http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
Realization of spin field-effect transistors consisting of van der Waals 2-dimensional materials
Afzal, Amir Muhammad 세종대학교 대학원 2019 국내박사
그래핀은 스핀트로닉스 첨단소자의 개발에 주요한 역할을 할 수 있는 소재이다. 그러나 그래핀의 스핀-궤도 상호작용이 작기 때문에 스핀홀 효과, 스핀 전계효과 트랜지스터 (FET), 의미있는 위상상태가 발현되기는 어렵다. 그래핀의 독특한 에너지 밴드 구조는 그래핀의 위상적인 구조에 기인한 것인데 스핀-궤도 상호작용을 증가시켜 밴드 구조를 변형시키려는 시도가 최근 계속되고 있다. 이차원 전이금속 칼코겐화합물을 그래핀에 접합시키면 그래핀의 전기적 특성을 변화시켜 스핀트로닉스에 활용될 수 있는 특성이 나타날 수 있다. 전이금속 칼코겐화합물은 강한 스핀-궤도 상호작용을 가지므로 이차원에서 스핀수송 특성을 연구하는 분야에 적용될 수 있고, 반데르발스 이종접합을 연구하는 분야 등 새로운 연구분야에 적용될 수 있다. 이 논문은 그래핀에서 스핀-궤도 상호작용과 스핀 세차운동을 전기적인 방법으로 제어할 수 있음을 보여주고, 나아가 스핀 FET 개발에 대한 연구를 보여준다. 특히, 다층 WS2 반도체와 2층 그래핀 사이의 접합면에서 상호작용은 고급 스핀소자를 구현하는데 필요한 흥미로운 특성을 만들어낸다. 논문의 첫번째 부분은 얇은 WS2 층으로 이루어진 박막들 사이에 2층 그래핀을 끼어 있는 구조에서 그래핀의 스핀-궤도 상호작용의 변화에 대해 연구한 것이다. 우리는 상, 하층 게이트로 이루어진 이중 게이트 소자를 제작하여 게이트 전압에 가하여 그래핀의 스핀-궤도 상호작용을 조절하였다. 얇은 WS2 층으로 이루어진 박막들 사이에 2층 그래핀의 스핀-궤도 상호작용은 일반 그래핀과 비교하여 아주 크게 증가하였으며 저온에서 weak antilocalization 현상을 관찰할 수 있었다. Weak antilocalization 현상을 분석하여 스핀 이완시간이 일반적인 그래핀보다 10배 감소한 것을 발견하였다. 이러한 발견들을 확인하기 위하여 Shubnikov-de Haas 진동 현상을 실험적으로 살펴보았으며 우리가 만든 구조에서 zero-field spin-splitting을 확인하였다. 두 가지 다른 방법(weak antilocalization, Shubnikov-de Haas 진동 현상)으로 측정한 스핀-궤도 상호작용의 크기는 매우 유사한 값으로 나타났으며 게이트 전압에 대한 의존성도 비슷하게 관찰되었다. WS2 층에 끼어 있는 2층 그래핀은 스핀-궤도 상호작용이 전기장으로 제어할 수 있으므로 스핀트로닉스 응용소자로서 혁신적인 시스템이 될 수 있음을 보여준다. 논문의 두번째 부분에서는 2층 그래핀 위에 WS2 층을 형성하여 스핀 FET를 제작한 것을 보여준다. 게이트 전압으로써 Rashba type 스핀-궤도 상호작용을 제어하여 스핀 세차운동을 조절하여 스핀 FET 작동을 보여주었다. 스핀-궤도 상호작용의 크기가 게이트 전압에 의존하여 변하는 것을 Hanle 신호를 측정하여 실험적으로 확인하였다. 아울러, 스핀 FET가 상온에서 작동하는 것으로 보여 주었다. 논문의 세번째 부분에서는 단층 그래핀 위에 Pb/Au (1 nm/10 nm) 층을 증착하여 스핀-궤도 상호작용의 크기가 변화되는 것을 관찰하였다. 이러한 시스템에서 스핀-궤도 상호작용의 크기가 80 meV로 증가되었으며, 매우 큰 스핀홀 효과를 불러일으키는 것을 non-local 신호 측정으로 확인하였다. 게이트 전압에 의존하는 weak localization 효과를 관찰하였고, 스핀-궤도 상호작용의 크기와 스핀 이완시간을 구해낼 수 있었다. 상온에서 관찰된 스핀홀효과는 스핀응용소자 개발에 새로운 전형을 제시해 줄 수 있으리라 생각한다. 게이트 전압에 의한 스핀-궤도 상호작용의 제어는 스핀 전류, 스핀 세차운동, 전자의 스핀 자유도를 조절할 수 있으므로 새로운 소자 응용의 장을 열 수 있으리라 기대한다. Graphene has an enormous potential in the development of advanced spintronic devices. It is a promising material with intrinsically small spin-orbital interaction (SOI), making the realization of some interesting and practically impossible phenomena such as the spin Hall effect (SHE), spin field effect transistor and non-trivial topological states. The electronic band structure of graphene is owing to interesting topological characteristics and strength to stimulate a systematic method to improve SOI extrinsically. The interfacial interactions of graphene sheet with semiconducting transition metal dichalcogenides (TMDs) cultivate the electronics and spintronics properties drastically, making it an intriguing candidate for spintronic applications. Transition metal dichalcogenide materials are studied to investigate such unexplored research avenues as spin transport behavior in 2-dimensional materials due to their strong spin-orbital interaction (SOI) and the proximity effect in van der Waals (vdW) heterostructures. This thesis demonstrates the realization of a spin field effect transistor (spin-FET) by successfully