Yasmin Halawani
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Yasmin Halawani has a total of 20 co-authors. Here are the top co-authors by number of shared publications:
- Baker Mohammad
Khalifa University, Abu Dhabi, United Arab Emirates
10 shared publications - Hani Saleh
Khalifa University
5 shared publications - Mahmoud Al-Qutayri
Department of Computer and Information Engineering, System on Chip Center, Khalifa University, Abu Dhabi, United Arab Emirates
5 shared publications - Eman Hassan
Khalifa University
2 shared publications - Said F. Al-Sarawib
School of Electrical and Electronic Engineering, The University of Adelaide, SA, Australia
2 shared publications - Dima Kilani
Khalifa University
1 shared publication
Latest Publications
- 21 May 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Artificial Intelligence
- Vol. 7 (1) , 157-171
TransHD: Spatial Transformer Features Extraction for HDC Synergetic Learning
Artificial intelligence (AI) relies on pattern recognition and classification algorithms to achieve accurate and efficient decision-making. Convolutional neural networks (CNNs) are the state-of-the-art for processing 2D image data, but their high computational and data requirements limit their use in resource-constrained environments. Hyperdimensional computing (HDC) offers a lightweight alternative, excelling in 1D tasks, but struggles to achieve competitive accuracy for 2D image classification. Hybrid frameworks combining HDC with CNNs have been proposed to improve accuracy but inherit the computational demands of deep learning, making them unsuitable for edge and IoT devices. To address this, we propose TransHD, a framework integrating lightweight spatial transformer networks (STNs) with HDC to enhance image classification performance while maintaining efficiency. TransHD achieves up to 9% higher accuracy than base HDC models on MNIST and Fashion-MNIST using only 30% of the training data, and reduces computational complexity by 2.5x through optimized STN feature map usage. On resource-constrained platforms like the Raspberry Pi 4, TransHD accelerates inference times by 3.4x and improves energy efficiency by 3.2x, with an accuracy trade-off of approximately 3% compared to CNNs. This study demonstrates the potential of combining HDC with STNs to develop efficient AI solutions for IoT and edge computing, where low energy consumption and computational efficiency are critical.
- 21 May 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Artificial Intelligence
- Vol. 7 (1) , 157-171
TransHD: Spatial Transformer Features Extraction for HDC Synergetic Learning
Artificial intelligence (AI) relies on pattern recognition and classification algorithms to achieve accurate and efficient decision-making. Convolutional neural networks (CNNs) are the state-of-the-art for processing 2D image data, but their high computational and data requirements limit their use in resource-constrained environments. Hyperdimensional computing (HDC) offers a lightweight alternative, excelling in 1D tasks, but struggles to achieve competitive accuracy for 2D image classification. Hybrid frameworks combining HDC with CNNs have been proposed to improve accuracy but inherit the computational demands of deep learning, making them unsuitable for edge and IoT devices. To address this, we propose TransHD, a framework integrating lightweight spatial transformer networks (STNs) with HDC to enhance image classification performance while maintaining efficiency. TransHD achieves up to 9% higher accuracy than base HDC models on MNIST and Fashion-MNIST using only 30% of the training data, and reduces computational complexity by 2.5x through optimized STN feature map usage. On resource-constrained platforms like the Raspberry Pi 4, TransHD accelerates inference times by 3.4x and improves energy efficiency by 3.2x, with an accuracy trade-off of approximately 3% compared to CNNs. This study demonstrates the potential of combining HDC with STNs to develop efficient AI solutions for IoT and edge computing, where low energy consumption and computational efficiency are critical.
