DOI QR코드

DOI QR Code

Low-Power Streamable AI Software Runtime Execution based on Collaborative Edge-Cloud Image Processing in Metaverse Applications

에지 클라우드 협동 이미지 처리기반 메타버스에서 스트리밍 가능한 저전력 AI 소프트웨어의 런타임 실행

  • Received : 2022.08.10
  • Accepted : 2022.09.19
  • Published : 2022.11.30

Abstract

As the interest in the 4th industrial revolution and metaverse increases, metaverse with multi edge structure is proposed and noted. Metaverse is a structure that can create digital doctor-like system through a large amount of image processing and data transmission in a multi edge system. Since metaverse application requires calculating performance, which can reconstruct 3-D space, edge hardware's insufficient calculating performance has been a problem. To provide streamable AI software in runtime, image processing, and data transmission, which is edge's loads, needs to be lightweight. Also lightweight at the edge leads to power consumption reduction of the entire metaverse application system. In this paper, we propose collaborative edge-cloud image processing with remote image processing method and Region of Interest (ROI) to overcome edge's power performance and build streamable and runtime executable AI software. The proposed structure was implemented using a PC and an embedded board, and the reduction of time, power, and network communications were verified.

최근 4차산업혁명과 함께 메타버스에 대한 관심이 증가하는 가운데, 메타버스를 구축하는 멀티 에지 기반의 구조가 제시된다. 메타버스는 멀티 에지 시스템에서 많은 양의 영상처리와 데이터 전송을 통해 디지털 의사와 같은 시스템을 만들어 낼 수 있는 구조이다. 하지만 에지는 부족한 연산능력이라는 제약이 있으므로, 런타임 스트리밍이 가능한 서비스제공을 위해서는, 에지에서만 이루어졌던 영상처리와 데이터 전송을 에지-클라우드 협동을 통해 처리하여 저전력 시스템을 구축해야한다. 많은 에지들이 연결되는 시스템에서는, 그 무엇보다도 에지 경량화를 통해 효율적인 전체 시스템의 경량화 및 소모전력의 감소를 이루어낼 수 있다. 본 논문에서는 원격영상 처리방법과 Region of Interest (RoI) 기법을 사용하여, AI 소프트웨어의 저전력 런타임 스트리밍이 가능해지는, 에지-클라우드 협동 메타버스 애플리케이션을 제안한다. 에지-클라우드 협동 메타버스의 구조는 PC와 임베디드 보드를 사용하여 구현하였으며, 본 논문의 후반부에서는 에지의 시간 감소와 그에따른 전력 소모, 네트워크 통신량 감소를 검증하였다.

Keywords

Acknowledgement

This study was supported by the BK21 FOUR project funded by the Ministry of Education, Korea (4199990113966, 10%), and the Basic Science Research Program through the National Research Foundation of Korea (NRF) funded by the Ministry of Education (NRF-2018R1A6A1A03025109, 10%, NRF-2022R1I1A3069260, 10%) and by Ministry of Science and ICT (2020M3H2A1078119). This work was partly supported by an Institute of Information and communications Technology Planning and Evaluation (IITP) grant funded by the Korean government (MSIT) (No. 2021-0-00944, Metamorphic approach of unstructured validation/verification for analyzing binary code, 40%) and (No. 2022-0-00816, OpenAPI-based hw/sw platform for edge devices and cloud server, integrated with the on-demand code streaming engine powered by AI, 20%) and (No. 2022-0-01170, PIM Semiconductor Design Research Center, 10%).

References

  1. P. Hui, "Keynote Speaker: The Hitchhiker's Guide to the Metaverse," in 2022 IEEE Conference on Virtual Reality and 3D User Interfaces Abstracts and Workshops (VRW), Christchurch, New Zealand, pp. 203-203, 2022.
  2. J. Gascon-Samson, K. Jung, and K. Pattabiraman, "Poster: Towards a Distributed and Self-Adaptable Cloud-Edge Middleware," in 2018 IEEE/ACM Symposium on Edge Computing (SEC),Washington: WS, USA, pp. 338-340, 2018.
  3. M. Kang and D, Park, "Abnormal System Operation Detection by Comparing QR Code-Encoded Power Consumption Patterns in Software Execution Control Flow," Journal of the Korea Institute of Information and Communication Engineering, vol. 25, no. 11, pp. 1581-1587, Nov. 2021. https://doi.org/10.6109/JKIICE.2021.25.11.1581
  4. M. Kang and D. Park, "Lightweight Microcontroller with Parallelized ECC-Based Code Memory Protection Unit for Robust Instruction Execution in Smart Sensors," Sensors, vol. 21, no. 16, pp. 5508, Aug. 2021. https://doi.org/10.3390/s21165508
  5. Z. Zhou, T. Zhao, W. Li, and A. Y. Zomaya, "Distributed Online Resource Scheduling for Mobile Edge Servers," in 2021 IEEE International Conference on Edge Computing (EDGE), Chicago: IL, USA, pp. 86-93, 2021.
  6. M. Kang, D. Park. "High Speed and Robust Processor based on Parallelized Error Correcting Code Module" Journal of the Korea Institute of Information and Communication Engineering, vol. 24, no.9, pp. 1180-1186, Sep. 2020. DOI: 10.6109/jkiice.2020.24.9.1180.
  7. J. Chang, M. Kang, and D. Park, "Accelerated SVM Algorithm for Sensors Fusion-based Activity Classification in Lightweighted Edge Devices," in 2022 IEEE International Conference on Consumer Electronics (ICCE), Las Vegas: NV, USA, pp. 1-4, 2022.
  8. S. Park and J. Cho, "ROI-based visualization of spatial information for a remote-controlled robot," in 2013 10th International Conference on Ubiquitous Robots and Ambient Intelligence (URAI), Jeju, Korea, pp. 234-235, 2013.
  9. A. Verdant, P. Villard, A. Dupret, and H. Mathias, "Architecture for a low power image sensor with motion detection based ROI," in 2007 14th IEEE International Conference on Electronics, Circuits and Systems, Marrakech, Morocco, pp. 1023-1026, 2007.