Current Search: networking (x)
View All Items
(61 - 80 of 210)
Pages
- Title
- NEURAL NETWORKS SATISFYING STONE-WEIESTRASS THEOREM AND APPROXIMATING SCATTERED DATABYKOHONEN NEURAL NETWORKS.
- Creator
-
Thakkar, Pinal, Mohapatra, Ram, University of Central Florida
- Abstract / Description
-
Neural networks are an attempt to build computer networks called artificial neurons, which imitate the activities of the human brain. Its origin dates back to 1943 when neurophysiologist Warren Me Cello and logician Walter Pits produced the first artificial neuron. Since then there has been tremendous development of neural networks and their applications to pattern and optical character recognition, speech processing, time series prediction, image processing and scattered data approximation....
Show moreNeural networks are an attempt to build computer networks called artificial neurons, which imitate the activities of the human brain. Its origin dates back to 1943 when neurophysiologist Warren Me Cello and logician Walter Pits produced the first artificial neuron. Since then there has been tremendous development of neural networks and their applications to pattern and optical character recognition, speech processing, time series prediction, image processing and scattered data approximation. Since it has been shown that neural nets can approximate all but pathological functions, Neil Cotter considered neural network architecture based on Stone-Weierstrass Theorem. Using exponential functions, polynomials, rational functions and Boolean functions one can follow the method given by Cotter to obtain neural networks, which can approximate bounded measurable functions. Another problem of current research in computer graphics is to construct curves and surfaces from scattered spatial points by using B-Splines and NURBS or Bezier surfaces. Hoffman and Varady used Kohonen neural networks to construct appropriate grids. This thesis is concerned with two types of neural networks viz. those which satisfy the conditions of the Stone-Weierstrass theorem and Kohonen neural networks. We have used self-organizing maps for scattered data approximation. Neural network Tool Box from MATLAB is used to develop the required grids for approximating scattered data in one and two dimensions.
Show less - Date Issued
- 2004
- Identifier
- CFE0000226, ucf:46262
- Format
- Document (PDF)
- PURL
- http://purl.flvc.org/ucf/fd/CFE0000226
- Title
- On the design and performance of cognitive packets over wired networks and mobile ad hoc networks.
- Creator
-
Lent, Marino Ricardo, Gelenbe, Erol, Engineering and Computer Science
- Abstract / Description
-
University of Central Florida College of Engineering Thesis; This dissertation studied cognitive packet networks (CPN) which build networked learning systems that support adaptive, quality of service-driven routing of packets in wired networks and in wireless, mobile ad hoc networks.
- Date Issued
- 2003
- Identifier
- CFR0001374, ucf:52931
- Format
- Document (PDF)
- PURL
- http://purl.flvc.org/ucf/fd/CFR0001374
- Title
- Harmony: An Architecture for Network Centric Heterogeneous Terrain Database Re-Generation.
- Creator
-
Graniela Ortiz, Benito, Proctor, Michael, Gonzalez, Avelino, Wiegand, Rudolf, Goldiez, Brian, Cox, Robert, University of Central Florida
- Abstract / Description
-
This research investigated an alternative modeling and simulation terrain database generation paradigm that rapidly harmonizes changes target formats throughout a distributed simulation system while accommodating bandwidth and processing time limitations. This dissertation proposes a (")distributed partial bi-directional terrain database re-generation(") paradigm, which envisions network based terrain database updates between reliable partners. The approach is very attractive as it reduces...
Show moreThis research investigated an alternative modeling and simulation terrain database generation paradigm that rapidly harmonizes changes target formats throughout a distributed simulation system while accommodating bandwidth and processing time limitations. This dissertation proposes a (")distributed partial bi-directional terrain database re-generation(") paradigm, which envisions network based terrain database updates between reliable partners. The approach is very attractive as it reduces the amount of processing and bandwidth required to distribute locally emergent changes throughout a distributed system by only updating the affected target format data elements. In the prototype theoretical architecture that implements the approach, agent theory and ontologies are used to interpret data changes in external target formats and implement the necessary transformations on a server internal terrain database generation system. These changes are then distributed to clients to achieve consistency between all correlated representations. Experimental findings with the prototype suggests smaller network utilization and processing times than conventional terrain database generation will experience while maintaining correlated heterogeneous terrain database representations overtime. This Bi-Directional Ontology-driven TDB Re-Generation Architecture has the potential to revolutionize the traditional terrain database generation pipeline paradigm.
Show less - Date Issued
- 2011
- Identifier
- CFE0004475, ucf:49315
- Format
- Document (PDF)
- PURL
- http://purl.flvc.org/ucf/fd/CFE0004475
- Title
- Federal, State and Local Law Enforcement Agency Interoperability Capabilities and Cyber Vulnerabilities.
- Creator
-
Trapnell, Tyrone, Caulkins, Bruce, Wiegand, Rudolf, Bockelman, Patricia, Canham, Matthew, University of Central Florida
- Abstract / Description
-
The National Data Exchange (N-DEx) System is the central informational hub located at the Federal Bureau of Investigation (FBI). Its purpose is to provide network subscriptions to all Federal, state and local level law enforcement agencies while increasing information collaboration across all domains. The National Data Exchange users must satisfy the Advanced Permission Requirements, confirming the terms of N-DEx information use, and the Verification Requirement (verifying the completeness,...
Show moreThe National Data Exchange (N-DEx) System is the central informational hub located at the Federal Bureau of Investigation (FBI). Its purpose is to provide network subscriptions to all Federal, state and local level law enforcement agencies while increasing information collaboration across all domains. The National Data Exchange users must satisfy the Advanced Permission Requirements, confirming the terms of N-DEx information use, and the Verification Requirement (verifying the completeness, timeliness, accuracy, and relevancy of N-DEx information) through coordination with the record-owning agency (Management, 2018). A network infection model is proposed to simulate the spread impact of various cyber-attacks within Federal, state and local level law enforcement networks that are linked together through the topologies merging with the National Data Exchange (N-DEx) System as the ability to manipulate the live network is limited. The model design methodology is conducted in a manner that creates a level of organization from the state level to the local level of law enforcement agencies allowing for each organizational infection probability to be calculated and entered, thus making the model very specific in nature for determining spread or outbreaks of cyber-attacks among law enforcement agencies at all levels. This research will enable future researchers to further develop a model that is capable of detecting weak points within an information structure when multiple topologies merge, allowing for more secure operations among law enforcement networks.