- Open Access
- 20 December 2021
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 9, 167353-167363
C3PU: Cross-Coupling Capacitor Processing Unit Using Analog-Mixed Signal for AI Inference
This paper presents a novel cross-coupling capacitor processing unit (C3PU) that supports analog-mixed signal in-memory computing to perform multiply-and-accumulate (MAC) operations. The C3PU consists of a capacitive unit, a CMOS transistor, and a voltage-to-time converter (VTC). The capacitive unit serves as a computational element that holds the multiplier operand and performs multiplication once the multiplicand is applied at the terminal. The multiplicand is the input voltage that is converted to a pulse width signal using a low power VTC. The transistor transfers this multiplication where a voltage level is generated. A demonstrator of$5 \times 4$ C3PU array that is capable of implementing 4 MAC units is presented. The design has been verified using Monte Carlo simulation in 65 nm technology. The$5 \times 4$ C3PU consumed energy of 66.4 fJ/MAC at 0.3 V voltage supply with an error of 5.7%. The proposed unit achieves lower energy and occupies a smaller area by$3.4\times $ and$3.6\times $ , respectively, with similar error value when compared to a digital-based$8 \times 4$ -bit fixed point MAC unit. The C3PU has been utilized through an iris flower classification utilizing an artificial neural network which achieved a 90% classification accuracy compared to ideal accuracy of 96.67% using MATLAB.- 20 December 2021
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 9, 167353-167363
Open AccessC3PU: Cross-Coupling Capacitor Processing Unit Using Analog-Mixed Signal for AI Inference
This paper presents a novel cross-coupling capacitor processing unit (C3PU) that supports analog-mixed signal in-memory computing to perform multiply-and-accumulate (MAC) operations. The C3PU consists of a capacitive unit, a CMOS transistor, and a voltage-to-time converter (VTC). The capacitive unit serves as a computational element that holds the multiplier operand and performs multiplication once the multiplicand is applied at the terminal. The multiplicand is the input voltage that is converted to a pulse width signal using a low power VTC. The transistor transfers this multiplication where a voltage level is generated. A demonstrator of$5 \times 4$ C3PU array that is capable of implementing 4 MAC units is presented. The design has been verified using Monte Carlo simulation in 65 nm technology. The$5 \times 4$ C3PU consumed energy of 66.4 fJ/MAC at 0.3 V voltage supply with an error of 5.7%. The proposed unit achieves lower energy and occupies a smaller area by$3.4\times $ and$3.6\times $ , respectively, with similar error value when compared to a digital-based$8 \times 4$ -bit fixed point MAC unit. The C3PU has been utilized through an iris flower classification utilizing an artificial neural network which achieved a 90% classification accuracy compared to ideal accuracy of 96.67% using MATLAB. - Open Access
- 29 April 2021
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 9, 70430-70442
Design Exploration of ReRAM-Based Crossbar for AI Inference
ReRAM-based crossbar designs utilizing mixed-signal implementation has gained importance due to their low power, small size, low cost, and high throughput especially for multiply-and-add operations in AI-related applications. This paper provides a framework with associated code for analyzing the impact of ReRAM device variation and post-training analog conductance quantization that has not been fully explored for pre-trained network accelerators. A detailed study with end-to-end implementation is presented, ranging from mapping the pre-trained DNN weights to quantized crossbar conductance values and into final classification with the presence of variation. Monte Carlo analysis was performed to better analyze the impact of the different parameters on the final accuracy without being device-specific. The work assumes different conductance value variations, different conductance dynamic ranges, and various device quantization levels (QLs). MNIST and CIFAR-10 data sets were used in this study for ANN and CNN, respectively. Results show that for simple ANN, the accuracy drop due to quantization was ~ 2% at 64-QLs. While for CNN, the decrease in classification accuracy was around 10% with the same number of levels. Moreover, weight variation might cause a ~ 5% and ~ 8% drop in classification accuracy for ANN with 5% and CNN with 3% variation, respectively. The study confirms that increasing the number of levels with small variation results in near-optimal accuracy. However, the increase in accuracy saturates at an upper limit. The amount of distortion propagated through the layers is different in the two cases. It is dependent on the complexity of input data and network structure, such as the size of the neurons in each layer, the number of layers, and the number of channels and filters at each stage. This contribution is the first to provide the framework to explore the implication that emphasizes on device-independent post-training DNN quantization and weight variation on classification accuracy. This helps explore design trade-offs especially for edge devices in cases where there is no access to the training set to the end-user due to security or cost issues for pre-trained networks.
