2,708 research outputs found
Development assistance gone wrong : why support services have failed to expand exports
No full textThis study shows that in developing countries with no more than partly favorable policies toward manufactured exports, outside assistance to services that promote and support manufactured exports has had little discernible impact on exports and has rarely been effective in expanding them. The principal reasons for this lack of impact appear to be the after effects of inward-looking development policies, neglect of assistance to enterprises in the production and supply aspects of exporting, insufficient donor concern about the direct impact of their assistance on exports, and reliance on an inappropriate delivery mechanism. Recommendations which suggest new guidelines for donor assistance, project components and new country policies are explained in a companion paper, WPS 544.Economic Theory&Research,Environmental Economics&Policies,Health Economics&Finance,ICT Policy and Strategies,Poverty Assessment
Signal processing and communication with solitons
No full textAlso issued as Thesis (Ph. D.)--Massachusetts Institute of Technology. Dept. of Electrical Engineering and Computer Science, 1996.Includes bibliographical references (p. 137-142).Supported in part by the Defense Advanced Research Projects Agency. N00014-89-J-1489 Supported in part by Lockheed Martin Sanders, and in part by the U.S. Navy Office of the Chief of Naval Research. N00014-93-1-0686 Supported in part by the U.S. Air Force Office of Scientific Research. AFSOR-91-0034Andrew C. Singer
A importância moral da dor e do sofrimento animal na ética de Peter Singer
No full textDissertação (mestrado) - Universidade Federal de Santa Catarina, Centro de Ciências Humanas, Programa de Pós-Graduação em Filosofia, Florianópolis, 2012.O objetivo desta dissertação é defender a importância moral da consideração da dor e do sofrimento de animais não-humanos. Isso se dá através do principio da igual consideração de interesses desenvolvido por Peter Singer. A senciência possibilita os animais a terem interesses, no mínimo, o interesse evitar a dor e o sofrimento. É por essa razão que devem ser incluídos nas decisões morais. São reconstruídas e analisadas as objeções de Peter Harrison, Carl Cohen, R.G. Frey e Lawrence C. Becker direcionadas ao princípio de Singer, e que criticam os pressupostos básicos, quais sejam, a capacidade de sentirem dor/sofrimento e de terem interesses, sobre os quais se fundamenta a inclusão dos animais nas considerações morais. Cada uma dessas objeções é analisada e criticada de modo a demonstrar suas limitações e inconsistências, juntamente com as implicações morais geradas para seres humanos. Na análise dessas críticas, reforça-se a importância e a consideração moral que deve ser conferida à dor e ao sofrimento dos animais. Após essa discussão teórica, é analisado um caso de âmbito prático: a pesquisa científica sobre o câncer humano através do modelo animal. Verifica-se, a partir do princípio de Singer, a imoralidade de tal procedimento realizado em animais sencientes devido à violação de seus interesses. Com isso, a dissertação enfatiza a exigência ética de abolir o uso de animais nessa prática em razão da incapacidade preditiva dos animais, mas principalmente devido à dor e ao sofrimento causado neles e também aos seres humanos, que ficam sujeitos aos erros, prejuízos e sofrimentos originados pelo intenso uso animal nas pesquisas. Nessa conclusão, se constata que a insistência no uso de animais nos experimentos compromete o cientista a preferir usar seres humanos, uma vez que isso gera mais benefícios e resultados mais seguros. A recusa moral ao uso de humanos em pesquisas implica, por outro lado, na recusa moral do uso de animais, ou seja, sua abolição.Abstract : The aim of this dissertation is to defend the moral importance of considering pain and suffering of nonhuman animals. This is achieved through The Principle of Equal Consideration of Interests developed by Peter Singer. The sentience enables nonhuman animals to have interests, at least the interest of avoiding pain and suffering. That is why it should be included in moral decisions. The objections of Peter Harrison, Carl Cohen, RG Frey and Lawrence C. Becker directed to the principle of Singer are reconstructed and analyzed, as they are criticizing the basic assumptions, i.e., the ability to feel pain/suffering and have interests, upon which is based the inclusion of animals in moral considerations. Each of these objections is analyzed and criticized in order to demonstrate their limitations and inconsistencies, simultaneously with its moral implications for humans. In the analysis of these criticisms, it reinforces the moral importance and considerations that should be given to pain and suffering of animals. After this theoretical discussion, a case study of practical scope is analyzed: animal testing for scientific research on human cancer. It is verified from the Singer's principle that such procedures performed on sentient animals are a violation of their interests and, therefore, immoral. Thus, the dissertation emphasizes the ethical demand to abolish the use of nonhuman animals in this practice due to their predictive inability, but mainly due to the pain and suffering caused to them and also to humans, who are subject to errors, injuries and suffering originated by the intense use of nonhuman animals on research. The conclusion verifies that the insistence on the use of nonhuman animals in experiments moves the scientist to prefer using humans in experiments since it generates greater benefit and more reliable results. The moral refusal to using humans in research implies the moral rejection of the use of animals in experiments and consequently, its abolition
