Abstract: With the rapid development of deep learning, neural network and deep learning algorithms play a significant role in various practical applications. Due to the high accuracy and good performance, Convolutional Neural Networks (CNNs) especially have become a research hot spot in the past few years. However, the size of the networks becomes increasingly large scale due to the demands of the practical applications, which poses a significant challenge to construct a high-performance implementation of deep learning neural networks. Meanwhile, many of these application scenarios also have strict requirements on the performance and low-power consumption of hardware devices. Therefore, it is particularly critical to choose a moderate computing platform for hardware acceleration of CNNs. This article aimed to survey the recent advance in Field Programmable Gate Array (FPGA)-based acceleration of CNNs. Various designs and implementations of the accelerator based on FPGA under different devices and network models are overviewed, and the versions of Graphic Processing Units (GPUs), Application Specific Integrated Circuits (ASICs) and Digital Signal Processors (DSPs) are compared to present our own critical analysis and comments. Finally, we give a discussion on different perspectives of these acceleration and optimization methods on FPGA platforms to further explore the opportunities and challenges for future research. More helpfully, we give a prospect for future development of the FPGA-based accelerator.
Abstract: Autofluorescence Imaging (AFI) is a technology for detecting early carcinogenesis of the gastrointestinal tract in recent years. Compared with traditional white light endoscopy (WLE), this technology greatly improves the detection accuracy of early carcinogenesis, because the colors of normal tissues are different from cancerous tissues. Thus, edge detection can distinguish them in grayscale images. In this paper, based on the traditional Sobel edge detection method, optimization has been performed on this method which considers the environment of the gastrointestinal, including adaptive threshold and morphological processing. All of the processes are implemented on our self-designed system based on the image sensor OV6930 and Field Programmable Gate Array (FPGA), The system can capture the gastrointestinal image taken by the lens in real time and detect edges. The final experiments verified the feasibility of our system and the effectiveness and accuracy of the edge detection algorithm.
Abstract: Circuit switched networks widely used until the end of the 20th century have been transformed into packages switched networks. Time Division Multiplexing over Internet Protocol (TDMoIP) is a system that enables Time Division Multiplexing (TDM) traffic to be carried over packet switched networks (PSN). In TDMoIP systems, devices that send TDM data to the PSN and receive it from the network must operate with the same clock frequency. In this study, it was aimed to implement clock synchronization process in Field Programmable Gate Array (FPGA) chips using time information attached to the packages received from PSN. The designed hardware is verified using the datasets obtained for the different carrier types and comparing the results with the software model. Field tests are also performed by using the real time TDMoIP system.
Abstract: The reliability of the power grid depends on the successful operation of thousands of protective relays. The failure of one relay to operate as intended may lead the entire power grid to blackout. In fact, major power system failures during transient disturbances may be caused by unnecessary protective relay tripping rather than by the failure of a relay to operate. Adequate relay testing provides a first defense against false trips of the relay and hence improves power grid stability and prevents catastrophic bulk power system failures. The goal of this research project is to design and enhance the relay tester using a technology such as Field Programmable Gate Array (FPGA) card NI 7851. A PC based tester framework has been developed using Simulink power system model for generating signals under different conditions (faults or transient disturbances) and LabVIEW for developing the graphical user interface and configuring the FPGA. Besides, the interface system has been developed for outputting and amplifying the signals without distortion. These signals should be like the generated ones by the real power system and large enough for testing the relay’s functionality. The signals generated that have been displayed on the scope are satisfactory. Furthermore, the proposed testing system can be used for improving the performance of protective relay.
