High-Throughput In-Memory Computing for Binary Deep Neural Networks With Monolithically Integrated RRAM and 90-nm CMOS
Open accessThe DOI matches its Crossref record; 47 of the 55 references it cites were found in a registry or an index, and 8 are not checked yet.
Abstract
Deep neural network (DNN) hardware designs have been bottlenecked by conventional memories, such as SRAM due to density, leakage, and parallel computing challenges. Resistive devices can address the density and volatility issues but have been limited by peripheral circuit integration. In this work, we present a resistive RAM (RRAM)-based in-memory computing (IMC) design, which is fabricated in 90-nm CMOS with monolithic integration of RRAM devices. We integrated a 128 × 64 RRAM array with CMOS peripheral circuits, including row/column decoders and flash analog-to-digital converters (ADCs), which collectively become a core component for scalable RRAM-based IMC for large DNNs. To maximize IMC parallelism, we assert all 128 wordlines of the RRAM array simultaneously, perform analog computing along the bitlines, and digitize the bitline voltages using ADCs. The resistance distribution of low-resistance states is tightened by an iterative write-verify scheme. Prototype chip measurements demonstrate high binary DNN accuracy of 98.5% for MNIST and 83.5% for CIFAR-10 data sets, with 24 TOPS/W and 158 GOPS. This represents 22.3× and 10.1× improvements in throughput and energy-delay product (EDP), respectively, compared with the state-of-the-art literature, which can enable intelligent functionalities for area-/energy-constrained edge computing devices.
Overview
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
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?
The source backs part of the claim, not all of it — or backs it for a different population, condition or comparison.
In the citing paper
“So far, several groups and companies have claimed the realization of hybrid CMOS/memristor implementations of ANNs53–61, —from now on, memristive ANNs— with performance that is superior to that of fully-CMOS counterparts.”
From Hardware implementation of memristor-based artificial neural networks (2024) · cites this paper as [61]
Experimental — accuracy not yet measuredThe source backs part of the claim, not all of it — or backs it for a different population, condition or comparison.
In the citing paper
“Some groups prefer to incorporate a transistor in series to each MIM cell to obtain a better control over the currents through the device (i.e., improve conductance controllability and minimize sneak path currents)53–55,60,61,102,113,167,194,218–222,).”
From Hardware implementation of memristor-based artificial neural networks (2024) · cites this paper as [61]
Experimental — accuracy not yet measured
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Citations
Cites
Deep Residual Learning for Image Recognition
He, K., Zhang, X., Ren, S., & Sun, J. (2016) · 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2016.90 · Cited by 235,474 · cited 1×
“DEEP neural networks (DNNs) have been very successful in large-scale recognition and classification tasks [1]–[5], while state-of-the-art deep learning algorithms tend to present very deep and large network models [1]–[3].”
BackgroundGradient-based learning applied to document recognition
Lecun, Y., Bottou, L., Bengio, Y., & Haffner, P. (1998) · Proceedings of the IEEE · doi:10.1109/5.726791 · Cited by 58,805 · cited 12×
“We scanned in 2,000 test input vectors from MNIST BNN benchmark, and obtained a 2-”
MethodInfluentialDensely Connected Convolutional Networks
Huang, G., Liu, Z., Van Der Maaten, L., & Weinberger, K. Q. (2017) · 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2017.243 · Cited by 48,566
Learning Multiple Layers of Features from Tiny Images
Krizhevsky, A. (2024) · doi:10.57702/zp44cu3g · Cited by 23,395
Mastering the game of Go with deep neural networks and tree search
Silver, D., Huang, A., Maddison, C. J., Guez, A., Sifre, L., van den Driessche, G., Schrittwieser, J., Antonoglou, I., Panneershelvam, V., Lanctot, M., et al. (2016) · Nature · doi:10.1038/nature16961 · Cited by 15,612 · cited 1×
“DEEP neural networks (DNNs) have been very successful in large-scale recognition and classification tasks [1]–[5], while state-of-the-art deep learning algorithms tend to present very deep and large network models [1]–[3].”
