Department: Electrical Engineering ![[clear]](close-x.png "Remove this limiter")
93 matches in the database.
These are records: 1 - 30.
[1] [2] [3] [4]
1.
Abdalla, Abdelnassir.
Torque Ripple Minimization in Direct Torque Control of Induction Machines.
Degree: MS, Electrical Engineering, 2005, University of Akron
► Direct torque control (DTC) of an induction motor fed by a voltage…
(more)
▼ Direct torque control (DTC) of an induction motor fed by a voltage source inverter is a simple scheme that does not need long computation time, can be implanted without mechanical speed sensors and is insensitive to parameter variations. In principle, the motor terminal voltages and currents are sampled and used to estimate the motor flux and torque. Based on estimates of the flux position and the instantaneous errors in torque and stator flux magnitude, a voltage vector is selected to restrict the torque and the flux errors within their respective torque and flux hysteresis bands. In the conventional DTC, the selected voltage vector is applied for the whole switching period regardless of the magnitude of the torque error. This can result in high torque ripple. A better drive performance can be achieved by varying the duty ratio of the selected voltage vector during each switching period according to the magnitude of the torque error and position of the stator flux. A duty ratio control scheme for an inverter-fed induction machine using DTC method is presented in this thesis. The use of the duty ratio control resulted in improved steady state torque response, with less torque ripple than the conventional DTC. Fuzzy logic control was used to implement the duty ratio controller. The effectiveness of the duty ratio method was verified by simulation using SIMULINK, and experimentally using TMS320F240 DSP Evaluation Board.
Advisors/Committee Members: Elbuluk, Malik E.
Keywords: power electronics, motor drives, induction machines, direct torque control
More Like This
2.
Abubakari, Hamza.
IEEE 1588 Style Synchronization over a Wireless Link.
Degree: MS, Electrical Engineering, 2008, University of Akron
► Networked embedded systems, which rely on inexpensive nodes (hardware) and interact in…
(more)
▼ Networked embedded systems, which rely on inexpensive nodes (hardware) and interact in a peer-to-peer manner over wireless links, offer new opportunities for systems architecture and design in a variety of domains. The local clocks in such nodes present relatively large clock offsets at the application level. Achieving time synchronizationacross such nodes, however, remains a challenge. The IEEE 1588 time synchronization protocol specifies how such synchronization can be achieved over wired networks. The wireless domain further exacerbates the problem of achieving time synchronization because of high packet losses and low bandwidth. This thesis presents the design and implementation of a technique for synchronizing clocks, over a wireless link, of a pair of resource-constrained nodes. This is a software-only implementation, in the sense that there is no special hardware required to support the technique. This design builds on a prior technique that compensated clock offset using a conventional digital filter and a Proportional-Integral (PI) controller using IEEE 1588 messages over a wired network. To mitigate the effects of packet-losses in the wireless environment this design tracks and compensates for skew, which is the rate of change of the offset. Because the skew was determined to be gaussian distributed, a linear Kalman filter was used to track the skew. This filtered skew was used as the tracking signal by the PI controller and the output of this controller was used to discipline the clock by modulating the clock rate to match that of the leader. Experimental results demonstrate that this technique, which used a Kalman filter and compensated for skew, performs better than the prior technique that used a conventional digital filter and compensated for offset. These results demonstrate that this technique is resilient against packet losses and achieves better accuracy and stability for both single-hop and multi-hop scenarios. In the future, this technique can serve as a foundation to improve determinism and predictability in networked embedded systems.
Advisors/Committee Members: Sastry, Shivakumar.
Subjects: Computer science; Electrical engineering
Keywords: clocks; SYNCHRONIZATION; preamble; MAC; offsets; messages; ï¬lter
More Like This
3.
Adams, Jay L.
Hankel Operators for Fractional-Order Systems.
Degree: PhD, Electrical Engineering, 2009, University of Akron
► This dissertation presents an algorithm utilizing four unitarily equivalent representations of the…
(more)
▼ This dissertation presents an algorithm utilizing four unitarily equivalent representations of the Hankel operator to estimate the Hankel singular values of fractional-order systems based on the Rayleigh-Ritz method. This algorithm is applied to estimate the first ten Hankel singular values for systems of the forms s/(s^0.5+1) for real, positive a and 1/(s^q+1) for 0 < q < 2. Error bounds are generated for each Hankel singular value estimate. The algorithm is also applied to two physical fractional-order systems, the resistor-terminated and inductor-terminated semi-infinite lines, to show how it can be used. Hankel norms are also approximated for conjugate-order systems of the form (1/2)/(s^q+1)+(1/2)/(s^q*+1) for a variety of complex q to highlight the flexibility of the algorithm in Hankel-norm computation for a wide variety of systems.
Advisors/Committee Members: Hartley, Tom T.
Subjects: Electrical engineering
Keywords: fractional-order calculus; fractional-order systems; Hankel singular values
More Like This
4.
Adhikari, Suraj.
Simulations of Nanofiber Antenna and Its Applications.
Degree: MS, Electrical Engineering, 2009, University of Akron
► Metal-coated nanofibers act as an antenna in the presence of electromagnetic fields.…
(more)
▼ Metal-coated nanofibers act as an antenna in the presence of electromagnetic fields. Due to the physical structure of nanofibers (i.e. radii in the range of nanometers, and length in the range of millimeters) they can be treated as excellent linear antenna. The application of the nano-antenna is in the early stages of development. A model for an infinitely long, finite conductivity carbon nanotube as an antenna can be found in the current literature. We consider finite length metal coated nanofibers which are relatively inexpensive to produce. This is a theoretical work to examine the feasibility and difficulties to implement metal-coated nano-antennas. The analysis presented here includes an exact formulation (well known as Hallen’s equation) which includes an exact kernel but is singular. Due to the presence of a singularity, use of approximate kernel, which is easy to implement but less accurate, is generally preferred. This thesis accurately calculates current distribution along a nanowire antenna by the use of the exact kernel. While doing so, the singularities present in it are analytically removed. Furthermore, simulations for current due to closely spaced nanowire antennas are performed. A large number of such fibers are thought to construct a nanofiber mat, on which current distribution can be calculated. The result is important for the calculation of the power profile over the mat due to the electromagnetic effect in general.
Advisors/Committee Members: Hariharan, Subramaniya.
Subjects: Electrical engineering
Keywords: Hallen's equation; current distribution; nanowire antenna; radiation patterns
More Like This
5.
Akho-Zahieh, Maryam Mahmoud.
Design and Analysis of Multicarrier Multicode Wavelet Packets Based CDMA Communication Systems with Multiuser Detection.
