J. Semicond. >  Just Accepted

RESEARCH HIGHLIGHTS

Memristor-based energy-efficient signal processing: recent progress and technology trend

Jingyuan Huang§, Yunrui Jiao§, Han Zhao, Xingchu Li, Bin Gao, He Qian, Jianshi Tang and Huaqiang Wu

+ Author Affiliations

 Corresponding author: Jianshi Tang, jtang@tsinghua.edu.cn

DOI: 10.1088/1674-4926/26020062CSTR: 32376.14.1674-4926.26020062

PDF

Turn off MathJax



[1]
Huo Q, Yang Y M, Wang Y M, et al. A computing-in-memory macro based on three-dimensional resistive random-access memory. Nat Electron, 2022, 5(7): 469 doi: 10.1038/s41928-022-00795-x
[2]
Zhao H, Liu Z W, Tang J S, et al. Energy-efficient high-fidelity image reconstruction with memristor arrays for medical diagnosis. Nat Commun, 2023, 14: 2276 doi: 10.1038/s41467-023-38021-7
[3]
Wang C, Ruan G J, Yang Z Z, et al. Parallel in-memory wireless computing. Nat Electron, 2023, 6(5): 381 doi: 10.1038/s41928-023-00965-5
[4]
Liu Z W, Mei J, Tang J S, et al. A memristor-based adaptive neuromorphic decoder for brain–computer interfaces. Nat Electron, 2025, 8(4): 362 doi: 10.1038/s41928-025-01340-2
[5]
Christensen D V, Dittmann R, Linares-Barranco B, et al. 2022 roadmap on neuromorphic computing and engineering. Neuromorph Comput Eng, 2022, 2(2): 022501 doi: 10.1088/2634-4386/ac4a83
[6]
Sheridan P M, Cai F X, Du C, et al. Sparse coding with memristor networks. Nat Nanotechnol, 2017, 12(8): 784 doi: 10.1038/nnano.2017.83
[7]
Liu Z W, Tang J S, Gao B, et al. Neural signal analysis with memristor arrays towards high-efficiency brain–machine interfaces. Nat Commun, 2020, 11: 4234 doi: 10.1038/s41467-020-18105-4
[8]
Li C, Hu M, Li Y N, et al. Analogue signal and image processing with large memristor crossbars. Nat Electron, 2018, 1(1): 52 doi: 10.1038/s41928-017-0002-z
[9]
Cai F X, Correll J M, Lee S H, et al. A fully integrated reprogrammable memristor–CMOS system for efficient multiply–accumulate operations. Nat Electron, 2019, 2(7): 290 doi: 10.1038/s41928-019-0270-x
[10]
Li J C, Tian J, Lin Y D, et al. Memristive floating-point Fourier neural operator network for efficient scientific modeling. Sci Adv, 2025, 11(25): eadv4446 doi: 10.1126/sciadv.adv4446
[11]
Zuo P S, Wang Q S, Luo Y B, et al. Precise and scalable analogue matrix equation solving using resistive random-access memory chips. Nat Electron, 2025, 8(12): 1222 doi: 10.1038/s41928-025-01477-0
[12]
Zhao H, Wang L, Zhou Y Z, et al. In situ spectral reconstruction based on a memristor chip for energy-efficient computational spectrometry. Nat Electron, 2026: 1222