controling the spin-orbital interaction and spin precession angle in graphene through electrical means. Specifically, the interfacial interactions between the bilayer graphene (BLG) and multilayer tungsten disulfide (ML-WS2) give rise to fascinating properties for the realization of advanced spintronic devices. In the first portion, we designed a bilayer graphene (BLG) encapsulated by thin layers of tungsten disulfide (WS2) to explore the interface-induced spin-orbit interaction. We successfully fabricated a dual-gated device, where the strength of spin-orbital interaction is tuned by gate voltages. The strength of such induced spin-orbital interaction in the graphene is intensely elevated, which leads to a robust weak antilocalization effect at low temperature. The quantitative analysis of the weak antilocalization effect exhibits that the spin relaxation time (τ_so ) is 10 times smaller in magnitude than in graphene on traditional substrates. To support these outcomes, we also studied Shubnikov-de Haas oscillations, which give unambiguous confirmation of the zero-field spin-splitting in our heterostructure device. The spin-orbit interaction constants estimated by two different types of measurement (the WAL effect and SdH oscillations), demonstrating close values as a function of gate voltage, supporting the self-consistency of this study’s experimental outcomes. The gate modulation of the spin-orbital interaction in bilayer graphene encapsulated by WS2 nano-flakes establishes an innovative system to explore the manipulation of spin-dependent transport through an electric field. In the second part, a BLG/ML-WS2 vdW heterostructure spin field-effect transistor (FET) is fabricated to demonstrate the gate modulation of Rashba-type SOI and spin precession angle. The gate modulation of Rashba-type SOI and spin precession has been confirmed using the Hanle measurement. The change in spin precession angle agrees well with the local and non-local signals of the BLG/ML-WS2 spin FET. The operation of a spin FET in the absence of a magnetic field at room temperature is successfully demonstrated. Further, we designed a graphene field effect transistor (FET) with lead (Pb) layer intercalated between graphene (Gr) and gold (Au) to stimulate the SOI and studied the effect on the strength of the SOI. The SOI in our system is dramatically elevated to 80 meV, which leads to a giant non-local signal. We observed a strong SHE with non-local signal up to ~ 180 Ω at room temperature without applying any external magnetic field (B) and ferromagnetic contacts. Further, we extract spin key parameter from length and width dependence of nonlocal measurement. To support these findings, we also measured the temperature and gate-dependent weak localization (WL) effect. The quantitative analysis of WL in graphene gives SOI and spin relaxation time. The calculated SOI by two different methods demonstrates close values. The observation of a giant SHE at room temperature is an important step to introducing a new plate form to explore the spin-based concept and applications. The gate control of spin precession at room temperature is an