BackgroundGoogle's Neural Machine Translation System: Bridging the Gap between Human and Machine Translation
Wu, Y., Schuster, M., Chen, Z., Le, Q. V., Norouzi, M., Macherey, W., Krikun, M., Cao, Y., Gao, Q., Macherey, K., et al. (2016) · arXiv · doi:10.48550/arxiv.1609.08144 · Cited by 5,319 · cited 1×
“This is especially true for embedded hardware applications, such as autonomous driving 6 , machine translation 7 , and smart wearable devices 8 , where severe constraints exist in performance, power, and area.”
BackgroundIn-Datacenter Performance Analysis of a Tensor Processing Unit
Jouppi, N. P., Young, C., Patil, N., Patterson, D., Agrawal, G., Bajwa, R., Bates, S., Bhatia, S., Boden, N., Borchers, A., et al. (2017) · Proceedings of the 44th Annual International Symposium on Computer Architecture · doi:10.1145/3079856.3080246 · Cited by 4,537
XNOR-Net: ImageNet Classification Using Binary Convolutional Neural Networks
Rastegari, M., Ordonez, V., Redmon, J., & Farhadi, A. (2016) · Lecture Notes in Computer Science · doi:10.1007/978-3-319-46493-0_32 · Cited by 4,408 · cited 7×
“Conventional binary RRAMs cannot effectively represent the positive and negative weight values (+1 and −1) in recent BNNs [8], [9] because the LRS and HRS values of binary RRAM devices are both positive.”
BackgroundMethodResultInfluentialA million spiking-neuron integrated circuit with a scalable communication network and interface
Merolla, P. A., Arthur, J. V., Alvarez-Icaza, R., Cassidy, A. S., Sawada, J., Akopyan, F., Jackson, B. L., Imam, N., Guo, C., Nakamura, Y., et al. (2014) · Science · doi:10.1126/science.1254642 · Cited by 4,147
Eyeriss: An Energy-Efficient Reconfigurable Accelerator for Deep Convolutional Neural Networks
Chen, Y.-H., Krishna, T., Emer, J. S., & Sze, V. (2017) · IEEE Journal of Solid-State Circuits · doi:10.1109/jssc.2016.2616357 · Cited by 3,247 · cited 13×
“An RRAM macro with multi-level sense amplifiers in 55nm CMOS logic process was recently reported 30 , but a relatively low accuracy of 81.83% accuracy for CIFAR-10 dataset was achieved with binary/ternary precision, and only 9 WLs are asserted simultaneously in the 256×512 sub-array, which limits further parallelism.”
BackgroundMethodInfluential
Cited by
Items 101–171 of 171
Kim, J.-H., Oh, Y. H., Kim, H., Lee, J. W., & Kim, J. (2023) · IEEE Transactions on Circuits & Systems II Express Briefs · doi:10.1109/tcsii.2023.3273290 · Cited by 5
“ANALOG CIM using nonvolatile memories such as ReRAM, MRAM, PCM, and FRAM [1], [2], [3], [4], [5], [6], [11], [12], [14], [15] has demonstrated its viability as an efficient edge inference compute by utilizing parallel bit-line (BL) processing.”
MethodKwon, D., Park, M., Kang, W.-M., Hwang, J., Koo, R., Bae, J., & Lee, J. (2023) · IEEE Transactions on Electron Devices · doi:10.1109/ted.2023.3287824 · Cited by 6
“Furthermore, vector-matrix multiplication (VMM) operations in HNNs are performed in an analog manner according to Kirchhoff’s laws, accelerating the inference speed of neural networks [11], [12], [13].”
MethodYeo, I., & Seo, J.-S. (2022) · Nature Electronics · doi:10.1038/s41928-022-00804-z · Cited by 5
Self-citationPrabhu, N. L., & Raghavan, N. (2022) · IEEE Access · doi:10.1109/access.2022.3219066 · Cited by 5
“The switching imperfection in RRAM due to stochastic distribution of oxygen vacancies is improved by performing a two-step write-verification scheme where the device is programmed-verified-reprogrammed through an iterative process to achieve the device improvement [18], [19], [20], [21], [22], [23].”
