Programmable Linear RAM: A New Flash Memory-based Memristor for Artificial Synapses and Its Application to Speech Recognition System

Gao, Shifan; Hu, Jian; Jun, Xiao; Zhang, Bo; Lee, Choonghyun; Zhao, Yi; Kong, Weiran; Yang, Grace L.; Qiu, Xiang; Yang, Chun; Zhang, Cheng; Li, Binhan; Gao, Chao; Jiang, Haochuan; Wang, Zhexian

doi:10.1109/iedm19573.2019.8993598

Cited by 14 publications

(13 citation statements)

References 5 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Signal transformation converts signals from one domain into another, enabling more effective signal processing. Based on memristor arrays, Discrete Fourier Transform (DFT), performing the critical timefrequency transformation, has been implemented in simulations [40,41] , and also experimentally employed for the feature extraction of voice signals for speech recognition [42] . The computing process of the memristorbased DFT is shown in Fig.…”

Section: Signal Transformationmentioning

confidence: 99%

“…N/A Signal transform DFT [40][41][42] E Time-frequency transformation [40] and speech recognition [42] N/A 10 in speed, 109.8 in power efficiency [40] DCT [5,45] E Image compression and processing [5] Energy efficiency: 119.7 TOPs 1 W 1 [5] N/A DWT [44] S Image compression [44] Energy: 6.4 nJ/image Time: 15 s/image [44] 11 in energy efficiency, 1.28 in speed [44] Signal encoding CS [46,48,49,51] E Image compression and reconstruction [51] Power dissipation: 16.2 mW/read [51] 50 in power consumption [51] SC [47,[54][55][56] E Sparse representation of natural images [47] Energy: 719 J/image Time: 0.036 s/image [47] 16 in energy consumption [47] Component analysis PCA [58][59][60] E Classification of breast cancer [60] Power dissipation: 0.27 W/feature [60] N/A ICA [62][63][64] E Blind image source separation [64] N/A N/A Classification and regression N/A SVM [66,67] S Wake-up system [66] Energy: 0.7 nJ for potentiation, 0.5 pJ for depression [66] N/A SLP [68,…”

Section: Filtering Of Mixed Signals Of Two Frequencies N/amentioning

confidence: 99%

See 1 more Smart Citation

Memristor-based signal processing for edge computing

Zhao

Liu

Tang

et al. 2022

Tsinghua Sci. Technol.

View full text Add to dashboard Cite

The rapid growth of the Internet of Things (IoTs) has resulted in an explosive increase in data, and thus has raised new challenges for data processing units. Edge computing, which settles signal processing and computing tasks at the edge of networks rather than uploading data to the cloud, can reduce the amount of data for transmission and is a promising solution to address the challenges. One of the potential candidates for edge computing is a memristor, an emerging nonvolatile memory device that has the capability of in-memory computing. In this article, from the perspective of edge computing, we review recent progress on memristor-based signal processing methods, especially on the aspects of signal preprocessing and feature extraction. Then, we describe memristor-based signal classification and regression, and end-to-end signal processing. In all these applications, memristors serve as critical accelerators to greatly improve the overall system performance, such as power efficiency and processing speed. Finally, we discuss existing challenges and future outlooks for memristor-based signal processing systems.

show abstract

Section: Signal Transformationmentioning

confidence: 99%

Section: Filtering Of Mixed Signals Of Two Frequencies N/amentioning

confidence: 99%

Memristor-based signal processing for edge computing

Zhao

Liu

Tang

et al. 2022

Tsinghua Sci. Technol.

View full text Add to dashboard Cite

show abstract

“…DFT is a widely used time/spatial-frequency analysis technology and plays a key role in many applications including denoising, imaging and communication. As the DFT coefficients matrix has both real and imaginary parts, conventionally, four independent memristor arrays of the same size N×N would be needed to map the matrix entries onto memristor conductance for the implementation of a N-point DFT 32,33 , as shown in Fig. 2e.…”

Section: Implementation Of Mir On Memristor Arraymentioning

confidence: 99%

“…1a), showing appealing advantages in terms of energy efficiency and speed compared to conventional hardware [27][28][29] . Besides ANNs, there have also been attempts to use memristor arrays for implementing classic signal processing algorithms 30 , such as finite impulse response (FIR) filter 31 and discrete Fourier transformation (DFT) 32,33 , which has the potential to significantly accelerate medical image reconstruction speed and reduce energy consumption. In both applications, the most computationally intensive computations are vector-matrix multiplication (VMM); however, their actual implementations on memristor arrays are quite different in two aspects.…”

