Search results
(121 - 140 of 1,076)
Pages
- Title
- A STEP TOWARD SUPPORTING LONG-RUNNING APPLICATIONS WITH REAL-TIME CONSTRAINTS ON HYBRID CLOUDS
- Creator
- Wu, Hao
- Date
- 2017, 2017-05
- Description
-
The advancement of computer and network technology has brought the world into a new computer cloud era. The ”pay-as-you-go” business model and...
Show moreThe advancement of computer and network technology has brought the world into a new computer cloud era. The ”pay-as-you-go” business model and the service oriented models allow users to have ”unlimited” resources if needed and free from infrastructure maintenance and software upgrades. Cloud services are currently among the top-ranked high growth areas in computing and are seeing an acceleration in enterprise adoption with the worldwide market predicted to reach more than $270b in 2020. According to Google, currently more than 95% of the web services are deployed on cloud.Many di↵erent types of applications are deployed on computer clouds. However, due to inherent performance uncertainty within computer clouds, as of today, applications with real-time and high QoS constraints still operate on traditional computer systems and are not able to benefit from elastic computer clouds.. The thesis focuses on both theoretical analysis and real system implementation on the problem of guaranteeing real-time application’s deadline requirement while minimizing the application’s execution cost on hybrid clouds. Four major problems have been addressed towards moving applications with real-time constraint on hybrid computer clouds. 1). A minimal slack time and minimal distance (MSMD) scheduling algorithm is developed to minimize the resources needed to guarantee an application’s end-to-end deadline requirement using computer clouds. 2). A VM Instance Hour Minimization (IHM) algorithm is developed to reduce the application’s execution cost for given schedules. The proposed IHM algorithm can be integrated with common scheduling algorithm used in the literature. In addition, we also evaluated the feasibility of utilizing spot instance to further reduce the application’s execution cost while not sacrificing QoS guarantees. 3). A reference model for virtual machine launching overhead is developed to predict both system utilization and timing overhead during the VM launching process. 4). A hybrid cloud management tool that integrates the developed algorithms and reference model is developed to support running long-running applications with real-time constraints on hybrid clouds.
Ph.D. in Computer Science, May 2017
Show less
- Title
- PERFORMANCE ANALYSIS AND OPTIMIZATION OF LARGE-SCALE SCIENTIFIC APPLICATIONS
- Creator
- Wu, Jingjin
- Date
- 2013, 2013-07
- Description
-
Scientific applications are critical for solving complex problems in many areas of research, and often require a large amount of computing...
Show moreScientific applications are critical for solving complex problems in many areas of research, and often require a large amount of computing resources in terms of both runtime and memory. Massively parallel supercomputers with ever increasing computing power are being built to satisfy the need of large-scale scientific applications. With the advent of petascale era, there is an enlarged gap between the computing power of supercomputers and the parallel scalability of many applications. To take full advantage of the massive parallelism of supercomputers, it is indispensable to improve the scalability of large-scale scientific applications through performance analysis and optimization. This thesis work is motivated by cell-based AMR (Adaptive Mesh Refinement) cosmology simulations, in particular, the Adaptive Refinement Tree (ART) application. Performance analysis is performed to identify its scaling bottleneck, a performance emulator is designed for efficient evaluation of different load balancing schemes, and topology mapping strategies are explored for performance improvements. More importantly, the exploration of topology mapping mechanisms leads to a generic methodology for network and multicore aware topology mapping, and a set of efficient mapping algorithms for popular topologies. These have been implemented in a topology mapping library – TOPOMap, which can be used to support MPI topology functions.
PH.D in Computer Science, July 2013
Show less
- Title
- NEURAL ADAPTIVE CONTROL STRATEGY FOR HYBRID ELECTRIC VEHICLES WITH PARALLEL POWERTRAIN
- Creator
- Gurkaynak, Yusuf
- Date
- 2011-04-20, 2011-05
- Description
-
In a hybrid electric vehicle (HEV) with parallel powertrain, the system can be controlled by splitting the required power between the electric...
Show moreIn a hybrid electric vehicle (HEV) with parallel powertrain, the system can be controlled by splitting the required power between the electric propulsion machine and internal combustion engine (ICE) to meet specific goals related to fuel consumption, efficiency, performance, and/or emissions. This power splitting scenario, which is of great hybridization importance, is in fact the control strategy or energy management of the hybrid vehicle. Performance of the system depends on the control strategy, which needs to be robust, stable, reliable, and independent from uncertainties. This Ph.D. research is focused on model based control strategies, which are proposed for parallel hybrid powertrains, showing significant advantages in performance and fuel economy. If a model based control strategy is used to develop the hybrid power management algorithm, the accuracy of the model data needs to be high for proper control. Therefore, this type of management method is parameter sensitive. Implementing system identification features into this algorithm reduces the effect. As a result, the proposed controller algorithm learns the existing component parameters while operating. Furthermore, combining the base controller with an online tuner, which simultaneously optimizes the controller for current conditions, will improve the performance of the power management. In addition, this Ph.D. thesis presents a novel neural adaptive equivalent consumption minimization strategy (ECMS) and applies it to a hybrid representative sport utility vehicle (SUV) with parallel powertrain. The ECMS is a model based optimal control strategy and is based on the minimization of both fuel consumption and battery charge usage by introducing the equivalent coefficient between them. Proper operation of the controller depends on the accuracy of the model. It also depends on the correct selection of the equivalent coefficient. In this Ph.D. thesis, specific neural network structures are proposed for both coefficient selections by drive cycle recognition and for precise model building by system identification. This thesis also presents a novel fast solution method of ECMS algorithm for real time applications.
