TinyML을 초저전력 시스템에 탑재한 온센서 AI를 이용한 공동주택의 층간 소음 문제 해결
Addressing Inter-floor Noise Issues in Apartment Buildings using On-Sensor AI Embedded with TinyML on Ultra-Low-Power Systems
0
조회
0
다운로드
0
인용
0
참고문헌
키워드
초록
본 논문은 딥러닝 모델이 포함된 TinyML(Tiny Machine Learning)를 초저전력 시스템에 탑재하여, 층간소음 문제를 실시간으로 처리하는 방법을 제시한다. 이 방법이 가능한 이유는 딥러닝 모델 경량화 기술로 인해 컴퓨팅 리소스가 작은 시스템도 자체적으로 추론을 수행 할 수 있기 때문이다. 기존에 층간소음 문제를 해결하기 위해 제시됐던 방법은 센서에서 수집한 데이터를 서버로보내어 데이터를 분석한 후에 처리하는 방법 이었다. 하지만 이러한 중앙 처리 방법은 구축 비용이 비싸고 복잡하며, 실시간 처리가 어려운 문제가 있다. 이러한 한계점을 본 논문에서는 TinyML 을 사용한 On-Sensor AI(Artificial Intelligent) 로 해결하였다. 본 논문에서 제시한 방법은 시스템 설치가 간단하고 저비용 이면서 문제를 실시간적으로 처리할 수 있다.
Abstract
In this paper, we proposes a method for real-time processing of inter-floor noise problems by embedding TinyML, which includes a deep learning model, into ultra-low-power systems. The reason this method is feasible is because of lightweight deep learning model technology, which allows even systems with small computing resources to perform inference autonomously. The conventional method proposed to solve inter-floor noise problems was to send data collected from sensors to a server for analysis and processing. However, this centralized processing method has issues with high costs, complexity, and difficulty in real-time processing. In this paper, we address these limitations by employing On-Sensor AI using TinyML. The method presented in this paper is simple to install, cost-effective, and capable of processing problems in real-time.
참고문헌 (21)
- "Statistics Korea 2021 Population and Housing Census Results"[URL]
- C. H. Jeong, "A Research On Current Condition And Improvement Measures of Noise Between Floors In Mass Housing", pp.2009
- K. H. Park, "Criminal Study on the Noise - Focused on the Floor Impact Noise Dispute", The Institute of Legal Studies Inha University, 18(3) (2015)
- S. W. Lee, "Claims for damages caused by noise between floors of multi-family housing", Korea Land Law Association, 36(1), pp.153~192 (2020)
- J. Y. Park, "A Study on the Product Development Strategy of Functional Flooring for Noise Insulation", Journal of Korea Design Knowledge, 33, pp.401~409 (2015)[DOI]
- M. H. Kim, "Noise Measurement and Floor Noise Monitoring System due to Direct Impact and Air Transmission", Proceedings of KIIT Conference, pp.124~127 (2018)
- J. C. Seo, "The Current Status of Apartment Houses and Noise between Floors", Elastomers and composites, 17(1), pp.34~50 (2016)
- J. W. Lee, "Reality and Status of Inter-floor Noise in Apartment Complexes", Journal of KSNVE, 23(3), pp.4~8 (2013)
- J. H. Jeong, "A study on the annoyance and disturbance of floor impact noise according to noise sensitivity based on questionnaire survey", The Journal of the Acoustical Society of Korea, 37(6), pp.428~436 (2018)[DOI]
- "A study on Management Plan for Conflict Mitigation of Apartment Noise" (2014)[URL]
- J. T. Kim, "A new key to solving inter-floor noise in apartment complexes, utilizing the Internet of Things(IoT)", Korean Resident Autonomy Society, 150, pp.62~64 (2018)
- J. Lim, "Noise reduction system algorithm for noise reduction between apartment", Journal of the Korea Institute of information and Communication Engineering, 24, pp.271~272 (2020)
- T. Y. Jung, "Distinction System for Sources of Interlayer Noise Using Arduino and Open IoT Platform", Journal of the Korea Institute of Communication and Information Sciences, pp.904~905 (2019)
- M. Y. Kim, "Design of Floors Noise Management System based on IoT", Journal of Korean Institute of Information Scientists and Engineers, pp.479~481 (2019)
- J. H. Ko, "Floor Noise Epicenter Recognition System for Exact Epicenter Identification of Floor Noise based on SVM", Journal of KIIT, 20(3), pp.95~105 (2022)[DOI]
- "Arduino Nano 33 BLE board"[URL]
- H. S. Yang, "Enhancement Strategies for Practical Application of Inter-floor Noise Alert Systems", Academic Journal of Noise·Vibration, 28(2), pp.25~29 (2018)
- "TensorFlow Lite for Microcontrollers"[URL]
- P. Warden, "TinyML", pp.95~180 (2019), O’ReillyMedia
- "Interpreter of TensorFlow Lite"[URL]
- "Arduino IDE"[URL]