Noncontact heart rate measurement via period detection methods of periodic signals
No full textConventional heart rate measurement techniques, including manual measurement by placing two fingers on the wrist or side of the neck, electrocardiogram (ECG) or pulse oximetry devices require direct physical contact to the subject and this has the potential to cause inconveniences in cases when the subject has skin irritations or severe burns, contagious disease or intensive care conditions. Hence, there is a need for noncontact heart rate measurement. Previous research on noncontact heart rate measurement utilizes either a complex hardware setup, such as laser illuminators or high-speed cameras, or expensive computational methods, such as independent component analysis. The aim of this work is to apply different signal processing methods utilized in pitch detection problems of periodic signals, namely zero-crossings, autocorrelation, maximum likelihood and Fourier-based methods, to the signal obtained from the average pixel values of the frames of a video recording of the subject’s face obtained using a standard webcam in order to find simple yet effective methods and compare them to previous works, under the same stability assumptions of the subject and illumination of the environment. For this purpose, first, the noncontact heart rate measurement problem is posed as a sinusoidal parameter estimation problem in order to analyze it from a sinusoidal parameter estimation problem point of view. Then, nonparametric methods that are used in period detection of periodic signal problems is presented and their performances are compared with each other as well as previous research. Algorithms are tested on both synthetically generated data and data obtained from input video files. Findings indicate that under stability conditions, inexpensive pitch detection techniques perform almost as well or better compared to computationally expensive methods. Finally, a different approach is taken from the oscillator devices perspective, and an oscillator circuit is used in a similar fashion to an injection locking problem in order to find the period of the input signal from the output of the oscillator circuit. In cases where the circuit parameters such as resistor and capacitor values are tuned accordingly based on the input signal, it is possible to extract heart rate information from the falling edge occurrences of output oscillator voltage with high accuracy.Submission published under a 24 month embargo labeled 'Closed Access', the embargo will last until 2018-12-01The student, Gizem Tabak, accepted the attached license on 2016-12-07 at 13:25.The student, Gizem Tabak, submitted this Thesis for approval on 2016-12-07 at 13:33.This Thesis was approved for publication on 2016-12-07 at 16:04.DSpace SAF Submission Ingestion Package generated from Vireo submission #10472 on 2017-02-28 at 14:43:22Made available in DSpace on 2017-03-01T17:02:07Z (GMT). No. of bitstreams: 2
TABAK-THESIS-2016.pdf: 11170291 bytes, checksum: 0e6af9f53532abee055763949c60e79c (MD5)
LICENSE.txt: 4208 bytes, checksum: f992913bdf69dd97aa050c9e1d228741 (MD5)
Previous issue date: 2016-12-07Embargo set by: Seth Robbins for item 98737
Lift date: 2019-03-01T17:02:22Z
Reason: Author requested closed access (OA after 2yrs) in Vireo ETD systemEmbargo set by: Seth Robbins for item 98737
Lift date: 2019-03-01T17:03:32Z
Reason: Author requested closed access (OA after 2yrs) in Vireo ETD systemEmbargo set by: Seth Robbins for item 98737
Lift date: 2019-03-01T17:05:02Z
Reason: Author requested closed access (OA after 2yrs) in Vireo ETD systemEmbargo set by: Seth Robbins for item 98737
Lift date: 2019-03-01T17:06:55Z
Reason: Author requested closed access (OA after 2yrs) in Vireo ETD systemLimited Restriction Lifted for Item 98737 on 2019-03-02T10:15:14Z
Message passing algorithms - methods and applications
No full textAlgorithms on graphs are used extensively in many applications and research areas. Such applications include machine learning, artificial intelligence, communications, image processing, state tracking, sensor networks, sensor fusion, distributed cooperative estimation, and distributed computation. Among the types of algorithms that employ some kind of message passing over the connections in a graph, the work in this dissertation will consider belief propagation and gossip consensus algorithms.
