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.

Multilayer Architectures Based on a-SiC:H Material: Tunable Wavelength Filters in Optical Processing Devices

The characteristics of tunable wavelength filters based on a-SiC:H multilayered stacked pin cells are studied both theoretically and experimentally. The optical transducers were produced by PECVD and tested for a proper fine tuning of the cyan and yellow fluorescent proteins emission. The active device consists of a p-i'(a-SiC:H)-n/p-i(a-Si:H)-n heterostructures sandwiched between two transparent contacts. Experimental data on spectral response analysis, current-voltage characteristics and color and transmission rate discrimination are reported. Cyan and yellow fluorescent input channels were transmitted together, each one with a specific transmission rate and different intensities. The multiplexed optical signal was analyzed by reading out, under positive and negative applied voltages, the generated photocurrents. Results show that the optimized optical transducer has the capability of combining the transient fluorescent signals onto a single output signal without losing any specificity (color and intensity). It acts as a voltage controlled optical filter: when the applied voltages are chosen appropriately the transducer can select separately the cyan and yellow channel emissions (wavelength and frequency) and also to quantify their relative intensities. A theoretical analysis supported by a numerical simulation is presented.

Optical processing devices based on a-SiCH multilayer architectures

Red, green and blue optical signals were directed to an a-SiC:H multilayered device, each one with a specific transmission rate. The combined optical signal was analyzed by reading out, under different applied voltages, the generated photocurrent. Results show that when a chromatic time dependent wavelength combination with different transmission rates irradiates the multilayered structure, the device operates as a tunable wavelength filter and can be used in wavelength division multiplexing systems for short range communications. An application to fluorescent proteins detection is presented. (C) 2010 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

Performance comparison of a VDFLL versus VDLL and scalar GNSS receiver architectures in Harsh scenarios

We analyze the advantages and drawbacks of a vector delay/frequency-locked loop (VDFLL) architecture regarding the conventional scalar and the vector delay-locked loop (VDLL) architectures for GNSS receivers in harsh scenarios that include ionospheric scintillation, multipath, and high dynamics motion. The VDFLL is constituted by a bank of code and frequency discriminators feeding a central extended Kaiman filter (EKF) that estimates the receiver's position, velocity, and clock bias. Both code and frequency loops are closed vectorially through the EKF. The VDLL closes the code loop vectorially and the phase loops through individual PLLs while the scalar receiver closes both loops by means of individual independent PLLs and DLLs.

Algorithm-oriented design of efficient many-core architectures applied to dense matrix multiplication

Recent integrated circuit technologies have opened the possibility to design parallel architectures with hundreds of cores on a single chip. The design space of these parallel architectures is huge with many architectural options. Exploring the design space gets even more difficult if, beyond performance and area, we also consider extra metrics like performance and area efficiency, where the designer tries to design the architecture with the best performance per chip area and the best sustainable performance. In this paper we present an algorithm-oriented approach to design a many-core architecture. Instead of doing the design space exploration of the many core architecture based on the experimental execution results of a particular benchmark of algorithms, our approach is to make a formal analysis of the algorithms considering the main architectural aspects and to determine how each particular architectural aspect is related to the performance of the architecture when running an algorithm or set of algorithms. The architectural aspects considered include the number of cores, the local memory available in each core, the communication bandwidth between the many-core architecture and the external memory and the memory hierarchy. To exemplify the approach we did a theoretical analysis of a dense matrix multiplication algorithm and determined an equation that relates the number of execution cycles with the architectural parameters. Based on this equation a many-core architecture has been designed. The results obtained indicate that a 100 mm(2) integrated circuit design of the proposed architecture, using a 65 nm technology, is able to achieve 464 GFLOPs (double precision floating-point) for a memory bandwidth of 16 GB/s. This corresponds to a performance efficiency of 71 %. Considering a 45 nm technology, a 100 mm(2) chip attains 833 GFLOPs which corresponds to 84 % of peak performance These figures are better than those obtained by previous many-core architectures, except for the area efficiency which is limited by the lower memory bandwidth considered. The results achieved are also better than those of previous state-of-the-art many-cores architectures designed specifically to achieve high performance for matrix multiplication.

