Open Access. Powered by Scholars. Published by Universities.®

Computer Sciences Commons™

Open Access. Powered by Scholars. Published by Universities.®

Air Force Institute of Technology

Discipline
Keyword
Publication Year
Publication
Publication Type
File Type

Articles 721 - 750 of 1277

Full-Text Articles in Computer Sciences

Fuzzy State Aggregation And Off-Policy Reinforcement Learning For Stochastic Environments, Dean C. Wardell, Gilbert L. Peterson May 2006

Fuzzy State Aggregation And Off-Policy Reinforcement Learning For Stochastic Environments, Dean C. Wardell, Gilbert L. Peterson

Faculty Publications

Reinforcement learning is one of the more attractive machine learning technologies, due to its unsupervised learning structure and ability to continually learn even as the environment it is operating in changes. This ability to learn in an unsupervised manner in a changing environment is applicable in complex domains through the use of function approximation of the domain’s policy. The function approximation presented here is that of fuzzy state aggregation. This article presents the use of fuzzy state aggregation with the current policy hill climbing methods of Win or Lose Fast (WoLF) and policy-dynamics based WoLF (PD-WoLF), exceeding the learning rate …

Go to article

An Interactive Relaxation Approach For Anomaly Detection And Preventive Measures In Computer Networks, Garrick A. Bell Apr 2006

An Interactive Relaxation Approach For Anomaly Detection And Preventive Measures In Computer Networks, Garrick A. Bell

Theses and Dissertations

It is proposed to develop a framework of detecting and analyzing small and widespread changes in specific dynamic characteristics of several nodes. The characteristics are locally measured at each node in a large network of computers and analyzed using a computational paradigm known as the Relaxation technique. The goal is to be able to detect the onset of a worm or virus as it originates, spreads-out, attacks and disables the entire network. Currently, selective disabling of one or more features across an entire subnet, e.g. firewalls, provides limited security and keeps us from designing high performance net-centric systems. The most …

Go to article

Factors Impacting Key Management Effectiveness In Secured Wireless Networks, Yongjoo Shin Mar 2006

Factors Impacting Key Management Effectiveness In Secured Wireless Networks, Yongjoo Shin

Theses and Dissertations

The use of a Public Key Infrastructure (PKI) offers a cryptographic solution that can overcome many, but not all, of the MANET security problems. One of the most critical aspects of a PKI system is how well it implements Key Management. Key Management deals with key generation, key storage, key distribution, key updating, key revocation, and certificate service in accordance with security policies over the lifecycle of the cryptography. The approach supported by traditional PKI works well in fixed wired networks, but it may not appropriate for MANET due to the lack of fixed infrastructure to support the PKI. This …

Go to article

Type Ii Quantum Computing Algorithm For Computational Fluid Dynamics, James A. Scoville Mar 2006

Type Ii Quantum Computing Algorithm For Computational Fluid Dynamics, James A. Scoville

Theses and Dissertations

An algorithm is presented to simulate fluid dynamics on a three qubit type II quantum computer: a lattice of small quantum computers that communicate classical information. The algorithm presented is called a three qubit factorized quantum lattice gas algorithm. It is modeled after classical lattice gas algorithms which move virtual particles along an imaginary lattice and change the particles’ momentums using collision rules when they meet at a lattice node. Instead of moving particles, the quantum algorithm presented here moves probabilities, which interact via a unitary collision operator. Probabilities are determined using ensemble measurement and are moved with classical communications …

Go to article

Formal Mitigation Strategies For The Insider Threat: A Security Model And Risk Analysis Framework, Jonathan W. Butts Mar 2006

Formal Mitigation Strategies For The Insider Threat: A Security Model And Risk Analysis Framework, Jonathan W. Butts

