In the current era of big-data computing, most non-engineer domain experts lack the skills needed to design fpga-based hardware accelerators to address big-data problems in their fields. This work presents bFlow, a de...
详细信息
ISBN:
(纸本)9781479924097
In the current era of big-data computing, most non-engineer domain experts lack the skills needed to design fpga-based hardware accelerators to address big-data problems in their fields. This work presents bFlow, a development environment that facilitates the assembly of such accelerators, specifically those targeting fpga-based hybrid computing platforms, such as the Convey HC series. This framework attempts to address the above problem by making use of an abstracted, graphical front-end more friendly to users without computer engineering backgrounds than traditional, HDL-based design environments, as well as by accelerating bitstream compilation by means of incremental implementation techniques. bFlow's performance, usability, and application to big-data life-science problems were tested by participants of an NSF-funded Summer Institute organized by the Virginia Bioinformatics Institute (VBI). In about one week, a group of four non-engineering participants made significant improvements to a reference Smith-Waterman implementation, adding functionality and scaling theoretical throughput by a factor of 32.
We demonstrate the restoration of audio signals corrupted by clicks and pops using techniques from sparse signal recovery and compressive sensing. The demonstration features real-time signal restoration using the appr...
详细信息
ISBN:
(纸本)9781467357623;9781467357609
We demonstrate the restoration of audio signals corrupted by clicks and pops using techniques from sparse signal recovery and compressive sensing. The demonstration features real-time signal restoration using the approximate message passing algorithm on an fpga prototyping board. To highlight the restoration performance of our implementation, we remove clicks and pops from old phonograph recordings in real time.
This paper presents the strategies to implement Residue Number System reverse converter based on the Look-Up Table (LUT) approach that is applicable for general moduli set. The approach makes use of partitioning to di...
详细信息
ISBN:
(纸本)9781467361996
This paper presents the strategies to implement Residue Number System reverse converter based on the Look-Up Table (LUT) approach that is applicable for general moduli set. The approach makes use of partitioning to divide the LUT entries into multiple small LUTs in parallel, where their outputs can be further selected to obtain the equivalent binary number. Pipelining architecture is also incorporated to improve the operation speed. These techniques are hence suitable for general moduli set with large moduli value. Implementation results based on fpga further demonstrate the feasibility and effectiveness of the proposed approach.
This paper presents the realization of the fpga extension to the multichannel, pixel readout ASIC called VIPIC, designed in the collaboration between Fermi National Laboratory and AGH UST for X-ray Photon Correlation ...
详细信息
ISBN:
(纸本)9781479905348
This paper presents the realization of the fpga extension to the multichannel, pixel readout ASIC called VIPIC, designed in the collaboration between Fermi National Laboratory and AGH UST for X-ray Photon Correlation Spectroscopy (XPCS) experiments at BNL (Brookhaven National Laboratory). The fpga was proposed to extend the digital backend of the chip and supply it with 32-bit long counters allowing effective operation in imaging mode, working in continuous readout mode at 50 MHz together with intelligent control of supporting digital lines, communication verification, the data compression and providing higher abstraction level for other systems communicating with it. The paper presents a new approach of building such systems including tools and methods allowing implementation of high-speed data processing without deep knowledge of fpga structure or Hardware Description Language (HDL) experience. The advantage of LabVIEW (TM) graphical system design for both fpga and Real-Time OS programming is presented for building testing setup including automatic procedures like threshold scans and automatic correction with user-defined algorithms.
Secure circuits are prone to a wide range of physical attacks. Among those are fault attacks based on modifying the circuit environment in order to change its behaviour or to induce faults into its computations. There...
详细信息
ISBN:
(纸本)9781479906642
Secure circuits are prone to a wide range of physical attacks. Among those are fault attacks based on modifying the circuit environment in order to change its behaviour or to induce faults into its computations. There are many common means used to inject such faults: laser shots, electromagnetic pulses, overclocking, chip underpowering, temperature increase, etc. In this paper we study the effect of negative power supply glitches on a fpga. The obtained faults were compared to faults injected by clock glitches. As a result, both power and clock glitch induced faults were found to be identical. Because clock glitches are related to timing constraint violations, we shall consider that both power and clock glitches share this common fault injection mechanism. We also further studied the properties of this fault injection means.
An fpga router is developed using both the PathFinder and A* algorithms. Instead of the popular routing channel and rack model that is widely adopted by the research community, a modified routing model based on the ar...
详细信息
On the Internet, most of media information is transmitted in plaintext. Some others can easily tamper or intercut the information, so there are threats to those information transmitted in plaintext. This paper designs...
详细信息
A CAM-based Information Detection Hardware System for fast exact pattern matching is implemented on a hardware system with fpga and ASIC. The system has a simple structure, does not employ any Central Processor Unit (...
详细信息
This paper discusses the use of Self-organizing map (SOM) for color-space image compression. Two types of compressions, i.e., color and space compressions, are carried out by SOM. The feasibility of the proposed compr...
详细信息
ISBN:
(纸本)9781467363617;978146736360+0
This paper discusses the use of Self-organizing map (SOM) for color-space image compression. Two types of compressions, i.e., color and space compressions, are carried out by SOM. The feasibility of the proposed compression method is verified by computer simulation. By applying both color and space compression, higher compression is achieved. In addition, the proposed compression system is implemented in hardware using a hardware SOM. The system is designed by using VHDL, and operation of the system is verified by fpga implementation. Experimental result shows that the maximum clock frequency of this system is 26 MHz, and a single image is compressed with 25.2 ms, which yields 39.7 fps.
A ROM-less direct digital synthesizer architecture is presented in this paper. This architecture eliminates the ROM-based phase to sine wave amplitude converter, which is a bottleneck for pushing clock frequencies int...
详细信息
ISBN:
(纸本)9781479900664
A ROM-less direct digital synthesizer architecture is presented in this paper. This architecture eliminates the ROM-based phase to sine wave amplitude converter, which is a bottleneck for pushing clock frequencies into the gigahertz range. The design consists of a 16-bit phase accumulator, a set of 18 band pass finite impulse response filters, a 12-bit digital to analog converter and a low pass filter to produce a sine wave with output frequencies ranging from 36 MHz to 72 MHz with a resolution of 3.05 KHz and a 55 dB spur free dynamic range. The same hardware can be used to achieve output frequency ranging from hertz to gigahertz and a 191 Hz resolution by changing the clock frequency. A resolution of 0.05 Hz can be achieved by using a 32-bit phase accumulator. This design was simulated in Xilinx system generator (Sysgen) and mapped on to Virtex-6 fpga. The analysis results of the Sysgen and fpga data show that the proposed design is an effective alternative.
暂无评论