System-on-a-Chip (SoC) based hardware acceleration in Register Transfer Level (RTL) design

Today, modern System-on-a-Chip (SoC) systems have grown rapidly due to the increased processing power, while maintaining the size of the hardware circuit. The number of transistors on a chip continues to increase, but current SoC designs may not be able to exploit the potential performance, especially with energy consumption and chip area becoming two major concerns. Traditional SoC designs usually separate software and hardware. Thus, the process of improving the system performance is a complicated task for both software and hardware designers. The aim of this research is to develop hardware acceleration workflow for software applications. Thus, system performance can be improved with constraints of energy consumption and on-chip resource costs. The characteristics of software applications can be identified by using profiling tools. Hardware acceleration can have significant performance improvement for highly mathematical calculations or repeated functions. The performance of SoC systems can then be improved, if the hardware acceleration method is used to accelerate the element that incurs performance overheads. The concepts mentioned in this study can be easily applied to a variety of sophisticated software applications. The contributions of SoC-based hardware acceleration in the hardware-software co-design platform include the following: (1) Software profiling methods are applied to H.264 Coder-Decoder (CODEC) core. The hotspot function of aimed application is identified by using critical attributes such as cycles per loop, loop rounds, etc. (2) Hardware acceleration method based on Field-Programmable Gate Array (FPGA) is used to resolve system bottlenecks and improve system performance. The identified hotspot function is then converted to a hardware accelerator and mapped onto the hardware platform. Two types of hardware acceleration methods – central bus design and co-processor design, are implemented for comparison in the proposed architecture. (3) System specifications, such as performance, energy consumption, and resource costs, are measured and analyzed. The trade-off of these three factors is compared and balanced. Different hardware accelerators are implemented and evaluated based on system requirements. 4) The system verification platform is designed based on Integrated Circuit (IC) workflow. Hardware optimization techniques are used for higher performance and less resource costs. Experimental results show that the proposed hardware acceleration workflow for software applications is an efficient technique. The system can reach 2.8X performance improvements and save 31.84% energy consumption by applying the Bus-IP design. The Co-processor design can have 7.9X performance and save 75.85% energy consumption.

Gesture Recognition Using Neural Networks Based on HW/SW Cosimulation Platform

Hardware/software (HW/SW) cosimulation integrates software simulation and hardware simulation simultaneously. Usually, HW/SW co-simulation platform is used to ease debugging and verification for very large-scale integration (VLSI) design. To accelerate the computation of the gesture recognition technique, an HW/SW implementation using field programmable gate array (FPGA) technology is presented in this paper. The major contributions of this work are: (1) a novel design of memory controller in the Verilog Hardware Description Language (Verilog HDL) to reduce memory consumption and load on the processor. (2) The testing part of the neural network algorithm is being hardwired to improve the speed and performance. The American Sign Language gesture recognition is chosen to verify the performance of the approach. Several experiments were carried out on four databases of the gestures (alphabet signs A to Z). (3) The major benefit of this design is that it takes only few milliseconds to recognize the hand gesture which makes it computationally more efficient.

Implementação de uma arquitetura fuzzy neural em hardware com treinamento online

Computational Intelligence Methods have been expanding to industrial applications motivated by their ability to solve problems in engineering. Therefore, the embedded systems follow the same idea of using computational intelligence tools embedded on machines. There are several works in the area of embedded systems and intelligent systems. However, there are a few papers that have joined both areas. The aim of this study was to implement an adaptive fuzzy neural hardware with online training embedded on Field Programmable Gate Array – FPGA. The system adaptation can occur during the execution of a given application, aiming online performance improvement. The proposed system architecture is modular, allowing different configurations of fuzzy neural network topologies with online training. The proposed system was applied to: mathematical function interpolation, pattern classification and selfcompensation of industrial sensors. The proposed system achieves satisfactory performance in both tasks. The experiments results shows the advantages and disadvantages of online training in hardware when performed in parallel and sequentially ways. The sequentially training method provides economy in FPGA area, however, increases the complexity of architecture actions. The parallel training method achieves high performance and reduced processing time, the pipeline technique is used to increase the proposed architecture performance. The study development was based on available tools for FPGA circuits.

