel allocation for call block avoidance is an important issue of resource management in cellular networks. In this paper, we propose a mechanism which considers the real-time traffic information (e.g., traffic flow and vehicle speed) and the user behaviors (e.g., call inter-arrival time and call holding time) to analyze the adaptable amount of communication calls in the specific cell for channel allocation. In conducting the experiments of Call Block Probabilities (CBP), we simulate two cases by the situations of the whole day and traffic accident. The simulation results show that the CBP proposed by our scheme in the case of the whole day can decrease 3.37% CBP, compared to that of SCA scheme. Moreover, the CBP proposed by our scheme in the case of traffic accident can decrease 30.23% CBP, compared to that of SCA scheme. Therefore, our proposed mechanism is more effective that can decrease the number of CBP.

碩士
國立臺灣大學
物理學研究所
90
In a recent study, Ohira and Sawatari [Phys. Rev. E 58, 193 (1998)] presented a simple model of computer network traffic, which exhibits a phase transition between a low-traffic and congestion phase when the packets creation rate is increased. We further investigate the dynamics of this model by simplifying it into a random walk. It is shown that the packets clogged up around the midpoint of the boundary trigger the phase transition. Besides, the universal features observed among lattices of different sizes are analyzed quantitatively: the similarity may result from the average lifetime (steps taken from the host to destination) varying in proportion to the square of the creation rate; the saturation-like behavior occurring in the later period of congestion phase seems to come from the quantization in the number of nodes with cumulated queue. Finally, the relevance between different routing strategies and the performance of traffic is also discussed.

碩士
國立成功大學
電腦與通信工程研究所
101
In the view of the recent amendments of the Personal Information Protection Act triggered a lot of discussion on the issue of the transfer the burden of proof. We will develop a Network Behavior Tracking Admin System (NBTAS) based on network forensics. That is to say all of the log file which NBTAS analyzed will be preserved and authenticated. The method of preserving and authenticating is Modified Hash-based Message Authentication Code (Modified HMAC, MH), which is the study will propose. MH is based on HMAC, but we will modify the generated rules of the secret key. In order to increase the difficulty of cracking the message authentication codes which the attacker wants to do. Therefore, we could effectively control and reduce the events by the people with bad intention. Besides, MH will integrate the concept of Hash Tree, to reduce the authentication time in the face of judicial litigation. To enhancement of detecting the user suspicious network behavior, we will setup up blacklist and map with log file. If one of the connection is matched the blacklist, then we will record it.

The extracellular matrix (ECM) plays a significant role in defining the mechanical properties of biological tissues. The proteins, proteoglycans, and glycosaminoglycans that constitute the ECM are arranged into highly organized structures (e.g. fibrils and networks). Cellular behavior is affected by the stiffness of the microenvironment and influenced by the composition and organization of the ECM. Mechanosensing of ECM stiffness by cells occurs at the fibrillar (mesoscale) level between the single molecule (microscale) and the bulk tissue (macroscale) levels. However, the mechanical behavior of ECM proteins at the mesoscale are not well defined. Thus, better understanding of the ECM building blocks responsible for functional tissue assembly is critical in order to recapitulate in vivo conditions. There is a need for the mechanical characterization of the ECM networks formed by proteins synthesized in vivo while in their native configuration.

To address this gap, my goals highlighted in this dissertation were to develop appropriate experimental and computational methodologies and investigate the 3D organization and mechanical behavior of ECM networks in situ. The ECM of developing mouse tissues was used as a model system, taking advantage of the low-density networks present at this stage. First, we established a novel decellularization technique that enhanced the visualization of ECM networks in soft embryonic tissues. Based on this technique, we then quantified tissue-dependent strain of immunostained ECM networks in situ. Next, we developed mesoscale and macroscale testing systems to evaluate ECM networks under tension. Our systems were used to investigate tendon mechanics as a function of development, calculating tangent moduli from stress - strain plots. Similarly, we characterized ECM network deformation while uniaxially loading embryonic tissues, since this testing modality is ideal for fibril and network mechanics. Taken together, this information can facilitate the fabrication of physiologically relevant scaffolds for regenerative medicine by establishing mechanical guidelines for microenvironments facilitate functional tissue assembly.

Infrastructure is developing at an extremely fast pace which includes construction of metros, underground storage places, railway bridges, caverns and tunnels. Very often these structures are found in or on the rock masses. Rock masses are seldom found in nature without joints or discontinuities. Jointed rocks are characterized by the presence of inherent discontinuities of varied sizes with different orientations and intensities, which can have significant effect on their mechanical response. Constructions involving jointed rocks often become challenging jobs for Civil Engineers as the instability of slopes or excavations in these jointed rocks poses serious concerns, sometimes leading to the failure of structures built on them. Experimental investigations on jointed rock masses are not always feasible and pose formidable problems to the engineers. Apart from the technical difficulties of extracting undisturbed rock samples, it is very expensive and time consuming to conduct the experiments on jointed rock masses of huge dimensions. The most popular methods of evaluating the rock mass behaviour are the Numerical methods. In this thesis, numerical modelling of jointed rock masses is carried out using computer program FLAC (Fast Lagrangian Analysis of Continua). The objective of the present study is to study the effect of various joint parameters on the response of jointed rock masses in static as well as seismic shaking conditions. This is achieved through systematic series of numerical simulations of jointed rocks in triaxial compression, in underground openings and in large rock slopes. This thesis is an attempt to study the individual effect of different joint parameters on the rock mass behaviour and to integrate these results to provide useful insight into the behaviour of jointed rock mass under various joint conditions. In practice, it is almost impossible to explore all of the joint systems or to investigate all their mechanical characteristics and implementing them explicitly in the model. In these cases, the use of the equivalent continuum model to simulate the behaviour of jointed rock masses could be valuable. Hence this approach is mainly used in this thesis. Some numerical simulations with explicitly modelled joints are also presented for comparison with the continuum modelling. The applicability of Artificial Neural Networks for the prediction of stress-strain response of jointed rocks is also explored. Static, pseudo-static and dynamic analyses of a large rock slope in Himalayas is carried out and parametric seismic analysis of rock slope is carried out with varying input shaking, material damping and shear strength parameters. Results from the numerical studies showed that joint inclination is the most influencing parameter for the jointed rock mass behaviour. Rock masses exhibit lowest strength at critical angle of joint inclination and the deformations around excavations will be highest when the joints are inclined at an angle close to the critical angle. However at very high confining pressures, the influence of joint inclination gets subdued. Under seismic base shaking conditions, the deformations of rock masses largely depend on the acceleration response with time, frequency content and duration rather than the peak amplitude or the magnitude of earthquake. All these aspects are discussed in the light of results from numerical studies presented in this thesis.