Ph.D. in Electrical Engineering, May 2011
Show less
- Title
- KILOMETER-SPACED GNSS ARRAY FOR IONOSPHERIC IRREGULARITY MONITORING
- Creator
- Su, Yang
- Date
- 2017, 2017-05
- Description
-
This dissertation presents automated, systematic data collection, processing, and analysis methods for studying the spatial-temporal...
Show moreThis dissertation presents automated, systematic data collection, processing, and analysis methods for studying the spatial-temporal properties of Global Navigation Satellite Systems (GNSS) scintillations produced by ionospheric irregularities at high latitudes using a closely spaced multi-receiver array deployed in the northern auroral zone. The main contributions include 1) automated scintillation monitoring, 2) estimation of drift and anisotropy of the irregularities, 3) error analysis of the drift estimates, and 4) multi-instrument study of the ionosphere. A radiowave propagating through the ionosphere, consisting of ionized plasma, may su↵er from rapid signal amplitude and/or phase fluctuations known as scintillation. Caused by non-uniform structures in the ionosphere, intense scintillation can lead to GNSS navigation and high-frequency (HF) communication failures. With specialized GNSS receivers, scintillation can be studied to better understand the structure and dynamics of the ionospheric irregularities, which can be parameterized by altitude, drift motion, anisotropy of the shape, horizontal spatial extent and their time evolution. To study the structuring and motion of ionospheric irregularities at the sub-kilometer scale sizes that produce L-band scintillations, a closely-spaced GNSS array has been established in the auroral zone at Poker Flat Research Range, Alaska to investigate high latitude scintillation and irregularities. Routinely collecting lowrate scintillation statistics, the array database also provides 100 Hz power and phase data for each channel at L1/L2C frequency. In this work, a survey of seasonal and hourly dependence of L1 scintillation events over the course of a year is discussed. To efficiently and systematically study scintillation events, an automated low-rate scintillation detection routine is established and performed for each day by screening the phase scintillation index. The spaced-receiver technique is applied to cross-correlated phase and power measurements from GNSS receivers. Results of horizontal drift velocities and anisotropy ellipses derived from the parameters are shown for several detected events. Results show the possibility of routinely quantifying ionospheric irregularities by drifts and anisotropy. Error analysis on estimated properties is performed to further evaluate the estimation quality. Uncertainties are quantified by ensemble simulation of noise on the phase signals carried through to the observations of the spaced-receiver linear system. These covariances are then propagated through to uncertainties on drifts. A case study of a single scintillating satellite observed by the array is used to demonstrate the uncertainty estimation process. The distributed array is used in coordination with other measuring techniques such as incoherent scatter radar and optical all-sky imagers. These scintillations are correlated with auroral activity, based on all-sky camera images. Measurements and uncertainty estimates made over a 30-minute period are made and compared to a collocated incoherent scatter radar, and show good agreement in horizontal drift speed and direction during periods of scintillation for cases when the characteristic velocity is less than the drift velocity. The methods demonstrated are extensible to other zones and other GNSS arrays of varying size, number, ground distribution, and transmitter frequency.
Ph.D. in Mechanical and Aerospace Engineeering, May 2017
Show less
- Title
- NETWORK CODING BASED COOPERATIVE PEER-TO-PEER REPAIR IN WIRELESS NETWORKS
- Creator
- Liu, Yu
- Date
- 2012-07-11, 2012-07
- Description
-
Multimedia Broadcast/Multicast Service (MBMS) in cellular networks has emerged recently as a promising distribution model to provide rich...
Show moreMultimedia Broadcast/Multicast Service (MBMS) in cellular networks has emerged recently as a promising distribution model to provide rich content distribution where a batch of content is broadcast to a large number of peers simultaneously. However, ensuring efficient error-free message delivery in such a scenario is a challenge, since packet loss is inevitable due to the time-varying nature of wireless transmissions, and the server is probably overwhelmed by floods of individual retransmission requests from peers. Cooperative Peer-to-Peer (P2P) information repair has been proposed to mitigate the packet loss among peers during Base Station (BS) broadcast, by allowing peers to cooperate on information exchange among themselves, rather than asking the BS to rebroadcast the lost packets for the peers. Network Coding, a fairly recent transmission paradigm with the potential network throughput improvement and high reliability advantage, has been widely recognized as a promising information dissemination approach for wireless networks. In this research, we study the network coding based cooperative P2P information repair in wireless networks. We first propose our initial work - a connected dominating set (CDS) based P2P information repair (PPIR) protocol with network coding which utilize the clustering idea, to minimize the total repair latency as well as alleviate the congestion and burden of BS’s downlink channels. Then the decision making problem for P2P repair with densely distributed nodes is studied and two approaches are provided. Later on, the NC based P2P information repair protocol with tunable parameter (NC-PIRTP) which evolved from PPIR protocol is proposed to further reduce transmission collisions and total repair latency. At last, P2P information repair under mobile network environment with pedestrian speed is studied and three efficient protocols are illustrated which are suitable to different specific cases. Extensive simulation results are provided for performance evaluation and comparisons, and to demonstrate the effectiveness and efficiency of our proposed protocols in terms of the total repair latency. Furthermore, an analytical model is developed, based on which theoretical results are derived. These results validate our protocol models and provide useful protocol design guideline for the cooperative P2P information repair problem in wireless networks.
Ph.D. in Electrical Engineering, July 2012
Show less
- Title
- POWER SYSTEM VOLTAGE STABILITY AND AGENT BASED DISTRIBUTION AUTOMATION IN SMART GRID
- Creator
- Nguyen, Cuong Phuc
- Date
- 2011-04-25, 2011-05
- Description
-
Our interconnected electric power system is presently facing many challenges that it was not originally designed and engineered to handle. The...