Autonomic hardware manager: uma arquitetura de hardware autonômico usando a solução de repositório ativo de componentes

This Thesis main objective is to implement a supporting architecture to Autonomic Hardware systems, capable of manage the hardware running in reconfigurable devices. The proposed architecture implements manipulation, generation and communication functionalities, using the Context Oriented Active Repository approach. The solution consists in a Hardware-Software based architecture called "Autonomic Hardware Manager (AHM)" that contains an Active Repository of Hardware Components. Using the repository the architecture will be able to manage the connected systems at run time allowing the implementation of autonomic features such as self-management, self-optimization, self-description and self-configuration. The proposed architecture also contains a meta-model that allows the representation of the Operating Context for hardware systems. This meta-model will be used as basis to the context sensing modules, that are needed in the Active Repository architecture. In order to demonstrate the proposed architecture functionalities, experiments were proposed and implemented in order to proof the Thesis hypothesis and achieved objectives. Three experiments were planned and implemented: the Hardware Reconfigurable Filter, that consists of an application that implements Digital Filters using reconfigurable hardware; the Autonomic Image Segmentation Filter, that shows the project and implementation of an image processing autonomic application; finally, the Autonomic Autopilot application that consist of an auto pilot to unmanned aerial vehicles. In this work, the applications architectures were organized in modules, according their functionalities. Some modules were implemented using HDL and synthesized in hardware. Other modules were implemented kept in software. After that, applications were integrated to the AHM to allow their adaptation to different Operating Context, making them autonomic.

Protótipo de rastreador solar de um eixo baseado em microcontrolador

The increase in the efficiency of photo-voltaic systems has been the object of various studies the past few years. One possible way to increase the power extracted by a photovoltaic panel is the solar tracking, performing its movement in order to follow the sun’s path. One way to activate the tracking system is using an electric induction motor, which should have sufficient torque and low speed, ensuring tracking accuracy. With the use of voltage source inverters and logic devices that generate the appropriate switching is possible to obtain the torque and speed required for the system to operate. This paper proposes the implementation of a angular position sensor and a driver to be applied in solar tracker built at a Power Electronics and Renewable Energies Laboratory, located in UFRN. The speed variation of the motor is performed via a voltage source inverter whose PWM command to actuate their keys will be implemented in an FPGA (Field Programmable Gate Array) device and a TM4C microcontroller. A platform test with an AC induction machine of 1.5 CV was assembled for the comparative testing. The angular position sensor of the panel is implemented in a ATMega328 microcontroller coupled to an accelerometer, commanded by an Arduino prototyping board. The solar position is also calculated by the microcontroller from the geographic coordinates of the site where it was placed, and the local time and date obtained from an RTC (Real-Time Clock) device. A prototype of a solar tracker polar axis moved by a DC motor was assembled to certify the operation of the sensor and to check the tracking efficiency.

Implementação de uma matriz de neurônios dinamicamente reconfigurável para descrição de topologias de redes neurais artificiais multilayer perceptrons

The Artificial Neural Networks (ANN), which is one of the branches of Artificial Intelligence (AI), are being employed as a solution to many complex problems existing in several areas. To solve these problems, it is essential that its implementation is done in hardware. Among the strategies to be adopted and met during the design phase and implementation of RNAs in hardware, connections between neurons are the ones that need more attention. Recently, are RNAs implemented both in application specific integrated circuits's (Application Specific Integrated Circuits - ASIC) and in integrated circuits configured by the user, like the Field Programmable Gate Array (FPGA), which have the ability to be partially rewritten, at runtime, forming thus a system Partially Reconfigurable (SPR), the use of which provides several advantages, such as flexibility in implementation and cost reduction. It has been noted a considerable increase in the use of FPGAs for implementing ANNs. Given the above, it is proposed to implement an array of reconfigurable neurons for topologies Description of artificial neural network multilayer perceptrons (MLPs) in FPGA, in order to encourage feedback and reuse of neural processors (perceptrons) used in the same area of the circuit. It is further proposed, a communication network capable of performing the reuse of artificial neurons. The architecture of the proposed system will configure various topologies MLPs networks through partial reconfiguration of the FPGA. To allow this flexibility RNAs settings, a set of digital components (datapath), and a controller were developed to execute instructions that define each topology for MLP neural network.