We begin by considering the marginalization problem on factor graphs, which is often solved or approximated with Sum-Product belief propagation (BP) over the edges of the factor graph. For the case of sensor networks, where the conservation of energy is of critical importance and communication overhead can quickly drain this valuable resource, we present techniques for specifically addressing the needs of this low power scenario. We create a number of alternatives to Sum-Product BP. The first of these is a generalization of Stochastic BP with reduced setup time. We then present Projected BP, where a subset of elements from each message is transmitted between nodes, and computational savings are realized in proportion to the reduction in size of the transmitted messages. Zoom BP is a derivative of Projected BP that focuses particularly on utilizing low bandwidth discrete channels. We give the results of experiments that show the practical advantages of our alternatives to Sum-Product BP.
We then proceed with an application of Sum-Product BP in sequential investment. We combine various insights from universal portfolios research in order to construct more sophisticated algorithms that take into account transaction costs. In particular, we use the insights of Blum and Kalai's transaction costs algorithm to take these costs into account in Cover and Ordentlich's side information portfolio and Kozat and Singer's switching portfolio. This involves carefully designing a set of causal portfolio strategies and computing a convex combination of these according to a carefully designed distribution. Universal (sublinear regret) performance bounds for each of these portfolios show that the algorithms asymptotically achieve the wealth of the best strategy from the corresponding portfolio strategy set, to first order in the exponent. The Sum-Product algorithm on factor graph representations of the universal investment algorithms provides computationally tractable approximations to the investment strategies. Finally, we present results of simulations of our algorithms and compare them to other portfolios.
We then turn our attention to gossip consensus and distributed estimation algorithms. Specifically, we consider the problem of estimating the parameters in a model of an agent's observations when it is known that the population as a whole is partitioned into a number of subpopulations, each of which has model parameters that are common among the member agents. We develop a method for determining the beneficial communication links in the network, which involves maintaining non-cooperative parameter estimates at each agent, and the distance of this estimate is compared with those of the neighbors to determine time-varying connectivity. We also study the expected squared estimation error of our algorithm, showing that estimates are asymptotically as good as centralized estimation, and we study the short term error convergence behavior.
Finally, we examine the metrics used to guide the design of data converters in the setting of digital communications. The usual analog to digital converters (ADC) performance metrics---effective number of bits (ENOB), total harmonic distortion (THD), signal to noise and distortion ratio (SNDR), and spurious free dynamic range (SFDR)---are all focused on the faithful reproduction of observed waveforms, which is not of fundamental concern if the data converter is to be used in a digital communications system. Therefore, we propose other information-centric rather than waveform-centric metrics that are better aligned with the goal of communications. We provide computational methods for calculating the values of these metrics, some of which are derived from Sum-Product BP or related algorithms. We also propose Statistics Gathering Converters (SGCs), which represent a change in perspective on data conversion for communications applications away from signal representation and towards the collection of relevant statistics for the purposes of decision making and detection. We show how to develop algorithms for the detection of transmitted data when the transmitted signal is received by an SGC. Finally, we provide evidence for the benefits of using system-level metrics and statistics gathering converters in communications applications.Submission original under an indefinite embargo labeled 'Open Access'. The submission was exported from vireo on 2015-07-22 without embargo termsThe student, Andrew Bean, accepted the attached license on 2015-04-16 at 08:50.The student, Andrew Bean, submitted this Dissertation for approval on 2015-04-16 at 09:34.This Dissertation was approved for publication on 2015-04-21 at 14:33.DSpace SAF Submission Ingestion Package generated from Vireo submission #7884 on 2015-07-22 at 10:32:30Made available in DSpace on 2015-07-22T22:16:55Z (GMT). No. of bitstreams: 3