Optical processor based on a-SiC technology for spectral data error control

The SiC optical processor for error detection and correction is realized by using double pin/pin a-SiC:H photodetector with front and back biased optical gating elements. Data shows that the background act as selector that pick one or more states by splitting portions of the input multi optical signals across the front and back photodiodes. Boolean operations such as exclusive OR (EXOR) and three bit addition are demonstrated optically with a combination of such switching devices, showing that when one or all of the inputs are present the output will be amplified, the system will behave as an XOR gate representing the SUM. When two or three inputs are on, the system acts as AND gate indicating the present of the CARRY bit. Additional parity logic operations are performed by use of the four incoming pulsed communication channels that are transmitted and checked for errors together. As a simple example of this approach, we describe an all optical processor for error detection and correction and then, provide an experimental demonstration of this fault tolerant reversible system, in emerging nanotechnology.

Biofunctional composite coating architectures based on polycaprolactone and nanohydroxyapatite for controlled corrosion activity and enhanced biocompatibility of magnesium AZ31 alloy

In this work a biofunctional composite coating architecture for controlled corrosion activity and enhanced cellular adhesion of AZ31 Mg alloys is proposed. The composite coating consists of a polycaprolactone (PCL) matrix modified with nanohydroxyapatite (HA) applied over a nanometric layer of polyetherimide (PEI). The protective properties of the coating were studied by electrochemical impedance spectroscopy (EIS), a non-disturbing technique, and the coating morphology was investigated by field emission scanning electron microscopy (FE-SEM). The results show that the composite coating protects the AZ31 substrate. The barrier properties of the coating can be optimized by changing the PCL concentration. The presence of nanohydroxyapatite particles influences the coating morphology and decreases the corrosion resistance. The biocompatibility was assessed by studying the response of osteoblastic cells on coated samples through resazurin assay, confocal laser scanning microscopy (CLSM) and scanning electron microscopy (SEM). The results show that the polycaprolactone to hydroxyapatite ratio affects the cell behavior and that the presence of hydroxyapatite induces high osteoblastic differentiation. (C) 2014 Elsevier B.V. All rights reserved.

Limiting Internal Supply Voltage Spikes in DC-DC Converters

Implementing monolithic DC-DC converters for low power portable applications with a standard low voltage CMOS technology leads to lower production costs and higher reliability. Moreover, it allows miniaturization by the integration of two units in the same die: the power management unit that regulates the supply voltage for the second unit, a dedicated signal processor, that performs the functions required. This paper presents original techniques that limit spikes in the internal supply voltage on a monolithic DC-DC converter, extending the use of the same technology for both units. These spikes are mainly caused by fast current variations in the path connecting the external power supply to the internal pads of the converter power block. This path includes two parasitic inductances inbuilt in bond wires and in package pins. Although these parasitic inductances present relative low values when compared with the typical external inductances of DC-DC converters, their effects can not be neglected when switching high currents at high switching frequency. The associated overvoltage frequently causes destruction, reliability problems and/or control malfunction. Different spike reduction techniques are presented and compared. The proposed techniques were used in the design of the gate driver of a DC-DC converter included in a power management unit implemented in a standard 0.35 mu m CMOS technology.

Stacked pin Devices for Imaging Applications

In this paper we present results on the optimization of device architectures for colour and imaging applications, using a device with a TCO/pinpi'n/TCO configuration. The effect of the applied voltage on the color selectivity is discussed. Results show that the spectral response curves demonstrate rather good separation between the red, green and blue basic colors. Combining the information obtained under positive and negative applied bias a colour image is acquired without colour filters or pixel architecture. A low level image processing algorithm is used for the colour image reconstruction.

Digital radiology and Picture Archiving and Communication System (PACS)

Plain radiography still accounts for the vast majority of imaging studies that are performed at multiple clinical instances. Digital detectors are now prominent in many imaging facilities and they are the main driving force towards filmless environments. There has been a working paradigm shift due to the functional separation of acquisition, visualization, and storage with deep impact in the imaging workflows. Moreover with direct digital detectors images are made available almost immediately. Digital radiology is now completely integrated in Picture Archiving and Communication System (PACS) environments governed by the Digital Imaging and Communications in Medicine (DICOM) standard. In this chapter a brief overview of PACS architectures and components is presented together with a necessarily brief account of the DICOM standard. Special focus is given to the DICOM digital radiology objects and how specific attributes may now be used to improve and increase the metadata repository associated with image data. Regular scrutiny of the metadata repository may serve as a valuable tool for improved, cost-effective, and multidimensional quality control procedures.

An Efficient Long Distance Echo Canceller

This paper describes an implementation of a long distance echo canceller, operating on full-duplex with hands-free and in real-time with a single Digital Signal Processor (DSP). The proposed solution is based on short length adaptive filters centered on the positions of the most significant echoes, which are tracked by time delay estimators, for which we use a new approach. To deal with double talking situations a speech detector is employed. The floating-point DSP TMS320C6713 from Texas Instruments is used with software written in C++, with compiler optimizations for fast execution. The resulting algorithm enables long distance echo cancellation with low computational requirements, suited for embbeded systems. It reaches greater echo return loss enhancement and shows faster convergence speed when compared to the conventional approach. The experimental results approach the CCITT G.165 recommendation levels.