- 29 April 2021
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 9, 70430-70442
Open AccessDesign Exploration of ReRAM-Based Crossbar for AI Inference
ReRAM-based crossbar designs utilizing mixed-signal implementation has gained importance due to their low power, small size, low cost, and high throughput especially for multiply-and-add operations in AI-related applications. This paper provides a framework with associated code for analyzing the impact of ReRAM device variation and post-training analog conductance quantization that has not been fully explored for pre-trained network accelerators. A detailed study with end-to-end implementation is presented, ranging from mapping the pre-trained DNN weights to quantized crossbar conductance values and into final classification with the presence of variation. Monte Carlo analysis was performed to better analyze the impact of the different parameters on the final accuracy without being device-specific. The work assumes different conductance value variations, different conductance dynamic ranges, and various device quantization levels (QLs). MNIST and CIFAR-10 data sets were used in this study for ANN and CNN, respectively. Results show that for simple ANN, the accuracy drop due to quantization was ~ 2% at 64-QLs. While for CNN, the decrease in classification accuracy was around 10% with the same number of levels. Moreover, weight variation might cause a ~ 5% and ~ 8% drop in classification accuracy for ANN with 5% and CNN with 3% variation, respectively. The study confirms that increasing the number of levels with small variation results in near-optimal accuracy. However, the increase in accuracy saturates at an upper limit. The amount of distortion propagated through the layers is different in the two cases. It is dependent on the complexity of input data and network structure, such as the size of the neurons in each layer, the number of layers, and the number of channels and filters at each stage. This contribution is the first to provide the framework to explore the implication that emphasizes on device-independent post-training DNN quantization and weight variation on classification accuracy. This helps explore design trade-offs especially for edge devices in cases where there is no access to the training set to the end-user due to security or cost issues for pre-trained networks.
- Open Access
- 6 April 2021
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 9, 57179-57187
SLID: Exploiting Spatial Locality in Input Data as a Computational Reuse Method for Efficient CNN
Convolutional Neural Networks (CNNs) revolutionized computer vision and reached the state-of-the-art performance for image processing, object recognition, and video classification. Even though CNN inference is notoriously compute-intensive, as convolutions account for >90% of the total operation tasks, the ability to tradeoff between accuracy, performance, power, and latency to meet target application makes it an open research topic. This paper proposes the Spatial Locality Input Data (SLID) method for computational reuse during the inference stage for a pre-trained network. The method exploits input data spatial locality via skipping partial processing of the multiply-and-accumulate (MAC) operations for adjacent data and equating its value to previously computed ones. SLID improves the throughput of resource-constrained devices (Internet-of-Things, edge devices) and accelerates computations during the inference phase by reducing the number of MAC operations. Such approximate computing schema does not require a similarity quantification step nor any modification for the training stage. The computational data reuse was evaluated on three well-known distinctive CNN structures and data sets with alternating layer selections: LeNet, CIFAR-10, and AlexNet. The computational data reuse method saves up to 34.9%, 49.84%, and 31.5% of MAC operations while reducing the accuracy by 8%, 3.7%, and 5.0% for the three models mentioned earlier, respectively. Besides, the proposed method saves on memory access by eliminating data fetching of skipped inputs. Furthermore, filter size, strides, and padding on the accuracy and savings of operations are analyzed. SLID is the first work to exploit the input spatial locality for savings on CNN convolution operations with minimal accuracy loss and without memory or computational overhead. This makes it a great option to support intelligence at the edge.