Implementação de uma matriz de neurônios dinamicamente reconfigurável para descrição de topologias de redes neurais artificiais multilayer perceptrons

The Artificial Neural Networks (ANN), which is one of the branches of Artificial Intelligence (AI), are being employed as a solution to many complex problems existing in several areas. To solve these problems, it is essential that its implementation is done in hardware. Among the strategies to be adopted and met during the design phase and implementation of RNAs in hardware, connections between neurons are the ones that need more attention. Recently, are RNAs implemented both in application specific integrated circuits's (Application Specific Integrated Circuits - ASIC) and in integrated circuits configured by the user, like the Field Programmable Gate Array (FPGA), which have the ability to be partially rewritten, at runtime, forming thus a system Partially Reconfigurable (SPR), the use of which provides several advantages, such as flexibility in implementation and cost reduction. It has been noted a considerable increase in the use of FPGAs for implementing ANNs. Given the above, it is proposed to implement an array of reconfigurable neurons for topologies Description of artificial neural network multilayer perceptrons (MLPs) in FPGA, in order to encourage feedback and reuse of neural processors (perceptrons) used in the same area of the circuit. It is further proposed, a communication network capable of performing the reuse of artificial neurons. The architecture of the proposed system will configure various topologies MLPs networks through partial reconfiguration of the FPGA. To allow this flexibility RNAs settings, a set of digital components (datapath), and a controller were developed to execute instructions that define each topology for MLP neural network.

Sistema di visione stereo su architettura ZYNQ

Lo scopo della tesi è creare un’architettura in FPGA in grado di ricavare informazioni 3D da una coppia di sensori stereo. La pipeline è stata realizzata utilizzando il System-on-Chip Zynq, che permette una stretta interazione tra la parte hardware realizzata in FPGA e la CPU. Dopo uno studio preliminare degli strumenti hardware e software, è stata realizzata l’architettura base per la scrittura e la lettura di immagini nella memoria DDR dello Zynq. In seguito l’attenzione si è spostata sull’implementazione di algoritmi stereo (rettificazione e stereo matching) su FPGA e nella realizzazione di una pipeline in grado di ricavare accurate mappe di disparità in tempo reale acquisendo le immagini da una camera stereo.

Streaming di immagini via ethernet con Zynq con sistemi operativi Standalone e Linux

Il seguente lavoro di tesi si inserisce all'interno di un progetto accademico volto alla realizzazione di un sistema capace elaborare immagini utilizzando una rete FPGA, acquisite da un sensore. Ogni scrittura di un nuovo frame in memoria RAM genera un interrupt. L'obiettivo della tesi è creare un sistema client/server che permetta il trasferimento del flusso di frame dalla ZedBoard a un PC e la visualizzazione a video. Il progetto eseguito sulla ZedBoard è proposto in due versioni: la prima in assenza di sistema operativo (Standalone) e una seconda implementata su Linux. Il progetto eseguito sul PC è compatibile con Linux e Windows. La visualizzazione delle immagini è implementata utilizzando la libreria OpenCV.