Show moreOur interconnected electric power system is presently facing many challenges that it was not originally designed and engineered to handle. The increased interarea power transfers, aging infrastructure, and old technologies, have caused many problems including voltage instability, widespread blackouts, slow control response, among others. These problems have created an urgent need to transform the present electric power system to a highly stable, reliable, efficient, and self-healing electric power system of the future, which has been termed “smart grid”. This dissertation begins with a discussion on the voltage stability issue in bulk transmission networks. A new continuation power flow tool for studying the impacts of generator merit order based dispatch on inter-area transfer capability and static voltage stability is presented. In using this tool, it is realized that all distribution systems are represented by only a single lumped load model. While this representation is acceptable in traditional power system analysis, it may not be valid in the future smart grid where the distribution system will be integrated with intelligent and quick control capabilities to mitigate voltage problems before they propagate into the entire system. Therefore before analyzing the operation of the whole smart grid, it is important to understand the distribution system first. The second part of this dissertation presents a new platform for studying and testing emerging technologies in advanced Distribution Automation (DA) within smart grids. Due to the key benefits over the traditional centralized approach, namely flexible deployment, scalability, and avoidance of single-point-of-failure, a new distributed approach is employed to design and develop all elements of the platform. The multi-agent system (MAS), which has the three key characteristics of autonomy, local view, and decentralization, is selected to implement the advanced DA functions. The intelligent agents utilize the communication network for cooperation and negotiation. Communication latency is modeled using a user-defined probability density function. Failure-tolerant communication strategies are developed for agent communications. Major elements of advanced DA are developed in a completely distributed way and successfully tested for several IEEE standard systems, including: Fault Detection, Location, Isolation, and Service Restoration (FLISR); Coordination of Distributed Energy Storage Systems (DES); Distributed Power Flow (DPF); Volt-VAR Control (VVC); and Loss Reduction (LR).
Ph.D. in Electrical Engineering, May 2011
Show less
- Title
- NON-LINEAR CREEP-RELAXATION CONSTITUTIVE MODEL FOR PLAIN CONCRETE WITH ANISOTROPIC DAMAGE
- Creator
- Teran Torres, Bernardo
- Date
- 2016, 2016-05
- Description
-
A thermodynamic constitutive model for plain concrete, and other quasi-brittle materials, is developed. The model accounts for the anisotropic...
Show moreA thermodynamic constitutive model for plain concrete, and other quasi-brittle materials, is developed. The model accounts for the anisotropic damage-induced and the aging viscoelasticity of the material. The model is based on the theory of solidi cation proposed by Baºant and the continuum damage mechanics theory. The material is considered to be a viscoelastic-damageable material. The Helmholtz free energy utilized in the formulation is treated by means of the representation theorem of coupled damage-strain tensors. The model is capable of analyzing time-dependent damage (tertiary creep) under constant loading and damage due to cyclic creep. Two simple but practical examples are presented in order to illustrate the applicability of the model. The determination of the functions and constants representing the material behavior as well as any experimental companion is proposed for the continuation of the research.
Ph.D. in Civil Engineering, May 2016
Show less
- Title
- SYNTHESIS OF NANOPLATE STRUCTURES IN NI-BASED ALLOYS VIA DISCONTINUOUS PRECIPITATION
- Creator
- Zhou, Yang
- Date
- 2017, 2017-07
- Description
-
Nanostructured materials are of increasing interest due to their potential for use in numerous applications including catalysis and filtration...
Show moreNanostructured materials are of increasing interest due to their potential for use in numerous applications including catalysis and filtration. Methods of synthesizing nanomaterials often include chemical techniques producing individual particles, which gives rise to issues of agglomeration and containment, and the processes are difficult to scale-up. This work is directed at addressing issues of the occurrence and kinetics of self-assembly and structural alignment of nanostructures, which involves the discontinuous precipitation transformation followed by selective dissolution. This synthesis method has been demonstrated using a Ni-49at.%Co-12at.%Al alloy. A complete discontinuous transformation in the Ni 49-12 alloy can be achieved in times as short as 30min. Thus, the large-scale production of such structures is feasible using conventional heat treatment facilities. The synthesis technique is generally applicable to any alloy system in which DP goes to completion and one phase can be selectively removed. The nanoplates are self-assembled, self-supported and well aligned if the precipitate is coherent with the matrix. In order to control the process it is necessary to understand the phase equilibria involving γ, γ′ and β phases in the Al-Co-Ni system. This has been investigated both experimentally and computationally. The isothermal sections at 1100°C and 800°C as well as a partial liquidus projection were determined which result in modifications to previously published work. Comparing the experimental results with the calculated results using Thermo-Calc (TCNI8), there is reasonable agreement. The predicted separation of the γ phase into a two-phase ferromagnetic and paramagnetic region has never been observed experimentally in this or other published work. Future work should explore the ferromagnetic and paramagnetic phase separation as predicted using Thermo-Calc to verify its occurrence in this system. The nanostructured material produced by complete DP transformation may have interesting mechanical properties and these have been studied to a limited extent in this work. The hardness tests show that with decreasing annealing temperature to 600°C, the hardness of certain samples increases significantly due to the precipitation of second phase. The tensile and creep properties of alloy samples with DP phase present were also investigated. The results imply that the occurrence of DP will have a favorable effect on the tensile strength of the sample while lowering the ductility at the same time. A method to generate serrated grain boundaries based on the DP transformation is proposed. Such structures are expected to increase the creep resistance. This was found to be the case in a limited temperature and loading range. Higher temperature or stress levels lead to DP occurrence and cause negative effects on the creep resistance compared to conventionally processed material. The Curie temperature in the Al-Co-Ni alloy system was also studied and determined using a combined magnetic TG and DSC method. A ternary contour map of the Curie temperature has been constructed. From the contour map, the Curie temperature was seen to decrease from high Co, low Al content samples to low Co, high Al samples bypassing a platform at mid Co content. A nonlinear surface fitting was made through a Exponential2D model, the function is helpful for the prediction of Curie temperature of γ phase in Al-Co-Ni alloy system. This method also provides a novel idea of detecting phase transformation and precipitation through thermo-magnetically analyzing the magnetic behavior of the alloys.
