VTechWorks Repository :: Browsing by Author "Jacobs, Ira"

Browsing by Author "Jacobs, Ira"

Now showing 1 - 20 of 116

An access layer protocol for parallel networks
Kumar, Rajesh (Virginia Tech, 1993-05-15)
Parallelism can be applied to local area networks to achieve higher data rates using existing hardware technologies. Parallelism can be employed at the different layers of the Open Systems Interconnection (OS1) reference model. This research proposes and analyzes a technique that permits the use of different degrees of parallelism at different protocol layers. A new protocol layer, called the access layer is defined. The protocol and functionality for this layer are defined. The definitions make provision for incorporating an error correction coding procedure known as cross channel coding. A software simulator was built for the proposed parallel network. The simulator has a detailed model of the access layer and was used to verify the functionality defined for the access layer and to estimate the performance of the parallel network. The simulation results indicate that although the access layer processes data in serial, it is not a bottleneck in the parallel system. Other insights obtained from the experiments are also presented.
Adaptive Beamforming Using a Microphone Array for Hands-Free Telephony
Campbell, David Kemp (Virginia Tech, 1999-03-16)
This thesis describes the design and implementation of a 4-channel microphone array that is an adaptive beamformer used for hands-free telephony in a noisy environment. The microphone signals are amplified, then sent to an A/D converter. The microprocessor board takes the data from the 4 channels and utilizes digital signal processing to determine the direction-of-arrival of the sources and create an output which 'steers' the microphone array to the desired look direction while trying to minimize the energy of interference sources and noise. All of the processing for this thesis will be done on a computer using MATLAB. The MUSIC algorithm is used for direction finding in this thesis. It is shown to be effective in estimating direction-of-arrival for 1 speech source and 2 speech sources that are spaced fairly apart, with significant results down to a -5 dB SNR even. The MUSIC algorithm requires knowledge of the number of sources a priori, requiring an estimator for the number of sources. Though proposed estimators for the number of sources were examined, an effective estimator was not encountered for the case where there are multiple speech sources. Beamforming methods are examined which utilize knowledge of the source direction-of-arrival from the MUSIC algorithm. The input is split into 6 subbands such that phase-steered beamforming would be possible. Two methods of phase-steered beamforming are compared in both narrowband and wideband scenarios, and it is shown that phase-steering the array to the desired source direction-of-arrival has about 0.3 dB better beamforming performance than the simple time-delay steered beamformer using no subbands. As the beamforming solution is inadequate to achieve desired results, a generalized sidelobe canceler (GSC) is developed which incorporates a beamformer. The sidelobe canceler is evaluated using both NLMS and RLS adaptation. The RLS algorithm inherently gives better results than the NLMS algorithm, though the computational complexity renders the solution impractical for implementation with today's technology. A testing setup is presented which involves a linear 4-microphone array connected to a DSP chip that collects the data. Tests were done using 1 speech source and a model of the car noise environment. The sidelobe canceler's performance using 6 subbands (phase-delay GSC) and using 1 band (time-delay GSC) with NLMS updating are compared. The overall SNR improvement is determined from the signal and noise input and output powers, with signal-only as the input and noise-only as the input to the GSC. The phase-delay GSC gives on average 7.4 dB SNR improvement while the time-delay GSC gives on average 6.2 dB SNR improvement.
Adaptive, Turbo-coded OFDM
Ilunga, Lou (Virginia Tech, 2005-06-30)
Wireless technologies, such as satellite, cellular, and wireless internet are now commercially driven by ever more demanding consumers, who are ready for seamless integration of communication networks from the home to the car, and into the office. There is a growing need to quickly transmit information wirelessly and accurately. Engineers have already combine techniques such as orthogonal frequency division multiplexing (OFDM) suitable for high data rate transmission with forward error correction (FEC) methods over wireless channels. In this thesis, we enhance the system throughput of a working OFDM system by adding turbo coding and adaptive modulation (AD). Simulation is done over a time varying, frequency selective Rayleigh fading channel. The temporal variations in the simulated wireless channel are due to the presence of Doppler, a sign of relative motion between transmitter and receiver. The wideband system has 48 data sub-channels, each is individually modulated according to channel state information acquired during the previous burst. The end goal is to increase the system throughput while maintaining system performance under a bit error rate (BER) of 10-2. The results we obtained are preliminary. The lack of resources prevented us from producing detailed graphs of our findings.