Aceleración de algoritmos de estimación de movimiento mediante OpenCL

El flujo óptico y la estimación de movimiento es área de conocimiento muy importante usado en otros campos del conocimiento como el de la seguridad o el de la bioinformática. En estos sectores, se demandan aplicaciones de flujo óptico que realicen actividades muy importantes con tiempos de ejecución lo más bajos posibles, llegando a tiempo real si es posible. Debido a la gran complejidad de cálculos que siguen a este tipo de algoritmos como se observará en la sección de resultados, la aceleración de estos es una parte vital para dar soporte y conseguir ese tiempo real tan buscado. Por lo que planteamos como objetivo para este TFG la aceleración de este tipo de algoritmos mediante diversos tipos de aceleradores usando OpenCL y de paso demostrar que OpenCL es una buena herramienta que permite códigos paralelizados con un gran Speedup a la par que funcionar en toda una diversa gama de dispositivos tan distintos como un GPU y una FPGA. Para lo anteriormente mencionado trataremos de desarrollar un código para cada algoritmo y optimizarlo de forma no especifica a una plataforma para posteriormente ejecutarlo sobre las diversas plataformas y medir tiempos y error para cada algoritmo. Para el desarrollo de este proyecto partimos de la teoría de dos algoritmos ya existentes: Lucas&Kanade monoescala y el Horn&Schunck. Además, usaremos estímulos para estos algoritmos muy aceptados por la comunidad como pueden ser el RubberWhale o los Grove, los cuales nos ayudarán a establecer la corrección de estos algoritmos y analizar su precisión, dando así un estudio referencia para saber cual escoger.

A Methodology to Emulate Single Event Upsets in Flip-Flops using FPGAs through Partial Reconfiguration and Instrumentation

This paper presents a methodology to emulate Single Event Upsets (SEUs) in FPGA flip-flops (FFs). Since the content of a FF is not modifiable through the FPGA configuration memory bits, a dedicated design is required for fault injection in the FFs. The method proposed in this paper is a hybrid approach that combines FPGA partial reconfiguration and extra logic added to the circuit under test, without modifying its operation. This approach has been integrated into a fault-injection platform, named NESSY (Non intrusive ErrorS injection SYstem), developed by our research group. Finally, this paper includes results on a Virtex-5 FPGA demonstrating the validity of the method on the ITC’99 benchmark set and a Feed-Forward Equalization (FFE) filter. In comparison with other approaches in the literature, this methodology reduces the resource consumption introduced to carry out the fault injection in FFs, at the cost of adding very little time overhead (1.6 �μs per fault).

Hardware Implementation of a Fault-Tolerant Hopfield Neural Network on FPGAs

This letter presents an FPGA implementation of a fault-tolerant Hopfield NeuralNetwork (HNN). The robustness of this circuit against Single Event Upsets (SEUs) and Single Event Transients (SETs) has been evaluated. Results show the fault tolerance of the proposed design, compared to a previous non fault- tolerant implementation and a solution based on triple modular redundancy (TMR) of a standard HNN design.

De la fusion du génie logiciel et d'une bibliothèque à source ouverte pour la modélisation/simulation de processus matériel et logiciel

Thèse numérisée par la Direction des bibliothèques de l'Université de Montréal.

Sincronização em OFDM

Esta dissertação apresenta o trabalho sobre sincronização de receção para sistemas OFDM. Tendo como objetivo a integração da arquitetura desenvolvida no projeto de investigação \CROWN - Co-operative Radio over Fibre for Wireless Networks" atualmente em curso no Instituto de Telecomunicações. Esta arquitetura de receção foi implementada numa plataforma de desenvolvimento baseada em dispositivos programáveis FPGA, recorrendo as ferramentas de desenvolvimento MatLab, System Generator e ISE. O sistema implementado tem a particularidade de ter um princípio de funcionamento assíncrono e recorre aos algoritmos de Van de Beek [1] e Carlos Ribeiro [2] para proceder a estimação e consequente sincronização. Ambos os algoritmos foram utilizados para estimação do CFO, tendo o algoritmo de Van de Beek sido também utilizado para estimação do início de trama. Foram realizadas análises do desempenho do sistema para diferentes condições, sendo o objectivo de analisar o desempenho dos estimadores implementados. A performance foi então analisada de acordo com BER resultante e do erro de estimação do início de trama e do valor do CFO. Para além da análise individual dos resultados, e também feita uma comparação da precisão de ambos os estimadores.

De la fusion du génie logiciel et d'une bibliothèque à source ouverte pour la modélisation/simulation de processus matériel et logiciel

Thèse numérisée par la Direction des bibliothèques de l'Université de Montréal.