Theses and Dissertations

The advancement of technology and reliance on information systems have fostered an environment of sharing and trust. The rapid growth and dependence on these systems, however, creates an increased risk associated with the insider threat. The insider threat is one of the most challenging problems facing the security of information systems because the insider already has capabilities within the system. Despite research efforts to prevent and detect insiders, organizations remain susceptible to this threat because of inadequate security policies and a willingness of some individuals to betray their organization. To investigate these issues, a formal security model and risk analysis …

Go to article

Multiframe Shift Estimation, Stephen A. Bruckart Mar 2006

Multiframe Shift Estimation, Stephen A. Bruckart

Theses and Dissertations

The purpose of this research was to develop a fundamental framework for a new approach to multiframe translational shift estimation in image processing. This thesis sought to create a new multiframe shift estimator, to theoretically prove and experimentally test key properties of it, and to quantify its performance according to several metrics. The new estimator was modeled successfully and was proven to be an unbiased estimator under certain common image noise conditions. Furthermore its performance was shown to be superior to the cross correlation shift estimator, a robust estimator widely used in similar image processing cases, according to several criteria. …

Go to article

An Estimation Theory Approach To Detection And Ranging Of Obscured Targets In 3-D Ladar Data, Charles R. Burris Mar 2006

An Estimation Theory Approach To Detection And Ranging Of Obscured Targets In 3-D Ladar Data, Charles R. Burris

Theses and Dissertations

The purpose of this research is to develop an algorithm to detect obscured images in 3-D LADAR data. The real data used for this research was gathered using a FLASH LADAR system under development at AFRL/SNJM. The system transmits light with a wavelength of 1.55 micrometers and produces 20 128 X 128 temporally resolved images from the return pulse separated by less than 2 nanoseconds in time. New algorithms for estimating the range to a target in 3-D FLASH LADAR data were developed. Results from processing real data are presented and compared to the traditional correlation receiver for extracting ranges …

Go to article

Bda Enhancement Methodology Using Situational Parameter Adjustments, Michael V. Carras Jr. Mar 2006

Bda Enhancement Methodology Using Situational Parameter Adjustments, Michael V. Carras Jr.

Theses and Dissertations

In the context of close ground combat, the perception of Battle Damage Assessment (BDA) is closely linked with a soldier’s engagement decisions and has significant effects on the battlefield. Perceived BDA is also one of the most complex and uncertain processes facing the soldier in live combat. As a result, the modeling and simulation community has yet to adequately model the perceived BDA process in combat models. This research effort examines the BDA process from a perception standpoint and proposes a methodology to collect the pertinent data and model this perception in the Army’s current force-on-force model, CASTFOREM. A subject …

Go to article

An Adaptable Energy-Efficient Medium Access Control Protocol For Wireless Sensor Networks, Justin T. Kautz Mar 2006

An Adaptable Energy-Efficient Medium Access Control Protocol For Wireless Sensor Networks, Justin T. Kautz

Theses and Dissertations

Wireless networks have become ubiquitous recently and therefore their usefulness has also become more extensive. Wireless sensor networks (WSN) detect environmental information with sensors in remote settings. One problem facing WSNs is the inability to resupply power to these energy-constrained devices due to their remoteness. Therefore to extend a WSN's effectiveness, the lifetime of the network must be increased by making them as energy efficient as possible. An energy efficient medium access control (MAC) can boost a WSN's lifetime. This research creates a MAC protocol called Adaptive sensor Medium Access Control (AMAC) which is based on Sensor Medium Access Control …

Go to article

Mitigating Distributed Denial Of Service Attacks In An Anonymous Routing Environment: Client Puzzles And Tor, Nicholas A. Fraser Mar 2006

Mitigating Distributed Denial Of Service Attacks In An Anonymous Routing Environment: Client Puzzles And Tor, Nicholas A. Fraser

Theses and Dissertations

Online intelligence operations use the Internet to gather information on the activities of U.S. adversaries. The security of these operations is paramount, and one way to avoid being linked to the Department of Defense (DoD) is to use anonymous communication systems. One such system, Tor, makes interactive TCP services anonymous. Tor uses the Transport Layer Security (TLS) protocol and is thus vulnerable to a distributed denial-of-service (DDoS) attack that can significantly delay data traversing the Tor network. This research uses client puzzles to mitigate TLS DDoS attacks. A novel puzzle protocol, the Memoryless Puzzle Protocol (MPP), is conceived, implemented, and …