Ph.D. in Materials Science and Engineering, July 2017
Show less
- Title
- MULTIMODAL NEUROIMAGING IN BLAST INDUCED TRAUMATIC BRAIN INJURY
- Creator
- Stout, Jeffrey
- Date
- 2015, 2015-12
- Description
-
Recent military conflicts in Iraq and Afghanistan have resulted in an increase in the number of blast related traumatic brain injury (TBI)...
Show moreRecent military conflicts in Iraq and Afghanistan have resulted in an increase in the number of blast related traumatic brain injury (TBI) cases, currently estimated in excess of 300,000. Current neuroimaging methods, including MRI and CT, are generally not sensitive enough to diagnose blast TBL This research represents several advancements in the understanding of long-term TBI related changes as well as imaging comorbid neurocognitive deficits and psychopathologies. The multi modal approach was taken to leverage the differing sensitivities of each imagmg modality to better identify neuroimaging changes associated with blast TBL T l-weighted MRI was utilized to assess cortical thickness reduction in the TBI subjects, resulting in right temporal lobe decreases that were associated with posttraumatic stress disorder (PTSD) and neurobehavioral symptom indices. Utilizing positron emission tomography (PET), several areas of metabolic change were associated with blast TBI, PTSD, repetitive head injury, and TBI severity. Functional MRI was used to assess working memory deficit, demonstrating relatively normal function in the mild TBI subject, except for at higher cognitive loads; whereas the moderate TBI subjects demonstrated a trend of dysfunction at most levels of cognitive load. Fractional anisotropy (FA) changes were assessed using diffusion tensor imaging, however there was no resulting statistical significance due to differences in the locations of single subject FA changes that could not be realized at the group level using standard analyses. Investigation into multimodal searchlight to integrate the sensitivities of each imaging modality into a single statistical measure, provided enhanced sensitivity to TBI-induced changes in neuroimaging versus single modal searchlight and identified several clusters of change most notably in the cerebellum, temporal lobes, brainstem, and internal capsule. Through a multimodal approach we have shown several areas of damage and dysfunction associated with blast TBI in subjects greater than one year post injury. This research is important due to the large numbers of subjects currently experiencing long-term post concussive symptoms. Our findings show that both TBI and PTSD playa large role in the changes identified on neuroimaging. This research can provide information to assist in the understanding , diagnosis , and treatment of blast TBI.
Ph.D. in Biomedical Engineering, December 2015
Show less
- Title
- VIBRATION IN TRACTION MOTORS FOR ELECTRIC AND HYBRID ELECTRIC VEHICLES
- Creator
- Yang, Zhi
- Date
- 2014, 2014-12
- Description
-
Due to increased fuel efficiency and lower cost/mile feature, electric vehicle (EV) and hybrid electric vehicle (HEV) are becoming more and...
Show moreDue to increased fuel efficiency and lower cost/mile feature, electric vehicle (EV) and hybrid electric vehicle (HEV) are becoming more and more popular. It is estimated that the sale of electric vehicles will reach 3.8 million by 2020, while hybrid vehicles will grow to 4% by 2020 from their current share of 2%. To meet this target, EV an HEV motors, the core energy conversion components, should not only satisfy specific requirements in performance and efficiency but also constrain vibration. This necessitates the analysis of vibration in traction motors for EV/HEV application. The primary objective of this dissertation is to characterize and compare the electromagnetic and vibrational behavior of typical traction motors (PMSM with distributed winding, PMSM with concentrated winding, IM, and SRM) over a wide torque speed range. For this purpose, weak-coupled analysis of electromagnetic force and structure are performed in ANSYS environment. The secondary aim of this dissertation is to develop a rotor position related variable switching frequency pulse width modulated (PWM) strategy to ameliorate the acoustic noise due to high frequency harmonic current. Switching frequency is modified online to adapt current ripple and vibration requirement, thus ameliorate the acoustic noise. It is expected that this strategy with variable switching frequency has the advantages of spreading the vibration spectrum and reducing switching losses. Experimentally verification is also performed. At the end of this dissertation, characterization of vibration behavior of switch reluctance machine with higher number of rotor poles than stator poles is performed.
Ph.D. in Electrical and Computer Engineering, December 2014
Show less
- Title
- CMOS POLAR DIGITAL POWER AMPLIFIER FOR HIGH DATA RATE WIRELESS COMMUNICATIONS
- Creator
- Zhu, Qiuyao
- Date
- 2016, 2016-12
- Description
-
Power amplifier (PA) is the most important circuit block in an RF transmitter. It typically consumes more than 80% of the power taken by the...