Analysis and Characterization of Fiber Nonlinearities with Deterministic and Stochastic Signal Sources
Lee, Jong-Hyung (Virginia Tech, 2000-02-10)
In this dissertation, various analytical models to characterize fiber nonlinearities have been applied, and the ranges of validity of the models are determined by comparing with numerical results. First, the perturbation approach is used to solve the nonlinear Schrödinger equation, and its range of validity is determined by comparing to the split-step Fourier method. In addition, it is shown mathematically that the perturbation approach is equivalent to the Volterra series approach. Secondly, root-mean-square (RMS) widths both in the time domain and in the frequency domain are modeled. It is shown that there exists an optimal input pulse width to minimize output pulse width based on the derived RMS models, and the functional form of the minimum output pulse width is derived. The response of a fiber to a sinusoidally modulated input which models an alternating bit sequence is studied to see its utility in measuring system performance in the presence of the fiber nonlinearities. In a single channel system, the sinusoidal response shows a strong correlation with eye-opening penalty in the normal dispersion region over a wide range of parameters, but over a more limited range in the anomalous dispersion region. The cross-phase modulation (CPM) penalty in a multi-channel system is also studied using the sinusoidally modulated input signal. The derived expression shows good agreement with numerical results in conventional fiber systems over a wide range of channel spacing, ∆f, and in dispersion-shifted fiber systems when ∆f > 100GHz. It is also shown that the effect of fiber nonlinearities may be characterized with stochastic input signals using noise-loading analysis. In a dense wavelength division multiplexed (DWDM) system where channels are spaced very closely, the broadened spectrum due to various nonlinear effects like SPM (self-phase modulation), CPM, and FWM (four-wave mixing) is in practice indistinguishable. In such a system, the noise-loading analysis could be useful in assessing the effects of broadened spectrum due to fiber nonlinearities on system performance. Finally, it is shown numerically how fiber nonlinearities can be utilized to improve system performance of a spectrum-sliced WDM system. The major limiting factors of utilizing fiber nonlinearities are also discussed.
Analysis and design of broadband single-mode multi-clad fibers
Lu, Liang-Ju (Virginia Polytechnic Institute and State University, 1989)
ln the last several years, considerable attention has been paid to the study of dispersion-flattened single-mode fibers which offer a high transmission capacity with low losses through a wide range of wavelengths. However, the existing designs are sensitive to bending and manufacturing tolerances, and are not truly single-mode at most wavelengths of interest. To remedy these problems a new series of broadband dispersion-flattened truly single-mode fiber designs are proposed. These fibers have both dispersion-shifted and dispersion-flattened features with low splice and bend losses. Results demonstrating a total dispersion of ±0.97 ps/km-nm over the entire spectral range between 1.31 μm to 1.66 μm are presented. Such dispersion-flattening is achieved while simultaneously maintaining a mode-field radius of 3 μm to 5 μm in the dispersion-flattened wavelength range. The most significant achievement is that the proposed muIti-clad fiber design is strictly single-mode and splice and bend losses are smaller than those of double-clad, triple-clad, and quadruple-clad fibers with the same value of dispersion. Ultralow dispersion fibers, whose chromatic dispersion and the first and second-order derivatives of the chromatic dispersion are zero at 1.5 μm or 1.55 μm, are described. This effectively increases the laser emission tolerance. Ultralow dispersion fibers open the way to wavelength multiplexing with currently available inexpensive multifrequency lasers, either in local or long distance networks. These fibers also have low splice and bend losses compared to double-clad, triple-clad, and quadruple-cIad fibers. An inverse waveguide synthesis program, which can trace multiple objective functions and optimize multiple parameters simultaneously, is developed. An objective function is applied, for the first time, to optimize the dispersion-flattened single-mode fiber index profile with respect to: (1) minimum dispersion, (2) the wavelengths of zero-dispersion, (3) maximum width of dispersion-flattened window, (4) maximum layer index difference less than 0.8%, and (5) layer thickness larger than 3.5 μm. The accuracy of chromatic dispersion calculations in dispersion-flattened fibers is evaluated. lt has been shown that the accuracy of approximate methods is influenced not only by the index differences, but also by their derivatives with respect to wavelength. The matrix method and direct numerical integration of the wave equation are used to compute the mode propagation constants, cutoff frequencies, field distributions, mode-field radius, and splice loss, and carry out production tolerance analysis for multi-clad step-index fibers and graded-index fibers, respectively. Detailed analysis and optimized fiber data are presented.
