Senior Theses - Electrical and Computer Engineering
http://hdl.handle.net/2142/30795
The best of ECE undergraduate researchFri, 27 Nov 2015 19:10:25 GMT2015-11-27T19:10:25ZLoss in Vertical-Cavity Source-Emitting Lasers as a Result of Impurities
http://hdl.handle.net/2142/88382
Loss in Vertical-Cavity Source-Emitting Lasers as a Result of Impurities
Tai, Charlene
Vertical-cavity source emitting lasers (VCSELs) are useful for optical communication light sources because of their low cost and operating power. An understanding of the cavity optical loss is necessary in order to best optimize the design and performance of the devices. Using a method which measures the sub-threshold emission spectra, the loss can be quantitatively calculated. With this, optical loss of VCSELs with varying doping levels can be evaluated. With a semiconductor parameter analyzer, light output power as a function of injection current is measured for each different set of VCSELs to find the threshold current. The resulting current density can be found for each laser mesa size and the point where loss becomes independent of the aperture size can be identified. Using and optical spectrum analyzer, the spectral separation between the cold-cavity fundamental mode and the first higher-order mode can then be found without thermal effects by measuring the spectral splitting for each VCSEL biased at an injection current of approximately 0.9 times the threshold. The spectral splitting of small diameter lasers can be used to calculate the optical loss using the Helmholtz wave equation with an imaginary refractive index. The field amplitude loss can be extracted from the imaginary part of the resulting wavenumber. Two VCSEL samples with different doping are measured in this study. The calculated size-dependent loss is found to be the same for the two samples and thus it seems that that effects of the different doping are not sufficient to affect the cavity loss. It is, however, clearly shown that the introduction of impurities to create a doped pn junction does indeed create optical loss.
vertical cavity surface emitting laser; semiconductor laser; optical loss; oxide-confined micro-cavity
Mon, 01 Dec 2014 00:00:00 GMThttp://hdl.handle.net/2142/883822014-12-01T00:00:00ZTai, CharleneNeutral wind field model estimation and optimal layout determination
http://hdl.handle.net/2142/88381
Neutral wind field model estimation and optimal layout determination
Yue, Chenshuo
The project introduces three different but related processes for neutral wind field estimation and optimization of sensor layout for a network of Fabry–Pérot interferometers, which are instruments for collecting wind speed and temperature of thermospheric wind fields. A mathematical model is presented to estimate the thermospheric neutral wind field. Algorithms are introduced to find the optimal layout for neutral wind field estimation without assuming anything about the original wind field. The fundamental theory behind the algorithm design is based in linear algebra. Further exploration of the optimal layout for simulation of the neutral wind field leads to the development of an alternative method. The fundamental principle underlying this new method comes from advanced linear algebra, regularization, which is a process to solve ill-conditioned or singular linear systems. The quadratic assumption of the neutral wind field form shows great performance if the wind field is very simple. The regularization method to simulate the neutral wind has obvious advantages if the neutral wind field is in different form, since the regularization method could be used to simulate the wind field with undetermined form. Also, the regularization method could be adjusted to simulate the wind field with various preferences, such as smoothness. The test on North American Thermosphere-Ionosphere Observation Network, a network of 5 FPIs in United States, demonstrated the performance of the optimal layout we currently generated by greedy algorithm. In general, the optimal layout shows a pretty good noise bearing ability. Also, with fewer measurements collected by the FPIs, we are still able to get a simulated neutral wind containing a lot of useful information.
thermospheric wind field simulation; sensor layout optimization
Mon, 01 Dec 2014 00:00:00 GMThttp://hdl.handle.net/2142/883812014-12-01T00:00:00ZYue, ChenshuoInvestigation on Double Negative Metamaterials
http://hdl.handle.net/2142/88380
Investigation on Double Negative Metamaterials
To, Oscar
Metamaterials, materials that exhibit electromagnetic properties that are not naturally achievable, are currently a hot topic in research. Of the domain of metamaterials, major research efforts have focused on double negative metamaterials (DNM), materials that demonstrate a negative index of refraction, due to their potential applications such as superlens and cloaking. Moreover, passive metamaterials--materials that do not require additional energy to exhibit exotic properties--have demonstrated qualities beneficial as a DNM. Due to the nature of DNM, the atypical electromagnetic properties can only be seen over a finite range of frequencies. Current experimental realizations of DNM have been demonstrated in infrared wavelengths and radio frequency to millimeter waves; there have been no DNMs experimentally validated over the visual frequency spectrum. In this thesis, we propose a design of a DNM that displays negative refraction at the visual spectrum and validate the design experimentally. The design utilizes a layer of close-packed, precisely size-controlled, uniform nanospheres deposited on a layer of thin metallic film. We model the reflectance and transmittance of the design through the 3D finite-difference time-domain (FDTD) and 3D discontinuous Galerkin time-domain (DGTD) method. The theory behind our design and the results of the simulations and experiments will be discussed here.