[13]
Jiao Y R, Zhao H, Tang J S, et al. A memristor-based energy-efficient compressed sensing accelerator with hardware–software co-optimization for edge computing. Natl Sci Rev, 2026, 13(1): nwaf499 doi: 10.1093/nsr/nwaf499
[14]
Jiang M R, Shan K Y, He C P, et al. Efficient combinatorial optimization by quantum-inspired parallel annealing in analogue memristor crossbar. Nat Commun, 2023, 14: 5927 doi: 10.1038/s41467-023-41647-2
[15]
Wang S Q, Luo Y B, Zuo P S, et al. In-memory analog solution of compressed sensing recovery in one step. Sci Adv, 2023, 9(50): eadj2908 doi: 10.1126/sciadv.adj2908
[16]
Yue W S, Zhang T, Jing Z K, et al. A scalable universal Ising machine based on interaction-centric storage and compute-in-memory. Nat Electron, 2024, 7(10): 904 doi: 10.1038/s41928-024-01228-7
[17]
Wang Z X, Song W H, Wang T, et al. Real-time signal processing enabled by fused networks on a memristor-based system on a chip. Sci Adv, 2025, 11(30): eadv3436 doi: 10.1126/sciadv.adv3436
[18]
Huang Y, He C Y, Ling Y Z, et al. Radiofrequency signal processing with a memristive system-on-a-chip. Nat Electron, 2025, 8(7): 587 doi: 10.1038/s41928-025-01409-y
[19]
Mannocci P, Zucchelli C, Andreoli I, et al. A fully integrated analogue closed-loop in-memory computing accelerator based on static random-access memory. Nat Electron, 2026, 9(2): 200 doi: 10.1038/s41928-025-01549-1
[20]
Chen J H, Zheng X J, Tang J S, et al. Microscopic modeling and optimization of NbOx Mott memristor for artificial neuron applications. IEEE Trans Electron Devices, 2022, 69(12): 6686 doi: 10.1109/TED.2022.3212325
[21]
Yuan R, Tiw P J, Cai L, et al. A neuromorphic physiological signal processing system based on VO(2) memristor for next-generation human-machine interface. Nat Commun, 2023, 14(1): 3695 doi: 10.1038/s41467-023-39430-4
[22]
Park W, Song H C, Kim E Y, et al. Frequency switching neuristor for realizing intrinsic plasticity and enabling robust neuromorphic computing. Adv Mater, 2025, 37(44): e02255 doi: 10.1002/adma.202502255
[23]
Li C, Graves C E, Sheng X, et al. Analog content-addressable memories with memristors. Nat Commun, 2020, 11: 1638 doi: 10.1038/s41467-020-15254-4
[24]
Pedretti G, Graves C E, Serebryakov S, et al. Tree-based machine learning performed in-memory with memristive analog CAM. Nat Commun, 2021, 12: 5806 doi: 10.1038/s41467-021-25873-0
Fig. 1.  (Color online) Evolution of memristor-based signal processing in the past decade.