Analysis and simulation of the Kerr effect in long haul in-line fiber amplifier transmission systems
Ma, Xiaobing (Virginia Tech, 1994-12-05)
The next generation of transoceanic submarine cable systems will use in-line fiber amplifiers to replace electrical regenerators. This new approach requires a better understanding of the fiber waveguide, especially the nonlinear characteristics. It has been demonstrated, both by numerical simulation and experiments, that the Kerr effect has the most significant degradation effects on these systems with a single or a few channels. In this dissertation, the numerical simulation is the principal approach but this is supplemented with some analytical studies. There are two phenomena that are directly associated with the Kerr effect: spectrum broadening and four wave mixing (FWM). The broadened signal spectrum enhances the dispersion effect and consequently increases the inter symbol interference. This distortion is significant when the dispersion is relatively large. U sing erbium doped fiber amplifiers, amplified spontaneous emission (ASE) noise is added to the signal The FWM effect between the noise and signal causes a loss in the signal power. In the close vicinity of the zero dispersion wavelength, the FWM effect is maximized due to the loose phase match condition. In order to reduce these degradation effects, dispersion compensation has recently been proposed. Using this approach, it is possible to optimize the system configuration to achieve the best performance. In this dissertation, the dispersion compensation distance, dispersion coefficient, channel bandwidth, bit rate and the degree of imperfect compensation are all subject to the optimization. The evaluation is obtained by the numerical simulation using the mean squared error (MSE), which can be derived as the difference between the wavefronts received by a back-to-back receiver and a receiver at the end of the channel. Although the MSE can't be related to the bit error directly, this research provides the insight into how dispersion and noise behave in the presence of the Kerr effect and points the direction for future experimental research.?
Analysis of ATM Call Detail Records and Recommendations for Standards
Wang, Xianrui Roger (Virginia Tech, 1999-06-23)
Data network resource management and capacity planning are critical for network design, operation, and management. Equipment vendors often provide good information for traffic management and control and associated tools, but this information and the tools are based on independent, individual switches or routers rather than the whole network. There is a critical need for tools to monitor general resource usage in a network as a whole. In this research, we develop a toolkit to collect ATM Call Detail Records (CDRs) from two types of ATM switches from IBM and FORE Systems. Data records collected by the toolkit can then be used to assess network resource utilization and traffic characteristics with the objective of predicting future needs, making proper network management decisions, and ultimately, assisting in the ability to provide reliable quality of service (QoS) in the network. In addition, we examine current call detail records and requirements for more comprehensive network management and make recommendations for a standardized CDR.
Analysis of the hardware requirements of a high speed computer interface required to utilize fiber distributed data interface
Tolley, Dan B. (Virginia Tech, 1990)
As the use of computers in the workplace becomes more commonplace, the levels of interconnection and interoperability increases. The desire to pass large amounts of data almost instantaneously is the basis of the high speed local area network (LAN). To meet the needs of these LANs, the American National Standards Institute has developed the Fiber Distributed Data Interface (FDDI) standard. This new LAN can provides high speed fiber optic based communication between computers. In meeting the computer/LAN interface requirements, new methods for data transfer will be required. Trade-offs between the reliability, architecture and buffer sizes must be developed. These concepts must include variables of data transfer widths, protocol processing, transfer architecture and packet length distributions. This dissertation addresses these hardware requirements in using the high speed computer interface known as the Fiber Data Distributed Interface.