metamaterial; nanospheres; FDTA; DGTD; double negative metamaterial
Mon, 01 Dec 2014 00:00:00 GMThttp://hdl.handle.net/2142/883802014-12-01T00:00:00ZTo, OscarMethod of Moments (MoM): Application for Solving Augmented Electric Field Integral Equation (AEFIE)
http://hdl.handle.net/2142/88379
Method of Moments (MoM): Application for Solving Augmented Electric Field Integral Equation (AEFIE)
Li, Binye
Surface integral equations (SIEs) are promising candidates for modeling circuits because they
reduce degrees of freedom by restricting physical unknowns on the surface, which simplifies
complex structures. However, there are still challenges related to achieving stability over a broad
frequency band. Specifically, the low frequency breakdown of electrical field integral equation
(EFIE) operator is discussed in this work. In order to solve or alleviate this problem, the
separation of irrotational and solenoidal current must be accomplished. A proposed method, the
Augmented Electrical Field Integral Equation (AEFIE), is intended to separate the current
element by introducing charge as another variable and relate irrotational current and the charge
vector. Finally, the method of moments (MoM) is applied to solve the integral equation by
projecting the current onto RWG basis and performing subspace projections to fill out the
integral equation operator matrix. For complicated circuit structure, MoM can be accelerated
using the fast multipole algorithm (FMA).
Computational Electromagnetics; EFIE; MoM
Mon, 01 Dec 2014 00:00:00 GMThttp://hdl.handle.net/2142/883792014-12-01T00:00:00ZLi, BinyeCharacterization of Dynamic Solar Insulation for Photovoltaic Systems
http://hdl.handle.net/2142/88377
Characterization of Dynamic Solar Insulation for Photovoltaic Systems
Serna, Rodrigo
This thesis describes the implementation of an experimental setup to
capture the maximum power point data of photovoltaic (PV) modules. The
experimental setup operated for a one and a half year period, and the data
collected is analyzed in this paper to determine if it is possible to maximize
power conversion by varying the step and interval parameters of maximum
power point tracking (MPPT) algorithms for each day. For each day, short
circuit current data is collected at a high frequency (2 kHz) and full sweeps
of the PV IV curve is done about every three seconds. The large amounts of
data are broken down and each sweep is parameterized and then the perturb
and observe algorithm is run over the data for different interval and step
parameter sets. The conclusion is that it is possible to maximize the power
conversion over the whole year, though further analysis is currently needed to
ascertain to what extent this is possible.
Photovoltaics; Maximum Power Point Tracking; Optimization
Mon, 01 Dec 2014 00:00:00 GMThttp://hdl.handle.net/2142/883772014-12-01T00:00:00ZSerna, RodrigoDesigning Radio Frequency Integrated Circuits (RFICs) Based on Low Loss Chip-Scale Lumped Elements
http://hdl.handle.net/2142/88376
Designing Radio Frequency Integrated Circuits (RFICs) Based on Low Loss Chip-Scale Lumped Elements
Li, Zijian
Over the years the thirst for achieving better filter designs with low insertion loss in RFICs has gradually become unquenchable. Today, cell phones manufactured from different brands all require distinct frequency ranges of the transmitted signal. In order to select the wanted frequencies within a certain range while efficiently rejecting and attenuating the unwanted frequencies outside that range, a band-pass filter is needed. It is imperative to choose the correct prototype of band-pass filter when designing. Typical band-pass filters can be classified into four groups: Butterworth, Chebyshev, Elliptic and Bessel. In this paper, Butterworth filter prototype serves as the main design technique.
This thesis revolves around the following general questions that concern the testing, modeling and designing of the band-pass filter:
1. What order of Butterworth filter should be chosen to minimize the insertion loss?
2. What is the difference between ideal reactive components and non-ideal reactive components at high frequencies?
3. Is there an internal relation between Quality Factor, pass band bandwidth and insertion loss? Can this be derived into a closed form equation?
4. What changes should be made when switching from the theoretical simulation to the hands-on PCB board fabrication?
5. Lastly, is there any mismatch between simulation and real measurement? If yes, why are there mismatches and how can they be fixed?
RF/microwave filter; insertion loss; Butterfield filter
Mon, 01 Dec 2014 00:00:00 GMThttp://hdl.handle.net/2142/883762014-12-01T00:00:00ZLi, ZijianDesigning Microstrip Band Filter at 2.4 GHz
http://hdl.handle.net/2142/88375
Designing Microstrip Band Filter at 2.4 GHz
Kim, Doyoun
Bandpass filter’s role in the systems of wireless communication has been highlighted in
current days. The filtration of the signals that are being transferred and received has to take place
at a specific frequency that contains attenuation of a certain amount and a specified bandwidth.