Show less - Date Issued
- 2019
- Identifier
- CFE0007543, ucf:52621
- Format
- Document (PDF)
- PURL
- http://purl.flvc.org/ucf/fd/CFE0007543
- Title
- Modeling Crowd Mobility and Communication in Wireless Networks.
- Creator
-
Solmaz, Gurkan, Turgut, Damla, Bassiouni, Mostafa, Guha, Ratan, Goldiez, Brian, University of Central Florida
- Abstract / Description
-
This dissertation presents contributions to the fields of mobility modeling, wireless sensor networks (WSNs) with mobile sinks, and opportunistic communication in theme parks. The two main directions of our contributions are human mobility models and strategies for the mobile sink positioning and communication in wireless networks.The first direction of the dissertation is related to human mobility modeling. Modeling the movement of human subjects is important to improve the performance of...
Show moreThis dissertation presents contributions to the fields of mobility modeling, wireless sensor networks (WSNs) with mobile sinks, and opportunistic communication in theme parks. The two main directions of our contributions are human mobility models and strategies for the mobile sink positioning and communication in wireless networks.The first direction of the dissertation is related to human mobility modeling. Modeling the movement of human subjects is important to improve the performance of wireless networks with human participants and the validation of such networks through simulations. The movements in areas such as theme parks follow specific patterns that are not taken into consideration by the general purpose mobility models. We develop two types of mobility models of theme park visitors. The first model represents the typical movement of visitors as they are visiting various attractions and landmarks of the park. The second model represents the movement of the visitors as they aim to evacuate the park after a natural or man-made disaster.The second direction focuses on the movement patterns of mobile sinks and their communication in responding to various events and incidents within the theme park. When an event occurs, the system needs to determine which mobile sink will respond to the event and its trajectory. The overall objective is to optimize the event coverage by minimizing the time needed for the chosen mobile sink to reach the incident area. We extend this work by considering the positioning problem of mobile sinks and preservation of the connected topology. We propose a new variant of p-center problem for optimal placement and communication of the mobile sinks. We provide a solution to this problem through collaborative event coverage of the WSNs with mobile sinks. Finally, we develop a network model with opportunistic communication for tracking the evacuation of theme park visitors during disasters. This model involves people with smartphones that store and carry messages. The mobile sinks are responsible for communicating with the smartphones and reaching out to the regions of the emergent events.
Show less - Date Issued
- 2015
- Identifier
- CFE0006005, ucf:51024
- Format
- Document (PDF)
- PURL
- http://purl.flvc.org/ucf/fd/CFE0006005
- Title
- Hydrologic controls on the natural drainage networks extracted from high-resolution topographic data.
- Creator
-
Hooshyar, Milad, Wang, Dingbao, Medeiros, Stephen, Singh, Arvind, Kibler, Kelly, Weishampel, John, University of Central Florida
- Abstract / Description
-
Drainage networks are important geomorphologic and hydrologic features which significantly control runoff generation. Drainage networks are composed of unchannelized valleys and channels. At valley heads, flow changes from unconfined sheet flow on the hillslope to confined flow in valley. Localized confined flow dominates in valleys as a result of convergent topography with positive curvature. Channels initiate at some distance down from the valley head, and the transition from unchannelized...
Show moreDrainage networks are important geomorphologic and hydrologic features which significantly control runoff generation. Drainage networks are composed of unchannelized valleys and channels. At valley heads, flow changes from unconfined sheet flow on the hillslope to confined flow in valley. Localized confined flow dominates in valleys as a result of convergent topography with positive curvature. Channels initiate at some distance down from the valley head, and the transition from unchannelized valley to channel is referred to as the channel head. Channel heads occur at a point where fluvial transport dominates over diffusive transport.From the hydrologic perspective, channels are categorized as perennial, intermittent, and ephemeral streams based on the flow durations. Perennial streams flow for the most of the time during normal years and are maintained by groundwater discharge. Intermittent (i.e. seasonal) streams flow during certain times of the year receiving water from surface sources such as melting snow or from groundwater. Lastly, ephemeral streams flow only in direct response to precipitation without continuous surface flow. In this dissertation, the hydrologic controls on the drainage networks extracted from high resolution Digital Elevation Models (DEMs) based on Light Detection and Ranging (LiDAR) are investigated. A method for automatic extraction of valley and channel networks from high-resolution DEMs is presented. This method utilizes both positive (i.e., convergent topography) and negative (i.e., divergent topography) curvature to delineate the valley network. The valley and ridge skeletons are extracted using the pixels' curvature and the local terrain conditions. The valley network is generated by checking the terrain for the existence of at least one ridge between two intersecting valleys. The transition from unchannelized to channelized sections (i.e., channel head) in each 1st-order valley tributary is identified independently by categorizing the corresponding contours using an unsupervised approach based on K-means clustering. The method does not require a spatially constant channel initiation threshold (e.g., curvature or contributing area). Moreover, instead of a point attribute (e.g., curvature), the proposed clustering method utilizes the shape of contours, which reflects the entire cross-sectional profile including possible banks. The method was applied to three catchments: Indian Creek and Mid Bailey Run in Ohio, and Feather River in California. The accuracy of channel head extraction from the proposed method is comparable to state-of-the-art channel extraction methods. Valleys extracted from DEMs may be wet (flowing) or dry at any given time depending on the hydrologic conditions. The temporal dynamics of flowing streams are vitally important for understanding hydrologic processes including surface water and groundwater interaction and hydrograph recession. However, observations of wet channel networks are limited, especially in headwater catchments. Near infrared LiDAR data provide an opportunity to map wet channel networks owing to the fine spatial resolution and strong absorption of light energy by water surfaces. A systematic method is developed to map wet channel networks by integrating elevation and signal intensity of ground returns. The signal intensity thresholds for identifying wet pixels are extracted from frequency distributions of intensity return within the convergent topography extent using a Gaussian mixture model. Moreover, the concept of edge in digital image processing, defined based on the intensity gradient, is utilized to enhance detection of small wet channels. The developed method was applied to the Lake Tahoe area based on eight LiDAR acquisitions during recession periods in five watersheds. A power-law relationship between streamflow and wetted channel length during recession periods was derived, and the scaling exponent (L?Q^0.38) is within the range of reported values from fieldwork in other regions.Several studies in the past focused on the relationship between drainage density (i.e., drainage length divided by drainage area) and long-term climate and reported a U-shape pattern. In this dissertation, this relationship was re-visited and the effect of drainage area on drainage density was investigated. Long-term climate was quantified by climate aridity indices which is the ratio between long-term potential evaporation and precipitation. 