BEAN-DISSERTATION-2015.pdf: 4507112 bytes, checksum: 0db1bf3ae36ae121eb22f165d091ba72 (MD5)
Dissertation.zip: 5256676 bytes, checksum: 3de119548a683895ae0d6f8254743454 (MD5)
LICENSE.txt: 4208 bytes, checksum: e7eed6f007eda280071015cc73238761 (MD5)
Previous issue date: 2015-04-2
Determination of critical cooling rates in metallic glass forming alloy libraries through laser spike annealing
No full textThe glass forming ability (GFA) of metallic glasses (MGs) is quantified by the critical cooling rate (RC). Despite its key role in MG research, experimental challenges have limited measured RC to a minute fraction of known glass formers. We present a combinatorial approach to directly measure RC for large compositional ranges. This is realized through the use of compositionally-graded alloy libraries, which were photo-thermally heated by scanning laser spike annealing of an absorbing layer, then melted and cooled at various rates. Coupled with X-ray diffraction mapping, GFA is determined from direct RC measurements. We exemplify this technique for the Au-Cu-Si system, where we identify Au56Cu27Si17 as the alloy with the highest GFA. In general, this method enables measurements of RC over large compositional areas, which is powerful for materials discovery and, when correlating with chemistry and other properties, for a deeper understanding of MG formation.Peer reviewe
Obtaining Thickness-Limited Electrospray Deposition for 3D coating
No full textThe electrospray process utilizes the balance of electrostatic forces and surface tension within a charged spray to produce charged microdroplets with a narrow dispersion in size. In electrospray deposition, each droplet carries a small quantity of suspended material to a target substrate. Past electrospray deposition results fall into two major categories: (1) continuous spray of films onto conducting substrates and (2) spray of isolated droplets onto insulating substrates. A cross-over regime, or a self-limited spray, has only been limitedly observed in the spray of insulating materials onto conductive substrates. In such sprays, a limiting thickness emerges where the accumulation of charge repels further spray. In this study, we examined the parametric spray of several glassy polymers to both categorize past electrospray deposition results and uncover the critical parameters for thickness-limited sprays. The key parameters for determining the limiting thickness were (1) field strength and (2) the spray temperature, related to (1) the necessary repulsive field and (2) the ability for the deposited materials to swell in the carrier solvent vapor and redistribute charge. These control mechanisms can be applied to the uniform or controllably varied microscale coating of complex 3D objects.Peer reviewe
Signal processing and communication with solitons
No full textThesis (Ph. D.)--Massachusetts Institute of Technology, Dept. of Electrical Engineering and Computer Science, 1996.Includes bibliographical references (p. 137-142).by Andrew Carl Singer.Ph.D
An MLSE receiver for electronic dispersion compensation of OC-192 fiber links
No full textA maximum-likelihood sequence estimation (MLSE) receiver is fabricated to combat dispersionlintersymbol interference (chromatic and polarization mode), noise (optical and electrical), and nonlinearities (e.g., fiber, receiver photodiode, or laser) in OC-192 metro and long-haul links. The MLSE receiver includes a variable gain amplifier with 40-dB gain range and 7.5-GHz 3-dB bandwidth, a 12.5-Gb/s 4-bit analog-to-digital converter, a dispersion-tolerant phase-locked loop, a 1:8 demultiplexer, and a digital 'equalizer implementing the MLSE algorithm. The MLSE receiver achieves more than 50% reach extension at signal-to-noise levels of interest as compared to conventional clock data recovery systems
Minimum-delay HF communications
