본문 바로가기 주메뉴 바로가기
22대 대한민국 국회 국회정보길라잡이 국회의원검색
회원가입 로그인
HOME

정보검색

소장정보 검색

국회도서관 홈으로 정보검색 소장정보 검색

결과 내 검색

동의어 포함

고급검색

인명/단체명

저자정보 상세정보
인명/단체명을 입력하세요.

키워드

  • 대표어

  • 외국어

-

목차보기

Title Page

Abstract

Contents

Publications 13

Ⅰ. Introduction 14

1.1. Motivations 14

1.2. Sonar sensing on smartwatches 15

1.3. Outlines and contributions 16

Ⅱ. Related Work 18

2.1. Finger identification input modality 18

2.2. Behavioral biometrics for user authentication 19

2.3. Sonar sensing 20

Ⅲ. System 21

3.1. Sonar system for unmodified smartwatches 21

3.2. Finger identification system 22

3.3. User authentication system 24

Ⅳ. Study 1: Using Sonar to Identify Fingers on a Smartwatch 25

4.1. Study 25

4.1.1. Participants 25

4.1.2. Design 25

4.1.3. Procedure 26

4.2. Behavioral observations 27

4.3. Finger identification performance 29

4.3.1. Preprocessing and classifier 29

4.3.2. Classification performance 30

Ⅴ. Study 2: Using Sonar to Improve Behavioral Biometrics on a Smartwatch 34

5.1. Study 34

5.1.1. Participants 34

5.1.2. Experiment protocol 34

5.2. User authentication performance 35

5.2.1. Preprocessing and multimodal classifier 35

5.2.2. Classification performance 37

5.3. Usability 38

Ⅵ. Discussion 39

6.1. Feasibility of sonar sensing on smartwatches 39

6.2. Limitations and future works 41

Ⅶ. Conclusion 43

References 44

List of Tables

Table 1. Accuracy (in %) for different sonar ranges, expressed in nSamples, the length of the ZC sequence used. nSeqs is set to 60/40. 29

Table 2. Accuracy (in %) for different data capture periods, expressed in terms of both nSeqs, the number of impulse estimations performed, and time (ms). In these results, nSamples is set... 29

Table 3. Accuracy (in %) for individual and LOOCV models, including mean (µ) and standard deviation (σ) from all participants. 31

Table 4. Mean of individual user authentication performances for each finger tap in both multimodal and sonar image classifiers. 36

Table 5. Usability data in terms of ratings of performance, comfort, and speed across thumb, index, and middle finger taps 38

List of Figures

Figure 1. Overview of SonarID: during a screen touch by the thumb, index, or middle finger, a speaker on one side of a smartwatch emits an ultrasonic sonar signal (a Zadoff-Chu (ZC)... 22

Figure 2. Examples of the sonar fingerprints, or impulse response estimations, generated during periodic smartwatch taps by each finger. Top-left shows no tapping, top-right thumb taps,... 23

Figure 3. SonarAuth system. A user taps their smartwatch and behavioral biometric features are extracted. In addition to standard touch and motion features, we capture the motions of... 24

Figure 4. Study interface and interaction. It shows target positions (a), the interface during a trial (b), the index finger pose used for taps to start a trial and during the fixation period (c),... 26

Figure 5. Study data showing movement-times (left), touch-times (center), and error counts (right) for touches with the thumb, index, and middle fingers. 27

Figure 6. Confusion matrices for SonarID classifiers (% accuracy). Left: general model; center: mean individual model; right: mean LOOCV model. 32

Figure 7. Examples of generated sonar image, showing the impulse response estimate from a smartwatch touch (A), and two examples of augmented data derived from the same signal using... 35

초록보기

 Smartwatches are used by millions of people for applications in health, finance, and communication. However, their diminutive screen size limits the expressivity and security encompassing touchscreen interaction. To address this issue, this thesis explored two novel sonar-based approaches, each for general and user-specific perspectives.

Our first exploration targets the realm of general interaction, specifically finger identification. Despite the recognized potential of finger identification for enhancing smartwatch expressivity, its implementation remains challenging, often relying on external devices (e.g., worn magnets) or explicit instructions. Addressing these limitations, this paper explores a novel approach to natural and unencumbered finger identification on an unmodified smartwatch: sonar. To do this, we adapt an existing finger-tracking smartphone sonar implementation—rather than extract finger motion, we process raw sonar fingerprints representing the complete sonar scene recorded during a touch. We capture data from 16 participants operating a smartwatch and use their sonar fingerprints to train a deep learning recognizer that identifies taps by the thumb, index, and middle fingers with an accuracy of up to 93.7%, sufficient to support meaningful application development.

We then pivot to a user-specific angle, specifically user authentication. While various user authentication technologies have been extensively explored in smartphone use scenarios, the applicability of these approaches to smartwatches is typically limited due to the small watch form factor. To improve authentication on smartwatches, we propose SonarAuth, a novel user authentication system for unmodified commercial smartwatches using behavioral biometrics derived from motion, touch, and around-device motions. We collected data from 24 participants from single touch to the watch screen with the thumb, index, and middle fingers. Using a multimodal deep learning classifier, we achieved a promising mean Equal Error Rate(EER) of 6.41% for user authentication based on a single thumb tap. We note that our system is usable and has good potential to be combined with other authentication modalities.

Through this holistic investigation, the thesis highlights the transformative capability of sonar sensing in unmodified smartwatches, forging a path for more intuitive and secure wearable interactions in the real world.

추천서가 (다양한 추천 자료를 만나보세요)

권호기사보기

권호기사 목록 테이블로 기사명, 저자명, 페이지, 원문, 기사목차 순으로 되어있습니다.
기사명 저자명 페이지 원문 기사목차

권호

기사명 원문보기 기사목차
닫기