The Journal of the Korea institute of electronic communication sciences (한국전자통신학회논문지)

Korea Institute of Electronic Communication Science (한국전자통신학회)
권호 표지
DOI QR코드

DOI QR Code

Design of High-Speed Sense Amplifier for In-Memory Computing

인 메모리 컴퓨팅을 위한 고속 감지 증폭기 설계

  • 김나현 (강원대학교 BIT의료융합학과) ;
  • 김정범 (강원대학교 전자공학과)
  • Received : 2023.07.23
  • Accepted : 2023.10.17
  • Published : 2023.10.31
Download PDF
⟨ Previous Next ⟩

Abstract

A sense amplifier is an essential peripheral circuit for designing a memory and is used to sense a small differential input signal and amplify it into digital signal. In this paper, a high-speed sense amplifier applicable to in-memory computing circuits is proposed. The proposed circuit reduces sense delay time through transistor Mtail that provides an additional discharge path and improves the circuit performance of the sense amplifier by applying m-GDI (: modified Gate Diffusion Input). Compared with previous structure, the sense delay time was reduced by 16.82%, the PDP(: Power Delay Product) by 17.23%, the EDP(: Energy Delay Product) by 31.1%. The proposed circuit was implemented using TSMC's 65nm CMOS process, while its feasibility was verified through SPECTRE simulation in this study.

감지 증폭기는 메모리 설계에 필수적인 주변 회로로서, 작은 차동 입력 신호를 감지하여 디지털 신호로 증폭하기 위해 사용된다. 본 논문에서는 인 메모리 컴퓨팅 회로에서 활용 가능한 고속 감지 증폭기를 제안하였다. 제안하는 회로는 추가적인 방전 경로를 제공하는 트랜지스터 Mtail을 통해 감지 지연 시간을 감소시키고, m-GDI(:modified Gate Diffusion Input)를 적용하여 감지 증폭기의 회로 성능을 개선하였다. 기존 구조와 비교했을 때 감지 지연 시간은 16.82% 감소하였으며, PDP(: Power Delay Product)는 17.23%, EDP(: Energy Delay Product)은 31.1%가 감소하는 결과를 보였다. 제안하는 회로는 TSMC의 65nm CMOS 공정을 사용하여 구현하였으며 SPECTRE 시뮬레이션을 통해 본 연구의 타당성을 검증하였다.

Keywords

Acknowledgement

본 논문은 2023년도 정부(산업통상자원부)의 재원으로 한국산업기술진흥원의 지원을 받아 수행된 연구임(P0017011, 2023년 산업혁신인재성장지원사업)

References

  1. A. Agrawal, A. Jaiswal, C. Lee and K. Roy, "X-SRAM: Enabling In-Memory Boolean Computations in CMOS Static Random Access Memories," IEEE Transactions on Circuits and Systems I: Regular Papers, vol. 65, no. 12, Dec. 2018, pp. 4219-4232. https://doi.org/10.1109/TCSI.2018.2848999
  2. J. P. Kulkarni, A. Goel, P. Ndai and K. Roy, "A Read-Disturb-Free, Differential Sensing 1R/1W Port, 8T Bitcell Array," IEEE Transactions on Very Large Scale Integration (VLSI) Systems, vol. 19, no. 9, Sept. 2011, pp. 1727-1730. https://doi.org/10.1109/TVLSI.2010.2055169
  3. S. Song and Y. Kim, "Novel In-Memory Computing Adder Using 8+T SRAM." Electronics, vol. 11, no. 6, Mar 2022.
  4. J. Chen, W. Zhao, Y. Wang, Y. Shu, W. Jiang and Y. Ha, "A Reliable 8T SRAM for High-Speed Searching and Logic-in-Memory Operations," IEEE Transactions on Very Large Scale Integration (VLSI) Systems, vol. 30, no. 6, June 2022, pp. 769-780. https://doi.org/10.1109/TVLSI.2022.3164756
  5. J. Song, Y. Wang, X. Tang, R. Wang and R. Huang, "A 16Kb Transpose 6T SRAM In-Memory-Computing Macro based on Robust Charge-Domain Computing," 2021 IEEE Asian Solid-State Circuits Conference (A-SSCC), Busan, Korea, 2021, pp. 1-3.
  6. C. Yu, T. Yoo, K. T. C. Chai, T. Kim and B. Kim, "A 65-nm 8T SRAM Compute-in-Memory Macro With Column ADCs for Processing Neural Networks," IEEE Journal of Solid-State Circuits, vol. 57, no. 11, Nov. 2022, pp. 3466-3476. https://doi.org/10.1109/JSSC.2022.3162602
  7. C. Hong and J. Kim, "Design of In-Memory Computing Adder Using Low-Power 8+T SRAM," J. of the Korea Institute of Electronic Communication Sciences, vol. 18, no. 2, Apr. 2023, pp. 291-298
  8. T. Na, S. Woo, J. Kim, H. Jeong and S. Jung, "Comparative Study of Various Latch-Type Sense Amplifiers," IEEE Transactions on Very Large Scale Integration (VLSI) Systems, vol. 22, no. 2, Feb. 2014, pp. 425-429. https://doi.org/10.1109/TVLSI.2013.2239320
  9. R. D. Kadhao, S. R. Kala, K. Y. B. Nithin, M. H. Vasantha and D. Dwivedi, "A 2.5 GHz, 1-Kb SRAM with Auxiliary Circuit Assisted Sense Amplifier in 65-nm CMOS Process," 2023 36th International Conference on VLSI Design and 2023 22nd International Conference on Embedded Systems (VLSID), Hyderabad, India, 2023, pp. 1-6.
  10. K. Soundrapandiyan, S. K. Vishvakarma and B. S. Reniwal, "Enabling Energy-Efficient In-Memory Computing With Robust Assist-Based Reconfigurable Sense Amplifier in SRAM Array," IEEE Journal on Emerging and Selected Topics in Circuits and Systems, vol. 13, no. 1, pp. 445-455.
  11. A. Morgenshtein, A. Fish and I. A. Wagner, "Gate-diffusion input (GDI): a power-efficient method for digital combinatorial circuits," 2002 IEEE transactions on very large scale integration (VLSI) systems, vol. 10, no. 5, Oct. 2022, pp. 566-581. https://doi.org/10.1109/TVLSI.2002.801578
  12. U. Ramadass and P. Dhavachelvan, "Modified gate diffusion input technique: a new technique for enhancing performance in full adder circuits" Procedia Technology, vol. 6, 2012, pp. 74-81. https://doi.org/10.1016/j.protcy.2012.10.010