Matching Items (5)
Filtering by
- Member of: Theses and Dissertations
Description
The Resistive Random Access Memory (ReRAM) is an emerging non-volatile memory
technology because of its attractive attributes, including excellent scalability (< 10 nm), low
programming voltage (< 3 V), fast switching speed (< 10 ns), high OFF/ON ratio (> 10),
good endurance (up to 1012 cycles) and great compatibility with silicon CMOS technology [1].
However, ReRAM suffers from larger write latency, energy and reliability issue compared to
Dynamic Random Access Memory (DRAM). To improve the energy-efficiency, latency efficiency and reliability of ReRAM storage systems, a low cost cross-layer approach that spans device, circuit, architecture and system levels is proposed.
For 1T1R 2D ReRAM system, the effect of both retention and endurance errors on
ReRAM reliability is considered. Proposed approach is to design circuit-level and architecture-level techniques to reduce raw Bit Error Rate significantly and then employ low cost Error Control Coding to achieve the desired lifetime.
For 1S1R 2D ReRAM system, a cross-point array with “multi-bit per access” per subarray
is designed for high energy-efficiency and good reliability. The errors due to cell-level as well
as array-level variations are analyzed and a low cost scheme to maintain reliability and latency
with low energy consumption is proposed.
For 1S1R 3D ReRAM system, access schemes which activate multiple subarrays with
multiple layers in a subarray are used to achieve high energy efficiency through activating fewer
subarray, and good reliability is achieved through innovative data organization.
Finally, a novel ReRAM-based accelerator design is proposed to support multiple
Convolutional Neural Networks (CNN) topologies including VGGNet, AlexNet and ResNet.
The multi-tiled architecture consists of 9 processing elements per tile, where each tile
implements the dot product operation using ReRAM as computation unit. The processing
elements operate in a systolic fashion, thereby maximizing input feature map reuse and
minimizing interconnection cost. The system-level evaluation on several network benchmarks
show that the proposed architecture can improve computation efficiency and energy efficiency
compared to a state-of-the-art ReRAM-based accelerator.
technology because of its attractive attributes, including excellent scalability (< 10 nm), low
programming voltage (< 3 V), fast switching speed (< 10 ns), high OFF/ON ratio (> 10),
good endurance (up to 1012 cycles) and great compatibility with silicon CMOS technology [1].
However, ReRAM suffers from larger write latency, energy and reliability issue compared to
Dynamic Random Access Memory (DRAM). To improve the energy-efficiency, latency efficiency and reliability of ReRAM storage systems, a low cost cross-layer approach that spans device, circuit, architecture and system levels is proposed.
For 1T1R 2D ReRAM system, the effect of both retention and endurance errors on
ReRAM reliability is considered. Proposed approach is to design circuit-level and architecture-level techniques to reduce raw Bit Error Rate significantly and then employ low cost Error Control Coding to achieve the desired lifetime.
For 1S1R 2D ReRAM system, a cross-point array with “multi-bit per access” per subarray
is designed for high energy-efficiency and good reliability. The errors due to cell-level as well
as array-level variations are analyzed and a low cost scheme to maintain reliability and latency
with low energy consumption is proposed.
For 1S1R 3D ReRAM system, access schemes which activate multiple subarrays with
multiple layers in a subarray are used to achieve high energy efficiency through activating fewer
subarray, and good reliability is achieved through innovative data organization.
Finally, a novel ReRAM-based accelerator design is proposed to support multiple
Convolutional Neural Networks (CNN) topologies including VGGNet, AlexNet and ResNet.
The multi-tiled architecture consists of 9 processing elements per tile, where each tile
implements the dot product operation using ReRAM as computation unit. The processing
elements operate in a systolic fashion, thereby maximizing input feature map reuse and
minimizing interconnection cost. The system-level evaluation on several network benchmarks
show that the proposed architecture can improve computation efficiency and energy efficiency
compared to a state-of-the-art ReRAM-based accelerator.