Degree: PhD, Electrical Engineering, 2006, University of Akron
► As the demand for transmitting variable and high data rates rises, the…
(more)
▼ As the demand for transmitting variable and high data rates rises, the need to develop more efficient wireless communication systems increases. In this dissertation, a new Wavelet Packets based Multicarrier Multicode Code Division Multiple Access (WP-MC/MCD-CDMA) system, which combines the properties of wavelet packets with the concept of MC, MCD and CDMA, is proposed and analyzed. Another new wavelet packets based system, denoted as Wavelet Packets Multicarrier Multiuser Code Division Multiple Access (WP-MC/MU-CDMA) system, which combines the properties of wavelet packets with the concept of MC and CDMA, is proposed and analyzed. The WP-MC/MCD-CDMA can be used for multirate services using multicode schemes. In WP-MC/MU-CDMA system, a decorrelating suboptimum detector is used in the detection process. This will suppress the multiple access interference, support a higher number of users and relaxes the power control requirements. In our systems, wavelet packets functions are used as subcarrier instead of a sinusoidal function. Thus, the proposed systems have high immunity against interferences caused by multipath and multicarriers. The systems are evaluated in Nakagami slow fading channel, and their performance are investigated in terms of signal-to-noise plus interference ratio, bit error rate and outage probability. Three diversity combiners, namely selective diversity, equal gain combining and maximal ratio combining are used at the receiver to improve the system performance. Results show that the new proposed systems are viable for future generation wireless systems.
Advisors/Committee Members: Ugweje, Okechukwu C.
Keywords: Wavelet Packets; Multicarrier; Multicode; CDMA; Multiuser; Diversity techniques
More Like This
6.
Alkhateeb, Osama.
Singularity-Free Boundary Methods for Electrostatics and Wave Scattering.
Degree: MS, Electrical Engineering, 2012, University of Akron
► Traditional boundary methods are widely used in applied electromagnetics and other areas…
(more)
▼ Traditional boundary methods are widely used in applied electromagnetics and other areas of physics and engineering but suffer from the singularity of the Green function kernels. The boundary difference method (BDM) implemented and applied in this thesis alleviates this drawback by replacing the singular kernels with Green's functions defined on a discrete lattice. Although its key ideas are available in the literature, BDM has not been extensively applied in electromagnetics. The main objective of this thesis is to demonstrate the accuracy of BDM for electrostatic and wave scattering problems, and, furthermore, to extend the method to problems with edges and corners that present serious computational challenges. Another novelty of the thesis is the treatment of perfect electric conductors in the BDM framework. BDM shares one key advantage of the traditional boundary methods: the dimensionality of the problem is reduced (from 3D to 2D or from 2D to 1D). The resulting dense matrices, as opposed to sparse ones arising in finite difference and finite element analysis, can be handled in a computationally efficient way using Fast Multipole acceleration or other existing techniques, with relatively minor modifications. The focus of the thesis, however, is on the development and construction of BDM rather than numerical solvers. Six case studies in the thesis demonstrate the effectiveness of BDM for electrostatic and wave scattering problems, including field singularities at corners and edges. The method is applied to both dielectric and perfectly conducting scatterers. Convergence rates of the method established via numerical experiments are in agreement with the theoretical predictions.
Advisors/Committee Members: Tsukerman, Igor.
Subjects: Electrical Engineering; Electromagnetics; Electromagnetism
More Like This
7.
Annavajjula, Vamsi Krishna.
A FAILURE ACCOMMODATING BATTERY MANAGEMENT SYSTEM WITH INDIVIDUAL CELL EQUALIZERS AND STATE OF CHARGE OBSERVERS.
Degree: MS, Electrical Engineering, 2007, University of Akron
► Lithium-ion batteries are the most commonly chosen power source for many portable…
(more)
▼ Lithium-ion batteries are the most commonly chosen power source for many portable applications. Advantages like high energy density, high nominal voltage, less maintenance, and low self discharge rate are the driving force behind this choice. Although they have many advantages, lithium-ion batteries have not been used in various applications because of the difficulty of using them well and keeping the individual cells balanced in a series-connected battery pack. This provides our motivation to develop a Battery Management System (BMS) with individual cell equalizers and state of charge (SOC) observers. The main purpose of a BMS is to monitor the cells in a battery pack to ensure proper operation and balance the voltage and charge in the cells in a battery pack in order to maximize the available energy. A BMS was developed for a lithium-ion battery pack with six cells connected in series. The BMS monitors individual cell parameters like voltage, temperature, and current to ensure proper operating conditions and logs this information in an external memory for further processing. Battery model equations are derived, which serve as an SOC observer, to predict and correct the charge stored in the cell. A novel dissipative equalization scheme was proposed to achieve cell equalization among the series-connected cells in terms of both voltage and charge. In contrast to the already published equalization schemes, the proposed scheme achieves equalization among cells in the battery pack in terms of both voltage and stored charge during charging and discharge. Also the proposed battery management system was implemented in hardware to demonstrate its operation. In the event that a cell in the series-connected battery pack fails, the proposed BMS with minor modifications can isolate the failed cell from the battery pack without disturbing the rest of the operation of the pack; this makes the proposed system failure accommodating. Experiments conducted using the implemented BMS show that a charging strategy that includes cell equalization in terms of voltage allows 31% more energy to be stored in the pack than does a simpler strategy that stops charging once the strongest cell in the battery pack reaches the maximum allowable cell voltage. A charging strategy that includes cell equalization in terms of both voltage and stored charge allows 39.33% more energy. The proposed cell equalization scheme during discharge results in an extraction of 82.87% more energy from the battery pack than does a simpler strategy that stops discharging once the weakest cell in the battery pack reaches the minimum allowable voltage.
Advisors/Committee Members: Carletta, Joan.
Keywords: Cell Equalization, Battery Management System, Lithium-ion cells, Lithium-ion batteries
More Like This
8.
Anozie, Chidi H.
Event-Triggered Design of Networked Embedded Automation Systems.
Degree: MS, Electrical Engineering, 2010, University of Akron
► Systems that interact with the physical world can be designed in either…
(more)
▼ Systems that interact with the physical world can be designed in either an event- triggered (ET) paradigm or a time-triggered (TT) paradigm. Most real-time systems today are designed in the TT-paradigm because it is easier to achieve predictable performance. However, when some critical aspect of the system changes, designs based on the TT-paradigm must be completely revised. Modern application require systems to be flexible and reliable. Since the event-triggered systems are fundamen- tally designed as a colletion of loosely coupled local entities, it was interesting to investigate the predictablity of ET-systems. This investigation focused on a ET-design for operating a MicroFactory, which is a physical simulation of a discrete assembly system. The design was im- plemented and evaluated using a collection of networked embedded microcontrollers. Common tasks in a discrete assembly environment were captured as a collection of primitives. These primitives were redesigned and implemented in the ET-paradigm. Experimental results collected reflect the performance at the level of a single node and at the level of the system. Results were studied in comparison with similar ones obtained using a design that was based on the TT-paradigm. In the future, principles and techniques for reconfigurability can be incorporated in the proposed framework.