The initial stage of designing a microstrip filter is the performance of the approximated
calculation. This in particular has to be based on the utilization of lumped essentials, for instance
capacitors and inductors. It is through Agilent Advanced Design System that is centered on
specific parameter calculation that the performance of filter computational verification is done.
The printed circuit board is utilized in the fabrication of the transformation of the filter structure
that contains a quarter-wave. The major objective of this research was to use the theoretical and
designed filter experimental values to determine the closeness between the fabricated
measurement and theoretical simulation.
fabrication of printed circuit board; microstrip containing quarter wave transformation and Bandpass filter
Mon, 01 Dec 2014 00:00:00 GMThttp://hdl.handle.net/2142/883752014-12-01T00:00:00ZKim, DoyounA Dynamic Spectral Light Modulation Unit with Optical Feedback Control for Optogenetic Stimulation
http://hdl.handle.net/2142/88361
A Dynamic Spectral Light Modulation Unit with Optical Feedback Control for Optogenetic Stimulation
Yu, Haichuan
With recent advances in optical stimulation delivery to optogenetically modified neurons, individual or groups of neurons can be targeted with incredible spatio-temporal accuracy. It is enough to excite the functional neuronal electrophysiology in its native scale, making possible the study of functional neural circuitry. However, excitation of multiple neurons and distinct combinations of neurons within a neuronal circuit with laser optical stimulation has not yet been demonstrated. This work presents an optical feedback control unit that fills this need. The unit synthesizes user input and optical information in a continuous feedback loop to develop a control mask for the laser excitation. The resulting unit can perform complex real-time patterning of laser stimulation, and can be added on to experiments in neuroscience with some adaptation to native equipment.
optogenetics, neuron, optical, spatial, temporal, feedback, control, light, sculpting, patterning, SLM
Thu, 01 May 2014 00:00:00 GMThttp://hdl.handle.net/2142/883612014-05-01T00:00:00ZYu, HaichuanCharacterization of Etching Process and Etch Rate of Aluminum Thin Film
http://hdl.handle.net/2142/87845
Characterization of Etching Process and Etch Rate of Aluminum Thin Film
Kang, "Ryan" Sungho
This thesis reports on characterization of etching process of aluminum thin film.
It offers sequential steps of optimized etching process as well as the etch rate of
aluminum thin film--one of the ubiquitous metals in MEMS and IC industries|
for on-going RF MEMS research. To gain high device performance, an effi cient
and accurate fabrication process is critical. Despite a number of previous studies
on etch rate characterization, unique tuning methods and fabrication recipes are
required for different instruments and laboratories. In this report, the etching
characterization is performed using PlasmaLab Master/Slave Reactive Ion Etching (RIE) System with
flows of boron trichloride (BCl3) and chlorine gas (Cl2).
Drawbacks of the tools and chemicals as well as corresponding adjustments are
also discussed. Along with aluminum etch rate, etch rate of photoresist under the
same parameters can also be evaluated in the same manner. The results of both
etch rates are covered in this paper.
micro-electro-mechanical systems (MEMS); RF MEMS; micro-fabrication etching; lithography; physical vapor disposition (PVD)
Mon, 01 Dec 2014 00:00:00 GMThttp://hdl.handle.net/2142/878452014-12-01T00:00:00ZKang, "Ryan" SunghoQuantization Error Tolerance in Hashed Audio Spectra
http://hdl.handle.net/2142/87844
Quantization Error Tolerance in Hashed Audio Spectra
Sivaraman, Aswin
Matching an input spectrum with a learned dictionary of spectral frames is
common in audio signal processing, especially for speech de-noising and one-word
speech recognition. For large disordered spectral dictionaries,
exhaustively
searching for nearest-neighbor spectra is computationally expensive.
The proposed methodology utilizes hierarchical clustering of winner-take-all
(WTA) semantic hashes of the spectral frames in the dictionary. We define
a custom Hamming distance metric between hash codes that is analogous
to the original error (cross entropy). After clustering the training data, we
evaluate the functionality of this framework by assessing signal-to-noise ratio (SNR) for test signal reconstruction, exploring the quantization effects
of truncating the hierarchical clustering tree (dendogram). By defining a
tolerance level for noise, we seek to considerably reduce the search space for
spectral frames and significantly improve spectrogram-matching speed. An
extended application of this work is reduced power consumption for active
listening devices ("Hey Siri", "Ok Google", etc.), as well as increased transmission
quality without forsaking device talktime. The proposed framework
proved to be sub par, but possible improvements to this research are
discussed.
Locality sensitive hashing; winner take all hashing; source reconstruction; quantization; hierarchical clustering; dendogram; k-means; k-medoids
Fri, 01 May 2015 00:00:00 GMThttp://hdl.handle.net/2142/878442015-05-01T00:00:00ZSivaraman, Aswin