MethodOptimization of Bilayer Resistive Random Access Memory Based on Ti/HfO2/ZrO2/Pt
Sun, Z., Wang, P., Li, X., Chen, L., Yang, Y., & Wang, C. (2024) · Materials · doi:10.3390/ma17081852 · Cited by 5
“…type of memory device, which uses the characteristics of transition metal oxide (TMO) film material of a resistive layer to change its internal electronic structure under the action of electric field and converts back and forth between high resistance and low resistance to store information [1,2].”
BackgroundA 10-bit Two-Stage Pipeline SAR ADC in 55nm CMOS for Compute-in-Memory Applications
Xu, X., Ma, Y., Yu, X., Meng, F., Yu, W.-H., & Zhao, L. (2024) · IEEE Asia Pacific Conference on Circuits and Systems (APCCAS) · doi:10.1109/apccas62602.2024.10808421 · Cited by 5
Mitigating Adversarial Attack for Compute-in-Memory Accelerator Utilizing On-chip Finetune
Huang, S., Jiang, H., & Yu, S. (2021) · IEEE Non-Volatile Memory Systems and Applications Symposium (NVMSA) · doi:10.1109/nvmsa53655.2021.9628739 · Cited by 5
Self-citationLee, S., Kim, H., Yoo, H.-N., Kwon, D., & Lee, J. (2022) · Neurocomputing · doi:10.1016/j.neucom.2022.05.030 · Cited by 5
RIBoNN: Designing Robust In-Memory Binary Neural Network Accelerators
Kundu, S., Malhotra, A., Raha, A., Gupta, S. K., & Basu, K. (2022) · IEEE International Test Conference (ITC) · doi:10.1109/itc50671.2022.00061 · Cited by 5
“It should be pointed out that although a new technique for BNN acceleration using the XNOR-RRAM bitcell has recently been proposed [6], we observed that the impact of faults on it are similar to the baseline discussed.”
MethodCompute-in-RRAM with Limited On-chip Resources
Lu, A., Peng, X., & Yu, S. (2021) · IEEE International Conference on Artificial Intelligence Circuits and Systems (AICAS) · doi:10.1109/aicas51828.2021.9458481 · Cited by 6
“Therefore, ADC sharing by multiplexing between columns is widely used to alleviate the area requirement with the sacrifice of throughput [3, 10].”
MethodSelf-citationVQT-CiM: Accelerating Vector Quantization Enhanced Transformer with Ferroelectric Compute-in-Memory
Huang, X., Du, H., Zhou, M., Yan, Z., Zhuo, C., & Yin, X. (2025) · ACM/IEEE Design Automation Conference (DAC) · doi:10.1109/dac63849.2025.11133264 · Cited by 4
“Several studies have proposed CiM implementations as neural network accelerators, primarily categorized into non-volatile memory (NVM)-based [5, 6] and SRAM-based [7, 8] designs.”
XNOR Operation of Binary Neural Networks Using Nanoelectromechanical Memory Switches
Park, G. T., Lee, J. W., Woo, J. S., & Choi, W. Y. (2024) · IEEE Transactions on Electron Devices · doi:10.1109/ted.2024.3486267 · Cited by 4
“However, owing to the low ON / OFF resistance ratio of memristor-based eNVMs, both the inference accuracy and energy efficiency degrade during XNOR operations [15], [16], [17], [18], [19], [20].”
Kim, D., Jeong, H., Kim, S., Yoon, H., & Lee, K. J. (2025) · IEEE Transactions on Circuits and Systems I Regular Papers · doi:10.1109/tcsi.2025.3603076 · Cited by 4
“To address the energy and throughput limitations of A-to-D conversion, several quantization techniques [16], [32], [33], [34] have been introduced.”
Sparse Vector-Matrix Multiplication Acceleration in Diode-Selected Crossbars
Jao, N., Ramanathan, A. K., Sampson, J., & Narayanan, V. (2021) · IEEE Transactions on Very Large Scale Integration (VLSI) Systems · doi:10.1109/tvlsi.2021.3114186 · Cited by 4
Dabbagh, A., Karamimanesh, M., Hassanli, K., & Abiri, E. (2024) · The Journal of Supercomputing · doi:10.1007/s11227-024-06700-x · Cited by 4
“Deep neural networks (DNNs) significantly advance the field of AI due to the accuracy and efficiency of models, scalability, and interpretability [1–3].”