- 6 April 2021
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 9, 57179-57187
Open AccessSLID: Exploiting Spatial Locality in Input Data as a Computational Reuse Method for Efficient CNN
Convolutional Neural Networks (CNNs) revolutionized computer vision and reached the state-of-the-art performance for image processing, object recognition, and video classification. Even though CNN inference is notoriously compute-intensive, as convolutions account for >90% of the total operation tasks, the ability to tradeoff between accuracy, performance, power, and latency to meet target application makes it an open research topic. This paper proposes the Spatial Locality Input Data (SLID) method for computational reuse during the inference stage for a pre-trained network. The method exploits input data spatial locality via skipping partial processing of the multiply-and-accumulate (MAC) operations for adjacent data and equating its value to previously computed ones. SLID improves the throughput of resource-constrained devices (Internet-of-Things, edge devices) and accelerates computations during the inference phase by reducing the number of MAC operations. Such approximate computing schema does not require a similarity quantification step nor any modification for the training stage. The computational data reuse was evaluated on three well-known distinctive CNN structures and data sets with alternating layer selections: LeNet, CIFAR-10, and AlexNet. The computational data reuse method saves up to 34.9%, 49.84%, and 31.5% of MAC operations while reducing the accuracy by 8%, 3.7%, and 5.0% for the three models mentioned earlier, respectively. Besides, the proposed method saves on memory access by eliminating data fetching of skipped inputs. Furthermore, filter size, strides, and padding on the accuracy and savings of operations are analyzed. SLID is the first work to exploit the input spatial locality for savings on CNN convolution operations with minimal accuracy loss and without memory or computational overhead. This makes it a great option to support intelligence at the edge.
- Open Access
- 16 February 2021
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 10, 97651-97664
Hyper-Dimensional Computing Challenges and Opportunities for AI Applications
Brain-inspired architectures are gaining increased attention, especially for edge devices to perform cognitive tasks utilizing its limited energy budget and computing resources. Hyperdimensional computing (HDC) paradigm is an emerging framework inspired by an abstract representation of neuronal circuits’ attributes in the human brain. That includes a fully holographic random representation, high-dimension vectors representing data, and robustness to uncertainty. The basic HDC pipeline consists of an encoding, training and comparison stages. The encoding algorithm maps different representations of inputs into a single class and stores them in the associative memory (AM) throughout the training stage. Later, during the inference stage, the similarity is computed between the query vector, which is encoded using the same encoding model, and the stored classes in the AM. HDC has shown promising results for 1D applications using less power, and lower latency than state-of-the-art digital neural networks (DNN). While in 2D applications, convolutional neural network (CNN) still achieves higher classification accuracy at the expense of more computations. In this paper, a comprehensive study on the HDC paradigm, main algorithms, and its implementation is presented. Moreover, the main state-of-the-art HDC architectures for 1D and 2D applications are highlighted. The article also analyzes two competing paradigms, namely, HDC and CNN, in terms of accuracy, complexity, and the number of operations. The paper concluded by highlighting challenges and recommendations for future directions on the HDC framework.
- 16 February 2021
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 10, 97651-97664
Open AccessHyper-Dimensional Computing Challenges and Opportunities for AI Applications
Brain-inspired architectures are gaining increased attention, especially for edge devices to perform cognitive tasks utilizing its limited energy budget and computing resources. Hyperdimensional computing (HDC) paradigm is an emerging framework inspired by an abstract representation of neuronal circuits’ attributes in the human brain. That includes a fully holographic random representation, high-dimension vectors representing data, and robustness to uncertainty. The basic HDC pipeline consists of an encoding, training and comparison stages. The encoding algorithm maps different representations of inputs into a single class and stores them in the associative memory (AM) throughout the training stage. Later, during the inference stage, the similarity is computed between the query vector, which is encoded using the same encoding model, and the stored classes in the AM. HDC has shown promising results for 1D applications using less power, and lower latency than state-of-the-art digital neural networks (DNN). While in 2D applications, convolutional neural network (CNN) still achieves higher classification accuracy at the expense of more computations. In this paper, a comprehensive study on the HDC paradigm, main algorithms, and its implementation is presented. Moreover, the main state-of-the-art HDC architectures for 1D and 2D applications are highlighted. The article also analyzes two competing paradigms, namely, HDC and CNN, in terms of accuracy, complexity, and the number of operations. The paper concluded by highlighting challenges and recommendations for future directions on the HDC framework.