The adjusted effect of maternal body mass index, energy and macronutrient intakes during pregnancy, and gestational weight gain on body composition of full-term neonates

Objective - To evaluate the effect of prepregnancy body mass index (BMI), energy and macronutrient intakes during pregnancy, and gestational weight gain (GWG) on the body composition of full-term appropriate-for-gestational age neonates. Study Design - This is a cross-sectional study of a systematically recruited convenience sample of mother-infant pairs. Food intake during pregnancy was assessed by food frequency questionnaire and its nutritional value by the Food Processor Plus (ESHA Research Inc, Salem, OR). Neonatal body composition was assessed both by anthropometry and air displacement plethysmography. Explanatory models for neonatal body composition were tested by multiple linear regression analysis. Results - A total of 100 mother-infant pairs were included. Prepregnancy overweight was positively associated with offspring weight, weight/length, BMI, and fat-free mass in the whole sample; in males, it was also positively associated with midarm circumference, ponderal index, and fat mass. Higher energy intake from carbohydrate was positively associated with midarm circumference and weight/length in the whole sample. Higher GWG was positively associated with weight, length, and midarm circumference in females. Conclusion - Positive adjusted associations were found between both prepregnancy BMI and energy intake from carbohydrate and offspring body size in the whole sample. Positive adjusted associations were also found between prepregnancy overweight and adiposity in males, and between GWG and body size in females.

Sobrevivência: jogo interactivo para estudo de arquitecturas de agentes autónomos

Trabalho de projeto para obtenção do grau de Mestre em Engenharia Informática e de Computadores

Does the intake of LC-PUFAS during pregnancy affect the body composition of term neonates? Preliminary results

Background: The effect of the intake of polynsaturated long chain fatty acids (LCPUFAs) during pregnancy on fetal body composition has been assessed by studies using mostly neonatal anthropometry. Their results have been inconsistent, probably because neonatal anthropometry has several validity limitations. Air displacement plethismography (ADP) is a recently validated non-invasive method for assessing body composition in neonates. Objective: To determine the effect of the intake of LCPUFAs during pregnancy on the body composition of term neonates, measured by ADP. Methods: Cross-sectional study of a convenience sample of healthy full-term neonates and their mothers. The diet during pregnancy was assessed using a validated semi-quantitative food frequency questionnaire; Food Processor Plus® was used to convert food intake into nutritional values. Body composition was estimated by anthropometry and measured by ADP using Pea Pod™ Life Measurements Inc (fat mass - FM, fat-free mass and %FM) within the first 72h after birth. Univariate and multivariate analysis (linear regression model) were performed. Results: 54 mother-neonate pairs were included. Multivariate analysis adjusted to the maternal body mass index shows positive association between LCPUFAs intake and neonatal mid-arm circumference (= 0,610, p = 0,019) and negative association between n-6:n-3 ratio intake and neonatal %FM (= -2,744, p=0,066). Conclusion: To the best of our knowledge, this is the first study on this subject using ADP and showing a negative association between LCPUFAs n-6:n-3 ratio intake in pregnancy and neonatal %FM. This preliminary finding requires confirmation increasing the study power with a greater sample and performing interventional studies.

High Performance Multi-Standard Architecture for DCT Computation in H.264/AVC High Profile and HEVC Codecs

A new high performance architecture for the computation of all the DCT operations adopted in the H.264/AVC and HEVC standards is proposed in this paper. Contrasting to other dedicated transform cores, the presented multi-standard transform architecture is supported on a completely configurable, scalable and unified structure, that is able to compute not only the forward and the inverse 8×8 and 4×4 integer DCTs and the 4×4 and 2×2 Hadamard transforms defined in the H.264/AVC standard, but also the 4×4, 8×8, 16×16 and 32×32 integer transforms adopted in HEVC. Experimental results obtained using a Xilinx Virtex-7 FPGA demonstrated the superior performance and hardware efficiency levels provided by the proposed structure, which outperforms its more prominent related designs by at least 1.8 times. When integrated in a multi-core embedded system, this architecture allows the computation, in real-time, of all the transforms mentioned above for resolutions as high as the 8k Ultra High Definition Television (UHDTV) (7680×4320 @ 30fps).