Advisors/Committee Members: Sastry, Shivakumar.
Subjects: Computer Engineering; Electrical Engineering; Engineering; Experiments
Keywords: event-triggered design; networked embedded automation systems; time-triggered; TinyOS; sensors; actuators; component oriented
More Like This
9.
Archer, Branden.
Profiling Primitives of Networked Embedded Automation.
Degree: MS, Electrical Engineering, 2010, University of Akron
► Recent advances in wireless technologies now allow microcontrollers with low-power wireless radios…
(more)
▼ Recent advances in wireless technologies now allow microcontrollers with low-power wireless radios to integrate with sensor and actuator systems. Using such devices to perform automation tasks is referred to as Networked Embedded Automation (NEA). In NEA, devices perform local actions based on local sensing and communications with neighboring devices. Traditional automation systems use networks of Programmable Automation Controllers wired to each other to regulate system behavior. NEA represents a change in direction for automation systems as they are poised to leverage existing wireless sensor network technologies and distributed algorithm advances.A challenge with integrating NEA into automation systems is understanding predictability and safety issues that are inherent in these enhanced systems. This thesis proposes a set of primitives that are suitable for developing composable automation systems that are regulated by embedded devices interacting over wireless links. A node level architecture and design that helps realize these primitives is presented. Using a robust platform called FireFly, the temporal performance of these primitives is profiled. Experimental results demonstrate that the latency and jitter associated with these primitives are promising and useful for a class of automation systems. In the future, this work can be extended to larger scale systems with tighter temporal constraints.
Advisors/Committee Members: Sastry, Shivakumar.
Subjects: Engineering
Keywords: Networked Embedded Automation
More Like This
10.
Badawy, Mohamed O.
Modeling, Analysis and Experimental Implementation of the Parallel Power Processing Topology for Solar PV Applications.
Degree: MS, Electrical Engineering, 2012, University of Akron
► The main purpose of this thesis is to develop a small size,…
(more)
▼ The main purpose of this thesis is to develop a small size, low cost, and efficient maximum power point (MPP) tracker for individual photovoltaic (PV) panels to charge the storage devices in a standalone system. The proposed topology presented is a reversed buck boost converter enabling parallel power processing (PPP) with a power switch referenced to the common return. Small signal analysis of the proposed charging system is carried out to show the response of the system for any change in the duty ratio. The simulation and experimental results confirmed the validity of the model and the high efficiency system operation with MPP tracking. The use of the PPP with the split loads is also presented along with the simulation results. The results demonstrated an improvement in the size, efficiency and step down duty ratio of the used converter. A PPP based fast battery charger topology for electric vehicles is also presented as an additional application for the topology in this thesis. The proposed topology has high system efficiency as well as reduced voltage stress on decoupling capacitor. The enhancements of the proposed charging topology are evaluated and the results are demonstrated through an example design.
Advisors/Committee Members: Sozer, Yilmaz.
Subjects: Electrical Engineering
More Like This
11.
Bajic, Vladan.
DESIGN AND IMPLEMENTATION OF AN ADAPTIVE NOISE CANCELING SYSTEM IN WAVELET TRANSFORM DOMAIN.
Degree: MS, Electrical Engineering, 2005, University of Akron
► This thesis is focused on the analysis and performance comparison between two…
(more)
▼ This thesis is focused on the analysis and performance comparison between two methods of implementing adaptive filtering algorithms, namely the Normalized time domain Least Mean Squares (NLMS) algorithm, and the Wavelet transform domain LMS (WLMS) algorithm. A brief theoretical development of both methods is explained, as well as the advantages of performing the LMS class of algorithms in the wavelet transform domain. Then, both algorithms are implemented on the real time Digital Signal Porcesing (DSP) system used for audio signals processing. Two different wavelets, Daubechies 2 and Daubechies 3, are used for the comparison. Results are presented showing the performance of each program. The results are shown in both time and frequency domains. The time domain results show a very important characteristic of adaptive filters, that is, the convergence speed of the algorithms under different types of input signals. Also, the frequency domain results are shown for the same algorithms and same input signals. The frequency domain characteristics are analyzed after the adaptive filtering convergence. The noise produced by different algorithms was shown across the spectrum, and the distorting effects were analyzed. The trade-offs of convergence speed versus added noise were analyzed. Also, an analysis of different LMS algorithm implementations on a real time system was analyzed and shown. The overall results show the convergance speed improvement when using WLMS algorithms over the NLMS algorithm.
Advisors/Committee Members: Okechukwu, Ugweje C.
Keywords: Adaptive noise canceling; Wavelet domain LMS; Least mean squares; WLMS; Transform domain adaptive filters
More Like This
12.
Bakula, Casey J.
LOW-POWER PULSE-SHAPING FILTER DESIGN USING HARDWARE-SPECIFIC POWER MODELING AND OPTIMIZATION.
Degree: MS, Electrical Engineering, 2008, University of Akron
► This thesis presents a design process for decreasing the power consumption of…
(more)
▼ This thesis presents a design process for decreasing the power consumption of pulse-shaping filters (PSF), whose purpose is to limit radio transmissions to a specified channel. A hardware layout is assumed based on a low-power application, and a power model of the assumed hardware is developed that approximates the number of logic-stage output transitions that occur in the adder that computes the PSF output. The number of transitions is used as a cost function in a simulated annealing optimization routine that perturbs the ideal PSF coefficients in an effort to find a lower-cost alternative to the ideal PSF. The parameters that dictate the convergence properties of the simulated annealing routine were carefully tailored to this application. It is shown that, given enough time, the optimization routine will find a PSF that closely resembles the ideal PSF, but with considerably lower power consumption. Using a set of PSFs that would be typical for this application, savings between 18.4% and 74.5% of dynamic power consumption were achieved with insignificant distortion of the frequency response. It is observed that the amount of distortion can be controlled by changing the distance that the coefficients may be perturbed from their ideal values. The data presented here suggest that this distance can be very small and significant power savings can still be found, and a limit is suggested as to how large this distance can be before the distortion in the frequency response of the optimized PSF begins to outweigh the savings in power.
Advisors/Committee Members: Carletta, Joan.
Keywords: pulse-shaping filter; low-power; power modeling; simulated annealing; adder modeling; coefficient perturbation
More Like This
13.
Bhandari, Sarita.
Characterization and Modeling of NiZn and Li-based Batteries.