120 study sites across the United States with minimal human disturbance and a wide range of climate aridity index were selected based on the availability of LiDAR data. The drainage networks were delineated from LiDAR-based 1 m DEMs using the proposed curvature-based method. Despite the U-shaped relationship in the literature, our result shows a significant decreasing trend in the drainage density versus climate aridity index in arid regions; whereas no trend is observed in humid watersheds. This observation and its discrepancy with the reported pattern in the literature are justified considering the dynamics of the runoff erosive force and the resistance of vegetation and the climate controls on them. Our findings suggest that natural drainage networks in arid regions are more sensitive to the change in long-term climate conditions compared with drainage networks in humid climate. It was also found that drainage density has a decreasing trend with drainage area in arid regions; however, no trend was observed in humid regions. In a broader sense, the findings influence our understanding of the formation of drainage networks and the response of hydrologic systems to climate change. The formation and growth of river channels and their network evolution are governed by the erosional and depositional processes operating on the landscape due to movement of water. The branching angles, i.e., the angle between two adjoining channels, in drainage networks are important features related to the network topology and contain valuable information about the forming mechanisms of the landscape. Based on channel networks extracted from 1 m Digital Elevation Models of 120 catchments with minimal human impacts across the United States, we showed that the junction angles have two distinct modes with ?1 ? 49.5(&)deg; and ?2 ? 75.0(&)deg;. The observed angles are physically explained as the optimal angles that result in minimum energy dissipation and are linked to the exponent characterizing slope-area curve. Our findings suggest that the flow regimes, debris-flow dominated or fluvial, have distinct characteristic angles which are functions of the scaling exponent of the slope-area curve. These findings enable us to understand the geomorphologic signature of hydrologic processes on drainage networks and develop more refined landscape evolution models.
Show less - Date Issued
- 2017
- Identifier
- CFE0006604, ucf:51278
- Format
- Document (PDF)
- PURL
- http://purl.flvc.org/ucf/fd/CFE0006604
- Title
- Pervasive Secure Content Delivery Networks Implementation.
- Creator
-
Lugo-Cordero, Hector, Guha, Ratan, Stanley, Kenneth, Chatterjee, Mainak, Wu, Annie, Lu, Kejie, University of Central Florida
- Abstract / Description
-
Over the years, communication networks have been shifting their focus from providing connectivity in a client/server model to providing a service or content. This shift has led to topic areas like Service-Oriented Architecture (SOA), Heterogeneous Wireless Mesh Networks, and Ubiquitous Computing. Furthermore, probably the broadest of these areas which embarks all is the Internet of Things (IoT). The IoT is defined as an Internet where all physical entities (e.g., vehicles, appliances, smart...
Show moreOver the years, communication networks have been shifting their focus from providing connectivity in a client/server model to providing a service or content. This shift has led to topic areas like Service-Oriented Architecture (SOA), Heterogeneous Wireless Mesh Networks, and Ubiquitous Computing. Furthermore, probably the broadest of these areas which embarks all is the Internet of Things (IoT). The IoT is defined as an Internet where all physical entities (e.g., vehicles, appliances, smart phones, smart homes, computers, etc.), which we interact daily are connected and exchanging data among themselves and users. The IoT has become a global goal for companies, researchers, and users alike due to its different implementation and functional benefits: performance efficiency, coverage, economic and health. Due to the variety of devices which connect to it, it is expected that the IoT is composed of multiple technologies interacting together, to deliver a service. This technologies interactions renders an important challenge that must be overcome: how to communicate these technologies effectively and securely? The answer to this question is vital for a successful deployment of IoT and achievement of all the potential benefits that the IoT promises.This thesis proposes a SOA approach at the Network Layer to be able to integrate all technologies involved, in a transparent manner. The proposed set of solutions is composed of primarily the secure implementation of a unifying routing algorithm and a layered messaging model to standardizecommunication of all devices. Security is targeted to address the three main security concerns (i.e., confidentiality, integrity, and availability), with pervasive schemes that can be employed for any kind of device on the client, backbone, and server side. The implementation of such schemes is achieved by standard current security mechanisms (e.g., encryption), in combination with novel context and intelligent checks that detect compromised devices. Moreover, a decentralized content processing design is presented. In such design, content processing is handled at the client side, allowing server machines to serve more content, while being more reliable and capable of processing complete security checks on data and client integrity.
Show less - Date Issued
- 2017
- Identifier
- CFE0006620, ucf:51268
- Format
- Document (PDF)
- PURL
- http://purl.flvc.org/ucf/fd/CFE0006620
- Title
- Load-Balancing in Local and Metro-Area networks with MPTCP and OpenFlow.
- Creator
-
Jerome, Austin, Bassiouni, Mostafa, Yuksel, Murat, Zou, Changchun, Jin, Yier, University of Central Florida
- Abstract / Description
-
In this thesis, a novel load-balancing technique for local or metro-area traffic is proposed in mesh-style topologies. The technique uses Software Defined Networking (SDN) architecture with virtual local area network (VLAN) setups typically seen in a campus or small-to-medium enterprise environment. This was done to provide a possible solution or at least a platform to expand on for the load-balancing dilemma that network administrators face today. The transport layer protocol Multi-Path TCP ...
