A Methodology to Improve Linearity of Analog RRAM for Neuromorphic Computing
The DOI matches its Crossref record; 4 of the 4 references it cites were found in a registry or an index.
Abstract
The conductance tuning linearity is an important parameter of analog RRAM for neuromorphic computing. This work presents a novel methodology to improve the conductance tuning linearity of the filamentary RRAM. An electro-thermal modulation layer is designed and introduced to control the distribution of electric field and temperature in the filament region. For the first time, a HfOxbased RRAM is demonstrated with linear analog SET, linear analog RESET, 50ns speed, 10× analog tuning window, 100kΩ on-state resistance, and high temperature retention for multilevel states. The excellent performances of the analog RRAM devices enable high accuracy online learning in a neural network.
Overview
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
In the citing paper
“The RRAM device stack consists of a titanium nitride (TiN) bottom-electrode layer, a hafnium oxide (HfOx) switching layer, a tantalum oxide (TaOx) thermal-enhancement layer52 and a TiN top-electrode layer.”
From A compute-in-memory chip based on resistive random-access memory (2022) · cites this paper as [52]
In the citing paper
“Second-order memristors have been actively demonstrated in previous studies, showing that oxide-based memristors can be largely affected not only by the filament radius, but also by the device temperature.34,37”
Citations
Cites
Nanoscale Memristor Device as Synapse in Neuromorphic Systems
Jo, S. H., Chang, T., Ebong, I., Bhadviya, B. B., Mazumder, P., & Lu, W. (2010) · Nano Letters · doi:10.1021/nl904092h · Cited by 4,174
Woo, J., Moon, K., Song, J., Lee, S., Kwak, M., Park, J., & Hwang, H. (2016) · IEEE Electron Device Letters · doi:10.1109/led.2016.2582859 · Cited by 512
Improving Analog Switching in HfO x -Based Resistive Memory With a Thermal Enhanced Layer
Wu, W., Wu, H., Gao, B., Deng, N., Yu, S., & Qian, H. (2017) · IEEE Electron Device Letters · doi:10.1109/led.2017.2719161 · Cited by 282
Self-citationFully parallel write/read in resistive synaptic array for accelerating on-chip learning
Gao, L., Wang, I.-T., Chen, P.-Y., Vrudhula, S., Seo, J.-S., Cao, Y., Hou, T.-H., & Yu, S. (2015) · Nanotechnology · doi:10.1088/0957-4484/26/45/455204 · Cited by 141
Self-citation
Research works printed but not linked: 6
- park, 2013, IEDM, 625
- landon, 2015, APL, 107, 23108
- chen, 2017, IEDM, 135
- lee, 2015, IEDM, 91
- gao, 2017, IEDM, 91
- wu, 2017, IEDM, 274
Cited by
Items 201–224 of 224
Memristor Based Mixed-Accuracy Computation-in-Memory System
Wang, Z., Yu, R., & Gao, B. (2023) · IEEE International Conference on IC Design and Technology (ICICDT) · doi:10.1109/icicdt59917.2023.10332392
Self-citationWani, I. H., & Singh, A. (2023) · IEEE Silchar Subsection Conference (SILCON) · doi:10.1109/silcon59133.2023.10405237
“One practical approach to enhance linearity, symmetry, and the number of conductance states in RRAM for neuromorphic applications is to incorporate an electro-thermal modulation layer (ETML) or intermediate modulation layer (IML) [15].”
MethodKumar, D., Li, H., Singh, A., Rajbhar, M. K., Syed, A. M., Lee, H., & El‐Atab, N. (2024) · Research Explorer (The University of Manchester) · doi:10.48550/arxiv.2404.05701
AL2O3/ALOX Memristor with Nearly Ideal Synaptic Characteristics
He, Q., Wang, H., Bai, Y., Hu, J., Li, H., Ma, W., Xu, Y., Zhang, Y., & Yu, B. (2024) · China Semiconductor Technology International Conference (CSTIC) · doi:10.1109/cstic61820.2024.10531814
“Previous works have proposed various structures to optimize the formation of conductive filament [4-5], but these approaches often complicate the manufacturing process.”