Degree: MS, Electrical Engineering, 2012, University of Akron
► Battery modeling helps to understand the battery behavior, as it adds a…
(more)
▼ Battery modeling helps to understand the battery behavior, as it adds a bridge between the chemical and the electrical worlds. The aim of this thesis is to study the characteristics of the battery and suggest models that can approximate the voltage profile curve and the energy characteristic curve of different battery chemistries. In order to achieve this goal, batteries of different chemistries are monitored throughout their lifetime and the role of all the identified parameters on battery performance are observed. An observation is made that a cell operates in both linear and nonlinear regions, and the models of these regions are split accordingly. An electric circuit model is suggested to represent all the battery dynamics for the linear region. In addition, a mathematical model is used to represent the cell behavior in the nonlinear region. This thesis also focuses on determining the best mathematical model to represent the energy characteristic curve for different chemistry cells. Simulation results obtained by using the electric circuit model and the energy characteristic curve model are validated with the experimental results for NiZn cells and Li-based cells. Further, the models are validated by performing some error calculations, and confirming that the error is below the tolerable level.
Advisors/Committee Members: Hartley, Tom T.
Subjects: Electrical Engineering
Keywords: battery, modeling, NiZn, Li-ion
More Like This
14.
Bhattarai, Kripesh.
On the Use of a Digital Communication Channel for Feedback in a Position Control System.
Degree: MS, Electrical Engineering, 2012, University of Akron
► This thesis presents a study to understand how the features of the…
(more)
▼ This thesis presents a study to understand how the features of the communication protocol used to transmit feedback information on a discrete-time position control system feedback channel affect the overall quality of control. The features of several industrial standards for the transmission of feedback position information by absolute position encoders in a position control system have been identified. The communication channel used to feed back digital position information and the identified parameters are modeled and simulated to determine how the communications affects the performance of the control system, especially while communicating via long cables in the presence of noise. So, a framework has been developed in order to simulate combined operation of the communication channel and the control system. In this work, a series of simulations are performed for an example scenario of a blade pitch position control in a wind turbine. The quality of control is studied for this application when various error checking strategies are used, in systems with different cable lengths and different levels of noise. The results show that for systems communicating via longer cables, error checking is essential to achieve a high quality of control. For control systems that require fast sampling, a trade-off must be made between the bit rate, which determines the sample rate, and the degree of error checking. Recommendations for choosing which features to use are made based on the expected level of noise and cable length.
Advisors/Committee Members: Carletta, Joan E.
Subjects: Electrical Engineering
Keywords: control; communication; wind turbine; blade pitch; feedback; encoder
More Like This
15.
Bhetwal, Lalita.
Detection of Material Properties using Laser Optics.
Degree: MS, Electrical Engineering, 2011, University of Akron
► This thesis presents a method for detection of material properties such as…
(more)
▼ This thesis presents a method for detection of material properties such as refractive index, permittivity and conductivity of a remote object by using laser beams. The method is demonstrated for planar dielectric media in which properties of reflected and refracted wave information is known via Snell’s law. The reflection coefficient of a laser beam on an object is shown to depend on the angle of incidence, the beam frequency and material constants such as electrical permittivity and conductivity. By measuring the intensities of the incident and the reflected waves, reflection coefficients are determined. The reflection coefficient is then used to calculate the refractive index of the material at an accurate level. By varying the parameters such as the angle of incidence and the beam frequency, non-linear complex polynomial equations are obtained which are solved to estimate the unknown conductivity and permittivity of the medium. The results are generalized and applied to rough surfaces which are referred to as diffuse media. In this case, only partial information of the reflected field can be obtained. This partial information yields a scaling factor ‘s’, which we have termed as the diffuse parameter. Finally, by constructing appropriate numbers of nonlinear equations, the permittivity, the conductivity and the diffuse parameter are determined. The estimation of the material properties, made from the proposed method,are verified through experimental results. To achieve this, a series of experiments are carried out in which the angle of incidence is changed with fixed frequency of the incident wave and vice versa. Low power laser sources with wavelengths typically between 400 and 900 nm are used for these applications. Refractive index, electrical permittivity and conductivity of an aluminium sheet are measured through the experiment. The experimental results and the theoretical results are observed to vary by less than 10%. This work realized successfully that optical and electrical properties of materials can be calculated directly through laser applications.
Advisors/Committee Members: Hariharan, S. I.
Subjects: Electrical Engineering
Keywords: Laser Optics; refractive index; permittivity; conductivity
More Like This
16.
Bhuiya, Md. Omar F.
DESIGN AND OPTIMIZATION OF A STRIPLINE RESONATOR SENSOR FOR MEASUREMENT OF RUBBER THICKNESS IN A MOVING WEB.
Degree: MS, Electrical Engineering, 2006, University of Akron
► A stripline resonator sensor was designed for on-line measurement of rubber thickness.…
(more)
▼ A stripline resonator sensor was designed for on-line measurement of rubber thickness. The sensor was designed to provide non contact measurement of rubber thickness with adequate distance between MUT (material under test) and the sensor. The principle of operation of stripline resonator sensor was based on the interaction between microwaves and the matters. The interaction is a form of resonance characteristics of the device versus dielectric constant of the operating medium. After the initial design was completed, a software modeling was performed by using CONCERTO electromagnetic software. A test measurement was taken by changing the rubber thickness to different values and the shift of the resonant frequency was recorded. Once the testing was completed, a comprehensive optimization analysis was performed by changing the physical dimension of the sensor in order to improve the sensitivity. Finally a summary on error contribution from various sources was discussed and the suggestion was presented.
Advisors/Committee Members: Ida, Nathan.
Keywords: stripline sensr; resonator; thickness measuring sensor; thickness gauge; stripline resonator
More Like This
17.
Bishop, Daniel W.
GMSK Demodulation Methods and Comparisons.
Degree: MS, Electrical Engineering, 2008, University of Akron
► Modern phase modulated communication systems, such as GSM cellular phones,require the transmitters…
(more)
▼ Modern phase modulated communication systems, such as GSM cellular phones,require the transmitters to meet a number of performance parameters so that robust links can be achieved. A few of these parameters include: well tuned carrier frequency, low peak phase offset, and low RMS phase error. These parameters must be tested on all new cellular phones coming off the manufacturing line. The mass quantities of cellular phones being produced require that the tests be done quickly and accurately. The method for determining the frequency and phase performance of a GMSK transmitter requires a receiver to generate an ideal reference signal that contains the same symbols as the measured signal. The measured signal is subtracted from the ideal signal so that the errors generated from the transmitter can be easily determined. This work has two goals. First, it surveys methods of building robust GMSK receivers for testing the transmitter. Second, it proposes a new algorithm to improve performance. Building a GMSK receiver to test the transmitter requires three major components. These components are: the timing recovery mechanism, the carrier recovery mechanism, and the demodulator. The purpose of the timing recovery is to align the sampled data to symbols. This work surveys some previously derived methods such as cross correlation and autocorrelation methods. It then presents a new method that is an extension of the cross correlation method by adding a multiple linear regression technique. The purpose of the carrier recovery is to determine the carrier offset from the nominal frequency. Among previously derived methods we survey methods such as squaring the IQ data and using data aided carrier recovery. A new method derived from the multiple linear regression method is then presented. A maximum likelihood demodulator is essential for a robust receiver and this work discusses two methods. The integrate and dump method is by itself not a maximum likelihood detector, but with the proposed extension of the mini trellis it then becomes a pseudo-maximum likelihood detector. The Viterbi demodulator is a maximum likelihood detector and this work presents some optimizations for state reduction to decrease the number of computations and the amount of memory needed. We then compare the performance and accuracy of each of the component methods with simulations.