DownLoad: Full-Size Img PowerPoint
Fig. 2.  Conceptual roadmap of paradigm expansions in memristor-based signal processing. As memristor arrays scale from kilo-bit to mega-bit levels, the application landscape has evolved along three fundamental parts. (a) Temporal shift: from static signal processing (e.g., sparse coding[6]) to real-time spatiotemporal stream processing (e.g., video analytics[17]). Adapted from Ref. [6] and Ref.[17], respectively. (b) Complexity shift: from error-tolerant classification tasks (e.g., neural signal state identification[7]) to high-fidelity regression and inverse problems (e.g., computed tomography (CT) image reconstruction[2]). Adapted from Ref. [7] and Ref. [2], respectively. (c) Architecture shift: from task-specific accelerators with restricted topologies (e.g., dense crossbar arrays for combinatorial optimization[16]) to more versatile, reconfigurable computing engines capable of handling sparse graphs and diverse workloads. Adapted from Ref. [16].

DownLoad: Full-Size Img PowerPoint

Table 1.   Summary of key specifications and performance benchmarks of representative memristor-based signal processing systems.

Reference Array
size		 Technology
node		 Implemented algorithms Applications Computing accuracy Performance benchmark
Nat. Nano. 2017[6] 1 kb 40 nm LCA Sparse coding MSE:1.93×10−3/
2.23×10−3 (4bit)		 ~16× lower in energy
consumption
Nat. Elec. 2018[8] 8 kb 40 nm DCT and filtering Image compression and filtering STD: 0.46% (5~8bit) Energy efficiency:
119.7 TOPS/W
Nat. Elec. 2019[9] 5.5 kb 180 nm LCA Breast cancer identification Accuracy: 94.6%/96.8% Power dissipation:
0.27 μW/feature
Nat. Comm. 2020[7] 1 kb 130 nm FIR filtering Epilepsy-related neural signal analysis Accuracy: 93.46%/96.41% 400× higher in
power efficiency
Nat. Elec. 2022[1] 2 kb 55 nm Filtering MRI edge detection Accuracy: 90.54%/91.92% Energy efficiency
62.11 TOPS/W (IN1bit-W2bit)
Nat. Comm. 2023[14] 4 kb 180 nm Simulate annealing Combinatorial optimization Time to Solution: 10.8 μs 24.8 nJ /solution
Nat. Comm. 2023[2] 2 kb×8 130 nm DFT Medical image reconstruction PSNR: 22.38 dB/22.52 dB 153× higher in energy efficiency
Nat. Elec. 2023[3] 1 kb 180 nm Analog MVM MIMO bit error rate: 0/480 Energy efficiency
222 TOPS/W
Sci. Adv. 2023[15] 0.25 kb 45 nm LCA Image reconstruction of 2D image PSNR: 26.83 dB/30.71 dB ~0.1 mJ/iteration
Nat. Elec. 2024[16] 16 kb 40 nm Interaction-centric storage Combinatorial optimization Speed: 442–1450× 4.1 × 105–6.0 × 105× lower in energy consumption
Sci. Adv. 2025[10] 4 kb×8 130 nm FNO Scientific modeling (1D burger's eq/3D thermal conductivity) Accuracy: 99.6%/99.8% (1D) Energy efficiency
78 GOPS/W (FP32)
Nat. Elec. 2025[4] 128 kb 130 nm TRCA Brain-computer interface for drone control Accuracy: 85.17%/86.08% 1643× higher in power efficiency
Sci. Adv. 2025[17] 64 kb×10 65 nm DFT and MLP Real-time video processing and classification PSNR: 33.49 dB 49× higher in power efficiency
Nat. Elec. 2025[18] 64 kb×10 65 nm DFT and CNN RF spectrum analysis & transmitter identification & anomaly detection Accuracy: 90.44%/92.29% Energy efficiency
21 TOPS/W
Nat. Elec. 2025[11] 64 b&
1 Mb		 40 nm Inverse solver MIMO Residual error: 1×
10−7 (9 iters 24 bits)		 Energy efficiency
50 GOPS/W (N=16)
Nat. Elec. 2026[12] 576 kb 28 nm Pseudo-inverse calculation Computational spectrometry PSNR: 30.87 dB 126× higher in power efficiency
DownLoad: CSV
[1]
Huo Q, Yang Y M, Wang Y M, et al. A computing-in-memory macro based on three-dimensional resistive random-access memory. Nat Electron, 2022, 5(7): 469 doi: 10.1038/s41928-022-00795-x
[2]
Zhao H, Liu Z W, Tang J S, et al. Energy-efficient high-fidelity image reconstruction with memristor arrays for medical diagnosis. Nat Commun, 2023, 14: 2276 doi: 10.1038/s41467-023-38021-7
[3]
Wang C, Ruan G J, Yang Z Z, et al. Parallel in-memory wireless computing. Nat Electron, 2023, 6(5): 381 doi: 10.1038/s41928-023-00965-5
[4]