Show moreIn this thesis, a novel load-balancing technique for local or metro-area traffic is proposed in mesh-style topologies. The technique uses Software Defined Networking (SDN) architecture with virtual local area network (VLAN) setups typically seen in a campus or small-to-medium enterprise environment. This was done to provide a possible solution or at least a platform to expand on for the load-balancing dilemma that network administrators face today. The transport layer protocol Multi-Path TCP (MPTCP) coupled with IP aliasing is also used. The trait of MPTCP of forming multiple subflows from sender to receiver depending on the availability of IP addresses at either the sender or receiver helps to divert traffic in the subflows across all available paths. The combination of MPTCP subflows with IP aliasing enables spreading out of the traffic load across greater number of links in the network, and thereby achieving load balancing and better network utilization. The traffic formed of each subflow would be forwarded across the network based on Hamiltonian 'paths' which are created in association with each switch in the topology which are directly connected to hosts. The amount of 'paths' in the topology would also depend on the number of VLANs setup for the hosts in the topology. This segregation would allow for network administrators to monitor network utilization across VLANs and give the ability to balance load across VLANs. We have devised several experiments in Mininet, and the experimentation showed promising results with significantly better throughput and network utilization compared to cases where normal TCP was used to send traffic from source to destination. Our study clearly shows the advantages of using MPTCP for load balancing purposes in SDN type architectures and provides a platform for future research on using VLANs, SDN, and MPTCP for network traffic management.
Show less - Date Issued
- 2017
- Identifier
- CFE0006887, ucf:51705
- Format
- Document (PDF)
- PURL
- http://purl.flvc.org/ucf/fd/CFE0006887
- Title
- Networking and security solutions for VANET initial deployment stage.
- Creator
-
Aslam, Baber, Zou, Changchun, Turgut, Damla, Bassiouni, Mostafa, Wang, Chung-Ching, University of Central Florida
- Abstract / Description
-
Vehicular ad hoc network (VANET) is a special case of mobile networks, where vehicles equipped with computing/communicating devices (called (")smart vehicles(")) are the mobile wireless nodes. However, the movement pattern of these mobile wireless nodes is no more random, as in case of mobile networks, rather it is restricted to roads and streets. Vehicular networks have hybrid architecture; it is a combination of both infrastructure and infrastructure-less architectures. The direct vehicle...
Show moreVehicular ad hoc network (VANET) is a special case of mobile networks, where vehicles equipped with computing/communicating devices (called (")smart vehicles(")) are the mobile wireless nodes. However, the movement pattern of these mobile wireless nodes is no more random, as in case of mobile networks, rather it is restricted to roads and streets. Vehicular networks have hybrid architecture; it is a combination of both infrastructure and infrastructure-less architectures. The direct vehicle to vehicle (V2V) communication is infrastructure-less or ad hoc in nature. Here the vehicles traveling within communication range of each other form an ad hoc network. On the other hand, the vehicle to infrastructure (V2I) communication has infrastructure architecture where vehicles connect to access points deployed along roads. These access points are known as road side units (RSUs) and vehicles communicate with other vehicles/wired nodes through these RSUs. To provide various services to vehicles, RSUs are generally connected to each other and to the Internet. The direct RSU to RSU communication is also referred as I2I communication. The success of VANET depends on the existence of pervasive roadside infrastructure and sufficient number of smart vehicles. Most VANET applications and services are based on either one or both of these requirements. A fully matured VANET will have pervasive roadside network and enough vehicle density to enable VANET applications. However, the initial deployment stage of VANET will be characterized by the lack of pervasive roadside infrastructure and low market penetration of smart vehicles. It will be economically infeasible to initially install a pervasive and fully networked roadside infrastructure, which could result in the failure of applications and services that depend on V2I or I2I communications. Further, low market penetration means there are insufficient number of smart vehicles to enable V2V communication, which could result in failure of services and applications that depend on V2V communications. Non-availability of pervasive connectivity to certification authorities and dynamic locations of each vehicle will make it difficult and expensive to implement security solutions that are based on some central certificate management authority. Non-availability of pervasive connectivity will also affect the backend connectivity of vehicles to the Internet or the rest of the world. Due to economic considerations, the installation of roadside infrastructure will take a long time and will be incremental thus resulting in a heterogeneous infrastructure with non-consistent capabilities. Similarly, smart vehicles will also have varying degree of capabilities. This will result in failure of applications and services that have very strict requirements on V2I or V2V communications. We have proposed several solutions to overcome the challenges described above that will be faced during the initial deployment stage of VANET. Specifically, we have proposed: 1) a VANET architecture that can provide services with limited number of heterogeneous roadside units and smart vehicles with varying capabilities, 2) a backend connectivity solution that provides connectivity between the Internet and smart vehicles without requiring pervasive roadside infrastructure or large number of smart vehicles, 3) a security architecture that does not depend on pervasive roadside infrastructure or a fully connected V2V network and fulfills all the security requirements, and 4) optimization solutions for placement of a limited number of RSUs within a given area to provide best possible service to smart vehicles. The optimal placement solutions cover both urban areas and highways environments.
Show less - Date Issued
- 2012
- Identifier
- CFE0004186, ucf:48993
- Format
- Document (PDF)
- PURL
- http://purl.flvc.org/ucf/fd/CFE0004186
- Title
- Quantifying Trust and Reputation for Defense against Adversaries in Multi-Channel Dynamic Spectrum Access Networks.
- Creator
-
Bhattacharjee, Shameek, Chatterjee, Mainak, Guha, Ratan, Zou, Changchun, Turgut, Damla, Catbas, Necati, University of Central Florida
- Abstract / Description
-
Dynamic spectrum access enabled by cognitive radio networks are envisioned to drivethe next generation wireless networks that can increase spectrum utility by opportunisticallyaccessing unused spectrum. Due to the policy constraint that there could be no interferenceto the primary (licensed) users, secondary cognitive radios have to continuously sense forprimary transmissions. Typically, sensing reports from multiple cognitive radios are fusedas stand-alone observations are prone to errors...
