Content-based MIDI Music Retrieval and Computer-aided Composition Based on Musical Whistling Hung-Cche Shen* Chung-Nan Lee+ ABSTRACT In this dissertation, we have focused on the research of content-based MIDI music retrieval and computer-aided composition based on musical whistling. For MIDI music retrieval, a prototype system called ¡§Whistle-for-Music¡¨ is developed. This system enables users to retrieve MIDI format music by whistling a melodic fragment. It consists of three essential components which are query processing, MIDI preprocessing and an approximate search engine. For query processing, we have achieved a real-time and robust whistle-to-MIDI converter. For MIDI preprocessing, the extracted features from MIDI files contain individual, local and global melodic descriptions. In order to match a querying pattern with target, we extend an existing search engine into a fast approximate melodic matching engine. There is a systematic evaluation of ¡§Whistle-for-Music¡¨ system. The performed experiments include ¡§Query turnaround time¡¨, ¡§Unique prefix searching¡¨, ¡§Error density vs. match rank percent¡¨ and ¡§Statistics of global descriptions¡¨. The results show that careful measurement and objective comparisons can lead us to know the scaling trend about query and target. For computer-aided composition, a system called ¡§Whistle-to- Music¡¨ is presented. The ¡§Whistle-to-Music¡¨ system can ease the melody input and musical composition. Firstly, our ¡§Whistle-to-MIDI¡¨ transformation is to translate a whistled tune into a sequence of notes, which are defined by onset, duration, velocity and note pitch. This ¡§Whistle-to-MIDI¡¨ is a relative melody transcription since we apply ¡§pitch templates¡¨ and ¡§rhythm templates¡¨. Based on the given melodic motives, we propose the templates of melodic forms and harmonic tables for the tasks of motive development and automatic harmonization. We demonstrate that the proposed melodic forms templates can successfully produce various formal songs from one or few bass motives. The harmonic tables allow us to produce classic-style harmonization based on three theories. The process of ¡§Whistle-to-Music¡¨ system is described by giving examples of melody transcription, motive development and demonstrating the resulting music from automatic harmonization. In automatic harmonization, we demonstrate that the proposed harmonic tables can successfully produce a piece of well-formed harmonization from a variety of whistled tunes. * Author +Advisor

With the rapidly development of electronic commerce, the competitions between enterprises have been getting more and more turbulent. In order to win business in such a highly competitive environment, it has to not only implement customized marketing strategy but also have a timely response whenever there is any change in the business environment. Currently, in most companies, whenever the marketing people wants to implement a new sales promotion program, he has to communicate it with the MIS department and ask it to rewrite the Web system and pages. It is quite often a very time-consuming as well as strenuous process. In this research, we have explored all popular and important sales promotion models. Based on these models, an object-oriented Web system has been designed and developed. There is an interface by which the marketing people can set the new sales promotion program and then, the Web system can execute it immediately afterwards. The interface, allows the marketing people easily to select target customers, conditions for getting bonus and contents of bonus to set a new sales promotion program. That is, the marketing people can let the Web system execute a new sales promotion program whenever he wants without the help of system programmers. In addition, the whole design is object-oriented. It lets the system have very high portability.

Saltcedar is an invasive shrub that has moved into western United States riparian areas and is continuing to spread. Saltcedar is a phreatophyte that can utilize a saturated water table for moisture once established and is also highly tolerant of saline soil and water conditions. Literature has indicated that depth to the water table and salinity have a significant effect on growth and water use by saltcedar. Several studies were initiated to help develop a simulation model of saltcedar growth and water use based on the EPIC9200 simulation model. A study was initiated at the USDA-ARS Blackland Research Center Temple, Texas in the summer of 2002 to better understand the effects of water table depth and salinity on (1) saltcedar above and below ground biomass, root distribution, leaf area and (2) water use. Five different salinity levels (ranging from 0 ppm to 7500 ppm) and three different water table depths (0.5m, 1.0m, and 1.75m) were studied. Results indicated that increasing depth to the water table decreased saltcedar water use and growth. For the 0.5m water table depth, saltcedar water use during the 2002 growing season averaged 92.7 ml d-1 while the 1.75m depth averaged 56.6 ml d-1. Both root and shoot growth were depressed by increasing water table depth. Salinity had no effect on saltcedar growth or water use except at the 1250 ppm level, which used 110 ml of H2O d-1. This salinity had the highest water use indicating that this may be near the ecological optimum level of salinity for saltcedar. A predictive equation was developed for saltcedar water use using climatic data for that day, the previous day's climatic data, water table depth and salinity that included: previous day total amount of solar radiation, water table depth, previous day average wind speed, salinity, previous day total precipitation, previous day average vapor pressure, minimum relative humidity, previous day average wind direction, and maximum air temperature. Data from the field study and a potential growth study were integrated into the model. The model was parameterized for the Pecos River near Mentone, Texas. Predicted saltcedar water use was slightly lower than results reported by White et al. 2003.