No full textHF radio communications offer latency advantages over fiber-optics and are more economical than satellite links in long-distance communication applications. As a result, there is renewed interest in the field of HF from the high-frequency trading industry, as a means of transmitting financial data. Ideally, decisions should be made on received symbols as soon as they arrive, without waiting for other samples to accumulate. A significant problem is that the ionospheric HF channel is a dynamic medium that is considered one of the most challenging communication channels. In this thesis, we investigate alternative techniques towards reliable communication with minimum delay. We study HF channel models and communication link design tools, which indicate that our minimum-delay problem is feasible but difficult to solve. Extant HF modems are observed to add prohibitive amounts of latency to ensure robustness, motivating us to design receivers that provide good performance given a delay constraint. We adapt a coupled MAP-RLS receiver studied in literature to the HF channel to yield minimal decision-making delay. We introduce a multitrellis adaptive Viterbi algorithm (MAVA) that solves the problem of equalization for sparse and time-varying ISI channels, producing a robust MLSE receiver with several milliseconds of decision-making delay. Finally, we cascade this MLSE receiver with MAP detection to obtain high-fidelity low-delay performance, finding a practical solution to minimum-delay HF communications.Submission original under an indefinite embargo labeled 'Open Access'. The submission was exported from vireo on 2017-08-10 without embargo termsThe student, Toros Arikan, accepted the attached license on 2017-04-27 at 13:14.The student, Toros Arikan, submitted this Thesis for approval on 2017-04-27 at 15:32.This Thesis was approved for publication on 2017-04-27 at 17:44.DSpace SAF Submission Ingestion Package generated from Vireo submission #11116 on 2017-08-10 at 13:46:56Made available in DSpace on 2017-08-10T19:16:20Z (GMT). No. of bitstreams: 27
ARIKAN-THESIS-2017.pdf: 960089 bytes, checksum: c95d0f89bba7cc60a362e7c010f29ee8 (MD5)
Comm_system.jpg: 13917 bytes, checksum: 20c42fc07f27a30936e54e37b5df18a7 (MD5)
abs.tex: 1525 bytes, checksum: c0722b70b4f0acfaccecdac39bbeaa22 (MD5)
ack.tex: 518 bytes, checksum: 5141db8dfdb312ad71f5c5866ce2091c (MD5)
block_diagram_map_rls.png: 89591 bytes, checksum: 8098122a90faabab2141a472ffff8b50 (MD5)
block_diagram_mlse.png: 46635 bytes, checksum: f1a502b62c42a6dfe02d2f019af28ff2 (MD5)
block_diagram_mlse_map.png: 67058 bytes, checksum: 3c4436731971ce6491c356c78397e281 (MD5)
bpsk_vs_qpsk_mlse_1200.png: 42772 bytes, checksum: 8790f6db0279ec3d1bc2e284ca5f71cc (MD5)
channel_tracking_performance_corr_rls.png: 38506 bytes, checksum: fb4f99c5e41e7a2298cf69d9dab6658e (MD5)
channel_tracking_std_rls_longview.png: 38464 bytes, checksum: 352bc74f3f385e29bd2938fd1fd903b9 (MD5)
chapter-1-introduction.tex: 26329 bytes, checksum: 4117ba9fb45f1fb1c45109ea02826a3f (MD5)
chapter-2-minimum-delay-receiver-designs.tex: 40297 bytes, checksum: 279844c404f1c9e91f1b6fa587c7eaa9 (MD5)
chapter-3-results.tex: 20152 bytes, checksum: 0a0261f0d88ec1f67d71124ba4dd7a10 (MD5)
chapter-4-conclusions-and-future-work.tex: 9505 bytes, checksum: 1777535e22b64acec4e97189e166068b (MD5)
combined_vs_rls_1200.png: 40509 bytes, checksum: 673c11aa5bc62dfe520eeae615ba943f (MD5)
ecethesis.tex: 4976 bytes, checksum: b81155b338edbb9cbcb76b3e58478712 (MD5)
mlse_vs_map_1200.png: 32224 bytes, checksum: 87e34ccce030e13c6f10179900f6f9e4 (MD5)
multitrellis.png: 29351 bytes, checksum: 9fa83cd493dc9470626d532f35d7f17d (MD5)
perfect_vs_combined_1200.png: 37884 bytes, checksum: 8919de2316f3a7efefde7ccc331b2d5b (MD5)
perfect_vs_delayed_vs_rls_1200.png: 34340 bytes, checksum: 1d4fc124c0745f1488ecc0ca2c084ed4 (MD5)
perfect_vs_rls_1200.png: 42578 bytes, checksum: 301acc46063653e6479ea0f71a02fd99 (MD5)
rls_vs_rls_corr_1200.png: 26660 bytes, checksum: 64e5d7fc1d1269bfef312b1529ea88f5 (MD5)
thesisrefs.bib: 54659 bytes, checksum: 1ba3bbebbe2c525fa90b7b46c3293464 (MD5)
uiucecethesis09_2.cls: 22162 bytes, checksum: f7b508a2389c889fd082d4b2526767e7 (MD5)
voacap_dec_2011.png: 89310 bytes, checksum: 544770f4cca937f751945e4da9f0556e (MD5)
voacap_jan_2011.png: 70017 bytes, checksum: 38505cadbb6516d51995d263c3a47b23 (MD5)
LICENSE.txt: 4209 bytes, checksum: ff829f28ddebfbcf685ff67afafee948 (MD5)
Previous issue date: 2017-04-2
- …