An Analysis of W-fibers and W-type Fiber Polarizers
Paye, Corey (Virginia Tech, 2001-04-27)
Optical fibers provide the means for transmitting large amounts of data from one place to another and are used in high precision sensors. It is important to have a good understanding of the fundamental properties of these devices to continue to improve their applications. A specially type of optical fiber known as a W-fiber has some desirable properties and unique characteristics not found in matched-cladding fibers. A properly designed W- fiber supports a fundamental mode with a finite cutoff wavelength. At discrete wavelengths longer than cutoff, the fundamental mode experiences large amounts of loss. The mechanism for loss can be described in terms of interaction between the fiber's supermodes and the lossy interface at the fiber's surface. Experiments and computer simulations support this model of W-fibers. The property of a finite cutoff wavelength can be used to develop various fiber devices. Under consideration here is the fiber polarizer. The fiber polarizer produces an output that is linearly polarized along one of the fiber's principal axes. Some of the polarizer properties can be understood from the study of W-fibers.
Analysis, Design and Performance Evaluation of Optical Fiber Spectrum-Sliced WDM Systems
Arya, Vivek (Virginia Tech, 1997-06-03)
This dissertation investigates the design and performance issues of a recently demonstrated technique, termed as spectrum-slicing, for implementing wavelength-division-multiplexing (WDM) in optical fiber systems. Conventional WDM systems employ laser diodes operating at discrete wavelengths as carriers for the different data channels that are to be multiplexed. Spectrum-slicing provides an attractive low-cost alternative to the use of multiple coherent lasers for such WDM applications by utilizing spectral slices of a broadband noise source for the different data channels. The principal broadband noise source considered is the amplified spontaneous emission (ASE) noise from an optical amplifier. Each slice of the spectrum is actually a burst of noise that is modulated individually for a high capacity WDM system. The stochastic nature of the broadband source gives rise to excess intensity noise which results in a power penalty at the receiver. One way to minimize this penalty, as proposed and analyzed for the first time in this work, is to use an optical preamplifier receiver. It is shown that when an optical preamplifier receiver is used, there exists an optimum filter bandwidth which optimizes the detection sensitivity (minimizes the average number of photons/bit) for a given error probability. Moreover the evaluated detection sensitivity represents an order of magnitude ( > 10 dB) improvement over conventional PIN receiver-based detection techniques for such spectrum-sliced communication systems. The optimum is a consequence of signal energy fluctuations dominating at low values of the signal time bandwidth product (m), and the preamplifier ASE noise dominating at high values of m. Operation at the optimum bandwidth renders the channel error probability to be a strong function of the optical bandwidth, thus providing motivation for the use of forward error correction coding (FEC). System capacity (for BER = ) is shown to be 23 Gb/s without coding, and 75 Gb/s with a (255,239) Reed Solomon code. The effect of non-rectangular spectra on receiver sensitivity is investigated for both OOK and FSK transmission, assuming the system (de)multiplexer filters to be N'th order Butterworth bandpass. Although narrower filters are recommended for improving power budget, it is shown that system penalty due to filter shape may be kept < 1 dB by employing filters with N > 2. Moreover spectrum-sliced FSK systems using optical preamplifier receivers are shown, for the first time, to perform better in a peak optical power limited environment. Performance-optimized spectrum-sliced WDM systems have potential use in both local loop and long-distance fiber communication systems which require low-cost WDM equipment for high data rate applications.