Show moreDynamic spectrum access enabled by cognitive radio networks are envisioned to drivethe next generation wireless networks that can increase spectrum utility by opportunisticallyaccessing unused spectrum. Due to the policy constraint that there could be no interferenceto the primary (licensed) users, secondary cognitive radios have to continuously sense forprimary transmissions. Typically, sensing reports from multiple cognitive radios are fusedas stand-alone observations are prone to errors due to wireless channel characteristics. Suchdependence on cooperative spectrum sensing is vulnerable to attacks such as SecondarySpectrum Data Falsification (SSDF) attacks when multiple malicious or selfish radios falsifythe spectrum reports. Hence, there is a need to quantify the trustworthiness of radios thatshare spectrum sensing reports and devise malicious node identification and robust fusionschemes that would lead to correct inference about spectrum usage.In this work, we propose an anomaly monitoring technique that can effectively cap-ture anomalies in the spectrum sensing reports shared by individual cognitive radios duringcooperative spectrum sensing in a multi-channel distributed network. Such anomalies areused as evidence to compute the trustworthiness of a radio by its neighbours. The proposedanomaly monitoring technique works for any density of malicious nodes and for any physicalenvironment. We propose an optimistic trust heuristic for a system with a normal risk attitude and show that it can be approximated as a beta distribution. For a more conservativesystem, we propose a multinomial Dirichlet distribution based conservative trust framework,where Josang's Belief model is used to resolve any uncertainty in information that mightarise during anomaly monitoring. Using a machine learning approach, we identify maliciousnodes with a high degree of certainty regardless of their aggressiveness and variations intro-duced by the pathloss environment. We also propose extensions to the anomaly monitoringtechnique that facilitate learning about strategies employed by malicious nodes and alsoutilize the misleading information they provide. We also devise strategies to defend against a collaborative SSDF attack that islaunched by a coalition of selfish nodes. Since, defense against such collaborative attacks isdifficult with popularly used voting based inference models or node centric isolation techniques, we propose a channel centric Bayesian inference approach that indicates how much the collective decision on a channels occupancy inference can be trusted. Based on the measured observations over time, we estimate the parameters of the hypothesis of anomalous andnon-anomalous events using a multinomial Bayesian based inference. We quantitatively define the trustworthiness of a channel inference as the difference between the posterior beliefsassociated with anomalous and non-anomalous events. The posterior beliefs are updated based on a weighted average of the prior information on the belief itself and the recently observed data.Subsequently, we propose robust fusion models which utilize the trusts of the nodes to improve the accuracy of the cooperative spectrum sensing decisions. In particular, we propose three fusion models: (i) optimistic trust based fusion, (ii) conservative trust based fusion, and (iii) inversion based fusion. The former two approaches exclude untrustworthy sensing reports for fusion, while the last approach utilizes misleading information. Allschemes are analyzed under various attack strategies. We propose an asymmetric weightedmoving average based trust management scheme that quickly identifies on-off SSDF attacks and prevents quick trust redemption when such nodes revert back to temporal honest behavior. We also provide insights on what attack strategies are more effective from the adversaries' perspective.Through extensive simulation experiments we show that the trust models are effective in identifying malicious nodes with a high degree of certainty under variety of network and radio conditions. We show high true negative detection rates even when multiple malicious nodes launch collaborative attacks which is an improvement over existing voting based exclusion and entropy divergence techniques. We also show that we are able to improve the accuracy of fusion decisions compared to other popular fusion techniques. Trust based fusion schemes show worst case decision error rates of 5% while inversion based fusion show 4% as opposed majority voting schemes that have 18% error rate. We also show that the proposed channel centric Bayesian inference based trust model is able to distinguish between attacked and non-attacked channels for both static and dynamic collaborative attacks. We are also able to show that attacked channels have significantly lower trust values than channels that are not(-) a metric that can be used by nodes to rank the quality of inference on channels.
Show less - Date Issued
- 2015
- Identifier
- CFE0005764, ucf:50081
- Format
- Document (PDF)
- PURL
- http://purl.flvc.org/ucf/fd/CFE0005764
- Title
- MODIFICATIONS TO THE FUZZY-ARTMAP ALGORITHM FOR DISTRIBUTED LEARNING IN LARGE DATA SETS.
- Creator
-
Castro, Jose R, Georgiopoulos, Michael, University of Central Florida
- Abstract / Description
-
The Fuzzy-ARTMAP (FAM) algorithm is one of the premier neural network architectures for classification problems. FAM can learn on line and is usually faster than other neural network approaches. Nevertheless the learning time of FAM can slow down considerably when the size of the training set increases into the hundreds of thousands. We apply data partitioning and networkpartitioning to the FAM algorithm in a sequential and parallel settingto achieve better convergence time and to efficiently...
Show moreThe Fuzzy-ARTMAP (FAM) algorithm is one of the premier neural network architectures for classification problems. FAM can learn on line and is usually faster than other neural network approaches. Nevertheless the learning time of FAM can slow down considerably when the size of the training set increases into the hundreds of thousands. We apply data partitioning and networkpartitioning to the FAM algorithm in a sequential and parallel settingto achieve better convergence time and to efficiently train withlarge databases (hundreds of thousands of patterns).Our parallelization is implemented on a Beowulf clusters of workstations. Two data partitioning approaches and two networkpartitioning approaches are developed. Extensive testing of all the approaches is done on three large datasets (half a milliondata points). One of them is the Forest Covertype database from Blackard and the other two are artificially generated Gaussian data with different percentages of overlap between classes.Speedups in the data partitioning approach reached the order of the hundreds without having to invest in parallel computation. Speedups onthe network partitioning approach are close to linear on a cluster of workstations. Both methods allowed us to reduce the computation time of training the neural network in large databases from days to minutes. We prove formally that the workload balance of our network partitioning approaches will never be worse than an acceptable bound, and also demonstrate the correctness of these parallelization variants of FAM.
Show less - Date Issued
- 2004
- Identifier
- CFE0000065, ucf:46092
- Format
- Document (PDF)
- PURL
- http://purl.flvc.org/ucf/fd/CFE0000065
- Title
- A HYBRID SIMULATION METHODOLOGY TO EVALUATE NETWORK CENTRICDECISION MAKING UNDER EXTREME EVENTS.
