Matching Items (6)
Filtering by
- All Subjects: FPGA
- Genre: Academic theses
- Creators: Chakrabarti, Chaitali
- Creators: Li, Baoxin
- Status: Published
Description
This thesis report aims at introducing the background of QR decomposition and its application. QR decomposition using Givens rotations is a efficient method to prevent directly matrix inverse in solving least square minimization problem, which is a typical approach for weight calculation in adaptive beamforming. Furthermore, this thesis introduces Givens rotations algorithm and two general VLSI (very large scale integrated circuit) architectures namely triangular systolic array and linear systolic array for numerically QR decomposition. To fulfill the goal, a 4 input channels triangular systolic array with 16 bits fixed-point format and a 5 input channels linear systolic array are implemented on FPGA (Field programmable gate array). The final result shows that the estimated clock frequencies of 65 MHz and 135 MHz on post-place and route static timing report could be achieved using Xilinx Virtex 6 xc6vlx240t chip. Meanwhile, this report proposes a new method to test the dynamic range of QR-D. The dynamic range of the both architectures can be achieved around 110dB.
ContributorsYu, Hanguang (Author) / Bliss, Daniel W (Thesis advisor) / Ying, Lei (Committee member) / Chakrabarti, Chaitali (Committee member) / Arizona State University (Publisher)
Created2014
Description
Multidimensional (MD) discrete Fourier transform (DFT) is a key kernel algorithm in many signal processing applications, such as radar imaging and medical imaging. Traditionally, a two-dimensional (2-D) DFT is computed using Row-Column (RC) decomposition, where one-dimensional (1-D) DFTs are computed along the rows followed by 1-D DFTs along the columns. However, architectures based on RC decomposition are not efficient for large input size data which have to be stored in external memories based Synchronous Dynamic RAM (SDRAM). In this dissertation, first an efficient architecture to implement 2-D DFT for large-sized input data is proposed. This architecture achieves very high throughput by exploiting the inherent parallelism due to a novel 2-D decomposition and by utilizing the row-wise burst access pattern of the SDRAM external memory. In addition, an automatic IP generator is provided for mapping this architecture onto a reconfigurable platform of Xilinx Virtex-5 devices. For a 2048x2048 input size, the proposed architecture is 1.96 times faster than RC decomposition based implementation under the same memory constraints, and also outperforms other existing implementations. While the proposed 2-D DFT IP can achieve high performance, its output is bit-reversed. For systems where the output is required to be in natural order, use of this DFT IP would result in timing overhead. To solve this problem, a new bandwidth-efficient MD DFT IP that is transpose-free and produces outputs in natural order is proposed. It is based on a novel decomposition algorithm that takes into account the output order, FPGA resources, and the characteristics of off-chip memory access. An IP generator is designed and integrated into an in-house FPGA development platform, AlgoFLEX, for easy verification and fast integration. The corresponding 2-D and 3-D DFT architectures are ported onto the BEE3 board and their performance measured and analyzed. The results shows that the architecture can maintain the maximum memory bandwidth throughout the whole procedure while avoiding matrix transpose operations used in most other MD DFT implementations. The proposed architecture has also been ported onto the Xilinx ML605 board. When clocked at 100 MHz, 2048x2048 images with complex single-precision can be processed in less than 27 ms. Finally, transpose-free imaging flows for range-Doppler algorithm (RDA) and chirp-scaling algorithm (CSA) in SAR imaging are proposed. The corresponding implementations take advantage of the memory access patterns designed for the MD DFT IP and have superior timing performance. The RDA and CSA flows are mapped onto a unified architecture which is implemented on an FPGA platform. When clocked at 100MHz, the RDA and CSA computations with data size 4096x4096 can be completed in 323ms and 162ms, respectively. This implementation outperforms existing SAR image accelerators based on FPGA and GPU.