Go to article

A Monocular Vision Based Approach To Flocking, Brian Kirchner Mar 2006

A Monocular Vision Based Approach To Flocking, Brian Kirchner

Theses and Dissertations

Flocking is seen in nature as a means for self protection, more efficient foraging, and other search behaviors. Although much research has been done regarding the application of this principle to autonomous vehicles, the majority of the research has relied on GPS information, broadcast communication, an omniscient central controller, or some other form of "global" knowledge. This approach, while effective, has serious drawbacks, especially regarding stealth, reliability, and biological grounding. This research effort uses three Pioneer P2-AT8 robots to achieve flocking behavior without the use of global knowledge. The sensory inputs are limited to two cameras, offset such that the …

Go to article

A Real-Time Wireless Sensor Media Access Control (Mac) Protocol, Barry W. Park Mar 2006

A Real-Time Wireless Sensor Media Access Control (Mac) Protocol, Barry W. Park

Theses and Dissertations

Wireless sensor networks are rapidly becoming a platform for applications such as battlefield monitoring, intelligence gathering, environmental monitoring, and emergency response. Inherent in these applications is a priority and urgency of the information or messages. This means the messages must be delivered in a timely manner for them to be useful. This research assigns a message priority level and provides high-priority messages quicker access to the channel. Using MICA2 sensors and a modified Media Access Control (MAC) layer, real-time message End-to-End (ETE) delay was reduced by 50 percent. Coupled with this decrease in delay, these same real-time messages also had …

Go to article

Toward The Static Detection Of Deadlock In Java Software, Jose E. Fadul Mar 2006

Toward The Static Detection Of Deadlock In Java Software, Jose E. Fadul

Theses and Dissertations

Concurrency is the source of many real-world software reliability and security problems. Concurrency defects are difficult to detect because they defy conventional software testing techniques due to their non-local and non-deterministic nature. We focus on one important aspect of this problem: static detection of the possibility of deadlock - a situation in which two or more processes are prevented from continuing while each waits for resources to be freed by the continuation of the other. This thesis proposes a flow-insensitive interprocedural static analysis that detects the possibility that a program can deadlock at runtime. Our analysis proceeds in two steps. …

Go to article

Verification Of A Decision Level Fusion Algorithm Using A Proven Atr System And Measured Sar Data, James Douglas Thompson Mar 2006

Verification Of A Decision Level Fusion Algorithm Using A Proven Atr System And Measured Sar Data, James Douglas Thompson

Theses and Dissertations

Decision level fusion (DLF) algorithms combine outputs of multiple single sensors to make one confident declaration of a target. This research compares performance results of a DLF algorithm using measured data and a proven ATR system with results from simulated data and a modeled ATR system. This comparison indicates that DLF offers significant performance improvements over single sensor looks. However, results based on simulated data and a modeled ATR are slightly optimistic and overestimate results from measured data and a proven ATR system by nearly 10% over all targets tested.

Go to article

A Comparison Of Main Rotor Smoothing Adjustments Using Linear And Neural Network Algorithms, Nathan A. Miller Mar 2006

A Comparison Of Main Rotor Smoothing Adjustments Using Linear And Neural Network Algorithms, Nathan A. Miller

Theses and Dissertations

Helicopter main rotor smoothing is a maintenance procedure that is routinely performed to minimize airframe vibrations induced by non-uniform mass and/or aerodynamic distributions in the main rotor system. This important task is both time consuming and expensive, so improvements to the process have long been sought. Traditionally, vibrations have been minimized by calculating adjustments based on an assumed linear relationship between adjustments and vibration response. In recent years, artificial neural networks have been trained to recognize non-linear mappings between adjustments and vibration response. This research was conducted in order observe the character of the adjustment mapping of the Vibration Management …