StoX-Net: Stochastic Processing of Partial Sums for Efficient In-Memory Computing DNN Accelerators
Rogers, E. G., Mugdho, S. S., Gupte, K. K., & Wang, C. (2024) · IEEE International Conference on Rebooting Computing (ICRC) · doi:10.1109/icrc64395.2024.10937005 · Cited by 7
“To ensure robust data communication, an analog cross-bar IMC macro is connected with other components through digital-to-analog converters (DAC) at the input and ADC at the crossbar output [2], [10].”
Statistical computing framework and demonstration for in-memory computing systems
Zhang, B., Deaville, P., & Verma, N. (2022) · ACM/IEEE Design Automation Conference (DAC) · doi:10.1145/3489517.3530557 · Cited by 5
“The framework is demonstrated via a prototype MRAM-based IMC chip in 22nm FD-SOI, and PyTorch training/execution libraries.”
InfluentialColumnar Learning Networks for Multisensory Spatiotemporal Learning
Yoo, S., Park, Y., Wang, Z., Wu, Y., Medepalli, S., Thio, W. J.-C., & Lu, W. D. (2022) · Advanced Intelligent Systems · doi:10.1002/aisy.202200179 · Cited by 7
Possibilities and Limitations of Memristor Crossbars for Neuromorphic Computing
Telminov, O. A., & Gornev, E. (2022) · Scientific School Dynamics of Complex Networks and their Applications (DCNA) · doi:10.1109/dcna56428.2022.9923302 · Cited by 4
“The practical ceiling on the dimensionality of memristive crossbars is 128x64 elements [5, 6], which is insufficient for the implementation of modern neural networks.”
BackgroundXMA2: A crossbar-aware multi-task adaption framework via 2-tier masks
Zhang, F., Yang, L., Meng, J., Seo, J.-S., Cao, Y. K., & Fan, D. (2022) · Frontiers in Electronics · doi:10.3389/felec.2022.1032485 · Cited by 4
“Attracted by the high energy efficiency, various ReRAM-based neural network accelerators have been proposed (Mittal, 2019; Song et al., 2017; Yin et al., 2020; Eckert et al., 2018).”
BackgroundMethodSelf-citationFDSOI-Based Analog Computing for Ultra-Efficient Hamming Distance Similarity Calculation
Mema, A., Thomann, S., Genßler, P. R., & Amrouch, H. (2023) · IEEE Transactions on Circuits and Systems I Regular Papers · doi:10.1109/tcsi.2023.3267837 · Cited by 5
“However, analog circuits have been mainly researched in the context of emerging non-volatile memories like PCM, RRAM, or FeFET [3], [6].”
BackgroundSIMBA: A Skyrmionic In-Memory Binary Neural Network Accelerator
Miriyala, V. P. K., Vishwanath, K. R., & Fong, X. (2020) · National University of Singapore · Cited by 4
Zanotti, T., Pavan, P., & Puglisi, F. M. (2023) · Artificial intelligence · doi:10.5772/intechopen.110340 · Cited by 3
Hasler, J., & Basu, A. (2025) · Neuromorphic Computing and Engineering · doi:10.1088/2634-4386/ad9b4a · Cited by 4
Hemaram, S., Ahmed, S. T., Mayahinia, M., Münch, C., & Tahoori, M. B. (2023) · IEEE VLSI Test Symposium (VTS) · doi:10.1109/vts56346.2023.10140077 · Cited by 4
“Many recent works show that non-volatile memristive devices, specifically multi-level cells such as ReRAMs, are promising for implementing the CiM paradigm [5]–[9].”
BackgroundDesign-time Reference Current Generation for Robust Spintronic-based Neuromorphic Architecture
Ahmed, S. T., Mayahinia, M., Hefenbrock, M., Münch, C., & Tahoori, M. B. (2023) · ACM Journal on Emerging Technologies in Computing Systems · doi:10.1145/3625556 · Cited by 4
Modeling and Simulation of RRAM With Carbon Nanotube Electrode
Wang, D., Zhu, J.-H., Liu, Y., Wang, G., & Zhao, W. (2023) · IEEE Transactions on Nanotechnology · doi:10.1109/tnano.2023.3341414 · Cited by 3
“CMOS compatibility, simple structure, and high switch speed [9], [10], [11], [12].”