Show morePower amplifier (PA) is the most important circuit block in an RF transmitter. It typically consumes more than 80% of the power taken by the entire transmitter. Therefore, a highly efficient PA is the key to a successful RF front-end system. The polar transmitter architecture is studied herein in order to take advantage of the highly efficient switching-mode PA. However, due to the large expanded bandwidth from the nonlinear IQ to polar conversion and the sensitive amplitude/phase delay impairment, hardly any reported polar design is able to transmit high data rate wireless communication signals. In this work, an extensive research on the digital polar transmitter system for high data rate signals is presented. An integrated CMOS digital power am- plifier (DPA) design is demonstrated afterwards. This DPA consists of 9-bit fully thermometer-coded uniform cells to achieve high linearity for wide bandwidth OFDM signals. By analyzing the amplitude and phase paths impairment, which will cause both in-band and out-of-band distortions, a 960 MHz digital delay tuner is designed for precise amplitude and phase alignment. Furthermore, two digital pre-distortion algorithms for DPA are implemented and compared. Importantly, an on-chip DC- DC converter is included for direct battery connection and power control. A boosted cascode gate bias improves PA efficiency at the low power region. The proposed design is fabricated using a 55 nm RF CMOS technology. The DPA with several peripheral blocks occupies only 0.63 mm2 active silicon area. This DPA including the digital AM filtering achieves a peak output power of +21.9 dBm with 41% efficiency. It achieves EVM of 2.9% with 20 MHz IEEE 802.11ac compliance of 256-QAM OFDM signals, and also achieves EVM of 4.5% (CC0) / 4.8% (CC1) with 2 x 20 MHz 3GPP LTE-Advanced carrier aggregation compliance of 64-QAM OFDM signals. This highly linear DPA has demonstrated high flexibility, high efficiency, and small area. To the author's knowledge, this is the first reported DPA that meets either the linearity requirements of 256-QAM OFDM signals or the signal bandwidth of 40 MHz, paving the path for wideband high data rate wireless applications using digital polar architecture. At the same time, aiming at a higher average efficiency, a two-level class- G supply modulator is investigated to dynamically switch the DPA VDD. It has successfully demonstrated an average efficiency of 34.6% for this class-G modulated DPA in a complete circuit simulation using the IEEE 802.11b signal.
Ph.D. in Electrical Engineering, December 2016
Show less
- Title
- MEASUREMENT OF 013 IN THE DOUBLE CHOOZ EXPERIMENT
- Creator
- Yang, Guang
- Date
- 2016, 2016-12
- Description
-
Neutrino oscillation has been established for over a decade. The mixing angle 013 is one of the parameters that is most difficult to measure...
Show moreNeutrino oscillation has been established for over a decade. The mixing angle 013 is one of the parameters that is most difficult to measure due to its small value. Currently, reactor antineutrino experiments provide the best knowledge of 013, using the electron antineutrino disappearance phenomenon. The most compelling advantage is the high intensity of the reactor antineutrino rate. The Double Chooz experiment, located on the border of France and Belgium, is such an experiment, which aims to have one of the most precise 013 measurements in the world. Dou- ble Chooz has a single-detector phase and a double-detector phase. For the single- detector phase, the limit of the 013 sensitivity comes mostly from the reactor flux. However, the uncertainty on the reactor flux is highly suppressed in the double- detector phase. Oscillation analyses for the two phases have different strategies but need similar inputs, including background estimation, detection systematics evalua- tion, energy reconstruction and so on. The Double Chooz detectors are filled with gadolinium (Gd) doped liquid scintillator and use the inverse beta decay (IBD) sig- nal so that for each phase, there are two independent 013 measurements based on different neutron capturer (Gd or hydrogen). Multiple oscillation analyses are per- formed to provide the best 013 results. In addition to the 013 measurement, Double Chooz is also an excellent playground" to do diverse physics research. For example, a 252Cf calibration source study has been done to understand the spontaneous decay of this radioactive source. Further, Double Chooz also has the ability to do a sterile neutrino search in a certain mass region. Moreover, some new physics ideas can be tested in Double Chooz. In this thesis, the detailed methods to provide precise 013 measurement will be described and the other physics topics will be introduced.
Ph.D. in Pyhsics, December 2016
Show less
- Title
- IMAGE PROCESSING ALGORITHMS FOR PROSTATE CANCER LOCALIZATION WITH MULTISPECTRAL MAGNETIC RESONANCE IMAGING
- Creator
- Xin, Liu
- Date
- 2011-11, 2011-12
- Description
-
In this thesis, we develop a series of image processing algorithms to localize prostate cancer with multispectral magnetic resonance (MR)...
Show moreIn this thesis, we develop a series of image processing algorithms to localize prostate cancer with multispectral magnetic resonance (MR) images to guide biopsy, surgery and minimally invasive therapy. Besides, we develop a new method to for evaluation of image classification algorithms considering correlation between neighboring pixels. Prostate cancer is one of the most prevalent cancer types and one of the leading causes of cancer death among men in the United States. High-resolution MRI has shown higher accuracy than trans-rectal ultrasound (TRUS) to ascertain the presence of prostate cancer. In this work, three different types of MR techniques are employed to provide both morphological and functional information about the benign and malignant tissues of the prostate. These are T2-weighted (T2w) MRI, diffusionweighted imaging MRI (DWI) and dynamic contrasted enhanced MRI (DCE MRI). In the first chapter of this thesis, we briefly describe the fundamentals of different MR techniques, and the multispectral MR dataset used in our experiment. Then, we focus on two tasks of the prostate cancer localization problem: prostate gland extraction and prostate tumor localization. For each topic, we review the previous studies available in the literature, and present our methods with their advantages. Finally, the new image evaluation method considering correlation between pixels is presented. Our prostate segmentation method is fully unsupervised and extracts the prostate gland from DWI MRI in 3D by fusing the active contour model and shape prior information. For tumor localization, we develop an unsupervised approach which is based on fuzzy Markov random field (fuzzy MRF) model, a new scheme based on relative intensity values which can be combined with supervised segmentation classifiers to mimic the cancer localization procedures performed by human readers and a new feature named location map which incorporates the spatial inforx mation of the tumors to remove the need for manual peripheral zone extraction. The proposed image evaluation algorithm is based on receiver operating characteristics (ROC) analysis and it considers the correlation between neighboring pixels. This method could replace the conventional ROC analysis and offers a more accurate evaluation of the test image. Our algorithms are tested on 20 patients’ multispectral MR images, and the qualitative as well as quantitative experimental results demonstrate the efficacy of our segmentation methods and show that the proposed segmentation methods outperform the currently available used approaches. The evaluation method has been tested on computer simulated images and shows very promising results. The summary and future work is also described at the end of the thesis.