Advisors/Committee Members: Ida, Nathan.
Keywords: Frequency Offset; GSM; Training Sequences; phase error; offset; bits; Viterbi
More Like This
18.
Blinzler, Adam J.
Polarimeteric Power Spectral Density Analysis of Lung Cancer Cells.
Degree: MS, Electrical Engineering, 2012, University of Akron
► Lung cancer kills more than 150,000 people a year in the United…
(more)
▼ Lung cancer kills more than 150,000 people a year in the United States of America, according to the National Cancer Institute. Over 220,000 people will be diagnosed with the lung cancer in 2012 [1]. The most common cause of form of lung cancer is non-small cell lung cancer and it’s usually attributed to smoking tobacco or use of tobacco related products. Currently, the success rate of detecting and diagnosing early cancer stages is near zero while the 5 year survival rate after diagnosis is less than 20% [1]. New methodologies and modalities for imaging lung cells must be developed in order to increase detection and diagnosis of early stages of lung cancer in order to allow the use of more effective early stage treatments. Near infrared polarized light was used to investigate the power spectral density of the backscattered light from lung cancer samples of stage 2 Adenocarcinoma, Acinar Adenocarcinoma and Squamous cell carcinoma monoline cells. Experiments were performed on several samples with various polarizations. The backscattered light was normalized to investigate the differences in their Welch estimated power spectral density and the consistency of results the width of the backscattered power spectrum [2]. Welch estimated power spectral density was used to reduce the effects of harmonics on the result from a Fast Fourier Transform power spectral density. This study’s emerging research shows the viability of optical imaging techniques in lung cancer screening and diagnosis. The power spectral density, using the estimating technique developed by Welch, was used to analyze Stage 2 Squamous, Acinar Adenocarcinoma, and Adenocarcinoma monoline cancer cells that showed statistically significant ANOVA results. In the cases where the results were not statistically significant, the addition of a mixture of cancer types separated the results enough to be statistically significant and therefore provide a means of detection. This study provides solid ground work for Welch estimated power spectral density analysis to aid in lung cancer detection.
Advisors/Committee Members: Giakos, George.
Subjects: Biomedical Engineering; Biomedical Research; Electrical Engineering; Engineering; Optics
Keywords: FFT; PSD; Power Spectral Density; Fast Fourier Transoform; Lung Cancer; Polarimetry; NIR; Polarization
More Like This
19.
Blosco, John.
Using Receiver Squelch Techniques to Create Scalable Cellular Networks in Capacity Oriented IEEE 802.11 Deployments.
Degree: MS, Electrical Engineering, 2006, University of Akron
► With the recent growth of IEEE 802.11 devices in the industrial and…
(more)
▼ With the recent growth of IEEE 802.11 devices in the industrial and home markets, many companies and households have adopted wireless networks because of their ease of deployment and mobile connectivity. As the popularity and deployment of IEEE 802.11 devices continue to increase, the price of the devices will decrease. Users will demand higher dedicated throughput, resulting in the need for more Access Points (APs) in a deployment to increase its wireless capacity. This will create the need to effectively control the access points to provide maximum throughput performance of the entire system for large scale deployments. In this research, an inefficiency in the current IEEE 802.11 protocol resulting from recommended threshold levels for channel access is identified and analyzed. The cause of this inefficiency is first investigated in detail, and the variables affecting it are presented. The extent to which the throughput degradation affects current IEEE 802.11 systems is then considered, based on today’s channel access level settings. With analytical simulations, the system performance is first evaluated considering only access point interference in a non-shadowed environment. Client interference and signal shadowing is then considered to more accurately model a real world environment. Finally, the thesis conducts experiments in multiple environments, showing measured throughput improvement based on the threshold adjustments presented. The results are compared with a simulation model in order to account for the concepts and theory presented in the research.
Advisors/Committee Members: Ugweje, Okechukwu.
Keywords: CIR; Contention Cell; Co-Channel; CCA; throughput; Planned Cell; SOP
More Like This
20.
Bussa, Srikanth.
A FPGA Based Spectral Line and Pulsar Observing System for Radio Astronomy.
Degree: MS, Electrical Engineering, 2011, University of Akron
► This thesis presents the design of a radio astronomy spectral line and…
(more)
▼ This thesis presents the design of a radio astronomy spectral line and pulsar observing system. This system will be part of the 20-m telescope at NRAO (National Radio Astronomy Observatory, Green Bank, WV) and will be used by the radio SkyNet project for radio astronomy outreach. Radio astronomical observations using this system are done in the X-Band (8-10 GHz). The system can be broadly divided into three units consisting of a super heterodyne receiver, a FPGA based digital spectrometer and a data acquisition unit. The super heterodyne receiver amplifies, filters and down-converts the incoming radio sky frequency (RF) signal in the X-band to an intermediate frequency (IF) 500 MHz band signal, centered at 750 MHz. A tunable local oscillator selects the 500 MHz bandwidth in the band of sky frequencies. The IF signal is fed to the digital spectrometer. The high speed 8-bit analog to digital converter (ADC) in the spectrometer samples the IF signal at 1 Gs/s. The 8-bit ADC provides the high dynamic range required for the system. The digital outputs from the ADC are fed to the Xilinx Virtex V FPGA in ROACH (Reconfigurable Open Architecture Computing Hardware developed by CASPER group) board. A polyphase filter bank FFT spectrometer is implemented in the FPGA using CASPER (Collaboration for Astronomy Signal Processing tools and Electronics Research) technology. The spectra are accumulated for a specified integration time and dumped out into a host computer through a 10 Gigabit Ethernet (GbE) link. The characterization of the ADC and spectrometer are presented. The band pass stability of the spectrometer is measured to be 1000 s and hence any observations should measure a reference spectrum within this time. Since the front-end receiver was not available, the functionality of the spectrometer was tested using an L-Band (21cm) receiver system of the 43m radio telescope at NRAO. We have successfully detected HI emission from the Andromeda galaxy and signals from the Pulsar B0329+54.
Advisors/Committee Members: Ida, Nathan.
Subjects: Astronomy; Electrical Engineering
More Like This
21.
Cai, Jingfeng.
Decision Tree Pruning Using Expert Knowledge.