Section: Introductionmentioning

confidence: 99%

“…In comparison, the entries in the matrix for signal processing algorithms, e.g., DFT coefficients, are pre-calculated without any training, so they are more susceptible to those non-ideal device characteristics, which could lead to large errors in the output. Secondly, while the entries in the matrix for ANNs are usually real numbers, the matrices used in signal processing usually have both real and imaginary parts, and directly mapping them onto different memristor arrays (as proposed in literature 32,33 ) could result in large overhead in both energy consumption and area. Therefore, a new scheme to efficiently implement signal processing algorithms such as DFT on memristor arrays needs to be developed.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Energy-efficient high-fidelity image reconstruction with memristor arrays for medical diagnosis

Zhao

Liu

Tang

et al. 2022

Preprint

View full text Add to dashboard Cite

Medical imaging is an important tool to make accurate medical diagnosis and disease intervention. Current medical image reconstruction algorithms mainly run on Si-based digital processors with von Neumann architecture, which faces critical challenges to process massive amount of data for high-speed and high-quality imaging. Here, we present a memristive image reconstructor (MIR) to greatly accelerate image reconstruction with discrete Fourier transformation (DFT) by computing-in-memory (CIM) with memristor. To implement DFT on memristor arrays efficiently, we proposed a high-accuracy quasi-analogue mapping (QAM) method and generic complex matrix transfer (CMT) scheme, to improve the mapping precision and transfer efficiency, respectively. With these two strategies, we used MIR to demonstrate high-fidelity magnetic resonance imaging (MRI) and computed tomography (CT) image reconstructions, achieving software-equivalent qualities with peak signal-to-noise ratios (PSNR) of 40.88 dB and 22.38 dB, respectively. The reconstructed images were then segmented using a popular nnU-Net algorithm to further evaluate the reconstruction quality. For the MRI task, the final DICE scores were 0.979 and 0.980 for MIR and software, respectively; while for the CT task, the DICE scores were 0.977 and 0.985 for MIR and software, respectively. These results validated the feasibility of using memristor-reconstructed images for medical diagnosis. Furthermore, our MIR also exhibited more than 153× and 79× improvements in energy efficiency and normalized image reconstruction speed, respectively, compared to graphics processing unit (GPU). This work demonstrates MIR as a promising platform for high-fidelity image reconstruction for future medical diagnosis, and also largely extends the application of memristor-based CIM beyond artificial neural networks.

show abstract

Memristor‐Based Intelligent Human‐Like Neural Computing

Wang

Song

Chen

et al. 2022

Adv Elect Materials

View full text Add to dashboard Cite

In the field of a relatively dangerous working environment, NASA developed Valkyrie, a 44-degree-of-freedom, series elastic actuator-based robot. In addition, Valkyrie is designed to respond to disasters like nuclear disasters and advance human spaceflight in extraterrestrial planetary settings. [6] By implementing safety features and allowing remote intervention, Atkeson et al. enabled an Atlas humanoid robot to meet the standard in performing disaster response-related tasks involving physical contact with the environment. [7] Currently, to allow humanoid robots to feel and process the environmental information as human beings for perfectly implementing tasks, perceptual comprehension and computation efficiency are two key indexes. Nevertheless, it is challenging to achieve them. The former requires the installation of massive, diverse sensors in humanoids, which will inevitably slow down the processing speed under the current sensing-storage-process separated framework, i.e., von Neumann architecture. In other words, the current computer architecture builds up a barrier between sensing performance and computation efficiency. In this context, neuromorphic devices which can emulate the perceptual and computation functions of the biological nervous system have illustrated their potential to break the von Neumann barrier, attracting researchers' interests (Figure 1). Humanoid robots, intelligent machines resembling the human body in shape and functions, cannot only replace humans to complete services and dangerous tasks but also deepen the own understanding of the human body in the mimicking process. Nowadays, attaching a large number of sensors to obtain more sensory information and efficient computation is the development trend for humanoid robots. Nevertheless, due to the constraints of von Neumann-based structures, humanoid robots are facing multiple challenges, including tremendous energy consumption, latency bottlenecks, and the lack of bionic properties. Memristors, featured with high similarity to the biological elements, play an important role in mimicking the biological nervous system. The memristor-based nervous system allows humanoid robots to obtain high energy efficiency and bionic sensing properties, which are similar properties to the biological nervous system. Herein, this article first reviews the biological nervous system and memristor-based nervous system thoroughly, including the structures and also the functions. The applications of memristor-based nervous systems are introduced, the difficulties that need to be overcome are put forward, and future development prospects are also discussed. This review can hopefully provide an evolutionary perspective on humanoid robots and memristor-based nervous systems.

show abstract

Programmable Linear RAM: A New Flash Memory-based Memristor for Artificial Synapses and Its Application to Speech Recognition System

Cited by 14 publications

References 5 publications

Memristor-based signal processing for edge computing

Memristor-based signal processing for edge computing

Energy-efficient high-fidelity image reconstruction with memristor arrays for medical diagnosis

Memristor‐Based Intelligent Human‐Like Neural Computing

Contact Info

Product

Resources

About