interesting step in the field of spintronics. These outcomes may open a new platform for the manipulation of spin current, precession, and spin degree of freedom of electrons. Furthermore, an innovative ML heterostructure of BLG/ML-WS2 successfully developed as a spin FET to address the challenges of injection, detection, and gate modulation of spin precession angle in spintronics. First of all, the local spin valve (LSV) measurements were conducted at room temperature and 30 K to examine the spin valve behavior in the BLG/ML-WS2 spin FET on SiO2. We observed a distinct and noticeable spin signal and the change in resistance was (∆R=5.85 mΩ) at Vbg = 0 V. We also measured the local signal as a function of Vbg at different temperatures and B = 0 T for each magnetization alignment of the injector and detector. For all magnetization configurations, we observed clear oscillations of the local spin signal as a function of Vbg at room temperature and 30 K. Further, we also made a graphene/Pb/Au heterostructure field effect transistor. For the first time, we used lead (Pb) as an intercalated layer between the graphene and Au and measured SOI in heterostructure device in local and non-local measurement configurations. Intercalation of Pb thin layer between Au and graphene leads to momentous modification of the electronic structure and SOI of graphene. We observed a giant spin Hall effect (R_NL~180 Ω) in the absence of any external magnetic field and with out ferromagnetic leads. To further quantitatively examine the improvement in the magnitude of SOI in graphene with temperature and back gate voltage, we fit our magneto-conductivity data by using the theory of WL. The calculated values of SOI in graphene through non-local configuration and WL analysis give close results, supporting the self-consistency of this study’s experimental results. The effective improvement of SOI in the graphene-based system opens a new gateway to explore new areas of the spin field-effect transport (FET) phenomenon.
802.16e WiMAX 시스템에서 초기 Ranging Detection 능력 향상과 Time Synchronization 향상을 위한 알고리즘
Afzal, Muhammad Usman 한양대학교 대학원 2008 국내석사
An OFDM system performance highly depends on synchronization between the transmitter and the receiver. When a new SS or MSS is trying to enter a network, it is not synchronized. Initial ranging is proposed for IEEE 802.16e systems for synchronization. In 802.16e OFDMA system code division multiple access (CDMA) codes are used to improve the system efficiency in detecting the new user. A new MSS will transmit one of the predefined CDMA codes, which should be detected at the BS. The BS is not only to detect the new MSS but also to calculate its timing, frequency and power offset. Initial ranging is inherently affected by the channel selectivity and other users' interferences at low SNR. We present an enhanced ranging scheme using modified generalized chirp-like (GCL) sequences that improves ranging detection process using the combine multiple FFT blocks and cope with channel selectivity and other user’s interference at low SNR. Based on the ranging detection timing offset is calculated for synchronization.