Ph.D. in Electrical Engineering, December 2011
Show less
- Title
- SOFT ERROR TOLERANT LATCH CIRCUIT DESIGN FOR DEEPLY SCALED CMOS TECHNOLOGY
- Creator
- Nan, Haiqing
- Date
- 2012-01-25, 2012-05
- Description
-
As CMOS technology keeps scaling down, circuit designers face variety of challenges. Due to the scaling of supply voltage and node capacitance...
Show moreAs CMOS technology keeps scaling down, circuit designers face variety of challenges. Due to the scaling of supply voltage and node capacitance, digital circuits are more aware of noise and variations, which cause reliability issues such as soft error. Traditionally the soft error aware VLSI design is limited to applications which require high reliability and operated in high radiation environment such as avionics applications, medical equipments, space industry and military applications. However, with CMOS technology scales down to nanometer region, the VLSI circuits can also be affected by soft errors at ground level which features low radiation energy. In this thesis, totally 5 soft error tolerant latch designs are proposed including HLR-1, HLR-2, HLR-CG1, HLR-CG2, and HLR-CG3. All the proposed designs protect internal nodes as well as output node for soft error regardless the radiation energy. The proposed HLR-1 and HLR-2 latch circuits tolerate soft error for non-CG systems. Since the proposed HLR-1 and HLR-2 designs take advantages of floating node to tolerate soft error, these two designs cannot be applied with clock gating techniques and the minimum clock frequency of these two designs should be greater than 16MHz in order to maintain correct logic at the floating node. The power consumption and circuit delay between the proposed HLR-1 and HLR-2 designs are very close. The proposed HLR-1 design achieves a small amount of benefits in terms of power and delay compared with the proposed HLR-2 design. But the proposed HLR-2 circuit reduces area 3.5% compared to the proposed HLR-1 circuit. The proposed HLR-CG1, HLR-CG2 and HLR-CG3 latch designs fully tolerate soft error regardless of radiation energy for both CG and non-CG systems. Due to the auto correction mechanism embedded in the proposed HLR-CG1, HLR-CG2 and HLRCG3 designs, any soft error at any location will be automatically corrected without generating any floating nodes. The proposed HLR-CG3 features the smallest power consumption and delay but it has the largest area overhead compared to HLR-CG1 and HLR-CG2 circuits. The proposed HLR-CG1 design features the smallest area compared with HLRCG2 and HLR-CG3 designs. The design cost of HLR-CG2 design is between the proposed HLR-CG1 and HLR-CG3 designs. All the proposed designs achieve faster speed and smaller PDP compared to previous hardening techniques. Compared to the proposed HLR-1 design, previous designs increases power 3.77% on average, delay 272.74% on average, PDP 300.29% on average and decreases area 7.09% on average. Compared to the proposed HLR-2 design, previous designs increases power 3.77% on average, delay 272.40% on average, PDP 299.89% on average and decreases area 3.93% on average. Compared to the proposed HLR-CG1 design, previous designs increases area 19.65% on average, delay 213.14% on average, PDP 203.78% on average and decreases power 5.82% on average. Compared to the proposed HLR-CG2 design, previous designs increase area 6.49% on average, delay 193.28% on average, PDP 223.45% on average and power 6.51% on average. Compared to the proposed HLR-CG3 design, previous designs increases delay 272.18% on average, PDP 314.38% on average, power 8.01% on average and area 2.93% on average.
Ph.D. in Electrical and Computer Engineering, May 2012
Show less
- Title
- IMPACT OF WORK FLOW DISRUPTIONS, PROJECT CHARACTERISTICS AND FOREMEN CHARACTERISTICS ON IMPROVISATIONAL DECISIONS AND ACTIONS ON CONSTRUCTION SITES
- Creator
- Kleps, Stephen M.
- Date
- 2013, 2013-12
- Description
-
Disruptions on construction sites are a constant issue and unfortunately can have negative impacts on the workflow and productivity of the...
Show moreDisruptions on construction sites are a constant issue and unfortunately can have negative impacts on the workflow and productivity of the impacted crews. As a result of these disruptions, foremen must quickly generate improvised solutions to the disruptions to keep the work of their crew progressing. Yet the mechanics of how exactly this improvisational decision making is occurring has not been previously studied. To address this void in the literature, this study examined the factors that may influence a foremen's ability to improvise, such as the type of disruption, the jobsite characteristics, and the personality, experience, and education of the foremen, in relation to the level of improvisation occurring. To accomplish this, this study collected data from 50 foremen within the electrical construction industry to measure to what degree they improvised and with what speed they improvised in response to their daily disruptions, resulting in 244 recorded disruption incidents. In analyzing these incidents with multilevel modeling analysis techniques, it was determined that the type of disruption did not have as large of an effect on the degree and speed of the resulting improvisation as originally hypothesized. In addition, it was found that factors such as the level of cooperation on the job site, the time pressure on the foremen, and the level of collaboration on the job site influenced the degree and speed of the improvisation. Yet the factors that explained the largest total variance in both the degree and speed of improvisation were the personality traits of the foremen. Thus, these findings emphasize the importance of foremen on job sites, while also documenting how improvisation is occurring on job sites.
PH.D in Civil Engineering, December 2013
Show less
- Title
- DEVELOPMENT AND ANALYSIS OF A SECURE AND EFFICIENT VEHICULAR AD HOC NETWORK
- Creator
- Hao, Yong
- Date
- 2012-07-06, 2012-07
- Description
-
Vehicular ad hoc networks (VANETs) enable vehicles to communicate with each other by equipping every vehicle with an on board unit (OBU). Many...