- Open Access
- 23 June 2020
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 8, 117736-117745
FPGA-Based Memristor Emulator Circuit for Binary Convolutional Neural Networks
Binary convolutional neural networks (BCNN) have been proposed in the literature for resource-constrained IoTs nodes and mobile computing devices. Such computing platforms have strict constraints on the power budget, system performance, processing and memory capabilities. Nonetheless, the platforms are still required to efficiently perform classification and matching tasks needed in various applications. The memristor device has shown promising results when utilized for in-memory computing architectures, due to its ability to perform storage and computation using the same physical element. This is in addition to their nonlinear dynamic characteristics that can be used for various applications. Emulating the memristor operation through FPGAs implementation provides flexibility for rapid prototyping and exploration of the design space. In this paper, a FPGA-based memristor emulator circuit is implemented using a multi-bit XNOR gate as the main block for binary convolution which is followed by a memristor-based pooling layer. The BCNN layer is realized by bitwise XNOR-cell followed by wide NOR gate. The implementation has been successfully synthesized and verified using Xilinx Nexys4 FPGA with less than 1% utilization and 144.9 MHz frequency of operation. In order to test the functionality of the proposed cores, a digital implementation for the circuit was realized and verified experimentally. The experimental results show similar performance when compared to software simulation.
- 23 June 2020
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 8, 117736-117745
Open AccessFPGA-Based Memristor Emulator Circuit for Binary Convolutional Neural Networks
Binary convolutional neural networks (BCNN) have been proposed in the literature for resource-constrained IoTs nodes and mobile computing devices. Such computing platforms have strict constraints on the power budget, system performance, processing and memory capabilities. Nonetheless, the platforms are still required to efficiently perform classification and matching tasks needed in various applications. The memristor device has shown promising results when utilized for in-memory computing architectures, due to its ability to perform storage and computation using the same physical element. This is in addition to their nonlinear dynamic characteristics that can be used for various applications. Emulating the memristor operation through FPGAs implementation provides flexibility for rapid prototyping and exploration of the design space. In this paper, a FPGA-based memristor emulator circuit is implemented using a multi-bit XNOR gate as the main block for binary convolution which is followed by a memristor-based pooling layer. The BCNN layer is realized by bitwise XNOR-cell followed by wide NOR gate. The implementation has been successfully synthesized and verified using Xilinx Nexys4 FPGA with less than 1% utilization and 144.9 MHz frequency of operation. In order to test the functionality of the proposed cores, a digital implementation for the circuit was realized and verified experimentally. The experimental results show similar performance when compared to software simulation.
- 4 April 2019
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal on Emerging and Selected Topics in Circuits and Systems
- Vol. 9 (2) , 388-397
ReRAM-Based In-Memory Computing for Search Engine and Neural Network Applications
Resource-constrained computing devices such as those used in IoTs require low-power, high performance and small size to be enabled to operate efficiently. Resistive random access memory (ReRAM) is a promising technology for building novel in-memory computing architectures, due to its ability to perform storage and computation using the same physical element with low energy and high density. ReRAM-based search engines and neural network (NN) accelerators have grown significantly specially for IoT devices. In this paper, we propose a memristor-based Voltage-Resistance XNOR (VR-XNOR) cell. The advantages of this cell are demonstrated through building a reconfigurable Content-Addressable Memory (CAM) architecture that can support Binary-CAM (BCAM) and Ternary-CAM (TCAM) and enable approximate search operations. Moreover, the memristor-based VR-XNOR cell is utilized for binarized Convolutional Neural Networks (CNN) with focus on the convolution operation. This is achieved by replacing the convolution module with XNOR-based filter banks. Simulations of the proposed architectures for search engine and feature extraction were carried out using VTEAM model in Cadence Virtuoso Analog Design Environment. The proposed filter bank architecture achieves a 1-ns extraction cycle time over N filters and produces multiple output feature maps in a single processing cycle. The filter uses two memristor devices to realize each XNOR gate and has shown a significant reduction in the number of multiply-add operations.