Go to article

Flashlight: A Dynamic Detector Of Shared State, Race Conditions, And Locking Models In Concurrent Java Programs, Scott C. Hale Mar 2006

Flashlight: A Dynamic Detector Of Shared State, Race Conditions, And Locking Models In Concurrent Java Programs, Scott C. Hale

Theses and Dissertations

Concurrent Java programs are difficult to understand and implement correctly. This difficultly leads to code faults that are the source of many real-world reliability and security problems. Many factors contribute to concurrency faults in Java code; for example, programmers may not understand Java language semantics or, when using a Java library or framework, may not understand that their resulting program is concurrent. This thesis describes a dynamic analysis tool, named FlashLight, that detects shared state and possible race conditions within a program. FlashLight illuminates the concurrency within a program for programmers who are wholly or partially "in the dark" about …

Go to article

Evolving Self-Organized Behavior For Homogeneous And Heterogeneous Uav Or Ucav Swarms, Ian C. Price Mar 2006

Evolving Self-Organized Behavior For Homogeneous And Heterogeneous Uav Or Ucav Swarms, Ian C. Price

Theses and Dissertations

This investigation uses a self-organization (SO) approach to enable cooperative search and destruction of retaliating targets with swarms of homogeneous and heterogeneous unmanned aerial vehicles (UAVs). To facilitate specific system design, a facilitating SO algebraic framework is created that emphasizes scalability, robustness, and flexibility. This framework is then used to implement a UAV behavior architecture relying upon rules governing formation and target interaction. Sets of applicable behaviors are created by weighted summation of the rules where different weights act as distinct behavior archetypes. Appropriate behavior archetypes are based upon sense information distilled from the environment and a simple perceptron mapping. …

Go to article

Detecting Potential Insider Threats Through Email Datamining, James S. Okolica Mar 2006

Detecting Potential Insider Threats Through Email Datamining, James S. Okolica

Theses and Dissertations

No abstract provided.

Go to article

Metamorphism As A Software Protection For Non-Malicious Code, Thomas E. Dube Mar 2006

Metamorphism As A Software Protection For Non-Malicious Code, Thomas E. Dube

Theses and Dissertations

Most organizations are aware that threats from trusted insiders pose a great risk to their organization and are very difficult to protect against. Auditing is recognized as an effective technique to detect malicious insider activities. However, current auditing methods are typically applied with a one-size-fits-all approach and may not be an appropriate mitigation strategy, especially towards insider threats. This research develops a 4-step methodology for designing a customized auditing template for a Microsoft Windows XP operating system. Two tailoring methods are presented which evaluate both by category and by configuration. Also developed are various metrics and weighting factors as a …

Go to article

Cryptanalysis Of Pseudorandom Number Generators In Wireless Sensor Networks, Kevin M. Finnigin Mar 2006

Cryptanalysis Of Pseudorandom Number Generators In Wireless Sensor Networks, Kevin M. Finnigin

Theses and Dissertations

This work presents a brute-force attack on an elliptic curve cryptosystem implemented on UC Berkley's TinyOS operating system for wireless sensor networks. The attack exploits the short period of the pseudorandom number generator (PRNG) used by the cryptosystem to generate private keys. The attack assumes a laptop is listening promiscuously to network traffic for key messages and requires only the sensor node?s public key and network address to discover the private key. Experimental results show that roughly 50% of the address space leads to a private key compromise in 25 minutes on average. Furthermore, approximately 32% of the address space …

Go to article

Application Of Fuzzy State Aggregation And Policy Hill Climbing To Multi-Agent Systems In Stochastic Environments, Dean C. Wardell Mar 2006

Application Of Fuzzy State Aggregation And Policy Hill Climbing To Multi-Agent Systems In Stochastic Environments, Dean C. Wardell