ContributorsYu, Chi-Li (Author) / Chakrabarti, Chaitali (Thesis advisor) / Papandreou-Suppappola, Antonia (Committee member) / Karam, Lina (Committee member) / Cao, Yu (Committee member) / Arizona State University (Publisher)
Created2012
Description
Convolutional neural networks(CNNs) achieve high accuracy on large datasets but requires significant computation and storage requirement for training/testing. While many applications demand low latency and energy-efficient processing of the images, deploying these complex algorithms on the hardware is a challenging task. This dissertation first presents a compiler-based CNN training accelerator using DDR3 and HBM2 memory. An optimized RTL library is implemented to perform training-specific tasks and an RTL compiler is developed to generate FPGA-synthesizable RTL based on user-defined constraints. High Bandwidth Memory(HBM) provides efficient off-chip communication and improves the training performance. The impact of HBM2 on CNN training workloads is analyzed and compressively compared with DDR3. For training ResNet-20/VGG-like CNNs for the CIFAR-10 dataset, the proposed CNN training accelerator on Stratix-10 GX FPGA(DDR3) demonstrates 479 GOPS performance, and on Stratix-10 MX FPGA(HBM) shows 4.5/9.7 X energy-efficiency improvement compared to Tesla V100 GPU. Next, the FPGA online learning accelerator is presented. Adopting model segmentation techniques from Progressive Segmented Training(PST), the online learning accelerator achieved a 4.2X reduction in training latency. Furthermore, this dissertation presents an 8-bit floating-point (FP8) training processor which implements (1) Highly parallel tensor cores that maintain high PE utilization, (2) Hardware-efficient channel gating for dynamic output activation sparsity (3) Dynamic weight sparsity based on group Lasso (4) Gradient skipping based on FP prediction error. The 28nm prototype chip demonstrates significant improvements in FLOPs reduction (7.3×), energy efficiency (16.4 TFLOPS/W), and overall training latency speedup (4.7×) for both supervised training and self-supervised training tasks. In addition to the training accelerators, this dissertation also presents a CNN inference accelerator on ASIC(FixyNN) and FPGA(FixyFPGA). FixyNN consists of a fixed-weight feature extractor that generates ubiquitous CNN features and a conventional programmable CNN accelerator. In the fixed-weight feature extractor, the network weights are hard-coded into hardware and used as a fixed operand for the multiplication. Experimental results demonstrate FixyNN can achieve very high energy efficiencies up to 26.6 TOPS/W, and FixyFPGA achieves $2.34\times$ higher GOPS on ImageNet classification. In summary, this dissertation comprehensively discusses novel architectures of high-performance and energy-efficient ASIC/FPGA CNN inference/training accelerators.
ContributorsKolala Venkataramaniah, Shreyas (Author) / Seo, Jae-Sun (Thesis advisor) / Cao, Yu (Committee member) / Chakrabarti, Chaitali (Committee member) / Fan, Deliang (Committee member) / Arizona State University (Publisher)
Created2022
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
QR decomposition (QRD) of a matrix is one of the most common linear algebra operationsused for the decomposition of a square
on-square matrix. It has a wide range
of applications especially in Multiple Input-Multiple Output (MIMO) communication
systems. Unfortunately it has high computation complexity { for matrix size of nxn,
QRD has O(n3) complexity and back substitution, which is used to solve a system
of linear equations, has O(n2) complexity. Thus, as the matrix size increases, the
hardware resource requirement for QRD and back substitution increases signicantly.
This thesis presents the design and implementation of a
exible QRD and back substitution accelerator using a folded architecture. It can support matrix sizes of
4x4, 8x8, 12x12, 16x16, and 20x20 with low hardware resource requirement.
The proposed architecture is based on the systolic array implementation of the
Givens algorithm for QRD. It is built with three dierent types of computation blocks
which are connected in a 2-D array structure. These blocks are controlled by a
scheduler which facilitates reusability of the blocks to perform computation for any
input matrix size which is a multiple of 4. These blocks are designed using two
basic programming elements which support both the forward and backward paths to
compute matrix R in QRD and column-matrix X in back substitution computation.