- 4 April 2019
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal on Emerging and Selected Topics in Circuits and Systems
- Vol. 9 (2) , 388-397
ReRAM-Based In-Memory Computing for Search Engine and Neural Network Applications
Resource-constrained computing devices such as those used in IoTs require low-power, high performance and small size to be enabled to operate efficiently. Resistive random access memory (ReRAM) is a promising technology for building novel in-memory computing architectures, due to its ability to perform storage and computation using the same physical element with low energy and high density. ReRAM-based search engines and neural network (NN) accelerators have grown significantly specially for IoT devices. In this paper, we propose a memristor-based Voltage-Resistance XNOR (VR-XNOR) cell. The advantages of this cell are demonstrated through building a reconfigurable Content-Addressable Memory (CAM) architecture that can support Binary-CAM (BCAM) and Ternary-CAM (TCAM) and enable approximate search operations. Moreover, the memristor-based VR-XNOR cell is utilized for binarized Convolutional Neural Networks (CNN) with focus on the convolution operation. This is achieved by replacing the convolution module with XNOR-based filter banks. Simulations of the proposed architectures for search engine and feature extraction were carried out using VTEAM model in Cadence Virtuoso Analog Design Environment. The proposed filter bank architecture achieves a 1-ns extraction cycle time over N filters and produces multiple output feature maps in a single processing cycle. The filter uses two memristor devices to realize each XNOR gate and has shown a significant reduction in the number of multiply-add operations.
- 7 June 2018
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Very Large Scale Integration (VLSI) Systems
- Vol. 26 (12) , 2749-2758
Memristor-Based Hardware Accelerator for Image Compression
Memristor-based hardware accelerators are gaining an increased attention as a potential candidate to speed-up the vector-matrix operations commonly needed in many digital image processing tasks due to their area, speed, and energy efficiency. In this paper, a memristor-based image compression (MR-IC) architecture that exploits a lossy 2-D discrete wavelet transform is proposed. The architecture is composed of a computational memristor crossbar, an intermediate memory array that stores the row-transformed coefficients and a final memory that holds the compressed version of the original image. The computational memristor array performs in-memory computation on the initially stored transformation coefficients. Using the quantitative analysis approach, we demonstrate a 10x reduction in a number of operations compared with a conventional application-specific integrated circuit implementation. This translates to five orders of magnitude reduction in area, around 11x improvement in energy efficiency, and 1.28x speedup in computation time. Image quality metrics, such as peak signal-to-noise ratio (PSNR), structural similarity (SSIM) index, and complex wavelet-SSIM (CW-SSIM), are used to quantify the reduction in image quality due to lossy compression. The achieved metrics for conventional versus MR-IC are: PSNR 57.24 versus 33.29 dB, SSIM 0.9994 versus 0.8853, and CW-SSIM 1 versus 0.9983. Simulation results show that the 32 quantization levels proposed architecture provides significant improvements in energy, area, and performance compared to the 32 levels CMOS implementation with comparable CW-SSIM.