ContributorsMao, Manqing (Author) / Chakrabariti, Chaitali (Thesis advisor) / Yu, Shimeng (Committee member) / Cao, Yu (Committee member) / Orgas, Umit (Committee member) / Arizona State University (Publisher)
Created2019
Description
Field of cyber threats is evolving rapidly and every day multitude of new information about malware and Advanced Persistent Threats (APTs) is generated in the form of malware reports, blog articles, forum posts, etc. However, current Threat Intelligence (TI) systems have several limitations. First, most of the TI systems examine and interpret data manually with the help of analysts. Second, some of them generate Indicators of Compromise (IOCs) directly using regular expressions without understanding the contextual meaning of those IOCs from the data sources which allows the tools to include lot of false positives. Third, lot of TI systems consider either one or two data sources for the generation of IOCs, and misses some of the most valuable IOCs from other data sources.
To overcome these limitations, we propose iGen, a novel approach to fully automate the process of IOC generation and analysis. Proposed approach is based on the idea that our model can understand English texts like human beings, and extract the IOCs from the different data sources intelligently. Identification of the IOCs is done on the basis of the syntax and semantics of the sentence as well as context words (e.g., ``attacked'', ``suspicious'') present in the sentence which helps the approach work on any kind of data source. Our proposed technique, first removes the words with no contextual meaning like stop words and punctuations etc. Then using the rest of the words in the sentence and output label (IOC or non-IOC sentence), our model intelligently learn to classify sentences into IOC and non-IOC sentences. Once IOC sentences are identified using this learned Convolutional Neural Network (CNN) based approach, next step is to identify the IOC tokens (like domains, IP, URL) in the sentences. This CNN based classification model helps in removing false positives (like IPs which are not malicious). Afterwards, IOCs extracted from different data sources are correlated to find the links between thousands of apparently unrelated attack instances, particularly infrastructures shared between them. Our approach fully automates the process of IOC generation from gathering data from different sources to creating rules (e.g. OpenIOC, snort rules, STIX rules) for deployment on
the security infrastructure.
iGen has collected around 400K IOCs till now with a precision of 95\%, better than any state-of-art method.
To overcome these limitations, we propose iGen, a novel approach to fully automate the process of IOC generation and analysis. Proposed approach is based on the idea that our model can understand English texts like human beings, and extract the IOCs from the different data sources intelligently. Identification of the IOCs is done on the basis of the syntax and semantics of the sentence as well as context words (e.g., ``attacked'', ``suspicious'') present in the sentence which helps the approach work on any kind of data source. Our proposed technique, first removes the words with no contextual meaning like stop words and punctuations etc. Then using the rest of the words in the sentence and output label (IOC or non-IOC sentence), our model intelligently learn to classify sentences into IOC and non-IOC sentences. Once IOC sentences are identified using this learned Convolutional Neural Network (CNN) based approach, next step is to identify the IOC tokens (like domains, IP, URL) in the sentences. This CNN based classification model helps in removing false positives (like IPs which are not malicious). Afterwards, IOCs extracted from different data sources are correlated to find the links between thousands of apparently unrelated attack instances, particularly infrastructures shared between them. Our approach fully automates the process of IOC generation from gathering data from different sources to creating rules (e.g. OpenIOC, snort rules, STIX rules) for deployment on
the security infrastructure.
iGen has collected around 400K IOCs till now with a precision of 95\%, better than any state-of-art method.
ContributorsPanwar, Anupam (Author) / Ahn, Gail-Joon (Thesis advisor) / Doupe, Adam (Committee member) / Zhao, Ziming (Committee member) / Arizona State University (Publisher)
Created2017
Description
We present in this paper a method to compare scene classification accuracy of C-band Synthetic aperture radar (SAR) and optical images utilizing both classical and quantum computing algorithms. This REU study uses data from the Sentinel satellite. The dataset contains (i) synthetic aperture radar images collected from the Sentinel-1 satellite and (ii) optical images for the same area as the SAR images collected from the Sentinel-2 satellite. We utilize classical neural networks to classify four classes of images. We then use Quantum Convolutional Neural Networks and deep learning techniques to take advantage of machine learning to help the system train, learn, and identify at a higher classification accuracy. A hybrid Quantum-classical model that is trained on the Sentinel1-2 dataset is proposed, and the performance is then compared against the classical in terms of classification accuracy.