Liu Z W, Mei J, Tang J S, et al. A memristor-based adaptive neuromorphic decoder for brain–computer interfaces. Nat Electron, 2025, 8(4): 362 doi: 10.1038/s41928-025-01340-2
[5]
Christensen D V, Dittmann R, Linares-Barranco B, et al. 2022 roadmap on neuromorphic computing and engineering. Neuromorph Comput Eng, 2022, 2(2): 022501 doi: 10.1088/2634-4386/ac4a83
[6]
Sheridan P M, Cai F X, Du C, et al. Sparse coding with memristor networks. Nat Nanotechnol, 2017, 12(8): 784 doi: 10.1038/nnano.2017.83
[7]
Liu Z W, Tang J S, Gao B, et al. Neural signal analysis with memristor arrays towards high-efficiency brain–machine interfaces. Nat Commun, 2020, 11: 4234 doi: 10.1038/s41467-020-18105-4
[8]
Li C, Hu M, Li Y N, et al. Analogue signal and image processing with large memristor crossbars. Nat Electron, 2018, 1(1): 52 doi: 10.1038/s41928-017-0002-z
[9]
Cai F X, Correll J M, Lee S H, et al. A fully integrated reprogrammable memristor–CMOS system for efficient multiply–accumulate operations. Nat Electron, 2019, 2(7): 290 doi: 10.1038/s41928-019-0270-x
[10]
Li J C, Tian J, Lin Y D, et al. Memristive floating-point Fourier neural operator network for efficient scientific modeling. Sci Adv, 2025, 11(25): eadv4446 doi: 10.1126/sciadv.adv4446
[11]
Zuo P S, Wang Q S, Luo Y B, et al. Precise and scalable analogue matrix equation solving using resistive random-access memory chips. Nat Electron, 2025, 8(12): 1222 doi: 10.1038/s41928-025-01477-0
[12]
Zhao H, Wang L, Zhou Y Z, et al. In situ spectral reconstruction based on a memristor chip for energy-efficient computational spectrometry. Nat Electron, 2026: 1222
[13]
Jiao Y R, Zhao H, Tang J S, et al. A memristor-based energy-efficient compressed sensing accelerator with hardware–software co-optimization for edge computing. Natl Sci Rev, 2026, 13(1): nwaf499 doi: 10.1093/nsr/nwaf499
[14]
Jiang M R, Shan K Y, He C P, et al. Efficient combinatorial optimization by quantum-inspired parallel annealing in analogue memristor crossbar. Nat Commun, 2023, 14: 5927 doi: 10.1038/s41467-023-41647-2
[15]
Wang S Q, Luo Y B, Zuo P S, et al. In-memory analog solution of compressed sensing recovery in one step. Sci Adv, 2023, 9(50): eadj2908 doi: 10.1126/sciadv.adj2908
[16]
Yue W S, Zhang T, Jing Z K, et al. A scalable universal Ising machine based on interaction-centric storage and compute-in-memory. Nat Electron, 2024, 7(10): 904 doi: 10.1038/s41928-024-01228-7
[17]
Wang Z X, Song W H, Wang T, et al. Real-time signal processing enabled by fused networks on a memristor-based system on a chip. Sci Adv, 2025, 11(30): eadv3436 doi: 10.1126/sciadv.adv3436
[18]
Huang Y, He C Y, Ling Y Z, et al. Radiofrequency signal processing with a memristive system-on-a-chip. Nat Electron, 2025, 8(7): 587 doi: 10.1038/s41928-025-01409-y
[19]
Mannocci P, Zucchelli C, Andreoli I, et al. A fully integrated analogue closed-loop in-memory computing accelerator based on static random-access memory. Nat Electron, 2026, 9(2): 200 doi: 10.1038/s41928-025-01549-1
[20]
Chen J H, Zheng X J, Tang J S, et al. Microscopic modeling and optimization of NbOx Mott memristor for artificial neuron applications. IEEE Trans Electron Devices, 2022, 69(12): 6686 doi: 10.1109/TED.2022.3212325
[21]
Yuan R, Tiw P J, Cai L, et al. A neuromorphic physiological signal processing system based on VO(2) memristor for next-generation human-machine interface. Nat Commun, 2023, 14(1): 3695 doi: 10.1038/s41467-023-39430-4
[22]
Park W, Song H C, Kim E Y, et al. Frequency switching neuristor for realizing intrinsic plasticity and enabling robust neuromorphic computing. Adv Mater, 2025, 37(44): e02255 doi: 10.1002/adma.202502255
[23]
Li C, Graves C E, Sheng X, et al. Analog content-addressable memories with memristors. Nat Commun, 2020, 11: 1638 doi: 10.1038/s41467-020-15254-4
[24]
Pedretti G, Graves C E, Serebryakov S, et al. Tree-based machine learning performed in-memory with memristive analog CAM. Nat Commun, 2021, 12: 5806 doi: 10.1038/s41467-021-25873-0

