지원사업
학술연구/단체지원/교육 등 연구자 활동을 지속하도록 DBpia가 지원하고 있어요.
커뮤니티
연구자들이 자신의 연구와 전문성을 널리 알리고, 새로운 협력의 기회를 만들 수 있는 네트워킹 공간이에요.
논문 기본 정보
- 자료유형
- 학위논문
- 저자정보
- 지도교수
- Dong-Seong Kim
- 발행연도
- 2023
- 저작권
- 금오공과대학교 논문은 저작권에 의해 보호받습니다.
이용수0
이 논문의 연구 히스토리 (2)
- 이 논문의 후속연구가 궁금하신가요?
- 연관 학술논문 또는 학술발표를 통해 보다 발전된 연구결과를 확인하실 수 있습니다.
- 이 논문의 연구 히스토리 확인하기
초록· 키워드
오류제보하기
Preserving low energy usage in the Internet of Things, particularly in the battery-powered sensors of Wireless Sensor Networks (WSN), is critical. Failure to address this issue will produce data faulty across the network, which will contribute to increased data retransmission numbers that consume more energy and eventually reduce the network lifetime. Instead of relying solely on the sensor to continuously send the environmental data, digital twin-based data prediction method can significantly reduce the transmission number and maintain sensor network lifetime. However, implementing the digital twin-based solution for smart IoT is such a challenge. The existing solutions need to clearly define the system impact on implementing the smart IoT solutions due to inadequate results. This study discusses the digital twin-based smart IoT, where the virtual sensor continuously provides the sensor data with minimal involvement of the actual sensor device to reduce data transmission. To address the hardware impact, the proposed data prediction model of L-MLPDP is deployed and carefully configured with a pruning method to comply with resource-constrained devices. The digital system performance impact is explainable by providing various loss prediction results such as RMSE, MAE, and R2. Additionally, the system CPU usage, network performance, and network scalability test result provide digital twin system performance on the hardware.
목차
등록된 정보가 없습니다.
최근 본 자료
댓글(0)
0