A many-core co-processor for embedded parallel computing on FPGA

Single processor architectures are unable to provide the required performance of high performance embedded systems. Parallel processing based on general-purpose processors can achieve these performances with a considerable increase of required resources. However, in many cases, simplified optimized parallel cores can be used instead of general-purpose processors achieving better performance at lower resource utilization. In this paper, we propose a configurable many-core architecture to serve as a co-processor for high-performance embedded computing on Field-Programmable Gate Arrays. The architecture consists of an array of configurable simple cores with support for floating-point operations interconnected with a configurable interconnection network. For each core it is possible to configure the size of the internal memory, the supported operations and number of interfacing ports. The architecture was tested in a ZYNQ-7020 FPGA in the execution of several parallel algorithms. The results show that the proposed many-core architecture achieves better performance than that achieved with a parallel generalpurpose processor and that up to 32 floating-point cores can be implemented in a ZYNQ-7020 SoC FPGA.

A biosignal embedded system for physiological computing

Thesis submitted in the fulfilment of the requirements for the Degree of Master in Electronic and Telecomunications Engineering

Sistemas distribuídos: serviços Grid Computing

A necessidade de poder computacional é crescente nas diversas áreas de actuação humana, tanto na indústria, como em ambientes académicos. Grid Computing permite a ligação de recursos computacionais dispersos de maneira a permitir a sua utilização mais eficaz, fornecendo aos utilizadores um acesso simplificado ao poder computacional de diversos sistemas. Os primeiros projectos de Grid Computing implicavam a ligação de máquinas paralelas ou aglomerados de alto desempenho e alto custo, disponíveis apenas em algumas instituições. Contrastando com o elevado custo dos super-computadores, os computadores pessoais e a Internet sofreram uma evolução significativa nos últimos anos. O uso de computadores dispersos em uma WAN pode representar um ambiente muito interessante para processamento de alto desempenho. Os sistemas em Grid fornecem a possibilidade de se utilizar um conjunto de computadores pessoais de modo a fornecer uma computação que utiliza recursos que de outra maneira estariam omissos. Este trabalho consiste no estudo de Grid Computing a nível de conceito e de arquitectura e numa análise ao seu estado actual hoje em dia. Como complemento foi desenvolvido um componente que permite o desenvolvimento de serviços para Grids (Grid Services) mais eficaz do que o modelo de suporte a serviços actualmente utilizado. Este componente é disponibilizado sob a forma um plug-in para a plataforma Eclipse IDE.

Reconfigurable SiC Embedded Photonic Structures with Self Optical Bias Control

Multilayered heterostructures based on embedded a-Si:H and a-SiC:H p-i-n filters are analyzed from differential voltage design perspective using short- and long-pass filters. The transfer functions characteristics are presented. A numerical simulation is presented to explain the filtering properties of the photonic devices. Several monochromatic pulsed lights, separately (input channels) or in a polychromatic mixture (multiplexed signal) at different bit rates, illuminated the device. Steady-state optical bias is superimposed from the front and the back side. Results show that depending on the wavelength of the external background and impinging side, the device acts either as a short- or a long-pass band filter or as a band-stop filter. Particular attention is given to the amplification coefficient weights, which allow to take into account the wavelength background effects when a band or frequency needs to be filtered or the gate switch, in which optical active filter gates are used to select and filter input signals to specific output ports in wavelength division multiplexing (WDM) communication systems. This nonlinearity provides the possibility for selective removal or addition of wavelengths. A truth table of an encoder that performs 8-to-1 MUX function exemplifies the optoelectronic conversion.

Scalable Unified Transform Architecture for Advanced Video Coding Embedded Systems

A novel high throughput and scalable unified architecture for the computation of the transform operations in video codecs for advanced standards is presented in this paper. This structure can be used as a hardware accelerator in modern embedded systems to efficiently compute all the two-dimensional 4 x 4 and 2 x 2 transforms of the H.264/AVC standard. Moreover, its highly flexible design and hardware efficiency allows it to be easily scaled in terms of performance and hardware cost to meet the specific requirements of any given video coding application. Experimental results obtained using a Xilinx Virtex-5 FPGA demonstrated the superior performance and hardware efficiency levels provided by the proposed structure, which presents a throughput per unit of area relatively higher than other similar recently published designs targeting the H.264/AVC standard. Such results also showed that, when integrated in a multi-core embedded system, this architecture provides speedup factors of about 120x concerning pure software implementations of the transform algorithms, therefore allowing the computation, in real-time, of all the above mentioned transforms for Ultra High Definition Video (UHDV) sequences (4,320 x 7,680 @ 30 fps).

Computing the phase diagram of binary mixtures: A patchy particle case study

We investigate the phase behaviour of 2D mixtures of bi-functional and three-functional patchy particles and 3D mixtures of bi-functional and tetra-functional patchy particles by means of Monte Carlo simulations and Wertheim theory. We start by computing the critical points of the pure systems and then we investigate how the critical parameters change upon lowering the temperature. We extend the successive umbrella sampling method to mixtures to make it possible to extract information about the phase behaviour of the system at a fixed temperature for the whole range of densities and compositions of interest. (C) 2013 AIP Publishing LLC.