BackgroundAn Isotropic Shift-Pointwise Network for Crossbar-Efficient Neural Network Design
Guan, Z., Li, B., Ren, Y., Niu, M., Huang, H., Chesi, G., Yu, H., & Wong, N. (2024) · Design, Automation, and Test in Europe · doi:10.23919/date58400.2024.10546743
“To this end, compute-in-memory (CIM) has emerged as a promising solution, with resistive random-access memory (RRAM) [3] being a leading contender due to its high density, reduced power, and compatibility with monolithic 3D integration.”
BackgroundHeo, D., Bang, H., & Ko, J. H. (2024) · ACM/IEEE Design Automation Conference (DAC) · doi:10.1145/3649329.3656535
“Non-volatile memories such as RRAM [27], PCM [11], FeFET [9], and ECRAM [25] are integrated into this structure, functioning as neuromorphic devices (NDs) or memristors.”
Brief Review of Neuron Typology and Analysis of the Use of Memristor Crossbars
Tokarev, A. A., & Khorin, I. A. (2024) · Russian Microelectronics · doi:10.1134/s1063739724600766
“The memristor cell uses a TiN/TaO x /HfO x /TiN material stack [51] and exhibits the ability to continuously tune the conductance during both potentiation (SET) and depression (RESET) by modulating the electric field and heat [77].”
Demonstration of Ferroelectric Tunnel Field-Effect Transistor for Low Power Synapse Device
Go, S., Lee, S., Son, J., Lee, D. K., Noh, H., Park, J. Y., Kim, S., Ahn, H., Kim, S., & Kim, S. (2025) · IEEE Transactions on Nanotechnology · doi:10.1109/tnano.2025.3595532
“7 shows a benchmark of this work (i.e., FeTFET and FeFET) in terms of the training accuracy and the energy consumption comparing with [38], [39], [40], [41], [42], [43], [44], [45], [46].”
InfluentialA Hybrid Encryption Framework for Federated Learning Accelerated on RRAM
Li, L., Mou, X., Ren, J., Yang, Q., & Wang, Y. (2025) · International Symposium of Electronics Design Automation (ISEDA) · doi:10.1109/iseda65950.2025.11100980
Wang, C., Chen, Z. M., & Huang, S. (2025) · Integration · doi:10.1016/j.vlsi.2025.102543
Vo, H. Q. N., Chowdhury, M. T. R., Ramanan, P., Tütüncüoğlu, G., Yang, J., Qiu, F., & Yildirim, M. (2025) · arXiv (Cornell University) · doi:10.48550/arxiv.2509.21137
“This improvement is primarily attributed to the physics of the device, which allows lower voltage and duration of programming signals required to adjust the conductance state of the memory cell [58], which in turn enables more efficient analog matrix-vector operations.”
Fabrication and Characterization of Thin Film Devices for Use in Neuromorphic Computing
Christensen, A. (2025) · OSTI OAI (U.S. Department of Energy Office of Scientific and Technical Information)
Electric Field-Moderated Filament Control for Cryogenic RRAM Synapse
Kim, Y., & Woo, J. (2025) · IEEE Electron Device Letters · doi:10.1109/led.2025.3632969
“Among various emerging memories that can be artificial synapse candidates, resistive random-access memories (RRAMs) have garnered significant interest owing to its complementary metal-oxide-semiconductor compatibility, excellent scalability, and rapid switching characteristics [9], [10], [11].”