Calculations and Measurements of Raman Gain Coefficients of Different Fiber Types
Kang, Yuhong (Virginia Tech, 2002-12-09)
Fiber Raman amplification using the transmission line is a promising technology to increase the repeater distance as well as the capacity of the communication systems. Because of the growing importance of fiber Raman amplification, it is desired to predict the magnitude and shape of the Raman gain spectrum from the doping level and refractive index profiles of different fiber designs. This thesis develops a method to predict the Raman gain coefficients and spectra for a pure silica core fiber and two different types of GeO2-doped silica fibers given their index profiles. An essential feature of the model is the inclusion of the variation in Raman gain coefficient over the mode field due to the variation in the Ge concentration across the fiber core. The calculated Raman gain coefficients were compared with measurements of the peak Raman gain on a step-index GeO2-doped fiber and with published measurements from various sources. Agreement between the calculated and measured peak gain for the step-index fiber was excellent. There was qualitative agreement with published measurements but there were significant differences between the calculated and published gain coefficients, which are not understood. Part of the work sought a way of predicting Raman gain coefficients from a standard gain curve given only the fiber type and the effective area. This approach appears promising for moderately-doped fibers with the proper choice of effective area.
Capacity-approaching data transmission in MIMO broadcast channels
Jiang, Jing (Virginia Tech, 2004-06-30)
This dissertation focuses on downlink multi-antenna transmission with packet scheduling in a wireless packet data network. The topic is viewed as a critical system design problem for future high-speed packet networks requiring extremely high spectral efficiency. Our aim is to illustrate the interaction between transmission schemes at the physical layer and scheduling algorithms at the medium access control (MAC) layer from a sum-capacity perspective. Various roles of multiple antennas are studied under channel-aware scheduling, including diversity, beamforming and spatial multiplexing. At a system performance level, our work shows that downlink throughput can be optimized by joint precoding across multiple transmit antennas and exploiting small-scale fading of distributed multiple input and multiple output (MIMO) channels. There are three major results in this dissertation. First, it is shown that over a MIMO Gaussian broadcast channel, and under channel-aware scheduling, open-loop transmit antenna diversity actually reduces the achievable sum rate. This reveals a negative interaction between open-loop antenna diversity and the closed-loop multiuser diversity through scheduling. Second, a suboptimal dirty paper coding (DPC) approach benefits greatly from multiuser diversity by an efficient packet scheduling algorithm. Performance analysis of a suboptimal greedy scheduling algorithm indicates that, compared with the receiver-centric V-BLAST method, it can achieve a much larger scheduling gain over a distributed MIMO channel. Further, pre-interference cancellation allows for transmissions free of error propagation. A practical solution, termed Tomlinson-Harashima precoding (THP), is studied under this suboptimal scheduling algorithm. Similar to V-BLAST, a reordering is applied to minimize the average error rate, which introduces only a negligible sum-rate loss in the scenarios investigated. Third, for an orthogonal frequency division multiplexing (OFDM) system using MIMO precoding, it is shown that a DPC-based approach is readily applicable and can be easily generalized to reduce the peak-to-average power ratio (PAR) up to 5 dB without affecting the receiver design. Simulations show that in an interference-limited multi-cell scenario, greater performance improvement can be achieved by interference avoidance through adaptive packet scheduling, rather than by interference diversity or averaging alone. These findings suggest that, coordinated with channel-aware scheduling, adaptive multiplexing in both spatial and frequency domains provides an attractive downlink solution from a total capacity point of view.
Channel Impulse Response and Its Relationship to Bit Error Rate at 28 GHz
Miniuk, Mary (Virginia Tech, 2003-12-19)
Over the years, the Internet has become increasingly popular and people's dependence on it has increased dramatically. Whether it be to communicate to someone across the world, find blueprints, or check sports scores, the Internet has become a necessary resource for everyone. In emergency situations, this need increases further. After the terrorist attacks on the Pentagon, it took several days to restore communications. This is not an acceptable time frame when people's lives are at stake. Virginia Tech's Center for Wireless Telecommunication has developed a prototype of a rapidly deployable high bandwidth wireless communication system at 28 GHz (Local Multipoint Distribution Service frequency). This system provides a large bandwidth radio link to a disaster zone up to 5 km away and puts Ethernet speeds and 802.11 accesses to users within hours. Because of the possible variability in locations that the system can be deployed, it is necessary to find the most useable channel at the site as quickly as possible. In addition to 28GHz radio links, the system also has a built-in channel sounder that measures and captures the channel impulse response of the current channel. Until now, there has been limited research on the relationship between the channel impulse response and the usability of the channel quantified using bit error rate. This thesis examines several different channels captured by CWT's channel sounder and simulates the BER using Cadence's SPW with time-domain models of the channels. This thesis goes on further to show that BER greatly depends on the channel impulse response and the symbol rate.