BackgroundProcess and Runtime Variation Robustness for Spintronic-Based Neuromorphic Fabric
Ahmed, S. T., Mayahinia, M., Hefenbrock, M., Münch, C., & Tahoori, M. B. (2022) · IEEE European Test Symposium (ETS) · doi:10.1109/ets54262.2022.9810422 · Cited by 3
“However, existing solutions either only mitigate the process variations and cannot deal with run time variations [3]–[5], or they require costly on-the-fly re-mapping and re-training based on sensed temperature [6]–[8].”
BackgroundKingra, S. K., Parmar, V., Sharma, M. K., & Suri, M. (2022) · IEEE Transactions on Nanotechnology · doi:10.1109/tnano.2022.3193921 · Cited by 3
“synapses have been explored demonstrating error resilience to programming failures [12], [14].”
BackgroundShi, A., Zhang, Y., Han, L., Zhou, Z., Chen, Y., Liu, L., Shen, L., Huang, P. C., Liu, X., & Kang, J. (2024) · IEEE International Symposium on Circuits and Systems proceedings · doi:10.1109/iscas58744.2024.10558346 · Cited by 2
Choi, J., Park, H., Byun, Y., Seo, Y., & Kim, S. (2025) · The Journal of Chemical Physics · doi:10.1063/5.0298600 · Cited by 2
Lee, Y., Kim, S.-M., Kim, S., Shim, W., Kim, S., & Cho, S. (2025) · ACS Omega · doi:10.1021/acsomega.5c08019 · Cited by 2
Kim, J. I., So, M., Wang, W. J., Kim, T., Jeong, E. S., Sim, K., Park, H., Kim, S., Shin, Y.-G., Choi, J., et al. (2025) · Advanced Science · doi:10.1002/advs.202511801 · Cited by 2
Testbench on a Chip: A Yield Test Vehicle for Resistive Memory Devices
Upton, L. R., Lallement, G., Scott, M. D., Taylor, J., Radway, R. M., Rich, D., Nelson, M., Mitra, S., & Murmann, B. (2023) · International Symposium on Quality Electronic Design (ISQED) · doi:10.1109/isqed57927.2023.10129298 · Cited by 2
“(top-left) RRAM measurement setup with external read/write equipment [8]; (top-right) High-level depiction of early RRAM sample test chip designs.”
BackgroundOpen-RIMC: Open-Source RRAM-based IMC Co-Processor with Reconfigurable Logic Mapping
Parmar, V., Ray, A., Chithambara Moorthii, J., Mishra, R., Verma, D., Pandey, D., & Suri, M. (2023) · IEEE International Conference on Nanotechnology (IEEE-NANO) · doi:10.1109/nano58406.2023.10231247 · Cited by 3
“Recent works now focus on exploring inmemory computing (IMC) based architectures to perform BNN computations using SRAM [7], and NVM (non-volatile memory) technologies such as: Flash [8], RRAM [3], [9], [10], and MRAM [11].”
BackgroundThe effect of switching and cycle-to-cycle variations of RRAM on 4-bit encryption/decryption process
Nivetha, T., Bindu, B., & Kamsani, N. A. (2024) · Microelectronic Engineering · doi:10.1016/j.mee.2024.112244 · Cited by 2
Application of Business Intelligence Based on the Deep Neural Network in Credit Scoring
Feng, W., & Chen, M. (2022) · Security and Communication Networks · doi:10.1155/2022/2663668 · Cited by 2
RetractedSuper-steep synapses based on positive feedback devices for reliable binary neural networks
Kwon, D., Kim, H., Lee, K.-H., Hwang, J., Shin, W., Bae, J., Woo, S. Y., & Lee, J. (2023) · Applied Physics Letters · doi:10.1063/5.0131235 · Cited by 2
Zhai, X., Yin, W., Yang, Y., & Chen, W. (2022) · doi:10.1109/aces-china56081.2022.10064737 · Cited by 2
“Resistive random access memory (RRAM) is one of the emerging memories, which has great potential application for in-memory computing, on-chip storage and neuromorphic computing [1-3].”