A circuit design of Fast Fourier Transform for DVB-H system is presented in this thesis. This circuit is based on SDF (single path delay feedback) pipeline architecture with radix-2 computation element. We propose a novel method of timing scheduling that can share one complex multiplier for couple of stage and promote the utilization of complex multiplier to 100%, so we can improve the implementation with radix-2 computation. The number of bits is carefully selected by system simulation to meetthe requirements of DVB-H system. In addition, a memory table permutation deletion method for memory scheduling, which can reduce the size of memory storing twiddle factors tables. The circuit is carried out by CMOS 0.18£gm 1P6M process with core area 2.08 x 2.076 mm2. In the gate level simulation, the output data rate of this circuit is above 50MHz, so the circuit can meet the requirement of DVB-H system.

<p>Denna avhandling behandlar analog-digitalomvandling. I synnerhet behandlas postkorrektion av analog-digitalomvandlare (A/D-omvandlare). A/D-omvandlare är i praktiken behäftade med vissa fel som i sin tur ger upphov till distorsion i omvandlarens utsignal. Om felen har ett systematiskt samband med utsignalen kan de avhjälpas genom att korrigera utsignalen i efterhand. Detta verk behandlar den form av postkorrektion som implementeras med hjälp av en tabell ur vilken korrektionsvärden hämtas.</p><p>Innan en A/D-omvandlare kan korrigeras måste felen i den mätas upp. Detta görs genom att estimera omvandlarens överföringsfunktion. I detta arbete behandlas speciellt problemet att skatta kvantiseringsintervallens mittpunkter. Det antas härvid att en referenssignal finns tillgänglig som grund för skattningen. En skattare som baseras på sorterade data visas vara bättre än den vanligtvis använda skattaren baserad på sampelmedelvärde.</p><p>Nästa huvudbidrag visar hur resultatet efter korrigering av en A/D-omvandlare kan predikteras. Omvandlaren antas här ha en viss differentiell olinjäritet och insignalen antas påverkad av ett slumpmässigt brus. Ett postkorrektionssystem, implementerat med begränsad precision, korrigerar utsignalen från A/D-omvandlaren. Ett utryck härleds som beskriver signal-brusförhållandet efter postkorrektion. Förhållandet visar sig bero på den differentiella olinjäritetens varians, det slumpmässiga brusets varians, omvandlarens upplösning samt precisionen med vilken korrektionstermerna beskrivs.</p><p>Till sist behandlas indexering av korrektionstabeller. Valet av metod för att indexera en korrektionstabell påverkar såväl tabellens storlek som förmågan att beskriva och korrigera dynamiska fel. I avhandlingen behandlas i synnerhet tillståndsmodellbaserade metoder, det vill säga metoder där tabellindex bildas som en funktion utav flera på varandra följande sampel. Allmänt gäller att ju fler sampel som används för att bilda ett tabellindex, desto större blir tabellen, samtidigt som förmågan att beskriva dynamiska fel ökar. En indexeringsmetod som endast använder en delmängd av bitarna i varje sampel föreslås här. Vidare så påvisas hur valet av indexeringsbitar kan göras optimalt, och experimentella utvärderingar åskådliggör att tabellstorleken kan reduceras avsevärt utan att fördenskull minska prestanda mer än marginellt.</p><p>De teorier och resultat som framförs här har utvärderats med experimentella A/D-omvandlardata eller genom datorsimuleringar.</p> / <p>Analog-to-digital conversion and quantization constitute the topic of this thesis. Post-correction of analog-to-digital converters (ADCs) is considered in particular. ADCs usually exhibit non-ideal behavior in practice. These non-idealities spawn distortions in the converters output. Whenever the errors are systematic, it is possible to mitigate them by mapping the output into a corrected value. The work herein is focused on problems associated with post-correction using look-up tables. All results presented are supported by experiments or simulations.</p><p>The first problem considered is characterization of the ADC. This is in fact an estimation problem, where the transfer function of the converter should be determined. This thesis deals with estimation of quantization region midpoints, aided by a reference signal. A novel estimator based on order statistics is proposed, and is shown to have superior performance compared with the sample mean traditionally used.</p><p>The second major area deals with predicting the performance of an ADC after post-correction. A converter with static differential nonlinearities and random input noise is considered. A post-correction is applied, but with limited (fixed-point) resolution in the corrected values. An expression for the signal-to-noise and distortion ratio after post-correction is provided. It is shown that the performance is dependent on the variance of the differential nonlinearity, the variance of the random noise, the resolution of the converter and the precision of the correction values.</p><p>Finally, the problem of addressing, or indexing, the correction look-up table is dealt with. The indexing method determines both the memory requirements of the table and the ability to describe and correct dynamically dependent error effects. The work here is devoted to state-space--type indexing schemes, which determine the index from a number of consecutive samples. There is a tradeoff between table size and dynamics: more samples used for indexing gives a higher dependence on dynamic, but also a larger table. An indexing scheme that uses only a subset of the bits in each sample is proposed. It is shown how the selection of bits can be optimized, and the exemplary results show that a substantial reduction in memory size is possible with only marginal reduction of performance.</p>

Thesis (Ph. D.)--University of Texas at Austin, 2005. / Includes bibliographical references (leaves 210-223).