- 7 June 2018
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Very Large Scale Integration (VLSI) Systems
- Vol. 26 (12) , 2749-2758
Memristor-Based Hardware Accelerator for Image Compression
Memristor-based hardware accelerators are gaining an increased attention as a potential candidate to speed-up the vector-matrix operations commonly needed in many digital image processing tasks due to their area, speed, and energy efficiency. In this paper, a memristor-based image compression (MR-IC) architecture that exploits a lossy 2-D discrete wavelet transform is proposed. The architecture is composed of a computational memristor crossbar, an intermediate memory array that stores the row-transformed coefficients and a final memory that holds the compressed version of the original image. The computational memristor array performs in-memory computation on the initially stored transformation coefficients. Using the quantitative analysis approach, we demonstrate a 10x reduction in a number of operations compared with a conventional application-specific integrated circuit implementation. This translates to five orders of magnitude reduction in area, around 11x improvement in energy efficiency, and 1.28x speedup in computation time. Image quality metrics, such as peak signal-to-noise ratio (PSNR), structural similarity (SSIM) index, and complex wavelet-SSIM (CW-SSIM), are used to quantify the reduction in image quality due to lossy compression. The achieved metrics for conventional versus MR-IC are: PSNR 57.24 versus 33.29 dB, SSIM 0.9994 versus 0.8853, and CW-SSIM 1 versus 0.9983. Simulation results show that the 32 quantization levels proposed architecture provides significant improvements in energy, area, and performance compared to the 32 levels CMOS implementation with comparable CW-SSIM.
- 22 March 2018
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Very Large Scale Integration (VLSI) Systems
- Vol. 26 (12) , 2773-2780
Stateful Memristor-Based Search Architecture
Computer vision and recognition is emerging as one of the important pillars in artificial intelligence systems. It is a vital way to interpret the collected data and find matching patterns that will help in real-time decision making. CMOS-based search engines suffer from density and power limitations. Memristor is a feasible candidate that is capable of performing search within a stored structure (in-memory computing). This paper proposes the first memristor-based stateful search engine architecture based on a novel stateful heterogeneous memristive XOR gate. The design is suitable for 2-D media applications, such as image matching and pattern inspection. It performs bitwise comparison using the proposed XOR gate. The output states of all XOR gates are transferred into a single analog memristor value that is read via a digital comparator. The design assumes a single memristor device for each of the incoming data, template, and result bits. Each 2-D array of input, template, and output is reordered into a single 1-D array with 3 × (N × M) structure, where N represents the number of entry data and M is the number of bits per entry. This allows for a significantly higher storage density than conventional CMOSbased or other memristor-based search engines. Simulations of the proposed architecture demonstrate functionalities in search and compare modes using an LTSpice circuit simulator. The proposed architecture achieves a 3-ns search cycle time at 0.34 nJ/database at 1.5 V/1 GHz using 2N + 1 memristors.
- 22 March 2018
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Very Large Scale Integration (VLSI) Systems
- Vol. 26 (12) , 2773-2780
Stateful Memristor-Based Search Architecture
Computer vision and recognition is emerging as one of the important pillars in artificial intelligence systems. It is a vital way to interpret the collected data and find matching patterns that will help in real-time decision making. CMOS-based search engines suffer from density and power limitations. Memristor is a feasible candidate that is capable of performing search within a stored structure (in-memory computing). This paper proposes the first memristor-based stateful search engine architecture based on a novel stateful heterogeneous memristive XOR gate. The design is suitable for 2-D media applications, such as image matching and pattern inspection. It performs bitwise comparison using the proposed XOR gate. The output states of all XOR gates are transferred into a single analog memristor value that is read via a digital comparator. The design assumes a single memristor device for each of the incoming data, template, and result bits. Each 2-D array of input, template, and output is reordered into a single 1-D array with 3 × (N × M) structure, where N represents the number of entry data and M is the number of bits per entry. This allows for a significantly higher storage density than conventional CMOSbased or other memristor-based search engines. Simulations of the proposed architecture demonstrate functionalities in search and compare modes using an LTSpice circuit simulator. The proposed architecture achieves a 3-ns search cycle time at 0.34 nJ/database at 1.5 V/1 GHz using 2N + 1 memristors.
- 1 October 2014
- Published by Springer Nature in Applied Physics A
- Vol. 117 (3) , 1019-1023
Modeling and device parameter design to improve reset time in binary-oxide memristors
- 1 October 2014
- Published by Springer Nature in Applied Physics A
- Vol. 117 (3) , 1019-1023
Modeling and device parameter design to improve reset time in binary-oxide memristors