Characterization and modeling of crosstalk noise in digital systems and microwave applications
Teekaput, Prasit (Virginia Tech, 1990)
This dissertation presents the characterization and modeling of crosstalk noise based on the theory of coupled microstrip lines. An equivalent circuit model used in this work comprises of the addition of mutual inductances and mutual capacitances to the fundamental transmission line model. Characterization of crosstalk noise along adjacent lines, current-voltage characteristics, characteristic impedance, effective dielectric impedance, and maximum crosstalk are performed analytically. Computer simulations and computations of these parameters are also performed. The circuits are realized experimentally, and an investigation of crosstalk noise using time domain and frequency domain measurement techniques is conducted. The results illustrate that the computation matched closely the experimental data and explained the physical phenomena better.
Characterization of Wideband All-Fiber Waveplates
Sanghvi, Harsh (Virginia Tech, 2004-04-27)
Many commercial and laboratory applications of fiber optics require an ability to manipulate, change and control the state of polarization of light. This is usually accomplished with bulk-optic wave plates which are inherently narrowband, bulky, and often require careful manual tuning and adjustments. H.C. Huang has recently proposed that a variably spun birefringent fiber with spin rate slowly varying from zero to very fast (or vice versa) will transform the state of polarization (SOP) from linear to circular (and vice versa). The most remarkable feature of a Huang fiber is that the transformation from linear to circular polarization is independent of wavelength over entire single mode range of the fiber. In this thesis, using simulations and experiments we explore the properties of such a fiber element and investigate its similarities and differences as compared to a bulk-optic quarter-wave plate. In the simulations, we modeled the Huang fiber as a stack of birefringent waveplates and used this model verify the theoretically predicted polarization transformation behavior and the wideband nature of the Huang fiber. We analyzed the dependence of the polarization transformation by this device on various structural parameters and showed that while the fiber has loose tolerances with respect to the fiber length and the spin variation, it has a strong dependence on the maximum spin rate at the high-spun end. We investigated the PMD characteristics of the Huang fiber for short pulse applications. Using simulations we also verified that two such quarter wave-transforming fibers can be appropriately cascaded to obtain half-wave and full-wave transformation, analogous to bulk optics half wave and full wave plate respectively. In the experiments we studied the polarization transformation behavior of a Huang fiber sample when it is excited by different input states of polarization both from the un-spun end and the high-spun end. We found that the results from the experiments strongly support the simulation results. The experiment was performed at 1310 nm and 1550 nm to verify the wideband nature of the Huang fiber sample. We found that the Huang fiber indeed performs the prescribed state of polarization transformation over a wide band.
Co-Channel Interference In Bluetooth Piconets
Lynch, Jamel Pleasant (Virginia Tech, 2002-06-10)
Bluetooth™ is an emerging short-range RF wireless voice and data communication technology expected to spread widely in a couple of years. The open specification technology eliminates the need for cables to connect mobile phones, portable computers and countless other devices to each other from all different manufacturers. This thesis provides an overview of the emerging Bluetooth™ technology and investigates the performance of Bluetooth™ data networks in various network topologies simulated from actual usage scenarios. Using a typical office environment, the study examines the probability and effects of Co-Channel interference as Bluetooth™ ad-hoc networks are formed in adjacent offices. A computer aided simulation tool, MATLAB simulates a low to highly dense interfering Bluetooth™ environment which provides the parameters to evaluate the bluetooth co-channel interference and performance. Several metrics are identified to predict Bluetooth™ performance in a piconet after a collision has occurred: data through put, the probability of frequency collision, transmitter - receiver distance, and power received. Next, to predict Bluetooth performance we also need to define what constitutes a lost packet. Finally, a Bluetooth™ network simulation is developed to measure the metrics, given occurrence of the lost packets.