Theses and Dissertations

Reinforcement learning is one of the more attractive machine learning technologies, due to its unsupervised learning structure and ability to continually even as the operating environment changes. Applying this learning to multiple cooperative software agents (a multi-agent system) not only allows each individual agent to learn from its own experience, but also opens up the opportunity for the individual agents to learn from the other agents in the system, thus accelerating the rate of learning. This research presents the novel use of fuzzy state aggregation, as the means of function approximation, combined with the policy hill climbing methods of Win …

Go to article

Using Prospect Theory To Investigate Decision-Making Bias Within An Information Security Context, Neil J. Schroeder Dec 2005

Using Prospect Theory To Investigate Decision-Making Bias Within An Information Security Context, Neil J. Schroeder

Theses and Dissertations

Information security is an issue that has increased in importance over the past decade. In this time both practitioner and academic circles have researched and developed practices and process to more effectively handle information security. Even with growth in these areas there has been little research conducted into how decision makers actually behave. This is problematic because decision makers in the Department of Defense have been observed exhibiting risk seeking behavior when making information security decisions that seemingly violate accepted norms. There are presently no models in the literature that provide sufficient insight into this phenomenon. This study used Prospect …

Go to article

Early Warning And Prediction Of Internet Attacks And Exploits, Brian P. Zeitz Sep 2005

Early Warning And Prediction Of Internet Attacks And Exploits, Brian P. Zeitz

Theses and Dissertations

A safe, secure, and functional information network is vital in today’s Air Force net centric environment. Information is more critical today than it has ever been. As more operational functions are placed in cyber space and greater computing power becomes available to everyone, keeping these networks safe and secure is an almost unattainable task. Network security entails Intrusion Detection Security, but another form of security or “insecurity” is quickly gaining attention. Honeypots allow the black hat community to attack and penetrate non-production systems. By monitoring and studying these attacks, network defenders can develop better Information Assurance tactics, techniques and procedures …

Go to article

Operator State Estimation For Adaptive Aiding In Uninhabited Combat Air Vehicles, Christopher A. Russell Sep 2005

Operator State Estimation For Adaptive Aiding In Uninhabited Combat Air Vehicles, Christopher A. Russell

Theses and Dissertations

This research demonstrated the first closed-loop implementation of adaptive automation using operator functional state in an operationally relevant environment. In the Uninhabited Combat Air Vehicle (UCAV) environment, operators can become cognitively overloaded and their performance may decrease during mission critical events. This research demonstrates an unprecedented closed-loop system, one that adaptively aids UCAV operators based on their cognitive functional state A series of experiments were conducted to 1) determine the best classifiers for estimating operator functional state, 2) determine if physiological measures can be used to develop multiple cognitive models based on information processing demands and task type, 3) determine …

Go to article

Cooperative Reinforcement Learning Using An Expert-Measuring Weighted Strategy With Wolf, Kevin Cousin, Gilbert L. Peterson Sep 2005

Cooperative Reinforcement Learning Using An Expert-Measuring Weighted Strategy With Wolf, Kevin Cousin, Gilbert L. Peterson

Faculty Publications

Gradient descent learning algorithms have proven effective in solving mixed strategy games. The policy hill climbing (PHC) variants of WoLF (Win or Learn Fast) and PDWoLF (Policy Dynamics based WoLF) have both shown rapid convergence to equilibrium solutions by increasing the accuracy of their gradient parameters over standard Q-learning. Likewise, cooperative learning techniques using weighted strategy sharing (WSS) and expertness measurements improve agent performance when multiple agents are solving a common goal. By combining these cooperative techniques with fast gradient descent learning, an agent’s performance converges to a solution at an even faster rate. This statement is verified in a …

Go to article

A Comparison Of Generalizability For Anomaly Detection, Gilbert L. Peterson, Robert F. Mills, Brent T. Mcbride, Wesley T. Allred Aug 2005

A Comparison Of Generalizability For Anomaly Detection, Gilbert L. Peterson, Robert F. Mills, Brent T. Mcbride, Wesley T. Allred

