S-Space College of Engineering/Engineering Practice School (공과대학/대학원) Dept. of Electrical and Computer Engineering (전기·정보공학부) Theses (Ph.D. / Sc.D._전기·정보공학부)
Aerial Acoustic Communication Using Chirp Signal
처프 신호를 이용한 음파 통신 기법 연구
- 공과대학 전기·컴퓨터공학부
- Issue Date
- 서울대학교 대학원
- chirp signal; aerial acoustic communication; smart devices; software-based digital modem; application design
- 학위논문 (박사)-- 서울대학교 대학원 : 전기·컴퓨터공학부, 2014. 8. 최성현.
- Todays smart devices such as smartphones and tablet/wearable PCs are equipped with voice user interface (UI) in order to support intuitive command input from users. Speakers and microphones of the voice UI are generally used to play and record human voice and/or environmental sound, respectively. Accordingly, various aerial acoustic communication techniques have been introduced to utilize the voice UI as an additional communication interface beyond WiFi and/or Bluetooth. Smart devices are especially suitable for the aerial acoustic communication since the application processor (AP) of smart devices can process the sound to embed or fetch information in it. That is, smart devices work similar to software defined radio platform. The aerial acoustic communication is also very versatile as any audio interface can be utilized as a communication interface. In this dissertation, we propose an aerial acoustic communication technique using inaudible chirp signal as well as corresponding receiver architecture for smart devices. We additionally introduce the applications of the proposed communication technique in indoor environments.
We begin the receiver design for aerial acoustic communication by measuring the characteristics of indoor acoustic channel, composed of speaker, air-medium, and microphone. Our experimental research reveals that the indoor acoustic channel typically has long delay spread (approximately 40 msec), and it is very frequency-selective due to the frequency response of audio interfaces. We also show that legacy physical layer (PHY) modulation schemes such as phase/frequency shift keying (PSK/FSK) are likely to fail in this indoor acoustic channel, especially in long communication scenarios, due mainly to the instability of local oscillator and frequency selectivity of audio interfaces.
In order to resolve the above-mentioned problems, we use chirp signals for the aerial acoustic communication. The proposed acoustic receiver supports long-range communication independent of the device characteristics over the severely frequency-selective acoustic channel with large delay spread. The chirp signal has time-varying frequency with a specific frequency sweeping rate. The chirp signal was widely used for radar applications due to its capability of resolving multi-path propagation. However, this dissertation is the first study of adopting chirp signal in aerial acoustic communications for smart devices. The proposed receiver architecture of chirp binary orthogonal keying (BOK) can be easily implemented via fast Fourier transform (FFT) in smart devices application layer. Via extensive experimental results, we verify that the proposed chirp signal can deliver data at 16 bps up to 25 m distance in typical indoor environments, which is drastically extended compared to the few meters of previous research.
The data rate of 16 bps is enough to deliver short identification (ID) in indoor environments. The exemplary applications with this short ID can be multimedia content recognition and indoor location tracking. The low data rate, however, might be a huddle of the proposed system to be applied to the services that require high data rate. We design a backend server architecture in order to compensate for the low data rate and widen the application extent of the proposed receiver. The smart devices can send queries in order to refer to the backend server for additional information that is related with the received ID. We also propose an energy-efficient recording and processing method for the acoustic signal detection. Note that it would consume huge amount of energy if the smart devices contiguously sensed the acoustic signal for 24 hours. The smart devices instead control the sensing (i.e., recording) timing so that it is activated only when there exists chirp signal. This can drastically extend the battery lifetime by removing unnecessary signal processing. We also present two application examples of the proposed receiver, namely, (1) TV content recognition, and (2) indoor location tracking, including technical discussions on their implementations. Experiments and field tests validate the feasibility of the proposed aerial acoustic communication in practical environments.