Degree: PhD, Electrical Engineering, 2006, University of Akron
► Decision tree technology has proven to be a valuable way of capturing…
(more)
▼ Decision tree technology has proven to be a valuable way of capturing human decision making within a computer. It has long been a popular artificial intelligence(AI) technique. During the 1980s, it was one of the primary ways for creating an AI system. During the early part of the 1990s, it somewhat fell out of favor, as did the entire AI field in general. However, during the later 1990s, with the emergence of data mining technology, the technique has resurfaced as a powerful method for creating a decision-making program. How to prune the decision tree is one of the research directions of the decision tree technique, but the idea of cost-sensitive pruning has received much less investigation than other pruning techniques even though additional flexibility and increased performance can be obtained from this method. This dissertation reports on a study of cost-sensitive methods for decision tree pruning. A decision tree pruning algorithm called KBP1.0, which includes four cost-sensitive methods, is developed. The intelligent inexact classification is used for first time in KBP1.0 to prune the decision tree. Using expert knowledge in decision tree pruning is discussed for the first time. By comparing the cost-sensitive pruning methods in KBP1.0 with other traditional pruning methods, such as reduced error pruning, pessimistic error pruning, cost complexity pruning, and C4.5, on benchmark data sets, the advantage and disadvantage of cost-sensitive methods in KBP1.0 have been summarized. This research will enhance our understanding of the theory, design and implementation of decision tree pruning using expert knowledge. In the future, the cost-sensitive pruning methods can be integrated in other pruning methods, such as minimum error pruning and critical value pruning, and include new pruning methods in KBP. Using KBP to prune the decision tree and getting the rules from the pruned tree to help us build the expert system is another direction of our future work.
Advisors/Committee Members: Durkin, John.
Subjects: Computer Science
Keywords: Decision Tree; Expert System
More Like This
22.
Cajko, Frantisek.
Nano-Focusing of Light: Electromagnetic Analysis and Simulation.
Degree: PhD, Electrical Engineering, 2009, University of Akron
► Over the last decades, there has been an ever increasing interest in…
(more)
▼ Over the last decades, there has been an ever increasing interest in nano-focusing oflight and subwavelength resolution overcoming the classical diffraction limit. Examples of that are scanning near-field optical microscopy (SNOM) and “perfect lenses” with negative-index materials. Development of scanning techniques, better performing probes for SNOM and engineering of effective material parameters depends on numerical modeling more than ever before. More accurate models and precise simulations are required to obtain quantitative rather than just qualitative results. This dissertation discusses numerical challenges of nano-scale structure simulations with enhanced and strongly localized electric field distributions. In particular, the thesis focuses on the simulation of scattering-type apertureless SNOM in the mid-infrared and field distributions in plasmon-enhanced Raman spectroscopy in the visible range. Although the ideas of field enhancement are similar (sharp, optionally plasmon-coated, object causing a strong localized enhancement in the vicinity of an AFM tip), applicable models and the nature of computational and engineering challenges are different. For the plasmon-enhanced SNOM, the quasi-static and the full-wave FEM analyses are compared and a qualitative agreement is shown. The optical response of the AFM tip is shown to correlate with the amplitude of the local field distribution. This allows one to use dark field microscopy for tip testing. Several tip designs proposed in the literature were analyzed using the quasi-static approximation; parametric analysis and optimization were performed for selected tips. Numerical challenges due to the multi-scale nature of the problem and multiple scattering in scattering-type SNOM are exemplified in 3D simulations of a realistic cantilevered AFM tip in the mid-infrared. The finite element method (FEM) with adaptive meshing is shown to be a useful tool, but the computation resources of a standard PC must be stretched to their limits. Near and far fields were analyzed and an excellent agreement of the direct back-scattered field with experimental results observed. A substantial part of the dissertation deals with FLAME ” a generalized FD calculus. Numerical problems related to high-precision calculation of the coefficients of the scheme for fine grids are pointed out and overcome. A spurious space of solutions in the case of multiple possible schemes is discovered and remedies proposed. Numerical problems of FD for materials with a negative index of refraction are pointed out and the performance of FLAME is investigated. Advantages of FLAME over standard schemes for interfaces with negative index materials (NIM) are demonstrated on an example of a NIM slab in air.
Advisors/Committee Members: Tsukerman, Igor.
Subjects: Electrical engineering; Electromagnetism; Optics
Keywords: electromagnetic analysis; electrostatics; wave analysis; numerical modeling; negative refraction; NIM; apertureless SNOM; TERS; finite elements; FEM; flexible local approximation; FLAME
More Like This
23.
Chanda, Soumendu.
Powertrain Sizing and Energy Usage Adaptation Strategy for Plug-in Hybrid Electric Vehicles.
Degree: MS, Electrical Engineering, 2008, University of Akron
► An energy usage adaptation (EUA) strategy to manage the charge/discharge profile of…
(more)
▼ An energy usage adaptation (EUA) strategy to manage the charge/discharge profile of the energy storage system for plug-in hybrid vehicles is presented in this thesis. The objective of the EUA strategy is to bring the stored energy to a low level at the end of the daily drive cycle, and to limit the number of deep discharge cycles. The EUA algorithm first predicts the energy usage for a given day based on historical usage data. The predicted energy is then compared with the actual energy used and the battery energy available to set the SOC limits in the energy management algorithm. The EUA strategy has been tuned and tested using simulations of both a series and a series-parallel plug-in hybrid vehicle (model) with vehicle control algorithms developed for the purpose. The strategy is shown to improve the fuel economy of the vehicle and to reduce the cost per mile of operation by efficiently using the off board supplied energy. It also helps to extend the life of the battery by limiting the number of deep discharge cycles to no more than one per day. A well-to-wheel analysis of the designed plug-in hybrid is also done using the standard GREET model and through vehicle simulation to investigate the overall efficiency of plug-in hybrid vehicles. The well-to-wheel efficiency of the plug-in hybrids is found to be lower than those of the conventional gasoline and electric vehicles.
Advisors/Committee Members: Husain, Iqbal.
Subjects: Electrical engineering
Keywords: PLUG-IN HYBRID; ENERGY USAGE; drive cycle; hybrid vehicles; predicted energy
More Like This
24.
Cheerala, Chandana.
A Learning Automata Approach for Input-rate Control in Composable Conveyor Systems.
