DOI QR코드

DOI QR Code

Design of Lightweight Artificial Intelligence System for Multimodal Signal Processing

멀티모달 신호처리를 위한 경량 인공지능 시스템 설계

  • Received : 2018.08.03
  • Accepted : 2018.10.15
  • Published : 2018.10.31

Abstract

The neuromorphic technology has been researched for decades, which learns and processes the information by imitating the human brain. The hardware implementations of neuromorphic systems are configured with highly parallel processing structures and a number of simple computational units. It can achieve high processing speed, low power consumption, and low hardware complexity. Recently, the interests of the neuromorphic technology for low power and small embedded systems have been increasing rapidly. To implement low-complexity hardware, it is necessary to reduce input data dimension without accuracy loss. This paper proposed a low-complexity artificial intelligent engine which consists of parallel neuron engines and a feature extractor. A artificial intelligent engine has a number of neuron engines and its controller to process multimodal sensor data. We verified the performance of the proposed neuron engine including the designed artificial intelligent engines, the feature extractor, and a Micro Controller Unit(MCU).

최근 인간의 뇌를 모방하여 정보를 학습하고 처리하는 뉴로모픽 기술에 대한 연구는 꾸준히 진행되고 있다. 뉴로모픽 시스템의 하드웨어 구현은 다수의 간단한 연산절차와 고도의 병렬처리 구조로 구성이 가능하여, 처리속도, 전력소비, 저 복잡도 구현 측면에서 상당한 이점을 가진다. 또한 저 전력, 소형 임베디드 시스템에 적용 가능한 뉴로모픽 기술에 대한 연구가 급증하고 있으며, 정확도 손실 없이 저 복잡도 구현을 위해서는 입력데이터의 차원축소 기술이 필수적이다. 본 논문은 멀티모달 센서 데이터를 처리하기 위해 멀티모달 센서 시스템, 다수의 뉴론 엔진, 뉴론 엔진 컨트롤러 등으로 구성된 경량 인공지능 엔진과 특징추출기를 설계 하였으며, 이를 위한 병렬 뉴론 엔진 구조를 제안하였다. 설계한 인공지능 엔진, 특징 추출기, Micro Controller Unit(MCU)를 연동하여 제안한 경량 인공지능 엔진의 성능 검증을 진행하였다.

Keywords

References

  1. P. A. Merolla, J. V. Arthur, R. Alvarez-Icaza, A. S. Cassidy, J. Sawada, F. Akopyan, B. L. Jackson, N. Imam, C. Guo, Y. Nakamura, B. Brezzo, I. Vo, S. K. Esser, R. Appuswamy, B. Taba, A. Amir, M. D. Flickner, W. P. Risk, R. Manohar, and D. S. Modha, "A million spiking-neuron integrated circuit with a scalable communication network and interface," Science, vol. 345, no. 6197, 2014, pp. 668-673. https://doi.org/10.1126/science.1254642
  2. B. V. Benjamin, P. Gao, E. McQuinn, S. Choudhary, A. R. Chandrasekaran, J. M. Bussar, R. Alvarez-Icaza, J. V. Arthur, P. A. Merolla, and K. Boahen, "Neurogrid: A mixed-analog-digital multichip system for large-scale neural simulations," Proc. of the IEEE, vol. 102, no. 5, 2014, pp. 699-716. https://doi.org/10.1109/JPROC.2014.2313565
  3. K. Kim, "Optimal Structures of a Neural Network Based on OpenCV for a Golf Ball Recognition," J. of the Korea Institute of Electronic Communication Sciences, vol. 10, no. 2, Feb. 2015, pp. 267-273. https://doi.org/10.13067/JKIECS.2015.10.2.267
  4. Y. Lee and P. Moon, "A Comparison and Analysis of Deep Learning Framework," J. of the Korea Institute of Electronic Communication Sciences, vol. 12, no. 1, Feb. 2017, pp. 115-122. https://doi.org/10.13067/JKIECS.2017.12.1.115
  5. C. Schuman, T. Potok, R. Patton, J. Birdwell, M. Dean, G. Rose, and J. Slank, "A survey of neuromorphic computing and neural networks in hardware," arXiv preprint arXiv:1705.06963, 2017, pp. 1-8.
  6. S. Soman, Jayadeva, and M. Suri, "Recent trends in neuromorphic engineering," Big Data Analytics, vol. 1, no. 15, Dec. 2016, pp. 1-19.
  7. M. Suri, V. Parmar, A. Singla, R. Malviya, and S. Nair, "Neuromorphic hardware accelerated adaptive authentication system," Symp. Series on Computational Intelligence, Dec. 2015, pp. 1206-1213.
  8. T. Baltrusaitis, C. Ahuja, and L. Morency, "Multimodal Machine Learning: A Survey and Taxonomy," arXiv preprint arXiv:1705.09406, 2017, pp. 1-20.
  9. A. Benediktsson and J. Sveinsson. "Feature extraction for neural network classifiers," n Neurocomputation in Remote Sensing Data Analysis, 1997. pp. 97-104.
  10. A. Asuncion and D. Newman, "UCI Machine Learning Repository," 2007.