Faculty Publications

In security-related areas there is concern over the novel “zeroday” attack that penetrates system defenses and wreaks havoc. The best methods for countering these threats are recognizing “non-self” as in an Artificial Immune System or recognizing “self” through clustering. For either case, the concern remains that something that looks similar to self could be missed. Given this situation one could logically assume that a tighter fit to self rather than generalizability is important for false positive reduction in this type of learning problem. This article shows that a tight fit, although important, does not supersede having some model generality. This …

Go to article

Explicit Building Block Multiobjective Evolutionary Computation: Methods And Applications, Richard O. Day Jun 2005

Explicit Building Block Multiobjective Evolutionary Computation: Methods And Applications, Richard O. Day

Theses and Dissertations

This dissertation presents principles, techniques, and performance of evolutionary computation optimization methods. Concentration is on concepts, design formulation, and prescription for multiobjective problem solving and explicit building block (BB) multiobjective evolutionary algorithms (MOEAs). Current state-of-the-art explicit BB MOEAs are addressed in the innovative design, execution, and testing of a new multiobjective explicit BB MOEA. Evolutionary computation concepts examined are algorithm convergence, population diversity and sizing, genotype and phenotype partitioning, archiving, BB concepts, parallel evolutionary algorithm (EA) models, robustness, visualization of evolutionary process, and performance in terms of effectiveness and efficiency. The main result of this research is the development of …

Go to article

The Analysis Of A Link Between A Remote Local Area Network And Its Server Resources, Theresa D. Beaver Jun 2005

The Analysis Of A Link Between A Remote Local Area Network And Its Server Resources, Theresa D. Beaver

Theses and Dissertations

As the Air Force transitions to an expeditionary force, the service's ability to provide computer capabilities at remote locations becomes more and more paramount. One way to provide this support is to create a Local Area Network (LAN) in which the workstations are positioned at the deployed location while the servers are maintained at a Main Operating Base (MOB). This saves the military money, because it eliminates the need to purchase and deploy server equipment as well as eliminating the need to deploy personnel to set-up and maintain the servers. There is, however, a tradeoff. As the number of personnel …

Go to article

A Framework For Routing And Topological Decision-Making Within A Transformational Communications Architecture, Alexander I. Smith Jun 2005

A Framework For Routing And Topological Decision-Making Within A Transformational Communications Architecture, Alexander I. Smith

Theses and Dissertations

A Transformational Communications Architecture (TCA) has been proposed by the US government with the goal of creating "an Internet in the sky." This thesis presents a framework which places user requirements at the heart of routing and topological decision-making within the TCA. It provides new terminology to describe data transmissions and link characteristics and an approach to formulating cost functions to better meet user needs. The framework is intended to stimulate interest and focus efforts on how to meet military specific requirements by taking advantage of the TCA's diverse capabilities and topology. Cost functions using newly-defined parameters are simulated and …

Go to article

An Evolutionary Algorithm To Generate Hyper-Ellipsoid Detectors For Negative Selection, Joseph M. Shapiro, Gary B. Lamont, Gilbert L. Peterson Jun 2005

An Evolutionary Algorithm To Generate Hyper-Ellipsoid Detectors For Negative Selection, Joseph M. Shapiro, Gary B. Lamont, Gilbert L. Peterson

Faculty Publications

This paper introduces hyper-ellipsoids as an improvement to hyper-spheres as intrusion detectors in a negative selection problem within an artificial immune system. Since hyper-spheres are a specialization of hyper-ellipsoids, hyper-ellipsoids retain the benefits of hyper-spheres. However, hyper-ellipsoids are much more flexible, mostly in that they can be stretched and reoriented. The viability of using hyper-ellipsoids is established using several pedagogical problems. We conjecture that fewer hyper-ellipsoids than hyper-spheres are needed to achieve similar coverage of nonself space in a negative selection problem. Experimentation validates this conjecture. In pedagogical benchmark problems, the number of hyper-ellipsoids to achieve good results is significantly …

Go to article

First Prev Next Last