BackgroundA Write System for Compact RRAM Memory Arrays Based on F-1T1R
Caselli, M., & Boni, A. (2024) · IEEE International Symposium on Circuits and Systems proceedings · doi:10.1109/iscas58744.2024.10558418 · Cited by 1
“The design in 90-nm CMOS, described in [22], has 128 rows and 64 columns, and it achieves A R ≈ 91%, much higher than the 24.5% from our system, on the same array size.”
InfluentialA Novel Current Comparator Enabling Large RRAM Crossbars for BNNs and PUFs
Rajendran, G., Basak, D., Deb, S., Wang, S., & Chattopadhyay, A. (2024) · IFIP/IEEE International Conference on Very Large Scale Integration · doi:10.1109/vlsi-soc62099.2024.10767833 · Cited by 1
“…emerging non-volatile memory (NVM) device technologies, such as Resistive Random Access Memory (RRAM), Phase Change Memory (PCM), Spin-Transfer Toque (STT), and Spin-Orbit Torque (SOT) Magnetic Random Access Memories (MRAM) are mainly driven by their in-memory computing (IMC) applications [1], [2].”
High Resistance and Low Variation 8Kb SOT-MRAM Compute-in-Memory Array Based on 1T-1R Cell Structure
Li, J. H., Sun, S., Huang, W., Qu, Q. L., Bail, J. Y., Jin, H., Kang, W., Li, X., Liu, H. X., Wang, G. F., et al. (2024) · Non-Volatile Memory Technology Symposium · doi:10.1109/ieeeconf63530.2024.10830849 · Cited by 1
Dudkin, A. P., Ryndin, E. A., & Andreeva, N. V. (2025) · Russian Microelectronics · doi:10.1134/s106373972560044x · Cited by 2
Evolution of Neural Network Models and Computing-in-Memory Architectures
Kumar, A., Devi, M., Singh, S. K., Singh, K., & Mishra, D. (2025) · Archives of Computational Methods in Engineering · doi:10.1007/s11831-025-10481-8 · Cited by 3
“While artificial neural networks have been traditionally implemented using CMOS technology [ 29], their performance remains fundamentally limited by the memory-processing separation dictated by the conventional von Neumann architecture [ 4].”
Marium, S. M., & Chen, S. (2026) · Microelectronics · doi:10.3390/microelectronics2020010 · Cited by 1
Efficient training of neuromorphic electronics
Yang, S., Tang, J., Zhou, T., Linares-Barranco, B., Indiveri, G., Zheng, X., Azghadi, M. R., & Chen, B. (2026) · Nature Electronics · doi:10.1038/s41928-026-01631-2 · Cited by 2
Circuit and Training Techniques Compensating for Non-Ideal Effects in Memristor Neural Networks
Oh, S., An, J., & Min, K. (2022) · IEEE International Conference on Solid-State and Integrated Circuit Technology · doi:10.1109/icsict55466.2022.9963389 · Cited by 1
“The non-volatile memories mentioned earlier have demonstrated possible use of in-memory computing in many previous literatures [7], [8].”
Programmable mixed-signal circuits
Tappertzhofen, S. (2023) · SN Applied Sciences · doi:10.1007/s42452-023-05572-9 · Cited by 1
A high resolution and configurable 1T1R1C ReRAM macro for medical semantic segmentation
Su, J., Chen, Y., Feng, P., Jiang, Z., Li, Z., & Chen, G. (2024) · IEICE Electronics Express · doi:10.1587/elex.21.20240071 · Cited by 1
Deshmukh, S., Bende, A., Sanghai, D., Saraswat, V., Biswas, A., Kadam, A. A., Patil, S. V., Singh, A. K., Deshpande, V. V., & Ganguly, U. (2024) · IEEE Journal of the Electron Devices Society · doi:10.1109/jeds.2024.3508759 · Cited by 1
“Non-volatile memory technologies such as RRAM, PCM, and flash memory have been proposed for IMC-based ANN inference [33], [37].”