Silicon Based Nano-electronic Synaptic Device for Neuromorphic Hardware
Sikder, O. (2024) · IUScholarWorks (Indiana University) · doi:10.25394/pgs.26037304
유한요소법 시뮬레이션을 통한 리튬 이온 기반 전기화학 시냅스 소자 성능 평가 및 개선 연구
최진하 (2023) · Seoul National University Open Repository (Seoul National University)
Resistive Switching Systems for In-Memory Computation and Artificial Intelligence
Mukherjee, S. (2026) · Institute of Physics eBooks · doi:10.1088/978-0-7503-6169-9
Performance Analysis and Optimization of Fructose Memristor-Based Neuromorphic Systems
Uppaluru, H., Jiban, M. S. M., Riam, S. Z., Zhao, F., & Wang, J. (2026) · IEEE Journal on Emerging and Selected Topics in Circuits and Systems · doi:10.1109/jetcas.2026.3667550
Enhancing Post-Training Quantization Robustness in RRAM-Based CIM via σ-Clipping of Weights
Oh, B., Youn, S., Kim, H., & Kim, T. (2026) · ACS Applied Electronic Materials · doi:10.1021/acsaelm.6c00222
ReBIT: A ReRAM-Based In-Situ Training Accelerator with Robustness Against Stochasticity
Dang, P., Wang, W., He, Y., & Li, H. (2026) · Design, Automation, and Test in Europe · doi:10.23919/date69613.2026.11539502
“Simulations are based on TaOx/HfOx devices [7] fabricated with 32nm technology and incorporate the conductance update model proposed in [5].”
Fully Sputtered HfO y /HfO x RRAM: Improved Synaptic Behavior for Neuromorphic Training Accelerators
Kim, Y., Jeon, S., Lim, S., & Woo, J. (2026) · IEEE Transactions on Electron Devices · doi:10.1109/ted.2026.3714373
“It has also been reported that analog-type switching emerges only when the entire conductive filament (CF) evolves gradually rather than through localized abrupt formation, consistent with previous findings in previous works [8], [16].”
Moon, K., Kim, B. S., & Kim, H. J. (2026) · ACS Applied Electronic Materials · doi:10.1021/acsaelm.6c00637
Compact Model of Filament-Free Bulk RRAM
Zhou, Y., Zhou, Y., Kumar, A., Sun, F., Lo, Y., Cubukcu, E., & Kuzum, D. (2026) · IEEE Transactions on Electron Devices · doi:10.1109/ted.2026.3714752
“Large nonlinearity can significantly degrade learning accuracy in neural networks [14].”
Pyo, S. G. (2026) · Micromachines · doi:10.3390/mi17091032
“As one example of ongoing work in this area, Wu [76] et al. proposed a new electro-thermal methodology to improve the linearity of filamentary analog RRAM and demonstrated devices with excellent linearity for both SET and RESET operations.”
Energy-efficient Resistive In-memory Computing Architectures
Roy, S. K. (2024) · Illinois Digital Environment for Access to Learning and Scholarship (University of Illinois at Urbana-Champaign)
Metadata
- Authors
- Wu, W., Wu, H., Gao, B., Yao, P., Zhang, X., Peng, X., Yu, S., & Qian, H.
- Affiliations
- Tsinghua University (CN)
- Arizona State University (US)
- Venue
- 2018 IEEE Symposium on VLSI Technology
- Publisher
- IEEE
- Published
- 2018-06
- Pages
- 103-104
- OpenAlex
- W2899224990
- Cited by
- 197
- References
- 4
- Citations by year
- 2026: 8 · 2025: 20 · 2024: 26 · 2023: 27 · 2022: 37 · 2021: 26 · 2020: 38 · 2019: 13 · 2018: 2
- FWCI
- 31.50
- Type
- conference paper
- Language
- en
- Open access
- closed
- Locations
- Symposium on VLSI Technology (publishedVersion)
- Funders
- National Natural Science Foundation of China
- Beijing Innovation Center for Future Chip
- Topics
- Physical Sciences › Engineering › Electrical and Electronic Engineering › Advanced Memory and Neural Computing
- Physical Sciences › Engineering › Electrical and Electronic Engineering › Ferroelectric and Negative Capacitance Devices
- Physical Sciences › Engineering › Electrical and Electronic Engineering › Semiconductor materials and devices
- Keywords
- resistive random access memory, neuromorphic computing, hafnium oxide, multilevel memory, online learning, artificial neural networks
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.