Abstract: The use of customised soft-core processors in which instructions can be integrated into a system in application hardware is increasing in the Field Programmable Gate Array (FPGA) field. Specifically, the partial run-time reconfiguration of FPGAs in specialised processors for a particular domain can be very beneficial. In this report, the design and implementation for the customisation of a soft-core MIPS processor using an FPGA and partial reconfiguration (PR) of FPGA technology will be addressed to achieve efficient resource use. This can be achieved using a PR design flow that helps the design fit into a smaller device. Moreover, the impact of static power consumption could be reduced due to runtime reconfiguration. This will be done by configurable custom instructions implemented in the hardware as an extension on the MIPS CPU. The aim of this project is to investigate the PR of FPGAs for run-time adaptations of the instruction set of a soft-core CPU, including the integration of custom instructions and the exploration of the potential to use the MultiBoot feature available in Xilinx FPGAs to carry out the PR process. The system will be evaluated and tested on a Nexus 3 development board featuring a Xilinx Spartran-6 FPGA. The system will be able to load reconfigurable custom instructions dynamically into user programs with the help of the trap handler when the custom instruction is called by the MIPS CPU. The results of this experiment demonstrate that custom instructions in hardware can speed up a certain function and many instructions can be saved when compared to a software implementation of the same function. Implementing custom instructions in hardware is perfectly possible and worth exploring.
Abstract: This paper analyzes the experimental investigation of indirect field oriented control of Field Programmable Gate Array (FPGA) based five-phase induction motor drive. A detailed d-q modeling and Space Vector Pulse Width Modulation (SVPWM) technique of 5-phase drive is elaborated in this paper. In the proposed work, the prototype model of 1 hp 5-phase Voltage Source Inverter (VSI) fed drive is implemented in hardware. SVPWM pulses are generated in FPGA platform through Very High Speed Integrated Circuit Hardware Description Language (VHDL) coding. The experimental results are observed under different loading conditions and compared with simulation results to validate the simulation model.
Abstract: In this communication, we have made an attempt to design multiplier-less low-pass finite impulse response (FIR) filter with the aid of various mutation strategies of Differential Evolution (DE) algorithm. Impulse response coefficient of the designed FIR filter has been represented as sums or differences of powers of two. Performance of the proposed filter has been evaluated in terms of its frequency response and associated hardware cost. Supremacy of our approach has been substantiated by comparing our result with many of the existing multiplier-less filter design algorithms of recent interest. It has also been demonstrated that DE-optimized filter outperforms Genetic Algorithm (GA) based design by a large margin. Hardware efficiency of our algorithm has further been validated by implementing those filters on a Field Programmable Gate Array (FPGA) chip.
Abstract: Recently, the Field Programmable Gate Array (FPGA) technology offers the potential of designing high performance systems at low cost. The discrete wavelet transform has gained the reputation of being a very effective signal analysis tool for many practical applications. However, due to its computation-intensive nature, current implementation of the transform falls short of meeting real-time processing requirements of most application. The objectives of this paper are implement the Haar and Daubechies wavelets using FPGA technology. In addition, the Bit Error Rate (BER) between the input audio signal and the reconstructed output signal for each wavelet is calculated. From the BER, it is seen that the implementations execute the operation of the wavelet transform correctly and satisfying the perfect reconstruction conditions. The design procedure has been explained and designed using the stat-ofart Electronic Design Automation (EDA) tools for system design on FPGA. Simulation, synthesis and implementation on the FPGA target technology has been carried out.
Abstract: Designing and implementing intelligent systems has become a crucial factor for the innovation and development of better products of space technologies. A neural network is a parallel system, capable of resolving paradigms that linear computing cannot. Field programmable gate array (FPGA) is a digital device that owns reprogrammable properties and robust flexibility. For the neural network based instrument prototype in real time application, conventional specific VLSI neural chip design suffers the limitation in time and cost. With low precision artificial neural network design, FPGAs have higher speed and smaller size for real time application than the VLSI and DSP chips. So, many researchers have made great efforts on the realization of neural network (NN) using FPGA technique. In this paper, an introduction of ANN and FPGA technique are briefly shown. Also, Hardware Description Language (VHDL) code has been proposed to implement ANNs as well as to present simulation results with floating point arithmetic. Synthesis results for ANN controller are developed using Precision RTL. Proposed VHDL implementation creates a flexible, fast method and high degree of parallelism for implementing ANN. The implementation of multi-layer NN using lookup table LUT reduces the resource utilization for implementation and time for execution.