Resistive switching devices empower the AI era
Jie, Y. U., Li, C., Zhang, X., Liu, Q., & Liu, M. (2025) · Scientia Sinica Informationis · doi:10.1360/ssi-2024-0349 · Cited by 1
In-Memory Computing Architecture for Tsetlin Machine Inference using MTJ-based Crossbar
Kumar, A. S., Botsa, V., Ghazal, O. G., Boppu, S., & Shafik, R. (2025) · IEEE International Symposium on Circuits and Systems proceedings · doi:10.1109/iscas56072.2025.11043814 · Cited by 1
Exploiting Process Variations to Protect Machine Learning Inference Engine from Chip Cloning
Huang, S., Peng, X., Jiang, H., Luo, Y., & Yu, S. (2021) · IEEE International Symposium on Circuits and Systems proceedings · doi:10.1109/iscas51556.2021.9401659 · Cited by 1
Self-citationMerged Logic and Memory Fabrics for AI Workloads
Crafton, B., Spetalnick, S. D., & Raychowdhury, A. (2021) · Asia and South Pacific Design Automation Conference (ASP-DAC) · doi:10.1145/3394885.3431615
“In [22], RRAM cell-to-cell variation is minimized with a 6-step write-verify technique.”
BackgroundMethodMixed-Signal Compute and Memory Fabrics for Deep Neural Networks
Murmann, B. (2022) · Springer eBooks · doi:10.1007/978-3-030-91741-8_1
Programmable Mixed-Signal CircuitsPreprint
Tappertzhofen, S. (2023) · Research Square · doi:10.21203/rs.3.rs-3186408/v1
A low power time-domain computing-in-memory micro for binary neural networks
He, Y. (2023) · Digital Repository Service (Northeastern University) · doi:10.17760/d20487159
Compute-in-Memory Architecture
Jiang, H., Huang, S., & Yu, S. (2024) · Springer eBooks · doi:10.1007/978-981-97-9314-3_62
Self-citationRoy, S. K., & Shanbhag, N. R. (2024) · doi:10.1145/3676536.3676685 · Cited by 1
“In eNVM-based IMCs, the low compute SNDR [1–20] stems from the fundamental challenge in sensing low resistances in the presence of the analog non-idealities [44].”
An analytical framework for optimized design of mixed-signal accelerators with F-2T2R cells
Caselli, M., & Boni, A. (2025) · AEU - International Journal of Electronics and Communications · doi:10.1016/j.aeue.2025.155970
Error-Corrected RRAM-Based Compute-in-Memory Accelerator Using Voltage-Clamp Current Mirror
Liu, Y., Zhang, W., Song, X., Zhou, J., Li, J., Feng, H., Lu, W., Peng, C., Zhao, Q., Dai, C., et al. (2025) · IEEE Transactions on Device and Materials Reliability · doi:10.1109/tdmr.2025.3612983
“To reduce the area and power consumption of the ADC, Yin et al. [23] adopted a shared ADC to obtain the final convolution sum.”
ReAD: ReRAM-Based Computing-In-Memory Design and Optimization by AC-DC Mixed Signal
Chi, G., & Lin, I.-C. (2025) · IEEE International System-on-Chip Conference (SOCC) · doi:10.1109/socc66126.2025.11235394
“[14] claimed that in work [7], only 9 WLs are asserted simultaneously in the 256 × 512 sub-array in the 55nm CMOS logic process, resulting in a low throughput of 2.47 GOPS and inefficient parallelism.”
InfluentialAnalyzing Embedded Non-Volatile Memory and Technology Optimizations for In-Memory Computing
Deaville, P., Zhang, B., & Verma, N. (2025) · IEEE Journal of Solid-State Circuits · doi:10.1109/jssc.2025.3632212
“Such column combination amortizes readout circuitry energy and area, easing pitch matching to the memory array, while avoiding the need for column multiplexers, which can occupy significant area and introduce readout error due to the finite and non-linear multiplexer switch resistances [28].”
InfluentialLow Power Sram-Based in-Memory Computing with Self Write Feature
Parmar, P., Joshi, D., & Mishra, S. (2025) · IEEE International Conference on Distributed Computing, VLSI, Electrical Circuits and Robotics (DISCOVER) · doi:10.1109/discover66922.2025.11258973
“This makes it perfect for implementing computations within the memory itself, decreasing data transmission delays, and enhancing overall system performance, especially in applications like deep learning, associative memory, and neural networks [6]–[8].”
