전자통신동향분석 (Electronics and Telecommunications Trends)

한국전자통신연구원 (Electronics and Telecommunications Research Institute)
권호 표지
DOI QR코드

DOI QR Code

초저전력 엣지 지능형반도체 기술 동향

Trends in Ultra Low Power Intelligent Edge Semiconductor Technology

  • 발행 : 2018.12.01
PDF 다운로드
⟨ 이전 논문 다음 논문 ⟩

초록

In the age of IoT, in which everything is connected to a network, there have been increases in the amount of data traffic, latency, and the risk of personal privacy breaches that conventional cloud computing technology cannot cope with. The idea of edge computing has emerged as a solution to these issues, and furthermore, the concept of ultra-low power edge intelligent semiconductors in which the IoT device itself performs intelligent decisions and processes data has been established. The key elements of this function are an intelligent semiconductor based on artificial intelligence, connectivity for the efficient connection of neurons and synapses, and a large-scale spiking neural network simulation framework for the performance prediction of a neural network. This paper covers the current trends in ultra-low power edge intelligent semiconductors including issues regarding their technology and application.

키워드

HJTOCM_2018_v33n6_24_f0001.png 이미지

(그림 1) 시스코가 제안한 포그 컴퓨팅 개념

HJTOCM_2018_v33n6_24_f0002.png 이미지

(그림 2) 초저전력 엣지 지능형 반도체 개념

HJTOCM_2018_v33n6_24_f0003.png 이미지

(그림 3) SNN기반 대표적인 인공지능 반도체 코어인 IBM의 TrueNorth 16 chip Board

HJTOCM_2018_v33n6_24_f0004.png 이미지

(그림 4) 2018년 출시된 주요 모바일 AP

HJTOCM_2018_v33n6_24_f0005.png 이미지

(그림 5) 뇌 구조 모방 스파이킹 뉴럴 네트워크

HJTOCM_2018_v33n6_24_f0006.png 이미지

(그림 6) Address-Evet Representation

HJTOCM_2018_v33n6_24_f0007.png 이미지

(그림 7) DyNAPs의 트리/메쉬 복합 NoC의 구조

HJTOCM_2018_v33n6_24_f0008.png 이미지

(그림 8) Brian을 이용한 다양한 결과 출력 예제

<표 1> 스파이크 뉴럴 네트워크 시뮬레이터의 종류 및 특징 (○: 지원, △: 부분 지원)

HJTOCM_2018_v33n6_24_t0001.png 이미지

참고문헌

  1. T. Barnett Jr. et al., "Cisco Global Cloud Index 2015-2020," Cisco Knowledge Network (CKN) Session, Nove. 2016. https://www.cisco.com/c/dam/m/en_us/service-provider/ciscoknowledgenetwork/files/622_11_15-16-Cisco_GCI_CKN_2015-2020_AMER_EMEAR_NOV2016.pdf
  2. Qualcomm, "5G - Vision for the Next Generation of Connectivity," Mar. 2015. https://www.qualcomm.com/media/documents/files/whitepaper-5g-vision-for-the-next-generation-of-connectivity.pdf
  3. R. De La Mora et al., "Cisco IOx: Making Fog Real for IoT," Feb., 2014. https://blogs.cisco.com/digital/cisco-iox-making-fog-real-for-iot.
  4. R. Samuel et al., "Edge Processing: How Qualcomm is Helping Build the Industrial IoT," June 2017. https://www.qualcomm.com/news/onq/2017/06/28/edge-processing-how-qualcomm-helping-build-industrial-iot
  5. F. Akopyan et al., "TrueNorth: Design and Tool Flow of a 65mW 1 Million Neuron Programmable Neurosynaptic Chip," IEEE Trans.Comput.-Aided Des. Integr. Circuits Syst., vol. 34, no. 10, Oct. 2015, pp. 1537-1557. https://doi.org/10.1109/TCAD.2015.2474396
  6. M. Davies et al., "Loihi: A Neuromorphic Manycore Processor with On-Chip Learning," IEEE Micro, vol. 38, no. 1, Jan. Feb. 2018, pp. 82-99. https://doi.org/10.1109/MM.2018.112130359
  7. Qualcomm, "Introducing Qualcomm Zeroth Processors: Brain-Inspired Computing," OnQ Blog, Oct. 10, 2013. https://www.qualcomm.com/news/onq/2013/10/10/introducing-qualcomm-zeroth-processors-brain-inspired-computing
  8. Qualcomm, https://www.qualcomm.com/products/mobile-processors
  9. Samsung, https://www.samsung.com/semiconductor/minisite/exynos/
  10. Huawei, https://consumer.huawei.com/en/campaign/kirin980/
  11. Apple, "A12 Bionic," https://www.apple.com/kr/iphone-xs/a12-bionic/
  12. S. Moradi et al., "A Scalable Multicore Architecture with Heterogeneous Memory Structures for Dynamic Neuromorphic Asynchronous Processors (DYNAPs)," IEEE Trans. Biomedical Circuits Syst., vol. 12, no. 1, Feb. 2018, pp. 106-122. https://doi.org/10.1109/TBCAS.2017.2759700
  13. M. Sivilotti, "Wiring Considerations in Analog VLSI Systems with Application to Field-Programmable Networks," Ph.D. dissertation, Computation and Neural Systems, California Inst. Technol., Pasadena, CA, USA, 1991.
  14. M. Mahowald, "VLSI Analogs of Neuronal Visual Processing: A Synthesis of Form and Function," Ph.D. dissertation, Computation and Neural Systems, California Inst. Technol., Pasadena, CA, USA, 1992.
  15. G. Indiveri et al., "A Reconfigurable Neuromorphic VLSI Multi-chip System Applied to Visual Motion Computation," in Proc. Int. Conf. Microelectron. Neural, Fuzzy Bio-Inspired Syst., Granada, Spain, Apr. 1999, pp. 37-44.
  16. B.V. Benjamin et al., "Neurogrid: A Mixed-Analog-Digital Multichip System for Large-Scale Neural Simulations," Proc. IEEE, vol. 102, no. 5, May 2014, pp. 699-716. https://doi.org/10.1109/JPROC.2014.2313565
  17. S. Carrillo et al., "Scalable Hierarchical Network-on-Chip Architecture for Spiking Neural Network Hardware Implementations," IEEE Trans. Parallel Distrib. Syst., vol. 24, no. 12, Dec. 2013, pp. 2451-246. https://doi.org/10.1109/TPDS.2012.289
  18. E. Painkras et al., "SpiNNaker: A 1-W 18-Core System-on-chip for Massively-Parallel Neural Network Simulation," IEEE J. Solid-State Circuits, vol. 48, no. 8, Aug. 2013, pp. 1943-1953. https://doi.org/10.1109/JSSC.2013.2259038
  19. J. Park et al., "Hierarchical address event routing for reconfigurable large-scale neuromorphic systems," IEEE Trans. Neural Netw. Learn. Syst., vol. 28, no. 10, Oct. 2017, pp. 2408-2422. https://doi.org/10.1109/TNNLS.2016.2572164
  20. Cognitive Anteater Robotics Laboratory, "CARLsim: a GPU-Accelerated SNN Simulator," Mar. 2017. http://www.socsci.uci.edu/-jkrichma/CARLsim/
  21. The Neural Simulation Technology Initiative, http://nest-simulator.org/
  22. Brian, "The Brian Spiking Neural Network Simulator," http://briansimulator.org/