ContributorsMiller, Leslie (Author) / Spanias, Andreas (Thesis director) / Uehara, Glen (Committee member) / Barrett, The Honors College (Contributor) / Electrical Engineering Program (Contributor)
Created2023-05
Description
Convolutional Neural Network (CNN) has achieved state-of-the-art performance in numerous applications like computer vision, natural language processing, robotics etc. The advancement of High-Performance Computing systems equipped with dedicated hardware accelerators has also paved the way towards the success of compute intensive CNNs. Graphics Processing Units (GPUs), with massive processing capability, have been of general interest for the acceleration of CNNs. Recently, Field Programmable Gate Arrays (FPGAs) have been promising in CNN acceleration since they offer high performance while also being re-configurable to support the evolution of CNNs. This work focuses on a design methodology to accelerate CNNs on FPGA with low inference latency and high-throughput which are crucial for scenarios like self-driving cars, video surveillance etc. It also includes optimizations which reduce the resource utilization by a large margin with a small degradation in performance thus making the design suitable for low-end FPGA devices as well.
FPGA accelerators often suffer due to the limited main memory bandwidth. Also, highly parallel designs with large resource utilization often end up achieving low operating frequency due to poor routing. This work employs data fetch and buffer mechanisms, designed specifically for the memory access pattern of CNNs, that overlap computation with memory access. This work proposes a novel arrangement of the systolic processing element array to achieve high frequency and consume less resources than the existing works. Also, support has been extended to more complicated CNNs to do video processing. On Intel Arria 10 GX1150, the design operates at a frequency as high as 258MHz and performs single inference of VGG-16 and C3D in 23.5ms and 45.6ms respectively. For VGG-16 and C3D the design offers a throughput of 66.1 and 23.98 inferences/s respectively. This design can outperform other FPGA 2D CNN accelerators by up to 9.7 times and 3D CNN accelerators by up to 2.7 times.
FPGA accelerators often suffer due to the limited main memory bandwidth. Also, highly parallel designs with large resource utilization often end up achieving low operating frequency due to poor routing. This work employs data fetch and buffer mechanisms, designed specifically for the memory access pattern of CNNs, that overlap computation with memory access. This work proposes a novel arrangement of the systolic processing element array to achieve high frequency and consume less resources than the existing works. Also, support has been extended to more complicated CNNs to do video processing. On Intel Arria 10 GX1150, the design operates at a frequency as high as 258MHz and performs single inference of VGG-16 and C3D in 23.5ms and 45.6ms respectively. For VGG-16 and C3D the design offers a throughput of 66.1 and 23.98 inferences/s respectively. This design can outperform other FPGA 2D CNN accelerators by up to 9.7 times and 3D CNN accelerators by up to 2.7 times.
ContributorsRavi, Pravin Kumar (Author) / Zhao, Ming (Thesis advisor) / Li, Baoxin (Committee member) / Ren, Fengbo (Committee member) / Arizona State University (Publisher)
Created2020
Description
In this paper, I explore practical applications of neural networks for automated skin lesion identification. The visual characteristics are of primary importance in the recognition of skin diseases, hence, the development of deep neural network models proven capable of classifying skin lesions can potentially change the face of modern medicine by extending the availability and lowering the cost of diagnostic care. Previous work has demonstrated the effectiveness of convolutional neural networks in image classification in general, with even higher accuracy achievable by data augmentation techniques, such as cropping, rotating, and flipping input images, along with more advanced computationally intensive approaches. In this research, I provide an overview of Convolutional Neural Networks (CNN) and CNN implementation with TensorFlow and Keras API in context of image recognition and classification. I also experiment with custom convolutional neural network model architecture trained using HAM10000 dataset. The dataset used for the case study is obtained from Harvard Dataverse and is maintained by Medical University of Vienna. The HAM10000 dataset is a large collection of multi-source dermatoscopic images of common pigmented skin lesions and is available for academic research under Creative Commons Attribution-Noncommercial 4.0 International Public License. With over ten thousand dermatoscopic images of seven classes of benign and malignant skin lesions, the dataset is substantial for academic machine learning purposes for multiclass image classification. I discuss the successes and shortcomings of the model in respect to its application to the dataset.
ContributorsKaraliova, Natallia (Author) / Bansal, Ajay (Thesis director) / Gonzalez-Sanchez, Javier (Committee member) / Software Engineering (Contributor) / Barrett, The Honors College (Contributor)
Created2019-05