    • Search

    Advanced Search >>

    GET CITATION

    shu

    Export: BibTex EndNote

    Article Metrics

    Article views: 13 Times PDF downloads: 0 Times Cited by: 0 Times

    History

    Received: 19 February 2026 Revised: 21 April 2026 Online: Accepted Manuscript: 03 June 2026

    Catalog

      Email This Article

      User name:
      Email:*请输入正确邮箱
      Code:*验证码错误
      Send
      Article Navigation >  Journal of Semiconductors  > 2026  > Accepted Manuscript
      Jingyuan Huang, Yunrui Jiao, Han Zhao, Xingchu Li, Bin Gao, He Qian, Jianshi Tang, Huaqiang Wu. Memristor-based energy-efficient signal processing: recent progress and technology trend[J]. Journal of Semiconductors, 2026, In Press. doi: 10.1088/1674-4926/26020062 ****J Y Huang, Y R Jiao, H Zhao, X C Li, B Gao, H Qian, J S Tang, and H Q Wu, Memristor-based energy-efficient signal processing: recent progress and technology trend[J]. J. Semicond., 2026, accepted doi: 10.1088/1674-4926/26020062
      Citation:
      Jingyuan Huang, Yunrui Jiao, Han Zhao, Xingchu Li, Bin Gao, He Qian, Jianshi Tang, Huaqiang Wu. Memristor-based energy-efficient signal processing: recent progress and technology trend[J]. Journal of Semiconductors, 2026, In Press. doi: 10.1088/1674-4926/26020062 ****
      J Y Huang, Y R Jiao, H Zhao, X C Li, B Gao, H Qian, J S Tang, and H Q Wu, Memristor-based energy-efficient signal processing: recent progress and technology trend[J]. J. Semicond., 2026, accepted doi: 10.1088/1674-4926/26020062
      shu

      Memristor-based energy-efficient signal processing: recent progress and technology trend

      DOI: 10.1088/1674-4926/26020062
      CSTR: 32376.14.1674-4926.26020062

      • School of Integrated Circuits, Beijing Advanced Innovation Center for Integrated Circuits, BNRist, Tsinghua University, Beijing 100084, China

      More Information
      • Jingyuan Huang received his bachelor’s degree from Tsinghua University, Beijing, China, in 2022. He is currently pursuing the master’s degree at the School of Integrated Circuits, Tsinghua University, Beijing, China. His research interests include neuromorphic computing systems and computational optics
      • Yunrui Jiao received his bachelor’s degree from Tsinghua University, Beijing, China, in 2024. He is currently pursuing the master’s degree at the School of Integrated Circuits, Tsinghua University, Beijing, China. His research interests include neuromorphic computing systems and brain-computer interface
      • Prof. Jianshi Tang is currently an Associate Professor and Vice Dean of the School of Integrated Circuits at Tsinghua University, where he received his bachelor’s degree in 2008. He received his PhD degree from UCLA in 2014, and worked at IBM Research in 2015-2019. He has been awarded the First Prize in Natural Science of the Ministry of Education of China, MIT TR35 China, IEEE Brain Best Paper Award, etc. His current research mainly focuses on emerging memory and neuromorphic computing. He has authored over 200 papers, including Nature Electronics, Nature Nanotechnology, Nature Materials, IEDM, VLSI, etc
      • Corresponding author: jtang@tsinghua.edu.cn
      • Received Date: 2026-02-19
      • Revised Date: 2026-04-21
        • Available Online: 2026-06-03

      Catalog

        Journal of Semiconductors © 2017 All Rights Reserved   京ICP备05085259号-2

        /

        DownLoad:  Full-Size Img  PowerPoint
        Return
        Return