Degree: MS, Electrical Engineering, 2011, University of Akron
► Simple and effective control techniques are necessary to improve the Quality of…
(more)
▼ Simple and effective control techniques are necessary to improve the Quality of Service (QoS) in a variety of composable and reconfigurable networked systems. Traditional approaches to improve QoS are model-based and rely on representations of the state of the system, at multiple levels. This thesis presents a Learning Automata based approach to improve QoS that does not need to explicitly represent or estimate system state. A Learning Automaton (LA) is a model for adaptive decision-making that uses only stochastic feedback, also called reinforcement, from the environment and does not rely on detailed models or estimation of parameters. It learns to choose the optimal actions from a finite set of actions using such interactions with its environment. This investigation focused on a class of composable conveyor systems that represent a confluence of embedded and real-time systems, wireless communication, sensing and actuating devices, and networking technologies. In such systems, input-rate, i.e., the rate at which parts are injected into the system affects QoS. The input-rate controller presented in this thesis learns the "correct" rate of injection at an input based on noisy feedback on the latency and throughput of parts that moved through the system recently. When it is important to deliver each part at its output before its deadline is elapsed, an admission controller must be used to determine whether the part can be so delivered or not. Preliminary results on value of a probabilistic admission controller are also presented in this thesis. Simulation results confirm the performance of these controllers. These results motivate several investigations in the future to design better admission controllers, route controllers and congestion controllers for composable and networked systems that are based on simple adaptive mechanisms.
Advisors/Committee Members: Sastry, Shivakumar.
Subjects: Electrical Engineering; Engineering
Keywords: Input-rate control; Learning Automata; Composable Conveyor Systems; reconfigurable
More Like This
25.
Chippa, Mukesh Kumar.
Performance of Tree-Based Data Collection in Wireless Sensor Systems.
Degree: MS, Electrical Engineering, 2011, University of Akron
► The predictable collection of data from multiple nodes in a system is…
(more)
▼ The predictable collection of data from multiple nodes in a system is important problem for a variety of emerging applications. Several techniques that are reported in the literature in the context of wireless sensor networks are suitable for ad hoc topologies where energy minimization is a primary design objective. In engineered systems, the locations of the nodes are relatively static and do not change over time. However, faults in the system cause topology disruptions. This investigation focused on predictable data collection in such engineered systems. To improve the overall capacity of the system, the transmission power of the nodes is restricted so that only nodes in local neighborhoods can communicate with each other. In such a case, data from multiple nodes must be sent to one or more collection points, from which the data can be bridged to infrastructure systems. A tree-based strategy is eective in such systems. A tree formation protocol is necessary to establish and maintain such a tree. Because the nodes share a common wireless medium, it is also necessary to arbitrate access to the shared medium. This thesis presents experiment results that demonstrate the performance of the tree formation protocol and data collection over such a tree in a system of 18 nodes. Experimental results demonstrate that message losses in such an approach are largely because of congestion that occurs in the intermediate nodes. When the tree can be designed carefully to ensure that there is no congestion in the intermediate nodes, a message success rate of 100% can be achieved. The results of this work provide a baseline design and implementation of a practical tree formation protocol and data collection method. In the future this approach can be extended to construct optimal trees that provide desired Quality of Service.
Advisors/Committee Members: Sastry, Shivakumar.
More Like This
26.
Chretien, Ludovic.
POSITION SENSORLESS CONTROL OF NON-SALIENT PERMANENT MAGNET SYNCHRONOUS MACHINE.
Degree: MS, Electrical Engineering, 2006, University of Akron
► Permanent Magnet Synchronous Machines (PMSM) are being increasingly used because of their…
(more)
▼ Permanent Magnet Synchronous Machines (PMSM) are being increasingly used because of their advantages over other machines, which include compactness, high efficiency, and well developed drives. The control of sinusoidal PMSM is usually done in the synchronous frame, which requires the knowledge of the rotor position. The sensing of the position is done through optical encoders or synchro-resolvers that increases the cost of the drive and decreases the overall reliability of the system. The substitution of the position sensors by advanced algorithms embedded in the controls hardware and software has been investigated for the last couple of decades. This thesis presents a review of the most common position estimators, and a comparison has been made to determine which one is the best candidate for implementation in industrial applications. The sensorless method relying on back-EMF estimation has been found to be the most promising, and hence, is further investigated in detail. The sensorless drive is studied through simulations, from low speed to very high speed at different torque levels. Then, very low speed and starting of the PMSM is studied, and a non-intrusive method to select the direction of rotation is proposed, given some assumptions about the load. The sensorless starting and operation have been verified experimentally using a digital signal processor implementation. A solution to achieve speed reversal and four-quadrant operation has also been proposed. Finally, a sensitivity analysis of the impact of the position estimation error on the machine torque is presented. The implementation and associated analysis prove that the selected algorithm is highly suitable for position sensorless PMSM drives coupled to passive loads, with a decrease of less than 0.8% in the output torque compared to drives using a position encoder.
Advisors/Committee Members: Husain, Iqbal.
Keywords: PMSM; Position error; SENSORLESS; torque; iq; speed
More Like This
27.
Chu, I-Hsine (Jack).
CONTOUR GUIDED DISSEMINATION FOR NETWORKED EMBEDDED SYSTEMS.
Degree: MS, Electrical Engineering, 2006, University of Akron
► The emergence of new devices and technologies for networked embedded computing motivates…
(more)
▼ The emergence of new devices and technologies for networked embedded computing motivates new platforms for integrated computation, control, communication and coordination. These platforms are based on a large collection of loosely-coupled microcontrollers that communicate over low-power wireless links to achieve systemlevel objectives. The low-power transceiver in each node makes it necessary to propagate messages between non-neighboring nodes over multi-hop routes. Traditional methods for multi-hop communication, on wired networks, rely on unique network identifiers for each node and do not scale well as the number of nodes increases. These methods select one of the available paths between a source and a sink. Extensions to these methods, which are referred to as dissemination methods, have been reported in the sensor networks literature. These extensions are energy efficient and do not rely on network identifiers; however, not all of the available paths between a pair of nodes are fully utilized. This thesis presents a new approach for disseminating called contour guided dissemination. A contour is the union of all the shortest paths between a pair of nodes. When the network topology is regular, it is shown that a contour can be computed using local computations by each node. It is also shown that, despite the availability of multiple paths, it is not possible to effectively utilize the contour by uniformly spreading messages over the available paths. A heuristic that effectively utilizes the contour and simulation results to demonstrate the effectiveness of the contour guided dissemination is presented. The results of this thesis are the first steps towards effectively disseminating data in networked embedded systems with general topologies. These results suggest that some nodes must spread the data to exploit the available paths while other nodes must utilize only one of the desirable paths without spreading the data. Identifying such nodes in systems with general topologies, and integrating contour guided dissemination into system-level protocols are exciting avenues for future research.
Advisors/Committee Members: Sastry, Shivakumar.
Keywords: nq; node; CONTOUR; ni; xdif; messages
More Like This
28.
Dai, Jianhua.
Simulation of Multiobject Nanoscale Systems.
