검색 상세

Remote Detection of Ear Infection Using Antenna Impedance Variation

안테나 임피던스 변화를 이용한 중이염 진단 연구

  • 이용섭 (Yongsub Lee, 서강대학교 일반대학원)

초록 (요약문)

This dissertation presents two studies involving antennas and near-field analysis. Chapter 1 proposes a diagnostic study for otitis media using antenna impedance variation, while Chapter 2 analyzes Near Field to Far Field Transformation using an asymmetric autoencoder. Chapter 1. This study investigates the feasibility of remote detection of ear with infection, such as acute otitis media or otitis media with effusion, by analyzing variation of an antenna impedance. In cases of otitis media is not possible to be diagnosed with a conventional otoscopic examination, extensive testing, such as CT scans is required. It is not only time consuming but can cause risks of radiation. The development of a rapid diagnostic technique for otitis media can decrease time/monetary costs and resolve radiation issues. For the detection of ear infection, this study is based on utilizing impedance variation of an antenna, which changes depending on near-field medium. Furthermore, the time-domain reflectometry (TDR) was employed to examine the reflection characteristics of the antenna with time. Chapter 2. This study proposes a method for predicting the far-field radiation patterns of large antennas using an asymmetric autoencoder algorithm, addressing key challenges in traditional methods of evaluating electrical performance. Specifically, this approach leverages near-field data to predict far-field patterns while reducing the sampling area, relaxing Nyquist sampling constraints, and improving computational efficiency. The novelty of this research is as follows: asymmetric autoencoder demonstrated its ability to efficiently compress and reconstruct complex near-field data, enabling far-field radiation pattern predictions without relying on traditional EFIE calculations. Notably, the predicted results for key performance metrics, such as bandwidth and peak error, showed high levels of accuracy, with minimal deviation from the actual patterns. This research establishes a foundation for efficient and practical large antenna measurement, offering significant potential for future advancements in autoencoder-based methodologies. Key words: Acute otitis media (AOM), Otitis media with effusion (OME), Resonant cavity, Antipodal Vivaldi antenna (AVA), Reflection coefficient, Time-dependent reflection, Antenna impedance, Antenna near field coupling effect., Parabolic Antenna, Nyquist Theorem, Auto-Encoder, Multilayer Perceptron.

more

목차

1 Remote Detection of Ear Infection Using Antenna Impedance Variation 1
1.1 Introduction 1
1.2 Electromagnetic Analysis on Ear Structure 6
1.2.1 The External Auditory Canal 6
1.2.2 The Tympanic Cavity 9
1.3 Antenna Design and Simulation Result 11
1.3.1 Antipodal Vivaldi Antenna (AVA) with Slot 11
1.3.2 Simulation Result S11 and Resonant Frequency 13
1.4 Measurement and Result 21
1.4.1 Reflection Coefficient Measurement 23
1.4.2 Time Domain Reflectometry 25
1.5 Conclusion 29
2 Near Field to Far Field Transformation Using Asymmetric Auto-Encoder 31
2.1 Introduction 31
2.2 Structure of Asymmetric Auto-Encoder and Data Generation 35
2.2.1 Asymmetric Auto-Encoder 35
2.2.2 Data Generation 37
2.2.3 Far Field Estimation of 0.4 Sampled Near Field 39
2.3 Far Field Estimation of Cropped Near Field Data 41
2.4 Far Field Estimation of Down Sampled Near Field Data 46
2.5 Conclusion 51
Bibliography 53

more
반출 Meta View 목록