MUHAMMAD AFZAL UNIVERSITY OF SCIENCE AND TECHNOLOGY 2018 국내박사
Advances in fluorescent protein development over the last decade have led to finetuning of Aequorea victoria jellyfish color palette. For deep imaging of animal tissues, the optical window favorable for light penetration is in near-infrared wavelengths. Some of the red fluorescent proteins like mKate, Neptune, mNeptune1, mNeptune2 and mNeptune2.5 are reported as brighter fluorescence and improved maturation while mCardinal and mCardinal-V218E (mCardinal2) are developed for more red-shifted. We engineered mCardinal2 and its five mutants, T143C, T143G, C158A for green shift and C158D, M160E for large stokes shift. Several spectral and color changes are observed in each mutant. The produced red fluorescent protein mCardinal2 is further utilized to prove the validity of aptamer based conjugated protein and miRNA separation by acoustofluidic techniques. The work proves the importance and characteristic verification of the produced protein by acoustic separation of thrombin conjugated green fluorescent polystyrene particles from the red fluorescent protein mCardinal2 using fluorescence microscopy. Color variations with change in excitation and emission spectra and detailed crystal structures having His-tag may reveal significant interactions in RFP applications to expand variations in in-vivo imaging and cellular differentiation techniques. The acoustofluidic work finds its importance on a variety of biomolecules like DNA, miRNA and proteins. 지난 10 년간 형광 단백질의 발달로 Aequorea victoria 해파리 컬러 팔레트가 미세 조정되었습니다. 동물 조직의 깊은 이미징의 경우, 빛의 투과에 유리한 광학 창은 근적외선 파장에 있습니다. mKate, Neptune, mNeptune1, mNeptune2 및 mNeptune2.5 와 같은 적색 형광 단백질 중 일부는 밝은 형광 및 향상된 성숙으로보고되고 mCardinal & mCardinal-V218E (mCardinal2)는 더 많은 적색 이동을합니다. 우리는 mCardinal2 와 5 개의 돌연변이 체를 설계했으며 모두 His-tag 를 단백질 부분으로 포함하고 있습니다. T143C, T143G, C158A 는 녹색 변속을, C158D 및 M160E 는 대형 변속기를 시프트합니다. 몇 가지 스펙트럼 및 색상 변화가 각 돌연변이에서 관찰됩니다. 생성 된 적색 형광 단백질 인 mCardinal2 는 또한 음향 유체 기술에 의한 접합체 단백질 분리 (aptamer based conjugated protein separation)의 타당성을 입증하기 위해 활용됩니다. 작품은 형광 현미경을 사용하여 적색 형광 단백질 mCardinal2 에서 녹색 형광 폴리스티렌 입자의 트롬빈 공명의 음향 분리하여 생산 된 단백질의 중요성과 특성 검증을 증명합니다. 각 돌연변이에 대한 여기 및 방출 스펙트럼의 변화와 His- 태그를 갖는 상세한 결정 구조의 변화는 생체 내 이미징 및 세포 분화 기술의 변형을 확대하기 위해 RFP 애플리케이션에서 중요한 상호 작용을 나타낼 수있다. 음향 유체 작업은 DNA, miRNA 및 단백질과 같은 다양한 생체 분자에 대한 중요성을 발견했습니다.
A Numerical Study of Bubble Detachment from Solid Wall and Evaporative Growth
Muhammad Umar Afzal Graduate School of Pohang University of Science an 2012 국내석사
There are many studies available in literature which lead with enhancement of heat flux during the boiling process. Nucleate boiling is the only regime during boiling phenomena which is privileged by generation of bubbles and as a consequence of it show efficient heat transfer. Studies on nucleate boiling maintain to get attention of researchers because of its enormous potential. Poorly simulated bubbles produced somewhat misleading results to understand extremely complex simultaneous multi-transport phenomena of boiling. The effects of supplied heat to the bubble attached to the heated solid wall are investigated using numerical method. Easy access and availability towards commercial packages of "Computational Fluid Dynamics" make numerical study of bubble growth and detachment from solid heated wall quiet easier. In the present study, we used 'COMSOL Multiphysics' package with fairly realistic assumptions regarding bubble growth pattern. During the bubble growth, solutions on the shape of bubble are obtained for the governing equations which are the temperature field and the flow field. In this study we considered comparatively slower growth rate in a detailed manner. Several numerical results helped to better understand the multiphysics phenomena. First, it is found that bubble growth pattern shows week dependency on contact angle with solid wall and on the expanse of increased wall super heat larger growth rate can be achieved. Bubble diameter at moment of departure show linear decrease as the contact angle is reduced up to 20 degrees. Second, wall heat flux is found to be dominated by transient conduction mode of heat transfer. The heat fluxes are varied both temporally and spatially depending on the bubble shape and dynamics within domain of interest. Natural convection contribution is highest during the early stage of bubble growth. The highest rate in terms of energy utilization is achieved when the vapor bubble base diameter is nearly maximum. Time integrated values of heat fluxes revealed that 30% of energy is employed in vapor production, on the other hand 70% goes into superheating of liquid.