- Creator
-
Quijada, Sergio, Sepulveda, Jose, University of Central Florida
- Abstract / Description
-
Currently the network centric operation and network centric warfare have generated a new area of research focused on determining how hierarchical organizations composed by human beings and machines make decisions over collaborative environments. One of the most stressful scenarios for these kinds of organizations is the so-called extreme events. This dissertation provides a hybrid simulation methodology based on classical simulation paradigms combined with social network analysis for...
Show moreCurrently the network centric operation and network centric warfare have generated a new area of research focused on determining how hierarchical organizations composed by human beings and machines make decisions over collaborative environments. One of the most stressful scenarios for these kinds of organizations is the so-called extreme events. This dissertation provides a hybrid simulation methodology based on classical simulation paradigms combined with social network analysis for evaluating and improving the organizational structures and procedures, mainly the incident command systems and plans for facing those extreme events. According to this, we provide a methodology for generating hypotheses and afterwards testing organizational procedures either in real training systems or simulation models with validated data. As long as the organization changes their dyadic relationships dynamically over time, we propose to capture the longitudinal digraph in time and analyze it by means of its adjacency matrix. Thus, by using an object oriented approach, three domains are proposed for better understanding the performance and the surrounding environment of an emergency management organization. System dynamics is used for modeling the critical infrastructure linked to the warning alerts of a given organization at federal, state and local levels. Discrete simulations based on the defined concept of "community of state" enables us to control the complete model. Discrete event simulation allows us to create entities that represent the data and resource flows within the organization. We propose that cognitive models might well be suited in our methodology. For instance, we show how the team performance decays in time, according to the Yerkes-Dodson curve, affecting the measures of performance of the whole organizational system. Accordingly we suggest that the hybrid model could be applied to other types of organizations, such as military peacekeeping operations and joint task forces. Along with providing insight about organizations, the methodology supports the analysis of the "after action review" (AAR), based on collection of data obtained from the command and control systems or the so-called training scenarios. Furthermore, a rich set of mathematical measures arises from the hybrid models such as triad census, dyad census, eigenvalues, utilization, feedback loops, etc., which provides a strong foundation for studying an emergency management organization. Future research will be necessary for analyzing real data and validating the proposed methodology.
Show less - Date Issued
- 2006
- Identifier
- CFE0001243, ucf:46926
- Format
- Document (PDF)
- PURL
- http://purl.flvc.org/ucf/fd/CFE0001243
- Title
- A LIFE CYCLE SOFTWARE QUALITY MODEL USING BAYESIAN BELIEF NETWORKS.
- Creator
-
Beaver, Justin, Schiavone, Guy, University of Central Florida
- Abstract / Description
-
Software practitioners lack a consistent approach to assessing and predicting quality within their products. This research proposes a software quality model that accounts for the influences of development team skill/experience, process maturity, and problem complexity throughout the software engineering life cycle. The model is structured using Bayesian Belief Networks and, unlike previous efforts, uses widely-accepted software engineering standards and in-use industry techniques to quantify...
Show moreSoftware practitioners lack a consistent approach to assessing and predicting quality within their products. This research proposes a software quality model that accounts for the influences of development team skill/experience, process maturity, and problem complexity throughout the software engineering life cycle. The model is structured using Bayesian Belief Networks and, unlike previous efforts, uses widely-accepted software engineering standards and in-use industry techniques to quantify the indicators and measures of software quality. Data from 28 software engineering projects was acquired for this study, and was used for validation and comparison of the presented software quality models. Three Bayesian model structures are explored and the structure with the highest performance in terms of accuracy of fit and predictive validity is reported. In addition, the Bayesian Belief Networks are compared to both Least Squares Regression and Neural Networks in order to identify the technique is best suited to modeling software product quality. The results indicate that Bayesian Belief Networks outperform both Least Squares Regression and Neural Networks in terms of producing modeled software quality variables that fit the distribution of actual software quality values, and in accurately forecasting 25 different indicators of software quality. Between the Bayesian model structures, the simplest structure, which relates software quality variables to their correlated causal factors, was found to be the most effective in modeling software quality. In addition, the results reveal that the collective skill and experience of the development team, over process maturity or problem complexity, has the most significant impact on the quality of software products.
Show less - Date Issued
- 2006
- Identifier
- CFE0001367, ucf:46993
- Format
- Document (PDF)
- PURL
- http://purl.flvc.org/ucf/fd/CFE0001367
- Title
- EVOLUTIONARY OPTIMIZATION OF SUPPORT VECTOR MACHINES.
- Creator
-
Gruber, Fred, Rabelo, Luis, University of Central Florida
- Abstract / Description
-
Support vector machines are a relatively new approach for creating classifiers that have become increasingly popular in the machine learning community. They present several advantages over other methods like neural networks in areas like training speed, convergence, complexity control of the classifier, as well as a stronger mathematical background based on optimization and statistical learning theory. This thesis deals with the problem of model selection with support vector machines, that is...
Show moreSupport vector machines are a relatively new approach for creating classifiers that have become increasingly popular in the machine learning community. They present several advantages over other methods like neural networks in areas like training speed, convergence, complexity control of the classifier, as well as a stronger mathematical background based on optimization and statistical learning theory. This thesis deals with the problem of model selection with support vector machines, that is, the problem of finding the optimal parameters that will improve the performance of the algorithm. It is shown that genetic algorithms provide an effective way to find the optimal parameters for support vector machines. The proposed algorithm is compared with a backpropagation Neural Network in a dataset that represents individual models for electronic commerce.
Show less - Date Issued
- 2004
- Identifier
- CFE0000244, ucf:46251
- Format
- Document (PDF)
- PURL
- http://purl.flvc.org/ucf/fd/CFE0000244
- Title
- NON-SILICON MICROFABRICATED NANOSTRUCTURED CHEMICAL SENSORS FOR ELECTRIC NOSE APPLICATION.