This product presents the design of a useful multi-functional adjustable table. It will bring great convenience to people with disability who spend most of time in bed, so as to meet their daily needs. Since there are already many similar products on the market, we focused on updating the functions on table, which can help people reading books or using tablet PC on beds in a comfortable way. It can also support people to stand up and walk around. There is a C-clamp fixing the product consolidated to bed. Besides, the hydraulic and micro-computer systems are used to adjust the table into a proper height automatically. The whole structure was designed in Autodesk Inventor 2015, and the simulation, test, analysis experiments were made in Abaqus 6.1.2 to ensure its safety and stability. In addition, this design can be an incentive for the improvement of the auxiliary equipment, highlighting the importance of the development of mechanisms for the healthcare.

This thesis is devoted to use Lie group analysis to obtain all invariant solutions by constructing optimal system of one-dimensional subalgebras of the Lie algebra L5 for a nonlinear wave equation. I will show how the given symmetries ( Eq.2) are admitted by using partial differential equation (Eq.1), In addition to obtain the commutator table by using the same given symmetries. Subsequently, I calculate the transformations of the generators with the Lie algebra L5, which provide the 5-parameter group of linear transformations for the operators. Finally, I construct the invariant solutions for each member of the optimal system.

The main purpose of this thesis is to use modern goal-oriented adaptive methods of Lie group analysis to construct the optimal sys- tem of Black-Scholes equation. We will show in this thesis how to obtain all invariant solutions by constructing what has now become so popular, optimal system of sub-algebras, the main Lie algebra admit- ted by the Black-Scholes equation. First, we obtain the commutator table of already calculated symmetries of the Black-Scholes equation. We then followed with the calculations of transformation of the gen- erators with the Lie algebra L6 which provides one-parameter group of linear transformations for the operators. Here we make use of the method of Lie equations to solve the partial di®erential equations. Next, we consider the construction of optimal systems of the Black- Scholes equation where the method requires a simpli¯cation of a vector to a general form to each of the transformations of the generators. Further, we construct the invariant solutions for each of the op- timal system. This study is motivated by the analysis of Lie groups which is being taken to another level by ALGA here in Blekinge In- stitute Technology, Sweden. We give a practical and in-depth steps and explanation of how to construct the commutator table, the calcu- lation of the transformation of the generators and the construction of the optimal system as well as their invariant solutions. Keywords: Black-Scholes Equation, commutators, commutator table, Lie equa- tions, invariant solution, optimal system, generators, Airy equation, structure constant, / It was an accolade for us to work with Professor Nail.H. Ibrgimov. +46762600953

Estimating the quantity of water that reaches the water table following an infiltration event is vital for modeling and management of water resources. Estimating the time scale of groundwater recharge after a rainfall event is difficult because of the dependence on nonlinear soil characteristics and variability in antecedent conditions. Modeling the flow of water through the variably saturated zone is computationally intensive since it requires simulation of Richards' equation, a nonlinear partial differential equation without a closed-form analytical solution, with parametric relationships that are difficult to approximate. Hence, regional scale coupled (surface water - groundwater) hydrological models make simplistic assumptions about the quantity and timing of recharge following infiltration. For simplicity, such models assume the quantity of recharge to be a fraction of the total rainfall and the time to recharge the saturated groundwater is scaled proportionally to the depth to water table, in lieu of simulating computationally intensive flow in the variably saturated zone. In integrated or coupled (surface water - groundwater) regional scale hydrological models, better representation of the timing and quantity of groundwater recharge is required and important for water resources management. This dissertation presents a practical groundwater recharge estimation method and relationships that predict the timing and volume accumulation of groundwater recharge to moderate to deep water table settings. This study combines theoretical, empirical, and simulation techniques to develop a relatively simple model to estimate the propagation of the soil moisture wetting front through variably saturated soil. This model estimates the time scale and progression of recharge following infiltration for a specified depth to water table, saturated hydraulic conductivity and equilibrium moisture condition. High-resolution soil moisture data from a set of experiments conducted in a laboratory soil column were used to calibrate the HYDRUS-1D model. The calibrated model was used to analyze the time scale of recharge by varying soil hydraulic properties and simulating the application of rainfall pulses of varying volume and intensities. Modeling results were used to develop an equation that relates the non-dimensional travel time of the wetting front to excess moisture moisture content above equilibrium. This research indicates that for a soil with a known retention curve, the wetting front arrival time at a given depth can be described by a power law, where the power is a function of the saturated hydraulic conductivity. This equation relates the non-dimensional travel time of the wetting front to excess moisture content above the equilibrium moisture content. Since the equilibrium moisture content is dependent on the water retention curve, the powers in the equation governing the timing of recharge depend on the saturated hydraulic conductivity for a large variation in water retention curve. Also, the power law relates recharge (normalized by applied pulse volume) to time (normalized by the time of arrival of wetting front at that depth). The resulting equations predicted the model simulated normalized (relative) recharge with root mean square errors of less than 14 percent for the tested cases.