Abstract: This paper presents implementation of attitude controller for a small UAV using field programmable gate array (FPGA). Due to the small size constrain a miniature more compact and computationally extensive; autopilot platform is needed for such systems. More over UAV autopilot has to deal with extremely adverse situations in the shortest possible time, while accomplishing its mission. FPGAs in the recent past have rendered themselves as fast, parallel, real time, processing devices in a compact size. This work utilizes this fact and implements different attitude controllers for a small UAV in FPGA, using its parallel processing capabilities. Attitude controller is designed in MATLAB/Simulink environment. The discrete version of this controller is implemented using pipelining followed by retiming, to reduce the critical path and thereby clock period of the controller datapath. Pipelined, retimed, parallel PID controller implementation is done using rapidprototyping and testing efficient development tool of “system generator", which has been developed by Xilinx for FPGA implementation. The improved timing performance enables the controller to react abruptly to any changes made to the attitudes of UAV.
Abstract: In this paper, a new method of controlling position of AC Servomotor using Field Programmable Gate Array (FPGA). FPGA controller is used to generate direction and the number of pulses required to rotate for a given angle. Pulses are sent as a square wave, the number of pulses determines the angle of rotation and frequency of square wave determines the speed of rotation. The proposed control scheme has been realized using XILINX FPGA SPARTAN XC3S400 and tested using MUMA012PIS model Alternating Current (AC) servomotor. Experimental results show that the position of the AC Servo motor can be controlled effectively. KeywordsAlternating Current (AC), Field Programmable Gate Array (FPGA), Liquid Crystal Display (LCD).
Abstract: Higher-order Statistics (HOS), also known as
cumulants, cross moments and their frequency domain counterparts,
known as poly spectra have emerged as a powerful signal processing
tool for the synthesis and analysis of signals and systems. Algorithms
used for the computation of cross moments are computationally
intensive and require high computational speed for real-time
applications. For efficiency and high speed, it is often advantageous
to realize computation intensive algorithms in hardware. A promising
solution that combines high flexibility together with the speed of a
traditional hardware is Field Programmable Gate Array (FPGA). In
this paper, we present FPGA-based parallel architecture for the
computation of third-order cross moments. The proposed design is
coded in Very High Speed Integrated Circuit (VHSIC) Hardware
Description Language (VHDL) and functionally verified by
implementing it on Xilinx Spartan-3 XC3S2000FG900-4 FPGA.
Implementation results are presented and it shows that the proposed
design can operate at a maximum frequency of 86.618 MHz.
Abstract: Memory Errors Detection and Correction aim to secure the transaction of data between the central processing unit of a satellite onboard computer and its local memory. In this paper, the application of a double-bit error detection and correction method is described and implemented in Field Programmable Gate Array (FPGA) technology. The performance of the proposed EDAC method is measured and compared with two different EDAC devices, using the same FPGA technology. Statistical analysis of single-event upset (SEU) and multiple-bit upset (MBU) activity in commercial memories onboard the first Algerian microsatellite Alsat-1 is given.
Abstract: Falling has been one of the major concerns and threats
to the independence of the elderly in their daily lives. With the
worldwide significant growth of the aging population, it is essential
to have a promising solution of fall detection which is able to operate
at high accuracy in real-time and supports large scale implementation
using multiple cameras. Field Programmable Gate Array (FPGA) is a
highly promising tool to be used as a hardware accelerator in many
emerging embedded vision based system. Thus, it is the main
objective of this paper to present an FPGA-based solution of visual
based fall detection to meet stringent real-time requirements with
high accuracy. The hardware architecture of visual based fall
detection which utilizes the pixel locality to reduce memory accesses
is proposed. By exploiting the parallel and pipeline architecture of
FPGA, our hardware implementation of visual based fall detection
using FGPA is able to achieve a performance of 60fps for a series of
video analytical functions at VGA resolutions (640x480). The results
of this work show that FPGA has great potentials and impacts in
enabling large scale vision system in the future healthcare industry
due to its flexibility and scalability.