AUTOMATIC EVIDENCE ACQUISITION AND APPRAISAL TO SUPPORT EVIDENCE-BASED MEDICAL DECISION MAKING
Muhammad Afzal 경희대학교 대학원 2017 국내박사
An overwhelming amount of biomedical information is available in the form of text documents which can be retrieved intelligently to support the process of evidence-based decision making. However, there exists a large gap between the user space and evidentiary document space reaching a better clinical decision. Successful clinical practice demands best available evidence to find from the research literature for a better clinical action. However, clinicians face many barriers in order to access best available evidence. The main barriers are the lack of automated and reliable methods to finding and recognizing the best evidence among a huge array of evidentiary documents. Best evidence is the one which is relevant, is of high quality, and fits contextually. Current approaches are not sufficient to cover all the three, aforementioned, aspects of the best evidence. The existing approaches of query construction are largely information driven, nonstructured, and manual. Similarly, for the quality assessment, Boolean approaches are largely in practice. Even though modern approaches have shifted focus towards statistical approaches for quality assessment, however, dataset availability and reliability in addition to feature engineering are among the major challenges. Determining quality is not sufficient to establish an evidence is contextually fit for a user. Currently, user context is taken to the level of user goal and other aspects such as varied user role (physician, nurse, researcher) and environment (clinical setup, research unit) are majorly overlooked. The main goal of this thesis is to minimize human efforts getting best research evidence for better clinical decision making. This goal is achieved through satisfying the objectives of: to develop and evaluate methods/models for finding relevant evidentiary documents, to develop and evaluate methods/models for recognizing quality evidences, and to develop and evaluate methods/ models for evidence contextual fitness. To achieve these objectives, this thesis proposed a three-fold methodology for acquiring a best available evidence: relevant evidence acquisition, quality assessment, and contextual fitness determination. For relevant evidence acquisition, an automatic method called Task Aware PICO (Problem, Intervention, Comparison, and Outcome)-compliant question construction is proposed, which has two sub-parts: PICO compliancy and task awareness. PICO-compliant question construction involves knowledge of a clinical decision support system as a source of query contents. A mapping model called KAP (knowledge alignments to PICO) is constructed for correct mapping of different parts of a knowledge representation scheme to appropriate part of PICO. The mappings are achieved at two levels: structure level and concept level. For structure level mappings, a set of specialized models are proposed in order to cover the diversified knowledge representations such as Arden syntax medical logic module (MLM), production rules, and others. For concept level mappings, an algorithm called STI (salient term identification) is developed that identifies important terms for the final query on the basis of term matching using standard terminology system. PICO-compliant question is considered as the initial query, which is augmented with a clinical task information making it more concentrated on a specific user goal such as treatment, diagnosis, etiology, and prognosis. For clinical task awareness, the concepts belonging to the parts ”I” (Intervention) and ”C” (Comparison) are utilized to recognize the top level semantic category of a particular hierarchy in SNOMED CT. For quality assessment, a statistical-based quality assessment is proposed which is based on a classification model called quality recognition model (QRM). QRM is a support vector machine (SVM) based binary classification model which is trained on a dataset annotated by a team of professional experts. QRM utilizes two types of features: data features (title and abstract) and metadata features (medical subject heading (MeSH) terms and publication type). All of these features are engineered automatically by involving text processing functions of tokenization, stop words removal, case changing, stemming, and token filtration. For contextual fitness, this work proposed a method called context-aware evidence grading, which aggregates the user context with evidence context. The aggregation of user and evidence contexts are derived from the contextual matrices designed for user and evidence. The contextual matrices are initially constructed through two mechanisms: literature-based context acquisition and expert-driven context determination. Final grading of evidence is instantiated at three levels: high, medium, and low according to its fitness to the context. The proposed methodology presented in this thesis is evaluated at different levels by performing multiple experiments on different evaluation criteria. First of all, the correctness of automatically constructed PICO-compliant question is evaluated with four types of measurements: precision at ten retrieved documents (P10), mean precision (MP), total document reciprocal rank (TDRR), and mean reciprocal rank (MRR). Secondly, the QRM model performance is evaluated on a set of expert annotated evidentiary documents using 10-fold cross-validation technique. Thirdly, the evidence contextual fitness is duly verified from the physicians. Finally, the results obtained from these evaluations showed significant improvements in terms of accuracy and time efficiency. Moreover, this work has been realized for head and neck cancer treatment domain where its importance has been recognized by the physicians involved in evidence-based clinical practice.