- Creator
-
Gong, Jianwei, Chen, Quanfang, University of Central Florida
- Abstract / Description
-
A systematic investigation has been performed for "Electric Nose", a system that can identify gas samples and detect their concentrations by combining sensor array and data processing technologies. Non-silicon based microfabricatition has been developed for micro-electro-mechanical-system (MEMS) based gas sensors. Novel sensors have been designed, fabricated and tested. Nanocrystalline semiconductor metal oxide (SMO) materials include SnO2, WO3 and In2O3 have been studied for gas sensing...
Show moreA systematic investigation has been performed for "Electric Nose", a system that can identify gas samples and detect their concentrations by combining sensor array and data processing technologies. Non-silicon based microfabricatition has been developed for micro-electro-mechanical-system (MEMS) based gas sensors. Novel sensors have been designed, fabricated and tested. Nanocrystalline semiconductor metal oxide (SMO) materials include SnO2, WO3 and In2O3 have been studied for gas sensing applications. Different doping material such as copper, silver, platinum and indium are studied in order to achieve better selectivity for different targeting toxic gases including hydrogen, carbon monoxide, hydrogen sulfide etc. Fundamental issues like sensitivity, selectivity, stability, temperature influence, humidity influence, thermal characterization, drifting problem etc. of SMO gas sensors have been intensively investigated. A novel approach to improve temperature stability of SMO (including tin oxide) gas sensors by applying a temperature feedback control circuit has been developed. The feedback temperature controller that is compatible with MEMS sensor fabrication has been invented and applied to gas sensor array system. Significant improvement of stability has been achieved compared to SMO gas sensors without temperature compensation under the same ambient conditions. Single walled carbon nanotube (SWNT) has been studied to improve SnO2 gas sensing property in terms of sensitivity, response time and recovery time. Three times of better sensitivity has been achieved experimentally. The feasibility of using TSK Fuzzy neural network algorithm for Electric Nose has been exploited during the research. A training process of using TSK Fuzzy neural network with input/output pairs from individual gas sensor cell has been developed. This will make electric nose smart enough to measure gas concentrations in a gas mixture. The model has been proven valid by gas experimental results conducted.
Show less - Date Issued
- 2005
- Identifier
- CFE0000377, ucf:46328
- Format
- Document (PDF)
- PURL
- http://purl.flvc.org/ucf/fd/CFE0000377
- Title
- LEARNING HUMAN BEHAVIOR FROM OBSERVATION FOR GAMING APPLICATIONS.
- Creator
-
Moriarty, Christopher, Gonzalez, Avelino, University of Central Florida
- Abstract / Description
-
The gaming industry has reached a point where improving graphics has only a small effect on how much a player will enjoy a game. One focus has turned to adding more humanlike characteristics into computer game agents. Machine learning techniques are being used scarcely in games, although they do offer powerful means for creating humanlike behaviors in agents. The first person shooter (FPS), Quake 2, is an open source game that offers a multi-agent environment to create game agents (bots) in....
Show moreThe gaming industry has reached a point where improving graphics has only a small effect on how much a player will enjoy a game. One focus has turned to adding more humanlike characteristics into computer game agents. Machine learning techniques are being used scarcely in games, although they do offer powerful means for creating humanlike behaviors in agents. The first person shooter (FPS), Quake 2, is an open source game that offers a multi-agent environment to create game agents (bots) in. This work attempts to combine neural networks with a modeling paradigm known as context based reasoning (CxBR) to create a contextual game observation (CONGO) system that produces Quake 2 agents that behave as a human player trains them to act. A default level of intelligence is instilled into the bots through contextual scripts to prevent the bot from being trained to be completely useless. The results show that the humanness and entertainment value as compared to a traditional scripted bot have improved, although, CONGO bots usually ranked only slightly above a novice skill level. Overall, CONGO is a technique that offers the gaming community a mode of game play that has promising entertainment value.
Show less - Date Issued
- 2007
- Identifier
- CFE0001694, ucf:47201
- Format
- Document (PDF)
- PURL
- http://purl.flvc.org/ucf/fd/CFE0001694
- Title
- BLUETOOTH-BASE WORM MODELING AND SIMULATION.
- Creator
-
Xiang, Haiou, Zou, Cliff, University of Central Florida
- Abstract / Description
-
Bluetooth is one of the most popular technologies in the world in the new century. Meanwhile it attracts attackers to develop new worm and malicious code attacking Bluetooth wireless network. So far the growth of mobile malicious code is very fast and they have become a great potential threat to our society. In this thesis, we study Bluetooth worm in Mobile Wireless Network. Firstly we investigate the Bluetooth technology and several previously appeared Bluetooth worms, e.g. "Caribe","Comwar"...
Show moreBluetooth is one of the most popular technologies in the world in the new century. Meanwhile it attracts attackers to develop new worm and malicious code attacking Bluetooth wireless network. So far the growth of mobile malicious code is very fast and they have become a great potential threat to our society. In this thesis, we study Bluetooth worm in Mobile Wireless Network. Firstly we investigate the Bluetooth technology and several previously appeared Bluetooth worms, e.g. "Caribe","Comwar", and we find the infection cycle of a Bluetooth worm. Next, we develop a new simulator, Bluetooth Worm simulator (BTWS), which simulates Bluetooth worm' behaviors in Mobile wireless networks. Through analyzing the result, we find i) In ideal environment the mobility of Bluetooth device can improve the worm's propagation speed, but combining mobility and inquiry time issue would cause a Bluetooth worm to slow down its propagation under certain situation. ii) The number of initially infected Bluetooth devices mostly affects the beginning propagation speed of a worm, and energy issue can be ignored because the new technology can let Bluetooth device keeping work for a long time. iii) Co-channel interference and setting up monitoring system in public place can improve the security of Bluetooth wireless network.
Show less - Date Issued
- 2007
- Identifier
- CFE0001740, ucf:47313
- Format
- Document (PDF)
- PURL
- http://purl.flvc.org/ucf/fd/CFE0001740
- Title
- NETWORK INTRUSION DETECTION: MONITORING, SIMULATION ANDVISUALIZATION.