Show moreVehicular ad hoc networks (VANETs) enable vehicles to communicate with each other by equipping every vehicle with an on board unit (OBU). Many interesting and promising functionalities can be achieved in the VANETs, such as safety related application and data downloading application. In this thesis, we focus on the security and privacy provision as well as efficiency improvement of above two applications in the VANETs. In the safety related application, each vehicle periodically broadcasts messages including its current position, direction and velocity (which can be generated by a global positioning system (GPS) device) to inform its geographic data to its neighbors. Privacy is an important issue in VANETs. Meanwhile, some important security functionalities such as message authentication, integrity and non-repudiation should be integrated into the VANETs. In this thesis, we propose a distributed key management protocol based on group signature to provide security and privacy for vehicles. Distributed key management is expected to facilitate the revocation of malicious vehicles, verification efficiency, maintenance of the system and heterogeneous security policies, compared with the centralized key management assumed by the existing group signature schemes. In our framework, each road side unit (RSU) acts as the key distributor for the group, where a new issue incurred is that the semi-trust RSUs may be compromised. Therefore, we develop security protocols which are able to detect compromised RSUs and their malicious accomplices. Moreover, we address the issue of large computation overhead due to the group signature implementation. A practical cooperative message authentication protocol (CMAP) is thus proposed to alleviate the verification burden for vehicles. In the CMAP, on average, each vehicle just needs to verify a very small amount of received geographic messages. Compared with the existing probabilistic verification protocol, CMAP can save at least 40 % computation resource for vehicles. In the data downloading application, we propose a secure cooperative data downx loading framework for payment services in VANETs. In our framework, vehicles download data when they pass by an RSU and then share the data after they travel out of the RSU’s coverage. A fundamental issue of our framework is how vehicles share data with each other. Thus, we develop an application layer data sharing protocol (DSP) in which vehicles share their downloaded data one by one in sequence according to their positions. A better performance can be achieved by the proposed protocol because it is able to avoid medium access control (MAC) layer collisions and the hidden terminal effect. Analytical models are derived to quantitatively evaluate the impact of the distance between RSUs on the amount of data that vehicles can download in a drive through. The simulation results show that our protocol can download 87.4% more data for vehicles than the existing scheme “VC-MAC” when the distance between two consecutive RSUs reaches 10 kilometers. Moreover, we also address security and privacy issues in the process of data downloading and sharing. Both applicants’ exclusive access to the applied data and vehicles’ privacy are ensured by our framework. Compared with the communication overhead in the intuitive method, the communication overhead in our framework will be reduced to 50%. We also propose a security protocol to detect sybil attacks in privacy preserved VANETs. In the above two applications, vehicles’ location information is utilized to facilitate the efficiency. However, if malicious vehicles launch the sybil attack by forging several fake entities and claim they are at some certain positions. The overall performance of the applications will be compromised greatly. So, we propose a security protocol to detect sybil attacks by examining the rationality of vehicles’ positions. The attack detection utilizes the characteristics of communication. No extra hardware and little communication and computation overhead will be introduced to vehicles. Moreover, a smart attacker scenario in which a malicious vehicle may adjust its communication range to avoid detection and the malicious vehicles’ collusion scenario are also considered.
Ph.D. in Computer Engineering, July 2012
Show less
- Title
- STUDIES ON RING OPENING REACTIONS AND SYNTHETIC APPLICATIONS OF AZIRIDINIUM IONS
- Creator
- Sun, Xiang
- Date
- 2013, 2013-12
- Description
-
Aziridinium ions are valuable and important intermediates to prepare complex nitrogen containing compounds. Various enantiomerically enriched ...
Show moreAziridinium ions are valuable and important intermediates to prepare complex nitrogen containing compounds. Various enantiomerically enriched -haloamines were synthesized as precursors to generate aziridinium ions in situ. Optically active aziridinium perchlorates or triflates were prepared and characterized using 1H and 13C NMR. Regioselective and enantioselective ring opening reactions of aziridinium ions with diverse nucleophilies including cyanide, azide, amines, hydride and water were extensively studied. An efficient one-pot synthetic route to enantiomerically enriched alcohols and amine precursors was developed. Development of the adequate bifunctional chelators is a critical step for radioimmunotherapy (RIT) of cancer. The practical and scalable regiospecific ring opening reactions of aziridinium ions with macrocyclic amines were utilized for synthesis of a series of bifunctional ligands including 3p-C-NETA, C-NE3TA, 3p-C-NE3TA, 3p-C-DECA, 3p-C-DE4TA, an optically active (S)-C-NE3TA, and bifunctional DTPA analogues for RIT using 90Y and 177Lu. Preparation of bifunctional chelators that can rapidly form 64Cu complexes with high stability is an imminent goal for effective PET imaging. A series of NOTA and NE3TA-based chelators were prepared. NE3TA-based bile acid conjugates and NE3TA- based dual receptor-targeted conjugates were synthesized and possess potential for application of targeted PET imaging with 64Cu.
PH.D in Chemistry, December 2013
Show less
- Title
- FACTORS ASSOCIATED WITH COGNITIVE IMPAIRMENT IN WELLMOTIVATED CHRONIC NON-MALIGNANT PAIN PATIENTS EVALUATED FOR SPINAL CORD STIMULATION
- Creator
- Zalizniak, Kevin C.
- Date
- 2016, 2016-12
- Description
-
Cognitive impairment in individuals with chronic pain is frequently observed and clinically significant (McCracken, & Iverson, 2001). It has...