Afzal, Muhammad Shakeel 경북대학교 대학원 2019 국내박사
The acoustical transducers and their arrays are widely used in a broad range of applications in various modern-day systems including the most obvious use in underwater applications, acoustical testing of the materials and structure and biomedical applications. The single acoustical transducers generally have a limitation on the desired acoustical performance such as bandwidth, transmitting voltage response (TVR) and receiving sensitivity. The acoustical performance of an array highly depends on the size and orientation of array, constituent element and their mutual spacing and most importantly the design of the single transducer element of the array. The conventional fully sampled arrays composed of several elements have the drawbacks such as a higher cost, a complex structure, and various operational limitations. A sparse array with fewer elements then that of the conventional array is a prospective method to overcome these drawbacks. The structural and performance of one transducer element of the array play a crucial role in the overall performance of the array. In this study, the first three types of transducer structures are designed, and then a sparse array comprised of one representative transducer structure is optimized.The designed transducer structures were optimized for superior acoustical characteristics with the main objective of (delete) a wide bandwidth. First, a low-frequency bender transducer is designed for wider bandwidth and verified the validity of the design through experiments. The performance of the bender transducer was analyzed using the finite element method in detail to investigate the structural parameters for optimization. Then the optimal structure of the bender transducer was derived from achieving the wider bandwidth. Based upon the optimal design, a prototype transducer was fabricated to measure its acoustic properties and validate the design. Similarly, the second transducer structure for mid-frequency range was designed and optimized to achieve wideband characteristics. Secondly, multimode Tonpilz-type transducer configurations were analyzed using the finite element analysis (FEA), and the effects of all the structural design variables were thoroughly investigated. Then the structural optimization was carried out, and a fractional bandwidth (FBW) of 93.6% was achieved, which demonstrates the effectiveness of the optimized structure. The multilayered transducer structure which is generally used for high-frequency applications was the third designed transducer structure using the equivalent circuit method (ECM). The validity of the ECM was verified initially by comparing the results with those from the finite element analysis of the same structure. Furthermore, experimental testing of an actual multilayered transducer was carried out to verify the effectiveness of the ECM. The ECM presented in this study can predict the performance characteristics of the