Degree: PhD, Electrical Engineering, 2009, University of Akron
► The development of nanoscience and nanotechnology has important implications for advances of…
(more)
▼ The development of nanoscience and nanotechnology has important implications for advances of electronics, biology, medicine, photonics, and other areas. The growing knowledge in this field will lead to profound progress in the ways that materials, devices, and systems are understood and created. Numerical simulation is an indispensabe tool for understanding nanoscale systems, as our usual intuition may be misleading at the nanoscale. This dissertation focuses on two classes of numerical methods: the finite element method (FEM) and finite difference (FD) methods with their generalization known as the flexible local approximation method (FLAME). FEM is a versatile numerical method that is widely applied in all areas of engineering analysis. This method remains powerful for many physical nanoscale models, especially problems invloving complex geometries and inhomogeneous media, provided that the required number of finite elements is not too large. However, for a large number of objects, the complexity and the computational overhead of FE meshes and the related data structures become too high. Based on the simple Taylor expansions, FD method has significant advantage for geometrically simple problems. However, the accuracy of FD deteriorates for problems with geometrically complex boundaries and material interfaces not conforming to the FD grid lines. The Taylor expansion breaks down at material interface boundaries because the solution is not sufficiently smooth for such problems. FLAME is a generalized FD calculus recently developed. It replaces the Taylor expansion with a physically and mathematically more accurate local approximation. By this way, this method reduces or even eliminates the “staircase” noise at slanted or curved material interfaces. FLAME is first applied in the simulations of electrostatic and magnetostatic multiparticle problems. It shows higher accuracy both in two dimensions (2D) and three dimensions (3D) compared with the finite difference (FD) method and FEM. FLAME also exhibits flexibility in the interpolation of the potential, electric field, and the calculation of the force. For the problems in which components are in close proximity to each other, analytical/numerical bases and adaptive mesh algorithms are developed based on FLAME for better accuracy without increasing the complexity of the calculation. The FLAME method, including analytical/numerical bases and adaptive mesh algorithms, is also applied to wave scattering problems. The computational cost of FLAME in many cases is much lower than that of other methods at comparable levels of numerical accuracy. As a novel application of FLAME, this method is used to explore electrostatic interactions for macromolecules (e.g. protein molecules) in electrolytes. In the conventional model, the whole domain is divided into two layers: the inner macromolecular core and the outer solvent. The inner layer is governed by the Poisson equation with the existance of point charge, and the outer one is governed by the Poisson-Boltzmann equation due to the Boltzmann-like distribution of ions. Results show that this model has great accuracy for short-distance interaction. However, the accuracy for long-distance interaction is not as good as for short-distance interaction. To improve the whole accuracy, an interim layer with a low dielectric permittivity is introduced to simulate the region between macromolecular core and solvent. The simulation based on FLAME shows significant accuracy improvement compared with that of the conventional FD method. The accuracy in FLAME is high even for the area around point charge singularities. FEM is applied to a ferrofluid model that is of interest in magneticly driven assembly of micro- and nanoparticles. The ferrofluid particles are characterized by their volume density with a Boltzmann-like distribution function in the magnetic field. The problem is formulated in terms of the scalar , rather than vector, magnetic potential, which significantly reduces the computational cost. FEM is used for the problem of nano-focusing of light by a self-similar cascade of silver nanoparticles. The goal is to explore the electrodynamic effects affecting the very high local field enhancement. The results lead to appreciable corrections of field enhancement in real applications.
Advisors/Committee Members: Tsukerman, Igor.
Subjects: Electrical engineering; Electromagnetism; Engineering
Keywords: FLAME; FEM; nanoscale; electrostatic multiparticles; wavescattering; adaptive grid refinement; flexible approximation; macromolecules; Poisson-Boltzmann equation; proteins; magnetically self assembly; nanolense; electrodynamic resonances
More Like This
29.
Damecharla, Hima Bindu.
FPGA IMPLEMENTATION OF A PARALLEL EBCOT TIER-1 ENCODER THAT PRESERVES ENCODING EFFICIENCY.
Degree: MS, Electrical Engineering, 2006, University of Akron
► Embedded block coding with optimized truncation (EBCOT) Tier-1 is the most computationally…
(more)
▼ Embedded block coding with optimized truncation (EBCOT) Tier-1 is the most computationally intensive part of the JPEG2000 image coding standard. Past research on fast EBCOT Tier-1 hardware implementations has concentrated on cycle-efficient context formation, and achieved improved throughput at the cost of reduced encoding efficiency. In this thesis, a new fast EBCOT Tier-1 design called the Split Arithmetic Encoder (SAE) is presented. The proposed process exploits concurrency to obtain improved throughput with a lower penalty in encoding efficiency. As opposed to past research which concentrates on either context formation or arithmetic encoding, the design presented is for a complete EBCOT Tier-1 process, and considers system-level issues such as the need for buffering. A hardware architecture for the proposed process was developed, realized in VHDL, and synthesized for implementation on an Altera field programmable gate array. Functionality of the architecture is verified by simulating the system in Modelsim 6.0 for sixteen standard test images and comparing the resulting streams of symbol/context pairs and embedded bit-streams to expected outputs generated using a Matlab simulation. Results show, on average, a 55% improvement in processing time for the proposed architecture over a serial architecture for the set of sixteen standard test images.
Advisors/Committee Members: Carletta, Joan E.
Keywords: FPGA, EBCOT, JPEG2000, Context Formation, Arithmetic Encoding
More Like This
30.
Desai, Dileep Reddy.
Analog Non-Linear Multi-Variable Function Evaluation By Piece-wise Linear Approximation.
Degree: MS, Electrical Engineering, 2010, University of Akron
► A method for the evaluation of non-linear multi-variable functions using analog circuits…
(more)
▼ A method for the evaluation of non-linear multi-variable functions using analog circuits is developed. The non-linear multi-variable functions are approximated with piece-wise linear functions using a Chebyshev approximation technique. Then the obtained linear equations are implemented using current-mode analog circuitry consisting of current comparators and current mirrors. A quarter cycle of a sinusoidal function has been approximated with two lines with a maximum error of 2.75% of the full-scale output current range in the circuit simulations. A more complicated logarithmic function of two variables that is useful as a logarithmic number system subtractor is also approximated with two lines, with a maximum error of 4.26% in the circuit simulations. Implementation of the logarithmic number system subtractor requires a min-max current selector; this circuit sorts two input currents. The min-max current selector works for input currents over a 20µA range, from 70µA to 90µA, and has a maximum output current error of 10nA. It uses a current comparator that has a worst-case input-offset current of 5nA and a differential gain of 2.025V/nA. The logarithmic number system subtractor can be used for image processing applications such as finding the forward and backward differences of the neighboring pixels of an image.
Advisors/Committee Members: Carletta, Joan.
Subjects: Electrical engineering
Keywords: multi-variable function evaluation, analog circuits, LNS subtraction, min-max current selector
More Like This
[1] [2] [3] [4]