Podasca, A., & Das, A. (2025) · doi:10.1109/iccad66269.2025.11240848
“At the circuit level, techniques such as iterative write-verify schemes [30]–[32], reference-based differential sensing [33], and closed-loop programming [34]–[36] aim the device from LRS to HRS.”
Gautier, A. (2024) · theses.fr (ABES) · doi:10.70675/5827871fz5213z492ez9180z641a7bd710e5
Wu, Z.-Y., Ramesh, A. K., Bhukya, V. N., Wei, J., Sheu, S.-S., Hsin, Y.-C., & Tseng, Y. (2026) · IEEE Transactions on Electron Devices · doi:10.1109/ted.2026.3692177
Low Power HfOx/TaOx stacked memristors with nanocolumn electrode for neuromorphic computing
Yang, F., Zhao, X., Liu, J., Zhu, H., & Shu, Q. (2026) · Nanotechnology · doi:10.1088/1361-6528/ae7381
“RRAM is a new type of memory device, which uses the resistance change characteristics under the action of electric field and switch between the high resistance state (HRS) and low resistance state (LRS) to store information [3, 4].”
A Sparsity-Aware Reconfigurable Sensing-Quantization Circuit for RRAM-Based Analog Compute-in-Memory
Chen, Z., Ding, H., Yang, Y., Zhang, H., Li, J., Zhao, S., Fu, Y., Wang, Z., & Cai, Y. (2026) · IEEE International Symposium on Circuits and Systems proceedings · doi:10.1109/iscas66217.2026.11562807
Silicon-Proven 4Kb RRAM Array Design in 65 nm CMOS Technology Node
Abedin, M., Solanki, J., Cady, N. C., Liehr, M., & Beckmann, K. (2026) · IEEE International Midwest Symposium on Circuits and Systems (MWSCAS) · doi:10.1109/mwscas67364.2026.11680975
“Although there have been demonstrations of RRAM-based on-chip applications, complete design details such as unit cell design considerations, unit cell transistor body bias, enabling individual/multi-device selection methods, to replicate circuits are not always provided [7], [8].”
Multi-Operand In-Memory Computation Using Quad-Read 12T SRAM
Parmar, P., Joshi, D., & Mishra, S. (2026) · International Symposium on VLSI Design and Test (VDAT) · doi:10.1109/vdat72243.2026.11704613
Metadata
- Authors
- Yin, S., Sun, X., Yu, S., & Seo, J.-S.
- Affiliations
- Arizona State University (US)
- Georgia Institute of Technology (US)
- Venue
- IEEE Transactions on Electron Devices
- Published
- 2020-10
- Volume
- 67
- Issue
- 10
- Pages
- 4185-4192
- ISSN
- 0018-9383 (ISSN-L)
- 1557-9646
- OpenAlex
- W2974585810
- Cited by
- 171
- References
- 50
- Citations by year
- 2026: 8 · 2025: 26 · 2024: 35 · 2023: 30 · 2022: 38 · 2021: 24 · 2020: 9
- FWCI
- 8.68
- Type
- journal article
- Language
- en
- Open access
- green
- arxiv.org
- Licence
- Locations
- IEEE Transactions on Electron Devices (publishedVersion)
- arXiv (Cornell University) (submittedVersion)
- arXiv (Cornell University)
- Funders
- NSF-SRC-E2CDA (2018-NC-2762B)
- NSF (1652866, 1715443, 1740225)
- JUMP C-BRIC
- JUMP ASCENT (SRC Program sponsored by the Defense Advanced Research Projects Agency
- Topics
- Physical Sciences › Engineering › Electrical and Electronic Engineering › Advanced Memory and Neural Computing
- Physical Sciences › Engineering › Electrical and Electronic Engineering › Low-power high-performance VLSI design
- Physical Sciences › Engineering › Electrical and Electronic Engineering › Ferroelectric and Negative Capacitance Devices
- Keywords
- in-memory computing, resistive random access memory, binary neural networks, monolithic integration, 90 nm CMOS, analog-to-digital converter, energy-delay product, edge computing, neuromorphic hardware, MNIST, CIFAR-10
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.