multilayered transducer accurately and reliably in a shorter time. After designing the transducer structures, a fully sampled planar array of underwater Tonpilz transducers was designed to compute the pulse-echo performance parameters and derive the radiation pattern of the array. The structure of the sparse array was then optimized to have a performance equivalent to that of the fully sampled array with 50% fewer elements then the fully sampled array. The design method presented in this work is novel such that the sparse array was optimized by including the acoustic interactions between array elements in the analysis of the array beam pattern. The optimization was performed by considering the main performance parameters such as peak sidelobe level (PSLL) and main lobe beamwidth (MLBW). Finally, the design validation was carried out by comparing the analytical results with those obtained using the FEA, for the fully sampled and the sparse array. The analytically computed beam pattern showed an excellent agreement with those from the FEA which confirmed the efficacy of the designed sparse array by the inclusion of the mutual acoustic interactions. 음향 트랜스듀서와 이로 구성된 배열형 트랜스듀서는 수중, 재료와 구조의 음향 테스트 및 생체 의학등의 산업 분야에서 널리 사용되고 있다. 일반적으로 단일 음향 트랜스듀서로는 요구되는 대역폭, 송신 전압 응답 (TVR) 및 수신 감도등을 만족하지 못한다. 배열형 트랜스듀서의 음향 특성은 배열의 크기, 배열 형태, 단일 트랜스듀서들간의 간격, 단일 트랜스듀서의 음향 특성에 많은 영향을 받는다. 여러개의 단일 트랜스듀서들로 구성된 기존의 평면 배열 트랜스듀서는 값이 비싸며, 구조가 복잡하여 제작이 어렵다는 단점이 있다. Sparse 배열 트랜스듀서는 기존의 평면 배열 트랜스듀서보다 적은 개수의 단일 트랜스듀서로 구성되어 있으므로 이러한 단점을 보완할 수 있다. 배열형 트랜스듀서를 구성하는 단일 음향 트랜스듀서의 음향 특성은 전체 sparse 배열 트랜스듀서의 성능에 매우 많은 영향을 미친다. 본 연구에서는 3 가지 종류의 단일 트랜스듀서의 구조를 설계한 다음, 이들 중 하나의 대표적인 트랜스듀서로 구성된 sparse 배열 트래스듀서의 구조를 최적설계 하였다. 설계된 트랜스듀서의 구조는 더 넓은 대역폭을 가지기 위한 목적으로 우수한 음향 특성을 (delete) 가지도록 최적화 되었다. 첫 번째로, 저주파수용 벤더 트랜스듀서가 넓은 대역폭을 가지도록 설계하고 실험을 통해 설계 타당성을 검증하였다. 최적화를 위한 설계 변수를 선정하기 위해 유한요소법을 이용하여 벤더 트랜스듀서의 성능을 상세하게 분석하고 더 넓은 대역폭을 가지는 벤더 트래스듀서의 최적 구조를 도출하였다. 최적 설계에 기초하여, 벤더 트랜스듀서를 제작하여 음향 특성을 측정하고 설계의 타당성을 검증하였다. 유사한 방식으로 중간 범위 주파수용 두 번째 트랜스듀서의 구조가 광대역 특성을 가지도록 설계되었고 최적화되었다. 두 번째로, 다중 모드 Tonpilz 트랜스듀서의 음향 특성을 유한요소분석을 사용하여 분석하고, 모든 구조 설계 변수가 음향 특성에 미치는 영향을 분석하였다. 다음으로 구조 최적화를 수행하여 93.6 %의 비대역폭 (FBW)을 가지는 최적 구조를 도출하였고, 이를 통해 최적화된 구조의 효율성을 증명하였다. 일반적으로 고주파 응용 분야에 사용되는 공중 초음파 트랜스듀서의 구조는 등가 회로 방법 (ECM)을 사용하여 설계된 세 번째 트랜스듀서 구조이다. ECM의 타당성은 동일한 구조의 유한 요소 분석 결과와 비교함으로써 초기에 검증되었다. 또한 ECM의 유효성을 검증하기 위해 실제 초음파 트랜스듀서의 실험을 수행하였다. 본 연구에서 제시된 ECM은 초음파 트랜스듀서의 성능 특성을 짧은 시간 내에 정확하고 신뢰성 있게 예측할 수 있다. 각각의 단일 트랜스듀서들의 구조를 설계 한 다음, 전체 배열 트랜스듀서의 방사 패턴을 계산하기 위해 Tonpilz 트랜스듀서들로 구성된 평면 배열 트랜스듀서를 설계하였다. Sparse 어레이 트랜스듀서의 구조는 평면 배열 트랜스듀서보다 50 % 더 적은 단일 트랜스듀서로 구성되어 있으며, 평면 배열 트래스듀서와 동일한 성능을 가지도록 최적화되었다. 본 논문의 독창성은 전체 배열형 트랜스듀서의 빔 패턴 분석에 단일 트랜스듀서들간의 상호 음향 간섭을 고려하여 sparse 배열 트랜스듀서를 최적화한 것이다. 최적화는 부 로브의 최대 크기 (PSLL) 및 주 로브의 빔폭 (MLBW)과 같은 주요 성능을 고려하여 수행되었다. 마지막으로 sparse 배열과 평면 배열 트랜스듀서에 대해 FEA를 사용하여 얻은 결과와 이론식을 이용하여 계산된 결과를 비교하여 설계의 타당성을 검증하였다. 이론적으로 계산된 빔 패턴과 상호 음향 간섭이 고려된 sparse 어레이 트랜스듀서의 효능이 확인된 유한요소법으로 분석된 빔 패턴은 매우 잘 일치하였다.