Show moreCognitive impairment in individuals with chronic pain is frequently observed and clinically significant (McCracken, & Iverson, 2001). It has long been recognized that emotional factors contribute to both patient perception of impaired cognition and verifiable cognitive impairment on testing (Burt, Zembar, & Niederehe, 1995). However, scientific consensus is lacking regarding how specific emotions, such as depression, anxiety, and pain catastrophizing impact cognition in chronic pain patients. Research seeking to clarify such relationships has been hampered by methodological shortcomings, which include limited sample sizes, non-objective measures, and failure to examine multiple emotional dimensions in unique samples (McCracken and Vowels, 2014; Moriarty, McGuire, & Finn, 2011). The present study examined factors that might contribute to cognitive impairment in this population using a sample of 78 chronic pain patients evaluated for surgical candidacy for spinal cord stimulator (SCS) implantation at the University of Illinois at Chicago. Use of such a sample ensured patients were wellmotivated to perform to the best of their ability, so as to increase their chance of being cleared for such a highly desirable procedure. Additionally, the vast majority of participants passed well-validated objective measures of effort. Hypothesized associations between attentional function as measured objectively by the RBANS attention index and a number of predictor variables: depression and anxiety, subjective pain experience, pain catastrophizing, somatization, and engagement in pain behaviors were not found, and subsequent analyses of proposed mediating relationships could not be performed. However, fully a third (35.9 percent) of our well-motivated sample did not show clinically significant impairment (below 85, or 1 SD below the mean), as was expected. Thus, it is possible that a well-motivated sample may have been less likely than samples used in previous investigations to show cognitive impairment overall. Strengths and limitations of the study are discussed, as well as clinical and research implications.
Ph.D. in Psychology, December 2016
Show less
- Title
- SIZE AND VALUE RISK IN FINANCIAL FIRMS
- Creator
- Baek, Seungho
- Date
- 2013, 2013-12
- Description
-
Although the Fama and French’s three factor model is now the most popular replacement for CAPM in corporate finance and investment management,...
Show moreAlthough the Fama and French’s three factor model is now the most popular replacement for CAPM in corporate finance and investment management, the exclusion of financial firms can be questioned on both theoretical and empirical grounds. Financial firms are around 19 percent of the value of the U.S. stock market. The financial service industry is the major industry in many large U.S. cities including New York, Chicago, Los Angeles and Miami in that their GDPs of financial services for each city are more than 23 percent. Also, there is no theoretical reason for excluding financial firms. Modigliani and Miller [31] [32] suggest that leverage affects beta, but it does not invalidate the capital asset pricing model. It would therefore be more satisfying if the pricing model applied generally, rather than being restricted to nonfinancial corporations. From this consideration, this paper assesses the validity of size and value risk as common risk factors to measure of expected equity returns in financial companies based on the fact that Fama and French [11] [12] excluded financial firms in their study of the cross-section of expected stock returns. The findings from empirical asset pricing tests suggest that size and value risk premia commonly exist in both nonfinancial and financial firms even if two factors are less explicable in financial firms, that an interest rate risk premium (ΔL/L) which defined as a financial firm specific risk factor only appears in financial companies.
PH.D in Management Science, December 2012
Show less
- Title
- NON-GAUSSIAN STOCHASTIC DYNAMICS: MODELING, SIMULATION, QUANTIFICATION AND ASSIMILATION
- Creator
- Gao, Ting
- Date
- 2015, 2015-05
- Description
-
Motivated by real world applications, three topics - deterministic quantities, uncertainty quantification and data assimilation, are...
Show moreMotivated by real world applications, three topics - deterministic quantities, uncertainty quantification and data assimilation, are considered for non-Gaussian stochastic dynamics. More specifically, three problems are formulated to investigate non-Gaussian dynamics: (i) exit problem and time-dependent probability density; (ii) parameter and function estimation for stochastic differential equations driven by L´evy motion; and (iii) nonlinear data assimilation to infer transition phenomena. First, numerical algorithms are developed to study important metrics: mean exit time, escape probability and time-dependent probability density, which can be utilized to quantify dynamical behaviors of stochastic differential equations with non- Gaussian -stable L´evy motion. Moreover, detailed numerical analysis work is done to ensure the algorithms accurate, fast and stable considering the singular nature of the L´evy jump measure. Second, new approaches on parameter and function estimation in stochastic dynamical systems are devised. Taking advantage of observations on mean exit time, escape probability or probability density, model uncertainty can be quantified by some optimization methods. These methods are beneficial to systems for which mean exit time, escape probability or probability density are feasible to observe. Finally, nonlinear data assimilation on non-Gaussian models is studied. For continuous-discrete filtering, a recursive Bayesian approach is used, and for continuous filtering, Zakai equation is solved to provide the system state estimation. In both cases, time-dependent transition probability between metastable states are investigated. xiMotivated by real world applications, three topics - deterministic quantities, uncertainty quantification and data assimilation, are considered for non-Gaussian stochastic dynamics. More specifically, three problems are formulated to investigate non-Gaussian dynamics: (i) exit problem and time-dependent probability density; (ii) parameter and function estimation for stochastic differential equations driven by L´evy motion; and (iii) nonlinear data assimilation to infer transition phenomena. First, numerical algorithms are developed to study important metrics: mean exit time, escape probability and time-dependent probability density, which can be utilized to quantify dynamical behaviors of stochastic differential equations with non- Gaussian -stable L´evy motion. Moreover, detailed numerical analysis work is done to ensure the algorithms accurate, fast and stable considering the singular nature of the L´evy jump measure. Second, new approaches on parameter and function estimation in stochastic dynamical systems are devised. Taking advantage of observations on mean exit time, escape probability or probability density, model uncertainty can be quantified by some optimization methods. These methods are beneficial to systems for which mean exit time, escape probability or probability density are feasible to observe. Finally, nonlinear data assimilation on non-Gaussian models is studied. For continuous-discrete filtering, a recursive Bayesian approach is used, and for continuous filtering, Zakai equation is solved to provide the system state estimation. In both cases, time-dependent transition probability between metastable states are investigated.
Ph.D. in Applied Mathematics, May 2015
Show less