The proposed architecture has been mapped to Xilinx Zynq Ultrascale+ FPGA
(Field Programmable Gate Array), ZCU102. All inputs are complex with precision
of 40 bits (38 fractional bits and 1 signed bit). The architecture can be clocked at
50 MHz. The synthesis results of the folded architecture for dierent matrix sizes
are presented. The results show that the folded architecture can support QRD and
back substitution for inputs of large sizes which otherwise cannot t on an FPGA
when implemented using a
at architecture. The memory sizes required for dierent
matrix sizes are also presented.
on-square matrix. It has a wide range
of applications especially in Multiple Input-Multiple Output (MIMO) communication
systems. Unfortunately it has high computation complexity { for matrix size of nxn,
QRD has O(n3) complexity and back substitution, which is used to solve a system
of linear equations, has O(n2) complexity. Thus, as the matrix size increases, the
hardware resource requirement for QRD and back substitution increases signicantly.
This thesis presents the design and implementation of a
exible QRD and back substitution accelerator using a folded architecture. It can support matrix sizes of
4x4, 8x8, 12x12, 16x16, and 20x20 with low hardware resource requirement.
The proposed architecture is based on the systolic array implementation of the
Givens algorithm for QRD. It is built with three dierent types of computation blocks
which are connected in a 2-D array structure. These blocks are controlled by a
scheduler which facilitates reusability of the blocks to perform computation for any
input matrix size which is a multiple of 4. These blocks are designed using two
basic programming elements which support both the forward and backward paths to
compute matrix R in QRD and column-matrix X in back substitution computation.
The proposed architecture has been mapped to Xilinx Zynq Ultrascale+ FPGA
(Field Programmable Gate Array), ZCU102. All inputs are complex with precision
of 40 bits (38 fractional bits and 1 signed bit). The architecture can be clocked at
50 MHz. The synthesis results of the folded architecture for dierent matrix sizes
are presented. The results show that the folded architecture can support QRD and
back substitution for inputs of large sizes which otherwise cannot t on an FPGA
when implemented using a
at architecture. The memory sizes required for dierent
matrix sizes are also presented.
ContributorsKanagala, Srimayee (Author) / Chakrabarti, Chaitali (Thesis advisor) / Bliss, Daniel (Committee member) / Cao, Yu (Kevin) (Committee member) / Arizona State University (Publisher)
Created2020
Description
The rapid growth of Internet-of-things (IoT) and artificial intelligence applications have called forth a new computing paradigm--edge computing. Edge computing applications, such as video surveillance, autonomous driving, and augmented reality, are highly computationally intensive and require real-time processing. Current edge systems are typically based on commodity general-purpose hardware such as Central Processing Units (CPUs) and Graphical Processing Units (GPUs) , which are mainly designed for large, non-time-sensitive jobs in the cloud and do not match the needs of the edge workloads. Also, these systems are usually power hungry and are not suitable for resource-constrained edge deployments. Such application-hardware mismatch calls forth a new computing backbone to support the high-bandwidth, low-latency, and energy-efficient requirements. Also, the new system should be able to support a variety of edge applications with different characteristics. This thesis addresses the above challenges by studying the use of Field Programmable Gate Array (FPGA) -based computing systems for accelerating the edge workloads, from three critical angles. First, it investigates the feasibility of FPGAs for edge computing, in comparison to conventional CPUs and GPUs. Second, it studies the acceleration of common algorithmic characteristics, identified as loop patterns, using FPGAs, and develops a benchmark tool for analyzing the performance of these patterns on different accelerators. Third, it designs a new edge computing platform using multiple clustered FPGAs to provide high-bandwidth and low-latency acceleration of convolutional neural networks (CNNs) widely used in edge applications. Finally, it studies the acceleration of the emerging neural networks, randomly-wired neural networks, on the multi-FPGA platform. The experimental results from this work show that the new generation of workloads requires rethinking the current edge-computing architecture. First, through the acceleration of common loops, it demonstrates that FPGAs can outperform GPUs in specific loops types up to 14 times. Second, it shows the linear scalability of multi-FPGA platforms in accelerating neural networks. Third, it demonstrates the superiority of the new scheduler to optimally place randomly-wired neural networks on multi-FPGA platforms with 81.1 times better throughput than the available scheduling mechanisms.
ContributorsBiookaghazadeh, Saman (Author) / Zhao, Ming (Thesis advisor) / Ren, Fengbo (Thesis advisor) / Li, Baoxin (Committee member) / Seo, Jae-Sun (Committee member) / Arizona State University (Publisher)
Created2021