Color Face Recognition using Quaternionic Gabor Filters
Jones, Creed F. III (Virginia Tech, 2004-12-13)
This dissertation reports the development of a technique for automated face recognition, using color images. One of the more powerful techniques for recognition of faces in monochromatic images has been extended to color by the use of hypercomplex numbers called quaternions. Two software implementations have been written of the new method and the analogous method for use on monochromatic images. Test results show that the new method is superior in accuracy to the analogous monochrome method. Although color images are generally collected, the great majority of published research efforts and of commercially available systems use only the intensity features. This surprising fact provided motivation to the three thesis statements proposed in this dissertation. The first is that the use of color information can increase face recognition accuracy. Face images contain many features, some of which are only easily distinguishable using color while others would seem more robust to illumination variation when color is considered. The second thesis statement is that the currently popular technique of graph-based face analysis and matching of features extracted from application of a family of Gabor filters can be extended to use with color. A particular method of defining a filter appropriate for color images is used; the usual complex Gabor filter is adapted to the domain of quaternions.. Four alternative approaches to the extension of complex Gabor filters to quaternions are defined and discussed; the most promising is selected and used as the basis for subsequent implementation and experimentation. The third thesis statement is that statistical analysis can identify portions of the face image that are highly relevant — i.e., locations that are especially well suited for use in face recognition systems. Conventionally, the Gabor-based graph method extracts features at locations that are equally spaced, or perhaps selected manually on a non-uniform graph. We have defined a relevance image, in which the intensity values are computed from the intensity variance across a number of images from different individuals and the mutual information between the pixel distributions of sets of images from different individuals and the same individual. A complete software implementation of the new face recognition method has been developed. Feature vectors called jets are extracted by application of the novel quaternion Gabor filter, and matched against models of other faces. In order to test the validity of the thesis statements, a parallel software implementation of the conventional monochromatic Gabor graph method has been developed and side-by-side testing has been conducted. Testing results show accuracy increases of 3% to 17% in the new color-based method over the conventional monochromatic method. These testing results demonstrate that color information can indeed provide a significant increase in accuracy, that the extension of Gabor filters to color through the use of quaternions does give a viable feature set, and that the face landmarks chosen via statistical methods do have high relevance for face discrimination.
A Comparison Between Synchronous CDMA and Orthogonal Frequency Division Multiplexing (OFDM) for Fixed Broadband Wireless Access
Chheda, Shital Ratilal (Virginia Tech, 2002-04-25)
The growth of broadband Internet access has paved the way for the development of many new technologies. As the cost of implementing broadband access soars, the best alternative will be to use fixed wireless for these services. This thesis addresses the possibility of 3rd Generation (3G) mobile cellular wireless systems as the basis for fixed broadband wireless service. Two of the 3G technologies aimed at providing fixed broadband wireless access are Time Division Synchronous Code Division Multiple Access (TD-SCDMA) and Orthogonal Frequency Division Multiplexing (OFDM). This thesis aims to provide a preliminary study on using TD-SCDMA and OFDM for broadband wireless systems. Currently, there is not enough theory and information to establish the feasibility of using either of these technologies for broadband wireless access. First, the basic features and background on synchronous CDMA and OFDM are presented for the reader to better understand these technologies. Then, an example TD-SCDMA system is described, and some analytical and experimental results are presented. Finally, TD-SCDMA's technologies, along with this system's attributes, are compared analytically to that of Vector OFDM (VOFDM).