- Creator
-
Zhou, Mian, Lang, Sheau-Dong, University of Central Florida
- Abstract / Description
-
This dissertation presents our work on network intrusion detection and intrusion sim- ulation. The work in intrusion detection consists of two different network anomaly-based approaches. The work in intrusion simulation introduces a model using explicit traffic gen- eration for the packet level traffic simulation. The process of anomaly detection is to first build profiles for the normal network activity and then mark any events or activities that deviate from the normal profiles as...
Show moreThis dissertation presents our work on network intrusion detection and intrusion sim- ulation. The work in intrusion detection consists of two different network anomaly-based approaches. The work in intrusion simulation introduces a model using explicit traffic gen- eration for the packet level traffic simulation. The process of anomaly detection is to first build profiles for the normal network activity and then mark any events or activities that deviate from the normal profiles as suspicious. Based on the different schemes of creating the normal activity profiles, we introduce two approaches for intrusion detection. The first one is a frequency-based approach which creates a normal frequency profile based on the periodical patterns existed in the time-series formed by the traffic. It aims at those attacks that are conducted by running pre-written scripts, which automate the process of attempting connections to various ports or sending packets with fabricated payloads, etc. The second approach builds the normal profile based on variations of connection-based behavior of each single computer. The deviations resulted from each individual computer are carried out by a weight assignment scheme and further used to build a weighted link graph representing the overall traffic abnormalities. The functionality of this system is of a distributed personal IDS system that also provides a centralized traffic analysis by graphical visualization. It provides a finer control over the internal network by focusing on connection-based behavior of each single computer. For network intrusion simulation, we explore an alternative method for network traffic simulation using explicit traffic generation. In particular, we build a model to replay the standard DARPA traffic data or the traffic data captured from a real environment. The replayed traffic data is mixed with the attacks, such as DOS and Probe attack, which can create apparent abnormal traffic flow patterns. With the explicit traffic generation, every packet that has ever been sent by the victim and attacker is formed in the simulation model and travels around strictly following the criteria of time and path that extracted from the real scenario. Thus, the model provides a promising aid in the study of intrusion detection techniques.
Show less - Date Issued
- 2005
- Identifier
- CFE0000679, ucf:46484
- Format
- Document (PDF)
- PURL
- http://purl.flvc.org/ucf/fd/CFE0000679
- Title
- SESSION-BASED INTRUSION DETECTION SYSTEM TO MAP ANOMALOUS NETWORK TRAFFIC.
- Creator
-
Caulkins, Bruce, Wang, Morgan, University of Central Florida
- Abstract / Description
-
Computer crime is a large problem (CSI, 2004; Kabay, 2001a; Kabay, 2001b). Security managers have a variety of tools at their disposal – firewalls, Intrusion Detection Systems (IDSs), encryption, authentication, and other hardware and software solutions to combat computer crime. Many IDS variants exist which allow security managers and engineers to identify attack network packets primarily through the use of signature detection; i.e., the IDS recognizes attack packets due to their well...
Show moreComputer crime is a large problem (CSI, 2004; Kabay, 2001a; Kabay, 2001b). Security managers have a variety of tools at their disposal – firewalls, Intrusion Detection Systems (IDSs), encryption, authentication, and other hardware and software solutions to combat computer crime. Many IDS variants exist which allow security managers and engineers to identify attack network packets primarily through the use of signature detection; i.e., the IDS recognizes attack packets due to their well-known "fingerprints" or signatures as those packets cross the network's gateway threshold. On the other hand, anomaly-based ID systems determine what is normal traffic within a network and reports abnormal traffic behavior. This paper will describe a methodology towards developing a more-robust Intrusion Detection System through the use of data-mining techniques and anomaly detection. These data-mining techniques will dynamically model what a normal network should look like and reduce the false positive and false negative alarm rates in the process. We will use classification-tree techniques to accurately predict probable attack sessions. Overall, our goal is to model network traffic into network sessions and identify those network sessions that have a high-probability of being an attack and can be labeled as a "suspect session." Subsequently, we will use these techniques inclusive of signature detection methods, as they will be used in concert with known signatures and patterns in order to present a better model for detection and protection of networks and systems.
Show less - Date Issued
- 2005
- Identifier
- CFE0000906, ucf:46762
- Format
- Document (PDF)
- PURL
- http://purl.flvc.org/ucf/fd/CFE0000906
- Title
- EMPIRICAL MODELING OF A MARIJUANA EXPECTANCY MEMORY NETWORK IN CHILDREN AS A FUNCTION OF AGE AND MARIJUANA USE.
- Creator
-
Alfonso, Jacqueline, Dunn, Michael, University of Central Florida
- Abstract / Description
-
The present investigation modeled the expectancy memory organization and likely memory activation patterns of marijuana expectancies of children across age and marijuana use. The first phase of the study surveyed 142 children to obtain their first associate to marijuana use. From their responses, the Marijuana Expectancy Inventory for Children and Adolescents (MEICA) was developed. The second phase of the study administered the MEICA to a second sample of 392 children to model marijuana...
Show moreThe present investigation modeled the expectancy memory organization and likely memory activation patterns of marijuana expectancies of children across age and marijuana use. The first phase of the study surveyed 142 children to obtain their first associate to marijuana use. From their responses, the Marijuana Expectancy Inventory for Children and Adolescents (MEICA) was developed. The second phase of the study administered the MEICA to a second sample of 392 children to model marijuana expectancy organization and probable memory activation paths of marijuana users versus never-users. Results indicated that irrespective of age, adolescents who have used marijuana tend to emphasize positive-negative effects, whereas adolescents who have never used marijuana tend to emphasize psychological-physiological effects. Memory activation patterns also differed by marijuana use history such that users are more likely to begin their paths with short-term positive effects of marijuana, versus non-users who access long-term cognitive and physiological effects with more likelihood. This study is the first to examine specific marijuana outcome expectancies of children and adolescents as they relate to marijuana-using behavior. Implications for marijuana prevention and intervention programs, future research, and limitations of the current investigation are discussed.
Show less - Date Issued
- 2005
- Identifier
- CFE0000897, ucf:46629
- Format
- Document (PDF)
- PURL
- http://purl.flvc.org/ucf/fd/CFE0000897