Branching points in the low-temperature dipolar hard sphere fluid

In this contribution, we investigate the low-temperature, low-density behaviour of dipolar hard-sphere (DHS) particles, i.e., hard spheres with dipoles embedded in their centre. We aim at describing the DHS fluid in terms of a network of chains and rings (the fundamental clusters) held together by branching points (defects) of different nature. We first introduce a systematic way of classifying inter-cluster connections according to their topology, and then employ this classification to analyse the geometric and thermodynamic properties of each class of defects, as extracted from state-of-the-art equilibrium Monte Carlo simulations. By computing the average density and energetic cost of each defect class, we find that the relevant contribution to inter-cluster interactions is indeed provided by (rare) three-way junctions and by four-way junctions arising from parallel or anti-parallel locally linear aggregates. All other (numerous) defects are either intra-cluster or associated to low cluster-cluster interaction energies, suggesting that these defects do not play a significant part in the thermodynamic description of the self-assembly processes of dipolar hard spheres. (C) 2013 AIP Publishing LLC.

Embedded platform for ECG biometric recognition

Trabalho Final de Mestrado para obtenção do grau de Mestre em Engenharia de Electrónica e Telecomunicações

Embedded System for ECG Biometrics

Comunication in Internationa Conference with Peer Review First International Congress on Cardiovasular Technologies - CARDIOTECHNIX, Vilamoura, Portugal, 2013

Feature selection for clustering categorical data with an embedded modelling approach

Research on the problem of feature selection for clustering continues to develop. This is a challenging task, mainly due to the absence of class labels to guide the search for relevant features. Categorical feature selection for clustering has rarely been addressed in the literature, with most of the proposed approaches having focused on numerical data. In this work, we propose an approach to simultaneously cluster categorical data and select a subset of relevant features. Our approach is based on a modification of a finite mixture model (of multinomial distributions), where a set of latent variables indicate the relevance of each feature. To estimate the model parameters, we implement a variant of the expectation-maximization algorithm that simultaneously selects the subset of relevant features, using a minimum message length criterion. The proposed approach compares favourably with two baseline methods: a filter based on an entropy measure and a wrapper based on mutual information. The results obtained on synthetic data illustrate the ability of the proposed expectation-maximization method to recover ground truth. An application to real data, referred to official statistics, shows its usefulness.

Many-core approach to 2D-DCT calculation using an FPGA

Trabalho Final de Mestrado para obtenção do grau de Mestre em Engenharia de Electrónica e Telecomunicações

Mobile collaborative cloudless computing

Although the computational power of mobile devices has been increasing, it is still not enough for some classes of applications. In the present, these applications delegate the computing power burden on servers located on the Internet. This model assumes an always-on Internet connectivity and implies a non-negligible latency. The thesis addresses the challenges and contributions posed to the application of a mobile collaborative computing environment concept to wireless networks. The goal is to define a reference architecture for high performance mobile applications. Current work is focused on efficient data dissemination on a highly transitive environment, suitable to many mobile applications and also to the reputation and incentive system available on this mobile collaborative computing environment. For this we are improving our already published reputation/incentive algorithm with knowledge from the usage pattern from the eduroam wireless network in the Lisbon area.

BITalino: a novel hardware framework for physiological computing

Physical computing has spun a true global revolution in the way in which the digital interfaces with the real world. From bicycle jackets with turn signal lights to twitter-controlled christmas trees, the Do-it-Yourself (DiY) hardware movement has been driving endless innovations and stimulating an age of creative engineering. This ongoing (r)evolution has been led by popular electronics platforms such as the Arduino, the Lilypad, or the Raspberry Pi, however, these are not designed taking into account the specific requirements of biosignal acquisition. To date, the physiological computing community has been severely lacking a parallel to that found in the DiY electronics realm, especially in what concerns suitable hardware frameworks. In this paper, we build on previous work developed within our group, focusing on an all-in-one, low-cost, and modular biosignal acquisition hardware platform, that makes it quicker and easier to build biomedical devices. We describe the main design considerations, experimental evaluation and circuit characterization results, together with the results from a usability study performed with volunteers from multiple target user groups, namely health sciences and electrical, biomedical, and computer engineering. Copyright © 2014 SCITEPRESS - Science and Technology Publications. All rights reserved.

Trends Of CPU, GPU and FPGA for high-performance computing

Floating-point computing with more than one TFLOP of peak performance is already a reality in recent Field-Programmable Gate Arrays (FPGA). General-Purpose Graphics Processing Units (GPGPU) and recent many-core CPUs have also taken advantage of the recent technological innovations in integrated circuit (IC) design and had also dramatically improved their peak performances. In this paper, we compare the trends of these computing architectures for high-performance computing and survey these platforms in the execution of algorithms belonging to different scientific application domains. Trends in peak performance, power consumption and sustained performances, for particular applications, show that FPGAs are increasing the gap to GPUs and many-core CPUs moving them away from high-performance computing with intensive floating-point calculations. FPGAs become competitive for custom floating-point or fixed-point representations, for smaller input sizes of certain algorithms, for combinational logic problems and parallel map-reduce problems. © 2014 Technical University of Munich (TUM).

Numerical simulation of plasmonic effects in amorphous silicon induced by embedded aluminium nanoparticles

This work reports a theoretical study aimed to identify the plasmonic resonance condition for a system formed by metallic nanoparticles embedded in an a-Si: H matrix. The study is based on a Tauc-Lorentz model for the electrical permittivity of a-Si: H and a Drude model for the metallic nanoparticles. It is calculated the The polarizability of an sphere and ellipsoidal shaped metal nanoparticles with radius of 20 nm. We also performed FDTD simulations of light propagation inside this structure reporting a comparison among the effects caused by a single nanoparticles of Aluminium, Silver and, as a comparison, an ideally perfectly conductor. The simulation results shows that is possible to obtain a plasmonic resonance in the red part of the spectrum (600-700 nm) when 20-30 nm radius Aluminium ellipsoids are embedded into a-Si: H.