Coupled-waveguide Fabry-Perot resonator
Chang, Cheng-Chun (Virginia Tech, 1992-12-04)
Narrowband spectral filters find important applications in optical fiber communication systems, particularly in wavelength demultiplexers and single-frequency semiconductor lasers. Conventional Fabry-Perot resonators provide a narrow spectral width but lack the capability of mode discrimination. A new coupled-waveguide Fabry-Perot resonator made of two parallel waveguides with reflecting mirrors at the ends is proposed for application as narrowband tuned spectral filter in single-mode diode lasers and wavelength demultiplexers. The interference of counter propagating waves from reflection by end mirrors and the coupling of waves between the two parallel guides contribute to the operation of this resonator structure. Thus, the device exhibits the attributes of both Fabry-Perot resonator and directional coupler. The coupled-mode theory of parallel waveguides is employed to analyze the proposed structure. Spectral characteristics are derived from the governing coupled-mode equations and related boundary conditions. Two geometries consisting of identical waveguides, as well as nonidentical waveguides, are examined. The spectral characteristics of the proposed resonator demonstrate that significant improvement in mode discrimination capability and longitudinal mode spacing over the conventional Fabry-Perot resonator is achieved. Numerical results for several example cases are presented and the influence of various parameters on spectral properties are investigated.
Design and Analysis of Defect- and Fault-tolerant Nano-Computing Systems
Bhaduri, Debayan (Virginia Tech, 2007-02-19)
The steady downscaling of CMOS technology has led to the development of devices with nanometer dimensions. Contemporaneously, maturity in technologies such as chemical self-assembly and DNA scaffolding has influenced the rapid development of non-CMOS nanodevices including vertical carbon nanotube (CNT) transistors and molecular switches. One main problem in manufacturing defect-free nanodevices, both CMOS and non-CMOS, is the inherent variability in nanoscale fabrication processes. Compared to current CMOS devices, nanodevices are also more susceptible to signal noise and thermal perturbations. One approach for developing robust digital systems from such unreliable nanodevices is to introduce defect- and fault-tolerance at the architecture level. Structurally redundant architectures, reconfigurable architectures and architectures that are a hybrid of the previous two have been proposed as potential defect- and fault-tolerant nanoscale architectures. Hence, the design of reliable nanoscale digital systems will require detailed architectural exploration. In this dissertation, we develop probabilistic methodologies and CAD tools to expedite the exploration of defect- and fault-tolerant architectures. These methodologies and tools will provide nanoscale system designers with the capability to carry out trade-off analysis in terms of area, delay, redundancy and reliability. During execution, the next state of a digital system is only dependent on the present state and the digital signals propagate in discrete time. Hence, we have used Markov processes to analyze the reliability of nanoscale digital architectures. Discrete Time Markov Chains (DTMCs) have been used to analyze logic architectures and Markov Decision processes (MDPs) have been used to analyze memory architectures. Since structurally redundant and reconfigurable nanoarchitectures may consist of millions of nanodevices, we have applied state space partitioning techniques and Belief propagation to scale these techniques. We have developed three toolsets based on these Markovian techniques. One of these toolsets has been specifically developed for the architectural exploration of molecular logic systems. The toolset can generate defect maps for isolating defective nanodevices and provide capabilities to organize structurally redundant fault-tolerant architectures with the non-defective devices. Design trade-offs for each of these architectures can be computed in terms of signal delay, area, redundancy and reliability. Another tool called HMAN (Hybrid Memory Analyzer) has been developed for analyzing molecular memory systems. Besides analyzing reliability-redundancy trade-offs using MDPs, HMAN provides a very accurate redundancy-delay trade-off analysis using HSPICE. SETRA (Scalable, Extensible Tool for Reliability Analysis) has been specifically designed for analyzing nanoscale CMOS logic architectures with DTMCs. SETRA also integrates well with current industry-standard CAD tools. It has been shown that multimodal computational models capture the operation of emerging nanoscale devices such as vertical CNT transistors, instead of the bimodal Boolean computational model that has been used to understand the operation of current electronic devices. We have extended an existing multimodal computational model based on Markov Random Fields (MRFs) for analyzing structurally redundant and reconfigurable architectures. Hence, this dissertation develops multiple probabilistic methodologies and tools for performing nanoscale architectural exploration. It also looks at different defect- and fault-tolerant architectures and explores different nanotechnologies.

Browsing by Author "Jacobs, Ira"

Results Per Page

Sort Options